- Email: [email protected]
Quality control assurance of strontium-90 in foodstuffs by LSC
Applied Radiation and Isotopes ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Quality control assurance of strontium-90 in foodstuffs by LSC I. Lopes n, A. Mourato, J. Abrantes, G. Carvalhal, M.J. Madruga, M. Reis Instituto Superior Técnico, Campus Tecnológico e Nuclear, Laboratório de Proteção e Segurança Radiológica, Estrada Nacional 10 (ao km 139,7), Bobadela LRS 2695-066, Portugal
H I G H L I G H T S
Blank and background samples were analyzed according to internal quality control program established. Analysis of certified reference material was carried out to perform additional evaluation of the accuracy. Analysis of interlaboratory comparison sample showed Z-score and Z (ML) acceptable values. The Sr-90 activity measured in wild bilberry powder sample was in agreement with the reference value.
art ic l e i nf o
Keywords: Quality control Strontium-90 Foodstuffs Liquid scintillation
a b s t r a c t A method based on the separation of Sr-90 by extraction chromatography and beta determination by Liquid Scintillation Counting (LSC) technique was used for strontium analysis in food samples. The methodology consisted in prior sample treatment (drying and incineration) followed by radiochemical separation of Sr-90 by extraction chromatography, using the Sr-resin. The chemical yield was determined by gravimetric method, adding stable strontium to the matrix. Beta activity (Sr-90/Y-90) was determined using a low background liquid scintillation spectrometer (Tri-Carb 3170 TR/SL, Packard). The accuracy and the precision of the method, was performed previously through recovery trials with Sr-90 spiked samples, using the same type of matrices (milk, complete meals, meat and vegetables). A reference material (IAEA_321) was now used to measure the accuracy of the procedure. Participation in interlaboratory comparison exercises was also performed in order to establish an external control on the measurements and to ensure the adequacy of the method. & 2014 Elsevier Ltd. All rights reserved.
1. Introduction Because of its long physical and biological half-lives, Sr-90 is one of the most hazardous radionuclides and it may cause damage to the bone marrow (UNSCEAR, 2000). It undergoes β decay into Y-90 and both were released together to the environment, due to nuclear weapons testing in the atmosphere and from the nuclear fuel cycle. Immediately following a nuclear accident, the fresh fallout contains others radionuclides together with high radiostrontium activity ratio (Sr-89/Sr-90). The old fallout (several years after release) contains only beta emitter Sr-90 (T1/2¼ 28.5 years) and its daughter Y-90 (T1/ 2¼64.4 h). For these reason, the strontium levels in the environmental samples, with particular emphasis on foodstuffs, are of particular concern (Groska et al., 2012; Spasova et al., 2008). Radiochemistry methods for the determination of Sr-90 in environmental samples require complex techniques for strontium to be separated from the sample matrix and from other interfering
n
Corresponding author. E-mail address: [email protected] (I. Lopes).
radionuclides (Vajda and Kim, 2010; Kim et al., 2009). The Sr-90 and its daughter Y-90 must also be separated from the sample matrix prior to measurement. In case of environmental samples containing low levels of Sr-90, large amounts of sample are often required which makes sample processing not only more labor-intensive but also required more sensitive methods. By solid phase extraction, using a strontium specific resin (Sr-resin, Eichrom), Sr-90 can be isolated without interferences from other radionuclides (Lee et al., 2013). The Sr-resin is specific to Sr ions and enables rapid and simple column chromatographic separation of strontium from calcium, potassium and many other elements with of 3–8 mol l 1 HNO3 solution (Grahek et al., 2011). The determination of Sr-90 can be performed by measuring the radioactivity of either Sr-90, Y-90 or of both if the degree of secular equilibrium is known (Karacan, 2011). The beta measurements can be performed using liquid scintillation spectrometers or proportional counters (Popov et al., 2009). Although the background of liquid scintillation spectrometers are higher than that of proportional counters, liquid scintillation is favored due to the much better spectral resolution. In general, appropriate procedures and validated methods should be used, following the requirements specified in ISO/IEC
http://dx.doi.org/10.1016/j.apradiso.2014.01.022 0969-8043 & 2014 Elsevier Ltd. All rights reserved.
