Rapid Quantification of Radioactive Strontium-90 ... - ACS Publications

Jun 28, 2019 - ... in the concentration range of ppt–sub ppb as ingredients of the natural ..... Icap-ICP–MS: as an analytical system for 90Sr, we...
0 downloads 0 Views 2MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article Cite This: ACS Omega 2019, 4, 11276−11284

http://pubs.acs.org/journal/acsodf

Rapid Quantification of Radioactive Strontium-90 in Fresh Foods via Online Solid-Phase Extraction−Inductively Coupled Plasma− Dynamic Reaction Cell-Mass Spectrometry and Its Comparative Evaluation with Conventional Radiometry Makoto Furukawa,†,§,∥ Kyoko Takagi,⊥ Hisaya Matsunami,⊥ Yuko Komatsuzaki,¶ Tomohiko Kawakami,¶ Takuro Shinano,⊥,# and Yoshitaka Takagai*,†,‡ Downloaded via 193.56.73.186 on August 27, 2019 at 18:53:44 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



Faculty of Symbiotic Systems Science, Cluster of Science and Technology, Fukushima University, 1 Kanayagawa, Fukushima 960-1296, Japan ‡ Institute of Environmental Radioactivity, Fukushima University, 1 Kanayagawa, Fukushima 960-1296, Japan § PerkinElmer Japan Co., Ltd., 134 Godo, Hodogaya, Yokohama, Kanagawa 240-0005, Japan ∥ Faculty of Agriculture, University of Tokyo, 1−1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan ⊥ Fukushima Research Station, Tohoku Agricultural Research Center, National Agriculture and Food Research Organization, 50 Harajuku-minami, Arai, Fukushima 960-2156, Japan ¶ Kaken Inc., 1044 Horimachi, Mito, Ibaraki 310-0903, Japan S Supporting Information *

ABSTRACT: This paper describes a rapid quantification method for radioactive strontium (90Sr) in fresh foods (perishable foods) and has been comparatively evaluated with the common classical radiometric quantification method. Inductively coupled plasma−dynamic reaction cell-mass spectrometry with online solid-phase extraction (cascadeICP−MS) rapidly determines 90Sr in a pure water-based sample. Despite its advantages, its application to fresh foods (perishable foods) has not yet been reported; however, the analytical potential of this method for fresh foods must be evaluated. In this study, 90Sr was determined in 12 fresh foods via improved cascade-ICP−MS (Icas-ICP−MS). Addition and recovery tests were demonstrated using real samples of grape, apple, peach, Japanese pear, rice, buckwheat, soybean, spinach, shiitake mushroom, grass, sea squirt, and flounder. With a decomposed solution of Japanese pear, the measurement value coincided with the amount of spiked 90Sr. The reproducibility of the measurements was represented by relative standard deviations of 14.2 and 5.0% for spiked amounts of 20 and 200 Bq/kg, respectively (n = 10), and the recovery rates were 93.7 ± 7.1%. In this case, the limit of detection (LOD) was 2.2 Bq/kg (=0.43 pg/kg). These results were compared with the data obtained using a common classical radiometric quantification method (nitrate precipitation-low background gas flow counter (LBC) method) in the same samples. Both the methods showed equivalent performances with regard to reproducibility, precision, and LODs but different analysis times. Icas-ICP−MS required ∼22 min for analysis, whereas the nitrate precipitation-LBC method required 20 days, confirming that Icas-ICP−MS is the suitable method for analyzing 90Sr in fresh foods.



analysis, in particular, the rapid analysis of 90Sr, in fresh foods was performed.21,22 Radiometric methods23 are commonly used for determining 90 Sr concentration by employing a low background gas flow counter (LBC),24 a liquid scintillation counter (LSC),9 or a Cherenkov radiation counter.10 In radiometric analysis, coexisting radionuclides interfere with radiation from an analytical target; therefore, chemical separation is essential for removing coexisting radionuclides. In Japan, the Ministry of

