- Email: [email protected]
Development of radiochemical analysis strategies for decommissioning activities
Applied Radiation and Isotopes xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Development of radiochemical analysis strategies for decommissioning activities ⁎
D. Zapata-García , H. Wershofen Physikalisch-Technische Bundesanstalt (PTB), Bundesallee 100, 38116 Braunschweig, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Decommissioning Extraction chromatography Radiochemical analysis Concrete Steel Graphite
Radioactive waste generated in decommissioning activities need be classified according to their radioactive content. The radiological information required by national authorities includes diverse alpha and beta emitters, which can only be determined after a radiochemical separation process. This paper presents the work on the development of radiochemical methods for the simultaneous separation of several radionuclides in concrete, steel and graphite samples, on the basis of individual sample treatments which merge in a common radiochemical separation procedure based on extraction chromatography.
1. Introduction Currently there are 438 nuclear power plants in operation worldwide, and a further 149 that are shut down or are undergoing decommissioning, including 17 that have been fully decommissioned. These numbers become even bigger if fuel cycle facilities and research reactors are included (IAEA, 2015). Given that many of the nuclear facilities currently in operation will reach the end of their design lives within the next two decades, significant decommissioning activities are expected for several decades to come and these are one of the most challenging technological legacy issues that many countries will face in the near future. Most of the waste generated during decommissioning is not radiologically restricted and can be released into the environment or sent for recycling, but the rest will end up in different repositories, classified according to their radioactive content. Classification and control need to be done accurately in order to optimise the use of the limited repository space, but also to ensure that both the personnel involved and the public are not needlessly exposed to radiation and to minimise the environmental impact of such work. Thus, the characterisation of the radioactive materials present in nuclear facilities is a key technical aspect in the development of any decommissioning plan, but which proceeds also during and after the decommissioning process. The information required by national authorities, especially in the early stages of the decommissioning process, includes a great number of radionuclides, and characterisation is generally done combining several strategies, which range from activation calculations to sample taking and analysis in the laboratory. Whenever direct measurements can be
⁎
made, these are preferred because they produce reliable results and are faster and cheaper. However, radiochemical analysis is the only alternative when the radionuclides of interest are difficult to measure, as is the case of alpha and beta emitters, which need to be separated from the matrix before measurement. Although the costs involved in sample analysis are large, it is possible to make significant savings by the adoption of best available practices, such as the use of validated methods for on-site measurements and simultaneous determination of more than one radionuclide whenever possible (IAEA, 2014, 2010). In addition to this, the development of radiochemical methods which reduce the times of analysis from several to ideally one working day is one of the objectives many laboratories pursue nowadays, due to the time savings such improvements would entail. The samples that need processing within decommissioning works are also challenging, as they combine a complex matrix with an unknown distribution of the radionuclides. Among the variety of matrices probably are concrete, steel and graphite the ones of greatest interest (Cross et al., 2012). The present work deals with the development of procedures for the simultaneous determination of alpha and beta emitters in three matrices: concrete, steel and graphite. On the basis of individual sample treatment strategies based on the chemical properties of these matrices, the work focused in making these procedures merge in a common radiochemical analysis scheme based on extraction chromatography setups for the simultaneous separation of up to three radionuclides.
Corresponding author. E-mail address: [email protected] (D. Zapata-García).
http://dx.doi.org/10.1016/j.apradiso.2017.02.038 Received 16 September 2016; Received in revised form 13 February 2017; Accepted 20 February 2017 0969-8043/ © 2017 Elsevier Ltd. All rights reserved.
Please cite this article as: Zapata-García, D., Applied Radiation and Isotopes (2017), http://dx.doi.org/10.1016/j.apradiso.2017.02.038
Applied Radiation and Isotopes xxx (xxxx) xxx–xxx
D. Zapata-García, H. Wershofen
2. Materials and methods
9 M HCl and possibly interfering Th and Cm isotopes were stripped with 20 ml 4 M HCl. Finally, plutonium was stripped with 20 ml 0.1% NH4OCl in 2 M HCl. UTEVA resin cartridges were used for the extraction of uranium. After charging the sample, media was switched to HCl with 3 ml 9 M HCl and finally uranium was stripped with 20 ml 0.01 M HCl. Sr resin cartridges were used for the extraction of strontium. After charging the sample, the resin was rinsed with 10 ml 8 M HNO3 and strontium was stripped with 10 ml of 0.05 M HNO3. Multielement separations were performed using a setup of all three cartridges (TEVA, UTEVA, Sr) in tandem. Once the samples had gone through all three cartridges, these were separated and elution of each radionuclide was performed individually by applying the procedures described above. Plutonium and uranium were electroplated and measured by alphaparticle spectrometry while strontium was measured directly after elution without further treatment by high-resolution gamma spectrometry.
