Article pubs.acs.org/est
A Method of Measurement of 239Pu, 240Pu, 241Pu in High U Content Marine Sediments by Sector Field ICP−MS and Its Application to Fukushima Sediment Samples Wenting Bu,†,‡ Jian Zheng,‡,* Qiuju Guo,†,* Tatsuo Aono,§ Hirofumi Tazoe,⊥ Keiko Tagami,‡ Shigeo Uchida,‡ and Masatoshi Yamada⊥ †
State Key Laboratory of Nuclear Physics and Technology, School of Physics, Peking University, Beijing 100871, China Research Center of Radiation Protection, and §Fukushima Project Headquarters, National Institute of Radiological Sciences, Anagawa 4-9-1, Inage, Chiba 263-8555, Japan ⊥ Department of Radiation Chemistry, Institute of Radiation Emergency Medicine,Hirosaki University, 66-1 Hon-cho, Hirosaki, Aomori 036-8564, Japan ‡
ABSTRACT: An accurate and precise analytical method is highly needed for the determination of Pu isotopes in marine sediments for the long-term marine environment monitoring that is being done since the Fukushima Dai-ichi Nuclear Power Plant accident. The elimination of uranium from the sediment samples needs to be carefully checked. We established an analytical method based on anion-exchange chromatography and SF-ICP−MS in this work. A uranium decontamination factor of 2 × 106 was achieved, and the U concentrations in the final sample solutions were typically below 4 pg mL−1, thus no extra correction of 238U interferences from the Pu spectra was needed. The method was suitable for the analysis of 241Pu in marine sediments using large sample amounts (>10 g). We validated the method by measuring marine sediment reference materials and our results agreed well with the certified and the literature values. Surface sediments and one sediment core sample collected after the nuclear accident were analyzed. The characterization of 241Pu/239Pu atom ratios in the surface sediments and the vertical distribution of Pu isotopes showed that there was no detectable Pu contamination from the nuclear accident in the marine sediments collected 30 km off the plant site.
■
INTRODUCTION Plutonium has been released into the environment by nuclear activities of humans, such as nuclear weapon tests, nuclear power plant accidents, and nuclear fuel reprocessing. Most of the Pu released into the environment has entered the world’s oceans and large amounts of Pu in the marine environment have been incorporated into the sea bottom sediments because of its high particle affinity.1 Characterization of Pu in marine sediments has been widely studied not only for the purpose of environmental monitoring, but also for studies focusing on the environmental behavior of Pu and radioactive source identification.2−6 In the past, radioanalytical methods, that is, alpha spectrometry (for 238Pu and 239,240Pu) and liquid scintillation radiometry (for 241Pu), have been commonly used as the main techniques for the determination of Pu. However, when these techniques were used, complex and tedious sample preparations and long counting times were needed. Moreover, conventional alpha spectrometry lacks the ability to distinguish different Pu isotopes (239Pu and 240Pu). In recent years, the use of mass spectrometry as a tool to determine long-lived radionuclides has boomed worldwide. Among the different © 2013 American Chemical Society
mass spectrometry methods used for the determination of Pu, inductively coupled plasma mass spectrometry (ICP−MS) shows great advantages for both quantitative and isotope ratio measurements with respect to its easy sample preparation procedure, the relatively low cost of analysis, and high sensitivity.7 However, formation of the uranium hydride species 238 UH+ and 238UH2+ and the peak tailing of 238U+ severely hamper the accurate determination of 239Pu and 240Pu, especially for the samples with high U and low Pu concentrations. Other polyatomic interferences include 207 Pb16O2+, 208Pb16O2+, and 202Hg37Cl+. Uranium is the highest-numbered element naturally existing in significant quantities on earth. Marine sediments in reducing environments are enriched in uranium as compared to pelagic sediments, which means the 238U concentration increases with depth (e.g., from about 1.6 μg g−1 at the surface to about 3.2 μg g−1 at a depth of just 30 cm in the Okinawa Trough).8 To Received: Revised: Accepted: Published: 534
August 7, 2013 December 6, 2013 December 16, 2013 December 16, 2013 dx.doi.org/10.1021/es403500e | Environ. Sci. Technol. 2014, 48, 534−541
Environmental Science & Technology
Article
(m/Δm = 10000) of SF-ICP−MS, chemical separation of Pu is still strongly needed prior to mass analysis by SF-ICP−MS. In this work, we established an anion-exchange separation method using AG 1X8 and AG MP-1M resins for the determination of Pu isotopes (239Pu, 240Pu, 241Pu) in marine sediments by SF-ICP−MS as part of an ongoing project to assess the possible Pu contamination in the marine environment following the FDNPP accident. The merits of this method for U elimination were discussed. Then, we used marine sediment reference materials to validate the method and applied it to determine Pu characterization in surface sediment and sediment core samples collected in the western North Pacific off the Fukushima Prefecture coast after the FDNPP accident.