INTRODUCTION

Radioactive strontium-90 (90Sr) is a typical fission product of uranium with a half-life of 28.74 years, and it is a pure β-ray emitter (100% β−-decay).1 90Sr behaves similar to the homologous element Ca because of their similar chemical properties; therefore, osteogenic sarcomas can potentially arise because of the incorporation and accumulation of 90Sr in the growing bone and tissues when humans absorb 90Sr from food.2 Several reports3−5 have reported the transfer of 90Sr into foods such as seafood6 as well as animal,7 agricultural,8−11 and dairy products,12−15 which were contaminated by past nuclear incidents. The harmful influence on human health by ingesting such contaminated food has also been reported.16−20 Thus, 90Sr © 2019 American Chemical Society

Received: May 13, 2019 Accepted: June 13, 2019 Published: June 28, 2019 11276

DOI: 10.1021/acsomega.9b01381 ACS Omega 2019, 4, 11276−11284

ACS Omega

Article

Figure 1. Eight-channel flow map of the lab-on-valve system (double four-way switching valves) with a split-flow line system. [A] indicates the sample pass ways of a major volume of the sample introducing into the SPE column and minor part of the sample divided by the splitter introducing into ICP− MS for direct analysis of initial concentration of Sr. [B] indicates the sample pass ways of a preconcentrated Sr by SPE for introduction into ICP−MS.

Thus far, Icas-ICP−MS has been applied to a limited range of almost pure water-based samples of radioactively contaminated water originating from Fukushima Daiichi Nuclear Power Plants, related samples obtained from instruments for removing radioactive materials, and decomposed solutions of 90Srcontaminated soil obtained from environmental ponds. While the method has been applied to water-based samples, it is expected to be suitable for fresh foods (perishable foods) because of its rapidity. However, fresh food samples contain complex ingredients (with high matrix contents); therefore, the results must be thoroughly verified so that the quantitative values are correct when using the classical common method. Herein, Icas-ICP−MS was applied to model fresh food (typically 12 products) containing very low 90Sr levels. The common radiometric method was also used to quantify 90Sr concentrations in identical samples for comparison with the data obtained using the proposed method (Icas-ICP−MS) with respect to reproducibility, accuracy, LOD, and recovery. All model samples were obtained from commercially available products manufactured in Fukushima Prefecture. This paper describes original 90Sr concentrations in these fresh foods.

Education, Culture, Sports, Science and Technology in Japanese Government (MEXT) recommends using the LBC method as a standard analytical method for 90Sr determination. The method requires an ion-exchange and carbonate co-precipitation step. After reaching secular equilibrium (requires at least 2−3 weeks for ingrowth of 90Y), the separated 90Y is measured by LBC. Consequently, it takes more than 2−3 weeks after sampling to report the results. This nitrate precipitation-LBC procedure24 is recognized as the Japanese national official analytical method because it can determine lower 90Sr concentrations using the daughter nuclide 90Y, which has a higher energy relative to 90Sr. Despite its high sensitivity, nitrate precipitation-LBC24 requires over two weeks to perform the complex chemical separation and Y-generation (milking) steps, thereby making the sequential treating of samples difficult. Particularly, a rapid and sequential analysis of fresh foods is remarkably difficult because of the time constraints required to achieve commercial validity. Although rapid analysis of 90Sr has been extensively studied,25 only a few practical and versatile 90Sr detection methods for analyzing low concentration levels in a short time are reported. A recently developed method combining inductively coupled plasma−mass spectrometry (ICP−MS) with online solid-phase extraction (online SPE) and O2 reaction [in a dynamic reaction cell (DRC)] could measure 90Sr within 15−30 min in an allautomated analytical system.26 Thus, 90Sr inside an ICP−MS system could be isolated and purified in a stepwise manner by combining different types of separation methods; therefore, it was denoted as “cascade-ICP−MS”.26,27 This technique was applied to ICP−single MS and ICP−MS/MS.28 A method for one-shot simultaneous data acquisition allowing the quantification of 90Sr and the recovery percentage (R %) of Sr (split system),29 an internal standard corrected signal integration method,30 and a sensitivity improvement system using a N2−Ar mixed gas effect31 were reported for improving the analytical performance of cascade-ICP−MS. Several studies combining the above three and the original cascade-ICP−MS method achieved a limit of detection (LOD) of 0.3 Bq/L (=0.059 pg/L) in 30 min (as a standard solution).29−31 Herein, this combined methodology for 90Sr measurement is referred to as an improved Cascade-ICP−MS system (Icas-ICP−MS).