2.1. Reagents and samples High purity double deionised water and analytical grade reagents were used throughout this study. Commercial solutions of 242Pu (NIST), 232U (NIST) and 85Sr (PTB) were used as tracers. The activities added to each sample were 200 mBq in the case of actinides and 1 Bq in the case of 85Sr. Concrete samples were produced at National Physical Laboratory (NPL) using commercial components and commercial steel and graphite were used. 2.2. Equipment A fully automated electric fluxer K1 Pime (Katanax, Canada) and a START 1500 closed vessel microwave system (MLS GmbH, Germany) were used in this study. Radiometric measurements were done with alpha spectrometers Alpha Duo (Ortec) and high resolution gamma spectrometers IGC 34, PGT and GC5019-7500SL-R (Canberra).
3. Results and discussion The radiochemical separation scheme proposed in this work is based on the combined use of three extraction chromatography resins, each of them used for the separation of one single radionuclide. Methods used were developed in the laboratory on the basis of the multiple approaches found in the literature. (Eichrom Technologies, 2014; Maxwell et al., 2015, 2011; Tavcar et al., 2005; Wang et al., 2004) The radiochemical separation schemes described in this work were tested with spiked aqueous solutions, firstly with a single radionuclide using its specific resin, and afterwards with solutions spiked with all three radionuclides of interest using the tandem setup. Three replicates of each test were performed. Once the separation schemes had been tested with aqueous solutions, the work focused on the feasibility of such schemes for the analysis of matrices of interest in the field of decommissioning: concrete, steel and graphite. The spectra corresponding to plutonium and uranium measured by alpha-particle spectrometry showed no crossed interference of these elements, which provides confirmatory evidence that the proposed separation scheme is adequate. The results of the tests using aqueous solutions are presented in Fig. 1. Contrary to expectations, recovery values in the case of plutonium were higher for the multielemental tests than for the single radionuclide tests. Spectra were investigated for possible contamination of progeny present in the other tracer solutions, but no such effect was observed. Being TEVA the first cartridge in the tandem setup, this could not be attributed to the effect that other cartridges might induce and the results should be considered to show the variability in the performance of the resin applying the proposed separation scheme. In the case of uranium, recovery values showed great variability in the case of multielemental tests, with values ranging from 23% to 91%. This variability does not correlate with the results of the other two radionuclides. The possible uptake of uranium by the TEVA resin, though unexpected, was investigated. The results showed that the amount being retained by TEVA resin was below 0.8%, so the variability on the uranium recoveries cannot be based on this phenomenon. No other reason for this performance could be found. In the case of strontium, in contrast to actinides, the values of the two series of analyses do not show significant differences. Moreover, recovery values are the highest observed (ranging 85–95%), which can be explained on the basis of the specificity of this resin for strontium separations. In order to apply the proposed radiochemical separation scheme to the matrices of interest in this study, it was necessary to introduce sample treatment strategies previous to the radiochemical separation. The different treatment strategies were decided taking into account the sample chemical features in order to obtain a solution in nitric acid
2.3. Sample treatment For the analysis of aqueous solutions, tracers were added to 20 ml of concentrated HNO3 and evaporated to dryness on a sand bath. The final residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al (NO3)3. Different sample treatment procedures were applied to the different matrices considered in this work. Ground concrete samples (0.5 g) were spiked with tracers and fused at 1000 °C in Pt/Au crucibles with a mixture of LiBO2/Li2B4O7 adding LiBr as a wetting agent. The fusion protocol lasted 10 min and the fused material was dissolved with diluted HNO3 immediately after fusion. Silicates were removed by adding polyethylene glycol (PEG) and filtrating after leaving to polymerise for at least 12 h. The solution was then heated, concentrated HNO3 and HF added to eliminate any rest of silicate, evaporated to dryness once more and redissolved in 40% HNO3. Alkali metals were removed by means of Ca3(PO4)2 precipitation and the residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al(NO3)3. Steel samples (4.4 g) were dissolved with aqua regia to produce around 50 ml of solution. Tracers were added at this point and iron was removed by forming its complex with oxalate and coprecipitating the actinides and strontium with calcium oxalate at pH =1.7, before the precipitation of Fe(OH)3 occurs. Oxalate precipitate was treated twice with 40 ml aqua regia and evaporated to incipient dryness to destroy the oxalates and then redissolved with HNO3 and concentrated to dryness again. The final residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al(NO3)3. Graphite samples (0.2 g) were digested with H2SO4 and HNO3 at 260 °C for 30 min in a closed vessel microwave system. After digestion tracers were added and the solution obtained was concentrated to near dryness, 50 ml concentrated HNO3 added and concentrated to dryness again. The final residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al(NO3)3. 2.4. Radiochemical separation method Plutonium oxidation state in the solution was adjusted to Pu4+ before radiochemical separation by adding ferrous sulfamate followed by NaNO2. Radiochemical separation was performed using extraction chromatography cartridges (TrisKem International) on a vacuum box system. All cartridges were conditioned with 20 ml of 3 M HNO3 before adding the samples. TEVA resin cartridges were used for the extraction of plutonium. After charging the sample, media was switched to HCl by adding 4 ml of 2
Applied Radiation and Isotopes xxx (xxxx) xxx–xxx
D. Zapata-García, H. Wershofen
Fig. 2. Recovery values (%) obtained in the multielemental radiochemical separations of spiked concrete, steel and graphite. Error bars represent combined standard uncertainty.