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i
I. Lopes et al. / Applied Radiation and Isotopes ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
17025 international standard (ISO/IEC, 2005). The validation is the process used to confirm that the analytical procedure employed for a specific test is suitable for its intended use and should be as extensive as necessary to meet the needs of the given application. The methods need to be validated or revalidated, whenever the conditions for which the method has been validated changed (e.g., scope of the method, different matrix, instrument with different characteristics) (Drábová et al., 2012). The QC programs (internal and external) should include, among others, all necessary actions to assure if the equipment is calibrated and operating satisfactorily, to verify specified requirements such as accuracy and precision, and to detect potential faults in routine measurements, and should also include regular participation in interlaboratory comparison exercises. Results of validation trials using spiked samples were already reported (Lopes et al., 2010). The aim of this paper is to evaluate the accuracy of the method using a reference material and interlaboratory comparison samples.
2. Experimental The analytical method applied for Sr-90 determination in foodstuffs has been described elsewhere (Lopes et al., 2010). The method is based on the digestion of the sample and separation of strontium by extraction chromatography, using the Sr-resin. About 7 g of ashes were taken for analysis. During the digestion of the sample, stable strontium was added (E20 mg) to determine the chemical yield gravimetrically, and the mixture was heated to boiling and filtered. The Sr-90 was separated by oxalate and carbonate precipitations. Afterwards, the residue was dissolved with 3 mol l 1 HNO3 and loaded onto a 20 ml column filled with Sr- resin (100–150 μm), prerinsed with 3 mol l 1 HNO3. The column was rinsed first with 8 mol l 1 HNO3 followed by 3 mol l 1 HNO3 and the strontium retained in the column was stripped with 0.05 mol l 1 HNO3. After evaporation to dryness, the residue was dissolved with 8 ml of 0.1 mol l 1 HCl and mixed with 12 ml of Ultima Gold LLT cocktail (Perkin Elmer) in a glass scintillation vial. The measurements were performed by Liquid Scintillation Counting (LSC) technique using Packard Model Tri-Carb 3170 TR/SL spectrometer. The detection system was calibrated by measuring Sr-90/Y-90 standard sources, and using CCl4 as quenching agent. The procedure was also adapted for small columns (2 ml) when the amount of ash sample was not enough to perform the analysis. Fig. 1. Background control charts (Years: 2011 (a); 2012 (b)).
3. Results 3.1. Internal quality control Blank and background samples were analyzed, according to internal quality control program. A background sample was always measured with each set of 10 samples (Fig. 1) within each batch and analyses of blank samples were also performed (Fig. 2) for each batch. Control charts were used to help in the identification of possible systematic deviation from regular performance in time, for inspection of trends and to verify the stability of the counter. The average background counts ðXÞ was calculated, the standard deviation was used to set the warning (X 72s) and actions limits (X 7 3s), respectively, considering the first background measurements carried out during the calibration. For blank samples, the control chart was established at calibration date but the limit lines are recalculated for all plotted data. The equipment was operated under controlled room temperature conditions. Calibration of the LSC spectrometer is carried out once per year using Sr-90/Y-90 standards prepared from a certified standard solution (Amersham)
with increasing amounts of CCl4 (quenching agent) allowing the determination of the calibration curve (Quench Indicating Parameter not constant). The accuracy and the precision were determined during the validation trials, using Sr-90/Y-90 spiked samples, prepared from a mixture of several foodstuffs samples ashes (Lopes et al., 2010). The precision was estimated to be less than 4% and a good agreement between the added and the measured activities was obtained (the recovery percentages ranged from 70 to 80%). In routine work, replicates samples are not performed regularly since the amount of ashes obtained from the samples incineration is not enough to perform duplicate and/or triplicate samples. Furthermore, since the Sr-90 activity levels were below the detection limits, is not possible, in each batch, evaluate the precision of the procedure. Analysis of a certified reference material was also carried out, in order to perform additional evaluation of the accuracy for the validation of the method. The reference material (IAEA_321) used to verify the accuracy of the method was a milk powder collected from European
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i
I. Lopes et al. / Applied Radiation and Isotopes ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Table 1 Results of Sr-90 activity in IAEA_321 reference material. Reference material
Reference value (Bq kg 1)
Measured value (Bq kg 1 7 U; k¼ 2)
IAEA_321
3.16–3.44
3.427 0.21 (7 g sample ash) 2.807 0.89 (2 g sample ash)
Table 2 Comparison of expected value and measured Sr-90 activity in CSN/CIEMAT interlaboratory comparison sample. Radionuclide
Reference value (Bq kg 1 ash)
Measured value (Bq kg 1 ash 7 U; k ¼2)
Sr-90
47.7 7 8.6
47.8 7 6.6 (technician A) 41.9 7 5.7 (technician B)
Z
Z (ML)
0.01
0.01
0.67
0.66
Table 3 Results of Sr-90 activity in EC interlaboratory comparison wild bilberry powder.
