129I in sediment cores from the Celtic Sea by AMS through a microwave digestion process

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ Esta es la versión aceptada del artículo publicado en: This is an accepted manuscript of a paper published in: Nuclear Instruments and Methods in Physics Research. Section B: Beam Interactions with Materials and Atoms (2022): 02/12/2024 DOI: https://doi.org/10.1016/j.nimb.2022.08.016 Copyright: © 2022 Elsevier Ltd. All rights reserved. El acceso a la versión publicada del artículo puede requerir la suscripción de la revista. Access to the published version may require subscription. “This is an Accepted Manuscript of an article published by Elsevier in [Nuclear Instruments and Methods in Physics Research. Section B: Beam Interactions with Materials and Atoms] on [2022], available at: https://doi.org/10.1016/j.nimb.2022.08.016 ” 1 129I in sediment cores from the Celtic Sea by AMS through a microwave digestion process 1 Lérida-Toro, Victoria1*, Abascal, Unai1, Villa-Alfageme, María1,2, López-Gutiérrez, José María1,3; 2 Klar, Jessica4, Hicks, Natalie5, 3 1Centro Nacional de Aceleradores (Universidad de Sevilla, CSIC, Junta de Andalucía) (Spain), 4 2Dpto. Física Aplicada II, ETSIE, Universidad de Sevilla (Spain), 3Dpto. Física Aplicada I, EPS, 5 Universidad de Sevilla (Spain), 4Université de Perpignan UPVD - CEFREM (France), 5School of 6 Life Sciences, University of Essex (UK) 7 8 Highlights: 9 Keywords: Microwave digestion, Sediment core, Iodine-129, Accelerator Mass Spectrometry (AMS), 10 sedimentation rate 11 12 ABSTRACT 13 Radiochemical methods before measurement are indispensable to determine Iodine-129 by 14 Accelerator Mass Spectrometry (AMS) in environmental samples. In particular, in sediments, iodine 15 is associated with both inorganic and organic matter, and it presents in several chemical forms. 16 Therefore, the radiochemical procedure must be designed to extract all the different iodine species 17 from the matrix, whilst being as robust and time effective as possible. In this work, different 18 microwave digestion methods, together with the processing of the iodine species to be measured by 19 AMS, were tested and their performances are presented here. Two sediment cores, one with muddy 20 sediment (core A) and one with permeable sandy sediment (core I), collected in the Celtic Sea, were 21 2 used and the results were evaluated to determine an optimized iodine extraction. The method 1 consisted of a microwave digestion with concentrated nitric acid followed by a double liquid-liquid 2 extraction and the final precipitation of silver iodide (AgI), required to measure iodine by AMS. Back 3 extraction of iodine in a reducing solution during purification was carried out with hydrazine (NH2-4 NH2). Subsequent additional steps were added to optimize all iodine species extraction. First, 5 hydrochloric hydroxylamine (NH2OH·HCl) and sodium bisulfite (Na2S2O5) were used to homogenize 6 all iodine species before extraction through a redox process. Third method added sodium hypochlorite 7 (NaOCl) to remove any final organic matter after digestion. The final method incorporated sodium 8 hydroxide (NaOH) combined with NaOCl to facilitate reduction of iodine in solution and enhance 9 the NaOCl effect. The different methods were validated against an internal standard and the last 10 method was chosen, as it showed the most reproducible and accurate results. As practical application, 11 Iodine-129 concentration from both cores were compared, showing concentrations between 0.19 x 12 1012 at/kg and 7.16 x 1012 at/kg for core A and between 0.28 x 1012 at/kg and 2.40 x 1012 at/kg for 13 core I. Despite the 129I diffusion detected in the deeper layers, depth profiles accurately reproduced 14 Sellafield discharges, which is the nuclear fuel reprocessing plant closest to the cores. 15 1. INTRODUCTION 16 129I is a long-lived radioisotope (T1/2=15.7×106 years) naturally present in the environment. It is also 17 produced anthropogenically. Its main artificial sources are i) fallout from nuclear testing carried out 18 between 1945-1980 (43-150 kg 129I [1]), ii) nuclear accidents (for example, less than 2 kg 129I were 19 produced by Chernobyl accident [2]) and iii) discharges from Nuclear Fuel Reprocessing Plants 20 (NFRPs) [3,4]. In Europe, the greatest releases of 129I to the environment come from Sellafield 21 (United Kingdom) and La Hague (France) NFRPs, with well documented discharges. More than 6000 22 kg of 129I have been discharged to the atmosphere and the marine environment up to 2013 by liquid 23 and gaseous emissions from these facilities [5]. 