identify the Pu source and understand its migration in the marine environment, the vertical distribution of Pu in marine sediment cores can be investigated. However, in contrast with the distribution of U in marine sediments, Pu in the deeper layer sediments normally decreases rapidly to an extremely low level (0.3) and 241Pu/239Pu atom ratio (>0.1), in soil and litter samples collected around the FDNPP site.11 Schneider et al. recently further evidenced the Fukushima derived Pu contamination in the terrestrial environment.12 Before the accident, Pu in the marine environment in the western North Pacific could be attributed to global fallout and to close-in fallout transported by the oceanic currents from the Pacific Proving Ground (PPG).4,13−16 The 240Pu/239Pu atom ratio (0.30−0.33) from the FDNPP accident is higher than the global fallout ratio (0.18)17 but similar to the PPG close-in fallout ratio (0.30−0.36).14,18 However, the FDNPP accidentderived 241Pu/239Pu atom ratio (0.103−0.135)11 is almost 2 orders of magnitude higher than both the global fallout (0.0011)17 and the PPG close-in fallout (ca. 0.0020)19,20 values (241Pu decay corrected to 15 March 2011). Thus 241Pu in the marine sediments should be measured to better assess the impact of the accident on the Pu contamination in the Pacific. To measure 241Pu in marine sediments, larger amounts of sediment samples (>10 g) than those used to measure 239Pu and 240Pu (normally 1−2 g samples) are needed due to the low activity and short half-life of 241Pu, which, however, can lead to greater U interferences for the analysis. To eliminate uranium and its hydride interferences for ICP− MS detection, several solutions have been proposed. Zoriy et al.21 used heavy water as a solvent for the preparation of samples to decrease the formation of uranium hydrides. Tanner et al.22 and Vais et al.23 introduced CO2 and NH3 to the reaction cell of DRC ICP−MS to depress the U signal. However, commercial sector field ICP−MS (SF-ICP−MS), which is characterized by high sensitivity and low sample consumption, combined with a reaction cell is currently not available. To lower the formation of uranium hydride species in SF-ICP−MS, researchers have used different types of sample introduction systems.24−26 The intensity ratio of 238UH+/238U+ for SF-ICP−MS with different sample introduction systems ranged from 1 × 10−5 to 5 × 10−4.24 Considering that the concentration of uranium in the marine sediments is 6−9 orders of magnitudes higher than that of Pu and that the uranium hydride formations of 238UH+ and 238UH2+ cannot be resolved from 239Pu and 240Pu peaks even with the HR mode
■
EXPERIMENTAL SECTION Regents and Samples. High-purity water (18 MΩ cm−1) was prepared with a Millipore Milli-Q-Plus water purification system. All chemicals (HCl, HNO3, NaNO2, NH4I, H2O2, HBr) were of analytical grade and were used without any further purification. The two anion-exchange resins, AG 1X8 (100−200 mesh, Cl− form) and AG MP-1M (100−200 mesh, Cl− form) were obtained from Bio-Rad. 242Pu (CRM 130, plutonium spike assay and isotopic standard, New Brunswick Laboratory, USA) was used to spike the sediment