RESULTS AND DISCUSSION Evaluation of Analytical Performance of Icas-ICP−MS for 90Sr Analysis. The components of Icas-ICP−MS are based on previous reports.26,29−31 Figure 1 shows the outline and flow design. The detailed measurement protocol is described in the “Measurements” part of the Experimental section. The IcasICP−MS system for 90Sr analysis comprised [I] the cascadeICP−MS26 (i.e., online SPE−O2 reaction in the DRC− quadrupole mass filter), [II] a one-shot simultaneous data acquisition system quantifying both 90Sr and R % of Sr (split system),29 [III] an internal standard corrected signal integration method,30 and [IV] a sensitivity improvement system using a N2−Ar mixed gas effect.31 The experimental details of this system are included in the “Measurements” section. Figure 2 shows a 5 Bq/L 90Sr peak profile of the reference material solution (black line). As a reference, a blank peak (0 Bq/L of 90Sr, gray line) is also shown in Figure 2. The peak area depends on the 90Sr concentration, as shown in the inset of 11277

DOI: 10.1021/acsomega.9b01381 ACS Omega 2019, 4, 11276−11284

ACS Omega

Article

Figure 2. Quantifying the signal profile of 90Sr using the proposed method. Black peak line: the injection sample was a 50 mL solution containing 5 Bq/L radioactive 90Sr and 80 ppb natural isotope of Sr. Gray peak line: the sample was a 50 mL solution containing 80 ppb natural isotope of Sr. The intensities are a linear function of the 90Sr concentration vs the peak area (inset). The volume ratio between samples before and after preconcentration (50 → 0.6 mL) was 83.3 times. Eluate flow rate: 5 mL/min. The LOD was 0.35 Bq/L. Total measurement time was 22 min. The 5 Bq/L is equal to 1 pg/L.

Figure 4. Relation between the 88Sr intensity flowing in the split line and the peak area of the preconcentrated 88Sr eluated from the column line. The sample volume was 10 mL.

actual R % using the data in Figure 4 (i.e., the actual R % values were obtained by calculating the ratio between the actual and theoretical preconcentrated concentrations). The measured and spiked concentrations obtained using the proposed method were sufficiently consistent, as shown in Figure 5A. Additionally, the reproducibility was improved by

Figure 2. For a 50 mL sample injection, the LOD (3σ) was 0.35 Bq/L (equal to a mass concentration of 0.07 pg/L) and the measurement time was 22 min. This indicates sufficient removal of interfering components such as 90Zr and 74Ge, which were essentially contained in the concentration range of ppt−sub ppb as ingredients of the natural solution (note: isobar (m/z = 90; 74 Ge16O) was produced by reaction of 74Ge and oxygen). During this time, the final volume introduced into the ICP−MS after online SPE was 0.6 mL (±1.1%). This result was calculated from the eluate flow rate (5.0 mL/min), detected peak width (14.4 s ± 1.1%, n = 15), and sample dilution rate (an injected sample was diluted twice with the internal standard solution in the flow line) [namely, 5.0/60.0 (mL/s) × 14.4 s(±1.1%) × 0.50 = 0.6 mL]. Based on these values, the volume ratio between the initial solution (50 mL) and after preconcentration was 83.3 times. Stable 88Sr peaks before and after SPE were simultaneously obtained using a split method29 during one sample injection into this system. Figure 3A shows the 88Sr peak before SPE, that is,

Figure 5. Analytical performance of the proposed method. Panel [A] indicates the concordance of the 90Sr concentrations between the spiked certified reference material and the quantitative values obtained by the proposed method. Panel [B] presents the variation in the quantitative values with the proposed method. The RSD represents the relative standard deviation of the measurement (n = 3).