respect to the values obtained with spiked solutions, but variation for each radionuclide is different. Strontium is the element which shows best performance, with recovery values over 70% in all cases. Actinides, on the contrary, present lower recovery values in all the tests and in the case of steel analyses these do not even reach 50%. There are different factors which can help to understand these differences. On the one hand, the chemical approaches applied previous to radiochemical separation have different levels of complexity for the different samples and might have different degrees of effectiveness. Strontium shows better performance than actinides, which in the case of aqueous solutions was justified on the basis of the specificity of Sr resin with respect to TEVA and UTEVA. While this also applies in the case of the matrices studied here, differences among elements increase, probably because they show different behaviour during sample treatment due to their different chemical features. For example, in the coprecipitation stages of the analyses of concrete and steel, strontium, being an alkaline earth, quantitatively precipitates forming the same compound as calcium (oxalate or phosphate), while actinides are assumed to coprecipitate by means of other mechanisms. Apart from that, in the case of steel, during the treatment with NaNO2 for the adjustment of plutonium oxidation state, the solution became turbid and it needed filtering before going through the cartridges. This residue, despite its little amount, might have removed part of the radionuclides under study. On the other hand the amount of sample which could be processed was different for each treatment and this adds difficulty to the comparison of results. In the case of graphite and concrete, the amount of sample did not reach 1 g, while in the case of steel it was possible to work with 4.4 g. There is also the fact that matrix effects itself were different in each case and might have different impact in each radionuclide separation. These considerations provide arguments for the poor performance the method showed in the separation of actinides in steel and also give hints about the possible pitfalls on the application of straightforward methods to complex matrices, which will require further investigation.
Fig. 1. Recovery values (%) obtained for the three replicates for single-radionuclide (a) and multielemental (b) radiochemical separations working with spiked aqueous solutions. Error bars represent combined standard uncertainty.
medium which could be analysed applying the proposed radiochemical scheme. Concrete samples were fused with a mixture of lithium borates to initially achieve the dissolution of the sample, but this makes the removal of silicates necessary before starting with the radiochemical separation, as these tend to polymerise and end up blocking the resin cartridges otherwise. Removal of silicates was done by means of its polymerisation with PEG (Croudace et al., 1998). In the case of steel, samples were dissolved with aqua regia. Because iron is an interfering element in radiochemical separations utilizing extraction chromatography, the radionuclides of interest were separated from it by means of a calcium oxalate precipitation modifying a method in the literature (Boll et al., 1997). For the analysis of graphite, closed vessel microwave digestion followed by evaporation to dryness was the chosen treatment. The use of microwave system allows producing colourless and clear extracts by using only sulphuric and nitric acids, which is an advantage respect most of the treatments found in the literature, which require the use of perchloric acid (Buzzelli and Mosen, 1977; Pierson, 1993). The main disadvantage of this alternative is the small amount of sample that can be attacked in one vessel due to the high pressure produced during decomposition of graphite. However, most microwave digestions systems allow working with multiple vessels at once. All sample treatment procedures were designed to merge into the radiochemical analysis scheme previously developed and to be applied within a working day (8 h), although some of the stages could not meet this requirement (e.g. silicate precipitation in the analysis of concrete). The spectra corresponding to plutonium and uranium measured by alpha-particle spectrometry showed no crossed interference of these elements in any case. Moreover, in the analysis of concrete, the uranium spectra showed the lines of natural 238U and 234U, which is also a signal of the good performance of the method. The results of the study on the different matrices are presented in Fig. 2. These values indicate that the recovery values decrease with