samples as a yield tracer. Ocean sediment radionuclide standards, NIST-4357 and IAEA-368, were used to validate our analytical method. One soil standard reference material IAEA-soil-6 was also measured. Eight sediment core samples were collected in the western North Pacific 30 km off the FDNPP site during the four cruises of MR-11-05, KH-11-07, MR-12-02 and KT-13-1 from July 2011 to January 2013 (Figure 1), and the core was cut into 1 cm segments aboard the ship. The collection sites were as follows: ES4, 37°51′58″ N 143°34′31″ E; ES5, 37°47′41″ N 143°51′56″ E; FS1, 37°19′58″ N 142°10′03″ E; MC5, 37°35′01″ N 141°30′57″ E; MC1, 36°28′58″ N 141°29′56″ E; F1, 36°29′05″ N 141°30′01″ E; ES2, 37°03′59″ N 142°15′01″ E; K06, 37°20′00″ N 141°40′06″ E. Surface
Figure 1. Map showing the locations of the sediment samples collected in the western North Pacific after the FDNPP accident. 535
dx.doi.org/10.1021/es403500e | Environ. Sci. Technol. 2014, 48, 534−541
Environmental Science & Technology
Article
Table 1. U Decontamination of Different Separation Methods for the Analysis of Pu in Solid Environmental Samples 238
separation method
resin
sample
amounts (g)
DFa
anion exchange anion exchange anion exchange anion exchange anion exchange/extraction extraction extraction extraction extraction anion exchange
AG 1X8 SR resin disk AG 1X8+ AG 1X8 AG 1X8 + AG MP-1M Dowex 1X8/TEVA TNOA TEVA TEVA/UTEVA + TRU TEVA + Sr AG 1X8 + AG MP-1M
marine settling particles sediment sediment freshwater lake sediment sediment, soil soil sediment sediment, soil sediment, soil marine sediment
0.03−0.5 0.5−10 10 0.5−2 1−25 1 20 0.2−2 0.5−1 0.1−30
1−10 × 104 6.6 × 103 1.4 × 104 1 × 105 1−10 × 104 1.7−2 × 105 1−10 × 104 0.18−5.9 × 104 5.4 × 104 2 × 106
a
U in final solution (pg mL−1) 10−20 10 g) is needed for the analysis of 241Pu. By using our method for the determination of Pu in marine sediment and soil reference materials (NIST-4357, IAEA-368, and IAEA-soil-6), we found that 238U concentration in the final sample solution ranged from 1 to 4 pg mL−1 and the U decontamination factor was calculated as 2 × 106, more than 1 order of magnitude higher than that of the published analytical methods (Table 1). The mass spectra of a typical marine sediment sample determined by our analytical method and the spectrum of the operation blank are presented in Figure 2. By using our method, the detected 238U intensity was ca. 4 × 104, almost the same as that of the operational blank. As the 238 UH+/238U+ ratio for our analysis system was about 2 × 10−5, the contribution of 238UH+ to the 239Pu+ signal was less than 1 cps. Considering that the detected 239Pu intensities exceeded 3000, we concluded no subtractive corrections from 238U+ are needed for the analysis of Pu using our method. The chemical recovery of our method for the determination of Pu isotopes ranged from 45 to 76% with an average of 64%. This value is comparable with the chemical recoveries reported in the literature for the determination of Pu in sediment samples.2,5,36,37
Figure 2. Mass spectra of a sediment sample analyzed by our method and the operation blank: (a) sediment sample; (b) operational blank.