24 3 Due to its long half-life, the fact that is very conservative in water, and its sources are well established 1 temporally and spatially (due to the continuous releas in reported concentrations from NFRPs), 129I 2 has proved to be an excellent ocean tracer, especially in the Atlantic and Arctic Oceans [6–8]. 3 The concentration of 129I in the environmental and geological samples is normally too low to be 4 measured by radiometric techniques (<5×1014 at/g for solid samples, or <5×1012 at/kg for water 5 samples) [6]. For this reason, 129I in environmental levels is usually measured by Accelerator Mass 6 Spectrometry (AMS) [9], which is an ultrasensitive technique that combines conventional mass 7 spectrometry techniques with a particle accelerator so that particles are accelerated to energies much 8 higher than in other mass spectrometry techniques, obtaining very high sensitivities [10]. 9 In AMS the 129I/127I ratio is measured, then 127I is determined and 129I is obtained indirectly from this. 10 The first measurements step is a radiochemical procedure to separate iodine from the sample matrix, 11 purify it and finally precipitate it as silver iodide (AgI), which is the chemical form needed to measure 12 iodine by AMS. Radiochemical methods must be designed to efficiently extract all iodine species 13 from the matrix, using methods as simple as possible and guaranteeing a low-probability 14 contamination. 15 Extracting iodine from solid samples, such as marine sediments, can be done using different methods 16 which include alkaline fusion, oxidizing pyrolysis, or microwave acid digestion. In the alkaline fusion 17 method [11], decomposition of samples takes place in open crucibles using an oven set to above 18 250ºC, with slow temperature ramps up to this temperature. Oxidizing combustion [12] is a more 19 commonly used method for separation of iodine from solid sample, using quartz boats to hold the 20 samples through a special combustion system. 21 4 In this work, a microwave digestion procedure previously applied to solid materials [13], was adapted 1 and optimized for the extraction of iodine from sediments and the determination of 129I through AMS, 2 in a fast method with low risk of contamination. 3 The microwave digestion process requires special equipment that use PTFE closed vessels and allows 4 the simultaneous digestion of more than 10 samples. Key advantages of microwave digestion include 5 the shorter processing time if several samples are digested; the small amount of sample need; a low 6 probability of cross-contamination; and the reduced loss of volatile species as is typical with iodine 7 in an acidic environment. 8 After the digestion of the sample, the next step eliminates the remaining organic matter, avoiding any 9 loss due to volatilization, and the homogenization of all iodine species through a redox process. Next, 10 the iodine is purified through a double liquid-liquid extraction with a solvent and, finally, the 11 extracted iodine is precipitated as AgI. 12 Different modifications of former methods were evaluated, to streamline the process making it less 13 time consuming, and ensure the extraction of all iodine species was achieved, whilst increasing 14 reproducibility. Aliquots from two sediment cores from the Celtic Sea with contrasting characteristics 15 and a standardized sediment sample were used in the optimization of this new radiochemical method. 16 The final optimized version of this procedure was used for the measurement of the 129I concentration 17 in the both sediment cores to produce a full sediment profile of 129I in each. 18 2. MATERIAL AND METHODS 19 2.1. Samples description 20 Two sediment cores (A and I) from the Celtic Sea (Figure 1) were collected in March 2015 during 21 the DY030 sampling campaign of the new RRS Discovery. The sampling campaign was included in 22 5 the Shelf Sea Biogeochemistry (SSB) project supported by Natural Environment Research Council 1 from UK. A Core (51o 13’ 59’’N; 6o 7’ 59’’ W) was sampled on May the 11th and I Core (50o 36’ 0’’ 2 N; 7o 5’ 59’’ W), five days later, on May the 16th. Core A is 25 cm long and Core I is 21 cm. Both 3 cores (10 cm diameter) were sliced into segments of different thicknesses, 0.5 cm segments in the 4 most superficial part (top 2 cm), 1 cm segments in the middle (2-15 cm) and 2 cm segments in the 5 deepest layers to the bottom of the core. 6 2.2. Measurement technique 7 129I was measured by AMS using the 1MV Tandem AMS system (HVEE) in the Centro Nacional de 8 Aceleradores (CNA) in Sevilla, Spain. A detailed description of AMS system has been reported in 9 [14,15] and recent status and updates in [16,17]. 