Fig. 2. Blank control charts (Years: 2011 (a); 2012 (b)).
dairy-products processing plant. The material was produced from cow milk obtained from animals that had grazed on land contaminated with radioactive fallout resulting from the Chernobyl accident in 1986. The water content of the sample was determined (7.38%). After incineration at 660 1C the percentage of water was negligible (0.05%). Aliquots of 7 and 2 g were used for the Sr-90 analysis. The results presented in Table 1 are in agreement with the reference interval, ranging from 3.16 to 3.44 Bq kg 1. The relative deviations (%D) in both cases were 3.64% and 15.15%, respectively, less than 25% the acceptance laboratory criteria for the accuracy. The relative deviation was lower in the case where larger amount of sample was used. 3.2. External quality control A sample of ashes containing natural and artificial radionuclides with low levels of activity, prepared by University of Barcelona was analyzed. These ashes (15 kg) resulted from the calcination (600 1C) of complete meal samples, held in five laboratories, were milled and
En
Radionuclide Reference value (Bq kg 1 dry mass 7 U; k ¼2)
Measured value (Bq kg 1 dry mass 7 U; k¼ 2)
%D
Sr-90
1257 23
18 1.1
1537 8
homogenized. The activity control was made by gamma and alpha spectrometry. Aliquots (2 kg) were then fortified with artificial radionuclides (Am-241, Pu-238, Pu-239, Sr-90, Ni-63, Fe-55 and Co-60) and sub-samples of 200 g were sent to the participants (Gascó et al., 2011). The water content of the sample (0.18%) was determined before performing the analysis. Afterwards, the sample was incinerated and two aliquots of ashes were used for Sr-90 analysis. A lower chemical yield (28%) was obtained for the two aliquots, however the measured values (Table 2) are similar to the reference value with relative deviations (%D1 ¼0.2%; %D2 ¼12%) less than 25% (the acceptance laboratory criteria for the accuracy). The Z-score and Z (ML) parameters were calculated applying the same statistical treatment used by the interlaboratory comparison organizer (Gascó et al., 2011). The performance of the method is considered to be acceptable if the difference between the robust mean of the measured values and the recommended value isr 2. The data obtained showed Z-score and Z (ML) acceptable values (Table 2). EC interlaboratory comparison wild bilberry powder sample, containing high levels of Cs-137 and Sr-90, was also analyzed. Due to natural uptake in the environment (Chernobyl region) the radionuclides were metabolized by the plants and no spiking was applied. After the material has been processed to a free-flowing powder, it was sieved, homogenized and bottled in units of approximately 100 g (CRM_IRMM_426) (Meresova and Watjen, 2013). The water content was determined (3.52%) using a small aliquot (2.0 g) and an oven drying procedure (at 90 1C for 40 min). The remaining sample (97.8 g) was incinerated at 660 1C and only 2 g of ash sample was recovered. Consequently, the routine procedure could not be applied and the total amount of ashes was taken for analysis. Stable strontium was added (E5 mg) to determine the chemical yield gravimetrically. The volumes and amounts of reagents for sample digestion, oxalate/carbonate precipitations and column rinsed were adjusted according to the ash sample amount used. Instead of 20 ml chromatographic column, a small pre-packed column of 2 ml, filled with 0.7 g Sr- resin was used. The chemical
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i
I. Lopes et al. / Applied Radiation and Isotopes ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
Fig. 3. Results of EC interlaboratory comparison on Sr-90 in wild bilberry powder (Laboratory code number: 49).