Analysis of Certified Reference Materials. This analytical procedure based on SF-ICP−MS detection was validated by the analysis of certified reference materials (including two marine sediment reference materials NIST4357 and IAEA-368 and one soil reference material IAEA-soil6). For the marine reference materials, different sample weights (0.1−2.5 g) were used. The results we obtained are compared with the certified and previously reported values, as shown in Table 2. Information about the sample weight, 238 U concentration in the final sample solutions and the chemical recovery for each sample are also presented. The mean 239+240Pu activities for NIST-4357 (n = 4) and IAEA-368 (n = 3) were 10.15 ± 1.02 and 32.49 ± 1.08 mBq g−1, respectively, and agreed well with the certified values. Although the 240Pu/239Pu atom ratios for these reference materials were not certified, they were reported in the literature.37,38,40−42 In our study, the 240Pu/239Pu atom ratios for NIST-4357 and IAEA-368 were 0.233 ± 0.007 and 0.0321 ± 0.0003, respectively, which were comparable with the literature values. One soil reference sample IAEA-soil-6 was also analyzed to test the applicability of our method for the determination of Pu in soil samples. 239+240Pu activity of 1.00 ± 0.02 mBq g−1 and 240Pu/239Pu atom ratio of 0.192 ± 0.008 were detected. The obtained results agreed with the certified and the literature values as well.2,42 For the determination of 241Pu, using our method, the absolute detection limit calculated as three times of the standard deviation of the operational blanks (n = 10) was ca. 2 mBq. The 241Pu activity and 241Pu/239Pu atom ratio measured for NIST-4357 in this study ranged from 111.7 to 139.3 mBq g−1 (with an average of 119.5 ± 13.3 mBq g−1) and from 0.0123 to 0.0136 (with an average of 0.0131 ± 0.0010), respectively (241Pu decay corrected to 1 January 2000). While for IAEA-368, 241Pu was only detected in the sample with a weight of 2.43 g due to the low 241Pu activity in IAEA-368. The detected 241Pu activity and 241Pu/239Pu atom ratio were 10.27 ± 1.40 mBq g−1 and 0.00021 ± 0.00003, respectively. No certified values are available about the 241Pu information in these two marine sediment reference materials and there are very few reported literature values. Hernecek et al.38 reported the 241Pu 537
dx.doi.org/10.1021/es403500e | Environ. Sci. Technol. 2014, 48, 534−541
Environmental Science & Technology
Article
Table 2. Results of Reference Materials Measured by Our Method and the Certified and Information Values amount (g)
sample
239+240
241 Pu activity (mBq g−1)a,c
Pu activity (mBq g−1)a
NIST-4357 1 NIST-4357 2 NIST-4357 3 NIST-4357 4 mean (n = 4)b
0.49 0.39 0.64 2.42
10.06 ± 0.09 11.58 ± 0.12 9.76 ± 0.16 9.19 ± 0.13 10.15 ± 1.02
IAEA-368 1 IAEA-368 2 IAEA-368 3 mean (n = 3)b
0.13 0.14 2.43
32.08 ± 0.47 33.71 ± 0.44 31.68 ± 0.33 32.49 ± 1.08
114.4 139.3 112.5 111.7 119.5
± 8.2 ± 15.3 ± 9.6 ± 3.2 ± 13.3
10.27 ± 1.40
IAEA-soil-6 2.01 1.00 ± 0.02 certified and/or information values NIST-4357 9.2−13.3 IAEA-368 29.0−34.0 IAEA-soil-6 0.96−1.11
240
Pu/239Pu atom ratioa
0.223 ± 0.004 0.238 ± 0.004 0.237 ± 0.002 0.234 ± 0.004 0.233 ± 0.007 0.0325 ± 0.0013 0.0319 ± 0.0015 0.0321 ± 0.0004 0.0321 ± 0.0003
241
Pu/239Pu atom ratioa,c
238
0.0123 ± 0.0009 0.0136 ± 0.0015 0.0130 ± 0.0012 0.0136 ± 0.0004 0.0131 ± 0.0010
1.27 1.71 1.52 3.12
0.00021 ± 0.00003
0.192 ± 0.008 90.5 ± 5.6d 12 ± 2e
0.233−0.244f 0.030−0.042g 0.186−0.212h
U in final solution (pg mL−1)a ± ± ± ±
chemical recovery (%)
0.02 0.04 0.05 0.08
66.6 73.6 65.1 44.8
1.10 ± 0.03 1.20 ± 0.02 2.17 ± 0.04
67.0 75.9 57.5
1.24 ± 0.02
62.1
0.0132 ± 0.0007i 0.00019 ± 0.00004i