4. Conclusions The purpose of the present study was to investigate the feasibility of simultaneous radiochemical separations of plutonium, uranium and strontium on samples of interest in the field of decommissioning: concrete, steel and graphite. Straightforward sample treatment strategies were established which achieve to extract the elements of interest and eliminate the main interferences. Moreover, these strategies could be combined with a radiochemical separation scheme for the simultaneous separation of the radionuclides of interest. Despite the fact that the study was carried out with spiked samples 3
Applied Radiation and Isotopes xxx (xxxx) xxx–xxx
D. Zapata-García, H. Wershofen
Buzzelli, G., Mosen, A.W., 1977. Perchloric acid dissolution of graphite and pyrolytic carbon. Talanta 24, 383–385. http://dx.doi.org/10.1016/0039-9140(77)80024-6. Cross, M.T., Green, T.H., Adsley, I., 2012. Characterisation of radioactive materials in redundant nuclear facilities: key issues for the decommissioning plan. In: Laraia, M. (Ed.), Nuclear Decommissioning. Planning, Execution and International Experience. Woodhead Publishing, pp. 87–116 (http://dx.doi.org/10.1533/ 9780857095336.1.87). Croudace, I., Warwick, P., Taylor, R., Dee, S., 1998. Rapid procedure for plutonium and uranium determination in soils using a borate fusion followed by ion-exchange and extraction chromatography. Anal. Chim. Acta 371, 217–225. http://dx.doi.org/10. 1016/S0003-2670(98)00353-5. Eichrom Technologies, 2014. Americium, Neptunium, Plutonium, Thorium, Curium, Uranium, and Strontium in Water (ACW17VBS Rev.1.2). IAEA, 2010. A Procedure for the Rapid Determination of Pu Isotopes and Am-241 in Soil and Sediment Samples by Alpha Spectrometry, IAEA Analytical Quality in Nuclear Applications Series. International Atomic Energy Agency, Vienna. IAEA, 2014. A Procedure for the Sequential Determination of Radionuclides in Environmental Samples, IAEA Analytical Quality in Nuclear Applications Series. International Atomic Energy Agency, Vienna. IAEA, 2015. Nuclear technology review 2015 (Report), Nuclear Technology Review. Vienna. Maxwell, S.L., Culligan, B., Hutchison, J.B., McAlister, D.R., 2015. Rapid fusion method for the determination of Pu, Np, and Am in large soil samples. J. Radioanal. Nucl. Chem. 305, 599–608. http://dx.doi.org/10.1007/s10967-015-3992-x. Maxwell, S.L., Culligan, B.K., Kelsey-Wall, A., Shaw, P.J., 2011. Rapid radiochemical method for determination of actinides in emergency concrete and brick samples. Anal. Chim. Acta 701, 112–118. http://dx.doi.org/10.1016/j.aca.2011.06.011. Pierson, H.O., 1993. Handbook of Carbon, Graphite, Diamond and Fullerenes: Properties, Processing and Applications. Noyes Publications. http://dx.doi.org/10.1016/B978-08155-1339-1.50001-3. Tavcar, P., Jakopic, R., Benedik, L., 2005. Sequential determination of Am-241, Np-237, Pu radioisotopes and Sr-90 in soil and sediment samples. Acta Chim. Slov. Chim. Slov. 52, 60–66. Wang, J.-J., Chen, I.-J., Chiu, J.-H., 2004. Sequential isotopic determination of plutonium, thorium, americium, strontium and uranium in environmental and bioassay samples. Appl. Radiat. Isot. 61, 299–305. http://dx.doi.org/10.1016/j. apradiso.2004.03.025.
and not with real decommissioning materials, it did substantiate that these strategies yield good results in most cases and provide evidence of adequateness for its use in the analysis of such complex matrices. However, the findings showed significant variability in the recoveries of the separation. Further research should be done to investigate the sources for the variability observed and to study the sample treatment procedures in more detail in order to identify the processes which produce the loss of the elements of interest in order to minimise it. Additionally, variability due to composition changes, aging and other aspects should be taken into account by the application of the method to other samples. The methods proposed are promising alternatives to the methods usually applied in radioactivity laboratories. If these were to be combined with a technique providing fast measurements, accurate results fur such complex matrices would be achievable in a rapid way. Acknowledgements We thank NPL for providing the concrete and graphite samples. This work was carried out with funding by the European Union under the EMRP joint research project ENV54 MetroDecom. The EMRP is jointly funded by the EMRP participating countries within EURAMET and the European Union. References Boll, R.A., Schweitzer, G.K., Garber, R.W., 1997. An improved actinide separation method for environmental samples. J. Radioanal. Nucl. Chem. 220, 201–206. http://dx.doi. org/10.1007/BF02034856.