yield (7978%) obtained was similar to other yields, reported before (Lopes et al., 2010) and still within the range 70 to 80%. The reference and measured Sr-90 activity concentrations are presented in Table 3. Results of evaluation of EC interlaboratory comparison exercise on the determination of Sr-90, in wild bilberry powder, are presented in Fig. 3 (Meresova and Watjen, 2013). The solid lines indicate the reference activity concentrations and their corresponding standard deviations are plotted in dashed lines (representing720% range from the reference values). The percentage of relative deviation obtained in this case was %D¼ 18%. Evaluation based on the concept of En number was also performed. The En number takes into account the deviation of the activity concentration value reported by each laboratory from the reference value, and the combination of expanded uncertainties associated to them (Watjen et al., 2008). Experimental results are considered satisfactory if |En|r1 and warning if the 1o|En|r1.5. The En for this exercise was |En|¼1.1, meaning that further investigation should be taken into account, perhaps regarding the estimation of the uncertainties, given that %D was already within the acceptable criteria. 4. Conclusions Evaluation of the accuracy of the method using reference material and interlaboratory comparison samples was performed. The validation trials conducted previously have been complemented with this new data and have proven the adequacy of the method for the routine measurements of Sr-90 in foodstuffs, by LSC technique.
Grahek, Ž, Ivšćic, A.G., Krljan, N., Nodilo, M., 2011. Separation of Sr in combination of ion exchange and Sr resin with alcohol-nitric acid solution and rapid determination of 90Sr in wine and soil samples. J. Radioanal. Nucl. Chem 289, 437–449. Groska, J., Molnár, Z., Bokori, E., Vadjan, N., 2012. Simultaneous determination of 89
Sr and 90Sr comparison of methods and calculation techniques. J. Radioanal. Nucl. Chem. 291, 707–715. Gascó, C., Trinidad, J.A., Suárez, J.A., Llauradó, Valiño, M.F., 2011. Intercomparación analítica entre laboratorios de radiactividad ambiental. Evaluación de la intercomparación analítica de radionucleidos en muestras ambientales Campaña 2010–2011 (cenizas de alimentos). CSN/CIEMAT intercomparación 2010. ISO/IEC, 2005. International Standard ISO/IEC 17025; General Requirements for the Competence of Testing and Calibration Laboratories, 2nd edition. Karacan, F., 2011. The simple radiochemical determination of 90Sr in environmental solid samples by solvent extraction. J. Radioanal. Nucl. Chem. 288, 685–691. Kim, C.K., AL-Hamwi, A., Törvényi, Kis-benedek G., Sansone, U., 2009. Appl. Radiat. Isot. 67, 786–793. Lee, M.H., Park, T.H., Park, J.H, Song, K., K., Lee, M.S., 2013. Radiochemical separation of Pu, U, Am and Sr isotopes in environmental samples using extraction chromatographic resins. J. Radioanal. Nucl. Chem. 295, 1419–1422. Lopes, I., Madruga, M.J., Mourato, A., Abrantes, J., Reis, M., 2010. Determination of 90 Sr in Portuguese foodstuffs. J. Radioanal. Nucl. Chem. 286, 335–340. Meresova, J., Watjen, U., 2013. Evaluation of EC Comparison on the Determination of 40K, 90Sr, and 137Cs in Bilberry Powder (Draft). European Commission, Joint Research Centre, ISBN 978-92-79-22551-2. Popov, L., Mihailova, G., Hristova, I., Dimitrova, P., Tzibranski, R., Avramov, V., Naidenov, I., Stoenelova, B., 2009. Separation of strontium from calcium by the use of sodium hydroxide and its application for the determination of long-term background activity concentrations of 90Sr in 100 km area around Kozloduy Nuclear Power Plant (Bulgaria). J. Radioanal. Nucl. Chem. 279, 49–64. Spasova, Y., Wätjen, U., Altzitzoglou, T., 2008. European measurement comparison of 137Cs, 40K and 90Sr in milk powder. J. Radioanal. Nucl. Chem 277, 211–215. UNSCEAR, 2000. United Nations Scientific Committee on the Effects of Atomic Radiation and Effects of Ionizing Radiation, I, 1–654. Vajda, N., Kim, C.K., 2010. Determination of radiostrontium isotopes: a review of analytical methodology. Appl. Radiat. Isot. 68, 2306–2326. Watjen, U., Spasova, Y., Altzitzoglou, T., 2008. Measurement comparisons of radioactivity among European monitoring laboratories for the environment and food stuff. Appl. Radiat. Isot. 66, 742–749.
References Drábová, V., Strišovská, J., Dvorská, L., Kuruc, J., 2012. Validation of 90Sr separation method using molecular recognition product AnaLig Sr01 gel and extraction chromatography Sr_ resin. J. Radioanal. Nucl. Chem. 292, 1141–1145.