Uncertainties represent 1σ error. bAverage of the measured replicates ± standard deviation. cDecay of 241Pu corrected to 1 Jan. 2000. dData cited from Hernecek et al.38 eData cited from Alvarado et al.39 fData cited from Hernecek et al.38 and Yoshida et al.40 gData cited from Donard et al.,37 Kim et al.41 and Ohtsuka et al.42 hData cited from Muramatsu et al.2 and Ohtsuka et al.42 iData cited from Zhang et al.43 a
Table 3. Pu Activities and Atom Ratios in the Surface Sediments 30 km off the FDNPP Site sample
sample weight (g)
FS1
2.0 11.4 22.1 3.0 28.1 22.2 17.3 20.6 20.7 28.0 24.3
MC5 ES2 ES4 ES5 MC1 F1 K06 a
239+240
Pu activity (mBq g‑1)a 2.78 2.60 2.48 0.43 0.48 3.25 1.19 0.84 1.10 1.18 0.70
Uncertainties represent 1σ error. bDecay of
± ± ± ± ± ± ± ± ± ± ±
241
240
Pu/239Pu atom ratioa
0.05 0.03 0.03 0.01 0.01 0.02 0.01 0.02 0.02 0.01 0.01
0.227 0.222 0.223 0.236 0.244 0.221 0.195 0.175 0.236 0.239 0.238
± ± ± ± ± ± ± ± ± ± ±
0.007 0.002 0.003 0.011 0.005 0.002 0.003 0.002 0.005 0.002 0.003
241
Pu activity (mBq g‑1)a,b ND 3.52 3.13 ND 0.69 4.02 1.48 1.03 1.49 1.49 0.93
± 0.37 ± 0.44 ± ± ± ± ± ± ±
0.10 0.27 0.26 0.16 0.22 0.19 0.14
Pu/239Pu atom ratioa,b
241
ND 0.0015 0.0014 ND 0.0016 0.0013 0.0013 0.0012 0.0015 0.0014 0.0015
± 0.0002 ± 0.0003 ± ± ± ± ± ± ±
0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002
Pu was corrected to 15 March 2011; ND, not detected.
activity in NIST-4357 to be 90.1 ± 5.6 mBq g−1 and Alvarado et al.39 reported the 241Pu activity in IAEA-368 to be 12 ± 2 mBq g−1. Our results were comparable with their results. For the 241Pu/239Pu atom ratio, our results agreed well with the results reported by Zhang et al.43 (0.0132 ± 0.0007 and 0.00019 ± 0.00004 for NIST-4357 and IAEA-368, respectively). Analysis of Marine Sediments in the Pacific 30 km off the FDNPP Site. As mentioned in the Introduction section, information about 241Pu concentration and 241Pu/239Pu atom ratio and the vertical distribution of Pu isotopes in marine sediments found off the Fukushima Prefecture coast are important for identifying the Pu source and understanding the Pu migration behavior. We used our analytical method to measure Pu activities and Pu atom ratios in eight surface sediment samples and one sediment core sample collected in the western North Pacific 30 km from the FDNPP site after the nuclear accident. 239 Pu-240Pu−241Pu in Surface Sediment Samples. The analytical results of Pu isotopes in the surface sediment samples (0−3 cm) are summarized in Table 3. Because of the high U decontamination factor of the developed method, Pu activities and atom ratios in FS1 and MC5 measured with different sample amounts showed good consistency. 241Pu was
successfully detected in the surface sediments with sample amounts larger than 10 g. 238U concentrations in the final sample solutions were also less than 4 pg mL−1 even when ca. 30 g of marine sediment samples was analyzed. The relative uncertainties for our measured 241Pu activities in the surface sediments ranged from 6.7% to 17.6%. These results suggested that our analytical method was suitable for the determination of 241 Pu by using large amounts (>10 g) of marine sediment. It was found that the 240Pu/239Pu atom ratio in the marine sediments in the western North Pacific and its marginal seas before the FDNPP accident could be higher than the global fallout value (0.18) due to the influence of the PPG close-in fallout. The PPG presented a number of sites in Marshall Islands and a few other sites in the Pacific that were used by the United States to conduct a series of nuclear weapon testing from 1946 to 1962. Pu from the PPG nuclear test sites was characterized by a high 240Pu/239Pu atom ratio (0.30−0.36) and has been transported by the North Equatorial and Kuroshio currents as far as to the Japanese coast.10,13 Previously, we determined 239+240Pu activities and 240Pu/239Pu atom ratios in the surface sediments and sediment cores collected 30 km from the FDNPP site after the accident and suggested that there was no detectable Pu contamination from the accident in the 538