4
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Development of radiochemical analysis strategies for decommissioning activities ⁎
D. Zapata-García , H. Wershofen Physikalisch-Technische Bundesanstalt (PTB), Bundesallee 100, 38116 Braunschweig, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Decommissioning Extraction chromatography Radiochemical analysis Concrete Steel Graphite
Radioactive waste generated in decommissioning activities need be classified according to their radioactive content. The radiological information required by national authorities includes diverse alpha and beta emitters, which can only be determined after a radiochemical separation process. This paper presents the work on the development of radiochemical methods for the simultaneous separation of several radionuclides in concrete, steel and graphite samples, on the basis of individual sample treatments which merge in a common radiochemical separation procedure based on extraction chromatography.
1. Introduction Currently there are 438 nuclear power plants in operation worldwide, and a further 149 that are shut down or are undergoing decommissioning, including 17 that have been fully decommissioned. These numbers become even bigger if fuel cycle facilities and research reactors are included (IAEA, 2015). Given that many of the nuclear facilities currently in operation will reach the end of their design lives within the next two decades, significant decommissioning activities are expected for several decades to come and these are one of the most challenging technological legacy issues that many countries will face in the near future. Most of the waste generated during decommissioning is not radiologically restricted and can be released into the environment or sent for recycling, but the rest will end up in different repositories, classified according to their radioactive content. Classification and control need to be done accurately in order to optimise the use of the limited repository space, but also to ensure that both the personnel involved and the public are not needlessly exposed to radiation and to minimise the environmental impact of such work. Thus, the characterisation of the radioactive materials present in nuclear facilities is a key technical aspect in the development of any decommissioning plan, but which proceeds also during and after the decommissioning process. The information required by national authorities, especially in the early stages of the decommissioning process, includes a great number of radionuclides, and characterisation is generally done combining several strategies, which range from activation calculations to sample taking and analysis in the laboratory. Whenever direct measurements can be
⁎
made, these are preferred because they produce reliable results and are faster and cheaper. However, radiochemical analysis is the only alternative when the radionuclides of interest are difficult to measure, as is the case of alpha and beta emitters, which need to be separated from the matrix before measurement. Although the costs involved in sample analysis are large, it is possible to make significant savings by the adoption of best available practices, such as the use of validated methods for on-site measurements and simultaneous determination of more than one radionuclide whenever possible (IAEA, 2014, 2010). In addition to this, the development of radiochemical methods which reduce the times of analysis from several to ideally one working day is one of the objectives many laboratories pursue nowadays, due to the time savings such improvements would entail. The samples that need processing within decommissioning works are also challenging, as they combine a complex matrix with an unknown distribution of the radionuclides. Among the variety of matrices probably are concrete, steel and graphite the ones of greatest interest (Cross et al., 2012). The present work deals with the development of procedures for the simultaneous determination of alpha and beta emitters in three matrices: concrete, steel and graphite. On the basis of individual sample treatment strategies based on the chemical properties of these matrices, the work focused in making these procedures merge in a common radiochemical analysis scheme based on extraction chromatography setups for the simultaneous separation of up to three radionuclides.
Corresponding author. E-mail address: [email protected] (D. Zapata-García).
http://dx.doi.org/10.1016/j.apradiso.2017.02.038 Received 16 September 2016; Received in revised form 13 February 2017; Accepted 20 February 2017 0969-8043/ © 2017 Elsevier Ltd. All rights reserved.
Please cite this article as: Zapata-García, D., Applied Radiation and Isotopes (2017), http://dx.doi.org/10.1016/j.apradiso.2017.02.038
Applied Radiation and Isotopes xxx (xxxx) xxx–xxx
D. Zapata-García, H. Wershofen
2. Materials and methods
9 M HCl and possibly interfering Th and Cm isotopes were stripped with 20 ml 4 M HCl. Finally, plutonium was stripped with 20 ml 0.1% NH4OCl in 2 M HCl. UTEVA resin cartridges were used for the extraction of uranium. After charging the sample, media was switched to HCl with 3 ml 9 M HCl and finally uranium was stripped with 20 ml 0.01 M HCl. Sr resin cartridges were used for the extraction of strontium. After charging the sample, the resin was rinsed with 10 ml 8 M HNO3 and strontium was stripped with 10 ml of 0.05 M HNO3. Multielement separations were performed using a setup of all three cartridges (TEVA, UTEVA, Sr) in tandem. Once the samples had gone through all three cartridges, these were separated and elution of each radionuclide was performed individually by applying the procedures described above. Plutonium and uranium were electroplated and measured by alphaparticle spectrometry while strontium was measured directly after elution without further treatment by high-resolution gamma spectrometry.