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Quality control assurance of strontium-90 in foodstuffs by LSC I. Lopes n, A. Mourato, J. Abrantes, G. Carvalhal, M.J. Madruga, M. Reis Instituto Superior Técnico, Campus Tecnológico e Nuclear, Laboratório de Proteção e Segurança Radiológica, Estrada Nacional 10 (ao km 139,7), Bobadela LRS 2695-066, Portugal
H I G H L I G H T S
Blank and background samples were analyzed according to internal quality control program established. Analysis of certified reference material was carried out to perform additional evaluation of the accuracy. Analysis of interlaboratory comparison sample showed Z-score and Z (ML) acceptable values. The Sr-90 activity measured in wild bilberry powder sample was in agreement with the reference value.
art ic l e i nf o
Keywords: Quality control Strontium-90 Foodstuffs Liquid scintillation
a b s t r a c t A method based on the separation of Sr-90 by extraction chromatography and beta determination by Liquid Scintillation Counting (LSC) technique was used for strontium analysis in food samples. The methodology consisted in prior sample treatment (drying and incineration) followed by radiochemical separation of Sr-90 by extraction chromatography, using the Sr-resin. The chemical yield was determined by gravimetric method, adding stable strontium to the matrix. Beta activity (Sr-90/Y-90) was determined using a low background liquid scintillation spectrometer (Tri-Carb 3170 TR/SL, Packard). The accuracy and the precision of the method, was performed previously through recovery trials with Sr-90 spiked samples, using the same type of matrices (milk, complete meals, meat and vegetables). A reference material (IAEA_321) was now used to measure the accuracy of the procedure. Participation in interlaboratory comparison exercises was also performed in order to establish an external control on the measurements and to ensure the adequacy of the method. & 2014 Elsevier Ltd. All rights reserved.
1. Introduction Because of its long physical and biological half-lives, Sr-90 is one of the most hazardous radionuclides and it may cause damage to the bone marrow (UNSCEAR, 2000). It undergoes β decay into Y-90 and both were released together to the environment, due to nuclear weapons testing in the atmosphere and from the nuclear fuel cycle. Immediately following a nuclear accident, the fresh fallout contains others radionuclides together with high radiostrontium activity ratio (Sr-89/Sr-90). The old fallout (several years after release) contains only beta emitter Sr-90 (T1/2¼ 28.5 years) and its daughter Y-90 (T1/ 2¼64.4 h). For these reason, the strontium levels in the environmental samples, with particular emphasis on foodstuffs, are of particular concern (Groska et al., 2012; Spasova et al., 2008). Radiochemistry methods for the determination of Sr-90 in environmental samples require complex techniques for strontium to be separated from the sample matrix and from other interfering
n
Corresponding author. E-mail address: [email protected] (I. Lopes).
radionuclides (Vajda and Kim, 2010; Kim et al., 2009). The Sr-90 and its daughter Y-90 must also be separated from the sample matrix prior to measurement. In case of environmental samples containing low levels of Sr-90, large amounts of sample are often required which makes sample processing not only more labor-intensive but also required more sensitive methods. By solid phase extraction, using a strontium specific resin (Sr-resin, Eichrom), Sr-90 can be isolated without interferences from other radionuclides (Lee et al., 2013). The Sr-resin is specific to Sr ions and enables rapid and simple column chromatographic separation of strontium from calcium, potassium and many other elements with of 3–8 mol l 1 HNO3 solution (Grahek et al., 2011). The determination of Sr-90 can be performed by measuring the radioactivity of either Sr-90, Y-90 or of both if the degree of secular equilibrium is known (Karacan, 2011). The beta measurements can be performed using liquid scintillation spectrometers or proportional counters (Popov et al., 2009). Although the background of liquid scintillation spectrometers are higher than that of proportional counters, liquid scintillation is favored due to the much better spectral resolution. In general, appropriate procedures and validated methods should be used, following the requirements specified in ISO/IEC
http://dx.doi.org/10.1016/j.apradiso.2014.01.022 0969-8043 & 2014 Elsevier Ltd. All rights reserved.