dx.doi.org/10.1021/es403500e | Environ. Sci. Technol. 2014, 48, 534−541
Environmental Science & Technology
Article
investigated areas.15,16 However, information about the characterization of 241Pu, which could provide further evidence for the FDNPP Pu source identification, in the marine sediments before and after the Fukushima nuclear accident is limited. Lachner et al.20 reported the 241Pu/239Pu atom ratios in the sediments collected from the Bikini Atoll ranged from 0.0021 to 0.0025 (for 241Pu discussed here, decay corrected to 15 March 2011), which were comparable with the result (0.0018) in the hemp-palm leaves of Bontenchiku used in fishing gear of the Fifth Fukuryu-Maru, which was exposed to heavy fallout due to U.S. thermonuclear testing at Bikini Atoll in March 1954, as reported by Yamamoto et al.19 The 241Pu activities in the Fukushima surface sediments measured in this study ranged from 0.69 to 4.02 mBq g−1. These values were significantly lower than that (4.5−34.8 mBq g−1) of the soil and litter samples collected around the FDNPP site after the accident.11 For the marine sediment, Zheng and Yamada44 observed a 241 Pu activity of about 8 mBq g−1 in the deep layer (18−20 cm) sediment of Sagami Bay, which is located in the southwest of Fukushima, before the FDNPP accident and they suggested that the deposition of the PPG fallout was the main source for the 241Pu contamination. It is noted that even the highest 241Pu activity in the surface sediment 30 km off Fukushima coast was almost two times lower than that of the deep sediment of Sagami Bay. The 241Pu/239Pu atom ratios for the Fukushima surface sediments ranged from 0.0012 to 0.0016, 2 orders of magnitudes lower than that (0.103−0.135) released from the Fukushima accident.11 We plotted the Pu isotopic compositions in surface sediments and the literature values for other sources (Fukushima litter and soil samples, global fallout and PPG close-in fallout) in Figure 3. The Pu isotopic compositions observed in the Fukushima surface sediments were located on the mixing line between the global fallout and the PPG close-in fallout, not on the mixing line between the global fallout and the Fukushima source. The results clearly indicated that there was no detectable Pu contamination from the FDNPP accident in the marine sediments collected 30 km from the plant site. Vertical Distribution of Pu Isotopes in the ES2 Site Core Sediment Sample. The present analytical method is being routinely applied in our laboratory for the determination of Pu in the marine sediments off the Fukushima coast to assess the long-term effect of the FDNPP accident on the marine environment. In this study, a sediment core sample collected at the ES2 site in the western North Pacific after the nuclear accident was measured. The analytical results are presented in Figure 4. The 239+240Pu activity and 240Pu/239Pu atom ratio in ES2 core sediment sample ranged from 0.78 to 3.54 mBq g−1 and from 0.216 to 0.247, respectively. In a previous work, we summarized that the 239+240Pu activity and 240Pu/239Pu atom ratio in the marine sediment samples in the western North Pacific before the Fukushima accident could be lower than 5.81 mBq g−1 and 0.30, respectively.15 Both the 239+240Pu activity and the 240Pu/239Pu atom ratios in ES2 were in the range of the background data. As discussed before, 240Pu/239Pu atom ratios in the marine sediments in the western North Pacific before the FDNPP accident could be higher than the global fallout due to the effect of the PPG close-in fallout. 240Pu/239Pu atom ratios (>0.240) in the deeper layers (>4 cm) of ES2 core sediment sample were higher than that (0.216−0.232) of the upper layers (