2.1. Reagents and samples High purity double deionised water and analytical grade reagents were used throughout this study. Commercial solutions of 242Pu (NIST), 232U (NIST) and 85Sr (PTB) were used as tracers. The activities added to each sample were 200 mBq in the case of actinides and 1 Bq in the case of 85Sr. Concrete samples were produced at National Physical Laboratory (NPL) using commercial components and commercial steel and graphite were used. 2.2. Equipment A fully automated electric fluxer K1 Pime (Katanax, Canada) and a START 1500 closed vessel microwave system (MLS GmbH, Germany) were used in this study. Radiometric measurements were done with alpha spectrometers Alpha Duo (Ortec) and high resolution gamma spectrometers IGC 34, PGT and GC5019-7500SL-R (Canberra).
3. Results and discussion The radiochemical separation scheme proposed in this work is based on the combined use of three extraction chromatography resins, each of them used for the separation of one single radionuclide. Methods used were developed in the laboratory on the basis of the multiple approaches found in the literature. (Eichrom Technologies, 2014; Maxwell et al., 2015, 2011; Tavcar et al., 2005; Wang et al., 2004) The radiochemical separation schemes described in this work were tested with spiked aqueous solutions, firstly with a single radionuclide using its specific resin, and afterwards with solutions spiked with all three radionuclides of interest using the tandem setup. Three replicates of each test were performed. Once the separation schemes had been tested with aqueous solutions, the work focused on the feasibility of such schemes for the analysis of matrices of interest in the field of decommissioning: concrete, steel and graphite. The spectra corresponding to plutonium and uranium measured by alpha-particle spectrometry showed no crossed interference of these elements, which provides confirmatory evidence that the proposed separation scheme is adequate. The results of the tests using aqueous solutions are presented in Fig. 1. Contrary to expectations, recovery values in the case of plutonium were higher for the multielemental tests than for the single radionuclide tests. Spectra were investigated for possible contamination of progeny present in the other tracer solutions, but no such effect was observed. Being TEVA the first cartridge in the tandem setup, this could not be attributed to the effect that other cartridges might induce and the results should be considered to show the variability in the performance of the resin applying the proposed separation scheme. In the case of uranium, recovery values showed great variability in the case of multielemental tests, with values ranging from 23% to 91%. This variability does not correlate with the results of the other two radionuclides. The possible uptake of uranium by the TEVA resin, though unexpected, was investigated. The results showed that the amount being retained by TEVA resin was below 0.8%, so the variability on the uranium recoveries cannot be based on this phenomenon. No other reason for this performance could be found. In the case of strontium, in contrast to actinides, the values of the two series of analyses do not show significant differences. Moreover, recovery values are the highest observed (ranging 85–95%), which can be explained on the basis of the specificity of this resin for strontium separations. In order to apply the proposed radiochemical separation scheme to the matrices of interest in this study, it was necessary to introduce sample treatment strategies previous to the radiochemical separation. The different treatment strategies were decided taking into account the sample chemical features in order to obtain a solution in nitric acid
2.3. Sample treatment For the analysis of aqueous solutions, tracers were added to 20 ml of concentrated HNO3 and evaporated to dryness on a sand bath. The final residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al (NO3)3. Different sample treatment procedures were applied to the different matrices considered in this work. Ground concrete samples (0.5 g) were spiked with tracers and fused at 1000 °C in Pt/Au crucibles with a mixture of LiBO2/Li2B4O7 adding LiBr as a wetting agent. The fusion protocol lasted 10 min and the fused material was dissolved with diluted HNO3 immediately after fusion. Silicates were removed by adding polyethylene glycol (PEG) and filtrating after leaving to polymerise for at least 12 h. The solution was then heated, concentrated HNO3 and HF added to eliminate any rest of silicate, evaporated to dryness once more and redissolved in 40% HNO3. Alkali metals were removed by means of Ca3(PO4)2 precipitation and the residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al(NO3)3. Steel samples (4.4 g) were dissolved with aqua regia to produce around 50 ml of solution. Tracers were added at this point and iron was removed by forming its complex with oxalate and coprecipitating the actinides and strontium with calcium oxalate at pH =1.7, before the precipitation of Fe(OH)3 occurs. Oxalate precipitate was treated twice with 40 ml aqua regia and evaporated to incipient dryness to destroy the oxalates and then redissolved with HNO3 and concentrated to dryness again. The final residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al(NO3)3. Graphite samples (0.2 g) were digested with H2SO4 and HNO3 at 260 °C for 30 min in a closed vessel microwave system. After digestion tracers were added and the solution obtained was concentrated to near dryness, 50 ml concentrated HNO3 added and concentrated to dryness again. The final residue was dissolved with 10 ml of 6 M HNO3 and 10 ml of 1 M Al(NO3)3. 2.4. Radiochemical separation method Plutonium oxidation state in the solution was adjusted to Pu4+ before radiochemical separation by adding ferrous sulfamate followed by NaNO2. Radiochemical separation was performed using extraction chromatography cartridges (TrisKem International) on a vacuum box system. All cartridges were conditioned with 20 ml of 3 M HNO3 before adding the samples. TEVA resin cartridges were used for the extraction of plutonium. After charging the sample, media was switched to HCl by adding 4 ml of 2
Applied Radiation and Isotopes xxx (xxxx) xxx–xxx
D. Zapata-García, H. Wershofen
Fig. 2. Recovery values (%) obtained in the multielemental radiochemical separations of spiked concrete, steel and graphite. Error bars represent combined standard uncertainty.