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i
I. Lopes et al. / Applied Radiation and Isotopes ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
17025 international standard (ISO/IEC, 2005). The validation is the process used to confirm that the analytical procedure employed for a specific test is suitable for its intended use and should be as extensive as necessary to meet the needs of the given application. The methods need to be validated or revalidated, whenever the conditions for which the method has been validated changed (e.g., scope of the method, different matrix, instrument with different characteristics) (Drábová et al., 2012). The QC programs (internal and external) should include, among others, all necessary actions to assure if the equipment is calibrated and operating satisfactorily, to verify specified requirements such as accuracy and precision, and to detect potential faults in routine measurements, and should also include regular participation in interlaboratory comparison exercises. Results of validation trials using spiked samples were already reported (Lopes et al., 2010). The aim of this paper is to evaluate the accuracy of the method using a reference material and interlaboratory comparison samples.
2. Experimental The analytical method applied for Sr-90 determination in foodstuffs has been described elsewhere (Lopes et al., 2010). The method is based on the digestion of the sample and separation of strontium by extraction chromatography, using the Sr-resin. About 7 g of ashes were taken for analysis. During the digestion of the sample, stable strontium was added (E20 mg) to determine the chemical yield gravimetrically, and the mixture was heated to boiling and filtered. The Sr-90 was separated by oxalate and carbonate precipitations. Afterwards, the residue was dissolved with 3 mol l 1 HNO3 and loaded onto a 20 ml column filled with Sr- resin (100–150 μm), prerinsed with 3 mol l 1 HNO3. The column was rinsed first with 8 mol l 1 HNO3 followed by 3 mol l 1 HNO3 and the strontium retained in the column was stripped with 0.05 mol l 1 HNO3. After evaporation to dryness, the residue was dissolved with 8 ml of 0.1 mol l 1 HCl and mixed with 12 ml of Ultima Gold LLT cocktail (Perkin Elmer) in a glass scintillation vial. The measurements were performed by Liquid Scintillation Counting (LSC) technique using Packard Model Tri-Carb 3170 TR/SL spectrometer. The detection system was calibrated by measuring Sr-90/Y-90 standard sources, and using CCl4 as quenching agent. The procedure was also adapted for small columns (2 ml) when the amount of ash sample was not enough to perform the analysis. Fig. 1. Background control charts (Years: 2011 (a); 2012 (b)).
3. Results 3.1. Internal quality control Blank and background samples were analyzed, according to internal quality control program. A background sample was always measured with each set of 10 samples (Fig. 1) within each batch and analyses of blank samples were also performed (Fig. 2) for each batch. Control charts were used to help in the identification of possible systematic deviation from regular performance in time, for inspection of trends and to verify the stability of the counter. The average background counts ðXÞ was calculated, the standard deviation was used to set the warning (X 72s) and actions limits (X 7 3s), respectively, considering the first background measurements carried out during the calibration. For blank samples, the control chart was established at calibration date but the limit lines are recalculated for all plotted data. The equipment was operated under controlled room temperature conditions. Calibration of the LSC spectrometer is carried out once per year using Sr-90/Y-90 standards prepared from a certified standard solution (Amersham)
with increasing amounts of CCl4 (quenching agent) allowing the determination of the calibration curve (Quench Indicating Parameter not constant). The accuracy and the precision were determined during the validation trials, using Sr-90/Y-90 spiked samples, prepared from a mixture of several foodstuffs samples ashes (Lopes et al., 2010). The precision was estimated to be less than 4% and a good agreement between the added and the measured activities was obtained (the recovery percentages ranged from 70 to 80%). In routine work, replicates samples are not performed regularly since the amount of ashes obtained from the samples incineration is not enough to perform duplicate and/or triplicate samples. Furthermore, since the Sr-90 activity levels were below the detection limits, is not possible, in each batch, evaluate the precision of the procedure. Analysis of a certified reference material was also carried out, in order to perform additional evaluation of the accuracy for the validation of the method. The reference material (IAEA_321) used to verify the accuracy of the method was a milk powder collected from European
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i
I. Lopes et al. / Applied Radiation and Isotopes ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Table 1 Results of Sr-90 activity in IAEA_321 reference material. Reference material
Reference value (Bq kg 1)
Measured value (Bq kg 1 7 U; k¼ 2)
IAEA_321
3.16–3.44
3.427 0.21 (7 g sample ash) 2.807 0.89 (2 g sample ash)
Table 2 Comparison of expected value and measured Sr-90 activity in CSN/CIEMAT interlaboratory comparison sample. Radionuclide
Reference value (Bq kg 1 ash)
Measured value (Bq kg 1 ash 7 U; k ¼2)
Sr-90
47.7 7 8.6
47.8 7 6.6 (technician A) 41.9 7 5.7 (technician B)
Z
Z (ML)
0.01
0.01
0.67
0.66
Table 3 Results of Sr-90 activity in EC interlaboratory comparison wild bilberry powder.