increasing the 90Sr concentration in the sample, as shown in Figure 5B (i.e., RSD was 14.1 and 3.1% for 4 and 10 Bq/L of 90Sr, respectively). R % showed stable values (R %; 97 ± 5.8% (n = 18) regardless of the 90Sr concentration in the calibration range). Otherwise, to obtain LOD values for each food samples, certain different concentrations of certificated 90Sr were added to the food sample. The LOD values were obtained from these lines. The average LOD (n = 12, 3σ) was 0.35 ± 0.14 Bq/L, which was equal to 1.7 Bq/kg as a solid sample (=0.33 pg/kg as a mass concentration). Analytical Performance Evaluation: Comparison of 90 Sr Analysis via Nitrate Precipitation-LBC with Icas-ICP− MS. Samples [50 mL standard solution (n = 5 samples)] identical to those for Icas-ICP−MS measurement were subjected to analysis via nitrate precipitation-LBC, which is commonly used for official analysis. When 2 or 10 Bq/L of 90Sr was added, the measured values were 2.66 Bq/L (RSD 5.7%) and 12.6 Bq/L (RSD 2.6%), respectively. During this time, the LODs of 90Sr were 0.18 and 0.20 Bq/L. The Sr recovery rates were 97.1% (RSD 1.6%, n = 5) and 94.7% (RSD 0.5%, n = 5) respectively. A total of 20 days was required for the analysis. Table 1 lists a comparison of the analytical values obtained via Icas-ICP−MS and nitrate precipitation-LBC when analyzing 50 mL of standard solution containing 4 or 10 Bq/L of 90Sr. Results revealed that in the Icas-ICP−MS analysis, compared with nitrate precipitation−LBC analysis, the LOD, the coincidence

Figure 3. Signal profiles of Sr in the proposed method. The sample was a 50 mL solution containing 100 Bq/L radioactive 90Sr and 5 μg/L stable Sr. Panel [A] indicates the 88Sr signal from the bypass line. Panel [B] presents the signal of preconcentrated 88Sr and 90Sr from the online column.

the original solution flowed into the split line (bypass line). Figure 3B shows a peak of preconcentrated 88Sr eluated from the SPE column. Linearity is observed between the 88Sr intensity measured in the split line (i.e., 88Sr original concentration before SPE) and the preconcentrated 88Sr peak area (i.e., 88Sr concentration after SPE), as shown in Figure 4. The published cascade-ICP−MS method also shows almost complete R % (99%, RSD 1.2%)29 in measuring water-based reference material solutions. Based on this property, the proposed method gave the 11278

DOI: 10.1021/acsomega.9b01381 ACS Omega 2019, 4, 11276−11284

ACS Omega

Article

Table 1. Comparison of Analytical Performance between This Method and Conventional Radiometry analytical meansa b

quantitative value (concordance rate ) in 2.00 Bq/L measurement quantitative value (concordance rateb) in 10.0 Bq/L measurement detection limit (Bq/L)c recovery (R %)d reproducibility of 4.0 Bq/L measurements, (RSDe%, n = 3) reproducibility of 10.0 Bq/L measurements, (RSDe%, n = 3) analytical time to measure

Icas-ICP−MS

nitrate precipitation-LBC(radiometry)

2.06 (+3%) 11.1 (+11%) 0.35 97.0 ± 5.8% 14.1% 3.1% 22 min

2.66 (+33%) 12.6 (+26%) 0.20 94.7 ± 0.48% 5.7% 5.6% 20 days

a

Sample was 50 mL of aqueous solution containing the certified reference radioactive 90Sr (4.0 and 10.0 Bq/L). bConcordance rates were − (precise spiked concentration) c d × 100 . 3σ. The reproducibility (n = 18). calculated using the following equation: concordance rate (%) = (quantified concentration) (precise spiked concentration) e

{

}

RSD: relative standard deviation.

Table 2. 90Sr Concentrations in Agricultural and Marine Products as per the Icas-ICP−MS measurement value of 90Sr and the RSDb producta

no. 1

2

3

4

5

6

7

8

9

10

11

12

grape

apple

peach

Japanese pear

rice

buckwheat

soybean

spinach

shiitake mushroom

grass

sea squirt

flounder

spiked 90Sr (Bq/kg)

(Bq/kg)

0 20 200 0 20 200 0 20 200 0 20 200 0 20 200 0 20 200 0 20 200 0 20 200 0 20 200 0 20 200 0 20 200 0 20 200