respect to the values obtained with spiked solutions, but variation for each radionuclide is different. Strontium is the element which shows best performance, with recovery values over 70% in all cases. Actinides, on the contrary, present lower recovery values in all the tests and in the case of steel analyses these do not even reach 50%. There are different factors which can help to understand these differences. On the one hand, the chemical approaches applied previous to radiochemical separation have different levels of complexity for the different samples and might have different degrees of effectiveness. Strontium shows better performance than actinides, which in the case of aqueous solutions was justified on the basis of the specificity of Sr resin with respect to TEVA and UTEVA. While this also applies in the case of the matrices studied here, differences among elements increase, probably because they show different behaviour during sample treatment due to their different chemical features. For example, in the coprecipitation stages of the analyses of concrete and steel, strontium, being an alkaline earth, quantitatively precipitates forming the same compound as calcium (oxalate or phosphate), while actinides are assumed to coprecipitate by means of other mechanisms. Apart from that, in the case of steel, during the treatment with NaNO2 for the adjustment of plutonium oxidation state, the solution became turbid and it needed filtering before going through the cartridges. This residue, despite its little amount, might have removed part of the radionuclides under study. On the other hand the amount of sample which could be processed was different for each treatment and this adds difficulty to the comparison of results. In the case of graphite and concrete, the amount of sample did not reach 1 g, while in the case of steel it was possible to work with 4.4 g. There is also the fact that matrix effects itself were different in each case and might have different impact in each radionuclide separation. These considerations provide arguments for the poor performance the method showed in the separation of actinides in steel and also give hints about the possible pitfalls on the application of straightforward methods to complex matrices, which will require further investigation.
Fig. 1. Recovery values (%) obtained for the three replicates for single-radionuclide (a) and multielemental (b) radiochemical separations working with spiked aqueous solutions. Error bars represent combined standard uncertainty.
medium which could be analysed applying the proposed radiochemical scheme. Concrete samples were fused with a mixture of lithium borates to initially achieve the dissolution of the sample, but this makes the removal of silicates necessary before starting with the radiochemical separation, as these tend to polymerise and end up blocking the resin cartridges otherwise. Removal of silicates was done by means of its polymerisation with PEG (Croudace et al., 1998). In the case of steel, samples were dissolved with aqua regia. Because iron is an interfering element in radiochemical separations utilizing extraction chromatography, the radionuclides of interest were separated from it by means of a calcium oxalate precipitation modifying a method in the literature (Boll et al., 1997). For the analysis of graphite, closed vessel microwave digestion followed by evaporation to dryness was the chosen treatment. The use of microwave system allows producing colourless and clear extracts by using only sulphuric and nitric acids, which is an advantage respect most of the treatments found in the literature, which require the use of perchloric acid (Buzzelli and Mosen, 1977; Pierson, 1993). The main disadvantage of this alternative is the small amount of sample that can be attacked in one vessel due to the high pressure produced during decomposition of graphite. However, most microwave digestions systems allow working with multiple vessels at once. All sample treatment procedures were designed to merge into the radiochemical analysis scheme previously developed and to be applied within a working day (8 h), although some of the stages could not meet this requirement (e.g. silicate precipitation in the analysis of concrete). The spectra corresponding to plutonium and uranium measured by alpha-particle spectrometry showed no crossed interference of these elements in any case. Moreover, in the analysis of concrete, the uranium spectra showed the lines of natural 238U and 234U, which is also a signal of the good performance of the method. The results of the study on the different matrices are presented in Fig. 2. These values indicate that the recovery values decrease with