Fig. 2. Blank control charts (Years: 2011 (a); 2012 (b)).
dairy-products processing plant. The material was produced from cow milk obtained from animals that had grazed on land contaminated with radioactive fallout resulting from the Chernobyl accident in 1986. The water content of the sample was determined (7.38%). After incineration at 660 1C the percentage of water was negligible (0.05%). Aliquots of 7 and 2 g were used for the Sr-90 analysis. The results presented in Table 1 are in agreement with the reference interval, ranging from 3.16 to 3.44 Bq kg 1. The relative deviations (%D) in both cases were 3.64% and 15.15%, respectively, less than 25% the acceptance laboratory criteria for the accuracy. The relative deviation was lower in the case where larger amount of sample was used. 3.2. External quality control A sample of ashes containing natural and artificial radionuclides with low levels of activity, prepared by University of Barcelona was analyzed. These ashes (15 kg) resulted from the calcination (600 1C) of complete meal samples, held in five laboratories, were milled and
En
Radionuclide Reference value (Bq kg 1 dry mass 7 U; k ¼2)
Measured value (Bq kg 1 dry mass 7 U; k¼ 2)
%D
Sr-90
1257 23
18 1.1
1537 8
homogenized. The activity control was made by gamma and alpha spectrometry. Aliquots (2 kg) were then fortified with artificial radionuclides (Am-241, Pu-238, Pu-239, Sr-90, Ni-63, Fe-55 and Co-60) and sub-samples of 200 g were sent to the participants (Gascó et al., 2011). The water content of the sample (0.18%) was determined before performing the analysis. Afterwards, the sample was incinerated and two aliquots of ashes were used for Sr-90 analysis. A lower chemical yield (28%) was obtained for the two aliquots, however the measured values (Table 2) are similar to the reference value with relative deviations (%D1 ¼0.2%; %D2 ¼12%) less than 25% (the acceptance laboratory criteria for the accuracy). The Z-score and Z (ML) parameters were calculated applying the same statistical treatment used by the interlaboratory comparison organizer (Gascó et al., 2011). The performance of the method is considered to be acceptable if the difference between the robust mean of the measured values and the recommended value isr 2. The data obtained showed Z-score and Z (ML) acceptable values (Table 2). EC interlaboratory comparison wild bilberry powder sample, containing high levels of Cs-137 and Sr-90, was also analyzed. Due to natural uptake in the environment (Chernobyl region) the radionuclides were metabolized by the plants and no spiking was applied. After the material has been processed to a free-flowing powder, it was sieved, homogenized and bottled in units of approximately 100 g (CRM_IRMM_426) (Meresova and Watjen, 2013). The water content was determined (3.52%) using a small aliquot (2.0 g) and an oven drying procedure (at 90 1C for 40 min). The remaining sample (97.8 g) was incinerated at 660 1C and only 2 g of ash sample was recovered. Consequently, the routine procedure could not be applied and the total amount of ashes was taken for analysis. Stable strontium was added (E5 mg) to determine the chemical yield gravimetrically. The volumes and amounts of reagents for sample digestion, oxalate/carbonate precipitations and column rinsed were adjusted according to the ash sample amount used. Instead of 20 ml chromatographic column, a small pre-packed column of 2 ml, filled with 0.7 g Sr- resin was used. The chemical
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i
I. Lopes et al. / Applied Radiation and Isotopes ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
Fig. 3. Results of EC interlaboratory comparison on Sr-90 in wild bilberry powder (Laboratory code number: 49).
yield (7978%) obtained was similar to other yields, reported before (Lopes et al., 2010) and still within the range 70 to 80%. The reference and measured Sr-90 activity concentrations are presented in Table 3. Results of evaluation of EC interlaboratory comparison exercise on the determination of Sr-90, in wild bilberry powder, are presented in Fig. 3 (Meresova and Watjen, 2013). The solid lines indicate the reference activity concentrations and their corresponding standard deviations are plotted in dashed lines (representing720% range from the reference values). The percentage of relative deviation obtained in this case was %D¼ 18%. Evaluation based on the concept of En number was also performed. The En number takes into account the deviation of the activity concentration value reported by each laboratory from the reference value, and the combination of expanded uncertainties associated to them (Watjen et al., 2008). Experimental results are considered satisfactory if |En|r1 and warning if the 1o|En|r1.5. The En for this exercise was |En|¼1.1, meaning that further investigation should be taken into account, perhaps regarding the estimation of the uncertainties, given that %D was already within the acceptable criteria. 4. Conclusions Evaluation of the accuracy of the method using reference material and interlaboratory comparison samples was performed. The validation trials conducted previously have been complemented with this new data and have proven the adequacy of the method for the routine measurements of Sr-90 in foodstuffs, by LSC technique.