4. Conclusions The purpose of the present study was to investigate the feasibility of simultaneous radiochemical separations of plutonium, uranium and strontium on samples of interest in the field of decommissioning: concrete, steel and graphite. Straightforward sample treatment strategies were established which achieve to extract the elements of interest and eliminate the main interferences. Moreover, these strategies could be combined with a radiochemical separation scheme for the simultaneous separation of the radionuclides of interest. Despite the fact that the study was carried out with spiked samples 3
Applied Radiation and Isotopes xxx (xxxx) xxx–xxx
D. Zapata-García, H. Wershofen
Buzzelli, G., Mosen, A.W., 1977. Perchloric acid dissolution of graphite and pyrolytic carbon. Talanta 24, 383–385. http://dx.doi.org/10.1016/0039-9140(77)80024-6. Cross, M.T., Green, T.H., Adsley, I., 2012. Characterisation of radioactive materials in redundant nuclear facilities: key issues for the decommissioning plan. In: Laraia, M. (Ed.), Nuclear Decommissioning. Planning, Execution and International Experience. Woodhead Publishing, pp. 87–116 (http://dx.doi.org/10.1533/ 9780857095336.1.87). Croudace, I., Warwick, P., Taylor, R., Dee, S., 1998. Rapid procedure for plutonium and uranium determination in soils using a borate fusion followed by ion-exchange and extraction chromatography. Anal. Chim. Acta 371, 217–225. http://dx.doi.org/10. 1016/S0003-2670(98)00353-5. Eichrom Technologies, 2014. Americium, Neptunium, Plutonium, Thorium, Curium, Uranium, and Strontium in Water (ACW17VBS Rev.1.2). IAEA, 2010. A Procedure for the Rapid Determination of Pu Isotopes and Am-241 in Soil and Sediment Samples by Alpha Spectrometry, IAEA Analytical Quality in Nuclear Applications Series. International Atomic Energy Agency, Vienna. IAEA, 2014. A Procedure for the Sequential Determination of Radionuclides in Environmental Samples, IAEA Analytical Quality in Nuclear Applications Series. International Atomic Energy Agency, Vienna. IAEA, 2015. Nuclear technology review 2015 (Report), Nuclear Technology Review. Vienna. Maxwell, S.L., Culligan, B., Hutchison, J.B., McAlister, D.R., 2015. Rapid fusion method for the determination of Pu, Np, and Am in large soil samples. J. Radioanal. Nucl. Chem. 305, 599–608. http://dx.doi.org/10.1007/s10967-015-3992-x. Maxwell, S.L., Culligan, B.K., Kelsey-Wall, A., Shaw, P.J., 2011. Rapid radiochemical method for determination of actinides in emergency concrete and brick samples. Anal. Chim. Acta 701, 112–118. http://dx.doi.org/10.1016/j.aca.2011.06.011. Pierson, H.O., 1993. Handbook of Carbon, Graphite, Diamond and Fullerenes: Properties, Processing and Applications. Noyes Publications. http://dx.doi.org/10.1016/B978-08155-1339-1.50001-3. Tavcar, P., Jakopic, R., Benedik, L., 2005. Sequential determination of Am-241, Np-237, Pu radioisotopes and Sr-90 in soil and sediment samples. Acta Chim. Slov. Chim. Slov. 52, 60–66. Wang, J.-J., Chen, I.-J., Chiu, J.-H., 2004. Sequential isotopic determination of plutonium, thorium, americium, strontium and uranium in environmental and bioassay samples. Appl. Radiat. Isot. 61, 299–305. http://dx.doi.org/10.1016/j. apradiso.2004.03.025.
and not with real decommissioning materials, it did substantiate that these strategies yield good results in most cases and provide evidence of adequateness for its use in the analysis of such complex matrices. However, the findings showed significant variability in the recoveries of the separation. Further research should be done to investigate the sources for the variability observed and to study the sample treatment procedures in more detail in order to identify the processes which produce the loss of the elements of interest in order to minimise it. Additionally, variability due to composition changes, aging and other aspects should be taken into account by the application of the method to other samples. The methods proposed are promising alternatives to the methods usually applied in radioactivity laboratories. If these were to be combined with a technique providing fast measurements, accurate results fur such complex matrices would be achievable in a rapid way. Acknowledgements We thank NPL for providing the concrete and graphite samples. This work was carried out with funding by the European Union under the EMRP joint research project ENV54 MetroDecom. The EMRP is jointly funded by the EMRP participating countries within EURAMET and the European Union. References Boll, R.A., Schweitzer, G.K., Garber, R.W., 1997. An improved actinide separation method for environmental samples. J. Radioanal. Nucl. Chem. 220, 201–206. http://dx.doi. org/10.1007/BF02034856.
4