Grahek, Ž, Ivšćic, A.G., Krljan, N., Nodilo, M., 2011. Separation of Sr in combination of ion exchange and Sr resin with alcohol-nitric acid solution and rapid determination of 90Sr in wine and soil samples. J. Radioanal. Nucl. Chem 289, 437–449. Groska, J., Molnár, Z., Bokori, E., Vadjan, N., 2012. Simultaneous determination of 89
Sr and 90Sr comparison of methods and calculation techniques. J. Radioanal. Nucl. Chem. 291, 707–715. Gascó, C., Trinidad, J.A., Suárez, J.A., Llauradó, Valiño, M.F., 2011. Intercomparación analítica entre laboratorios de radiactividad ambiental. Evaluación de la intercomparación analítica de radionucleidos en muestras ambientales Campaña 2010–2011 (cenizas de alimentos). CSN/CIEMAT intercomparación 2010. ISO/IEC, 2005. International Standard ISO/IEC 17025; General Requirements for the Competence of Testing and Calibration Laboratories, 2nd edition. Karacan, F., 2011. The simple radiochemical determination of 90Sr in environmental solid samples by solvent extraction. J. Radioanal. Nucl. Chem. 288, 685–691. Kim, C.K., AL-Hamwi, A., Törvényi, Kis-benedek G., Sansone, U., 2009. Appl. Radiat. Isot. 67, 786–793. Lee, M.H., Park, T.H., Park, J.H, Song, K., K., Lee, M.S., 2013. Radiochemical separation of Pu, U, Am and Sr isotopes in environmental samples using extraction chromatographic resins. J. Radioanal. Nucl. Chem. 295, 1419–1422. Lopes, I., Madruga, M.J., Mourato, A., Abrantes, J., Reis, M., 2010. Determination of 90 Sr in Portuguese foodstuffs. J. Radioanal. Nucl. Chem. 286, 335–340. Meresova, J., Watjen, U., 2013. Evaluation of EC Comparison on the Determination of 40K, 90Sr, and 137Cs in Bilberry Powder (Draft). European Commission, Joint Research Centre, ISBN 978-92-79-22551-2. Popov, L., Mihailova, G., Hristova, I., Dimitrova, P., Tzibranski, R., Avramov, V., Naidenov, I., Stoenelova, B., 2009. Separation of strontium from calcium by the use of sodium hydroxide and its application for the determination of long-term background activity concentrations of 90Sr in 100 km area around Kozloduy Nuclear Power Plant (Bulgaria). J. Radioanal. Nucl. Chem. 279, 49–64. Spasova, Y., Wätjen, U., Altzitzoglou, T., 2008. European measurement comparison of 137Cs, 40K and 90Sr in milk powder. J. Radioanal. Nucl. Chem 277, 211–215. UNSCEAR, 2000. United Nations Scientific Committee on the Effects of Atomic Radiation and Effects of Ionizing Radiation, I, 1–654. Vajda, N., Kim, C.K., 2010. Determination of radiostrontium isotopes: a review of analytical methodology. Appl. Radiat. Isot. 68, 2306–2326. Watjen, U., Spasova, Y., Altzitzoglou, T., 2008. Measurement comparisons of radioactivity among European monitoring laboratories for the environment and food stuff. Appl. Radiat. Isot. 66, 742–749.
References Drábová, V., Strišovská, J., Dvorská, L., Kuruc, J., 2012. Validation of 90Sr separation method using molecular recognition product AnaLig Sr01 gel and extraction chromatography Sr_ resin. J. Radioanal. Nucl. Chem. 292, 1141–1145.
Please cite this article as: Lopes, I., et al., Quality control assurance of strontium-90 in foodstuffs by LSC. Appl. Radiat. Isotopes (2014), http://dx.doi.org/10.1016/j.apradiso.2014.01.022i