10 Iodine is extracted as I- through a Cs+ sputtering process as the source. The ions are selected according 11 to their mass and injected into accelerator using a low energy 90º injector magnet (LEM). Negative 12 ions are accelerated, and charge inversion takes place in the accelerator. Helium gas is used as stripper 13 gas at an optimum pressure of 1.8 × 10-2 mbar. This stripper gas has shown to give high transmission 14 for heavy ions in AMS [18]. To select charge state 2+ after the stripping process, terminal voltage is 15 set to approximately 700 kV. In these conditions, transmission is approximately 40%. After that, a 16 high-energy 90º analyzing magnet (HEM) selects 129I+2 ions and an 120º electrostatic defector (ESA) 17 performs the final charge, energy and mass analyses of the beam. The detector consists of a gas 18 ionization chamber where the energy of the ions is measured. 19 All samples, standards and blanks were measured for at least four different cycles in each 20 measurement. One standard is measured for every four or five samples and approximately one blank 21 for every twelve samples. The average value of measured blanks is calculated and subtracted to the 22 isotopic ratios obtained for samples and standards. The uncertainty associated with the average blank 23 6 isotopic ratio was estimated as a maximum between statistical errors and standard deviations of the 1 mean. Then, standard correction was carried out for all sample isotopic ratios. [19] 2 Table I shows the values of ratio 129I/127I for instrumental blanks and standards, samples and chemical 3 blanks. Instrumental blanks isotopic ratios are 129I/127I = 0.9-2 × 10−13. The standard used, Q2, was 4 prepared in our laboratory by isotopic dilution and has been determined to have an isotopic ratio of 5 ratio of 129I/127I = (1.117 ± 0.046) × 10−11 by ETH Zurich. 6 2.3. Radiochemical procedure 7 All reagents used were commercial products of analytical grade. Solutions were prepared using 8 ultrapure water (H2O-MQ). 9 The carrier solution was prepared by weighing 50 mg of solid iodine (Woodward Iodine) and 10 dissolving it in a reducing alkaline solution prepared with 1 g of sodium hydroxide (NaOH) lentils 11 and hydrazine hydrate (NH2-NH2) 80% dissolved in 50 mL ultrapure water. The iodine solution 12 concentration is calculated accurately, it must be 1 mg I- per ml. 13 At least one procedure blank was prepared for all sample series, in order to control and check for 14 contamination during the preparation process. Blanks were prepared following the same method as 15 the samples but without adding any sample material. 16 The general radiochemistry consisted essentially of extracting the iodine from the sample, previously 17 digested as described, using a double liquid-liquid extraction with a solvent and, finally, precipitating 18 the extracted iodine in the required AgI form. One first microwave oven based method was 19 implemented by Guzman et al. [13] and consisted of a microwave digestion with concentrated nitric 20 acid directly followed by double liquid-liquid extraction, first in an organic solvent and then in a 21 reducing solution. Back extraction of iodine in a reducing solution during purification was carried out 22 7 with sodium hydrogen sulfite (NaHSO3). Finally, the iodine was precipitated as AgI using silver 1 nitrate (AgNO3). In an intermediate step, SO32− and SO42− ions, added to the solution through the 2 NaHSO3, had to be eliminated. They were removed in an additional step by the addition of saturated 3 barium nitrate solution (Ba(NO3)2) followed by centrifugation several times, to avoid precipitation as 4 silver sulfate and silver sulfite (AgSO4 and AgSO3, respectively). 5 Previous tests showed that, generally, the elimination of SO32− and SO42− ions was not completely 6 efficient and the purity of the AgI was reduced due to the coprecipitation of AgSO4 and AgSO3, which 7 were responsible for lower measurement currents. In this work, to avoid the precipitation of these 8 sulfates and to make the procedure less time consuming, we have implemented a modification of that 9 former method to avoid the addition of sulfur and the extra step of removing SO42- and SO32- ions. 10 This way, in all the methods we present and evaluate here, NaHSO3 was substituted by NH2-NH2 as 11 reducing agent in the iodine back extraction during purification. This first method (denoted as method 12 A), that only included reducing reagent substitution, did not provide robust results, as they were not 13 reproducible and showed high variability. Sediments are complex matrices that contain organic and 14 inorganic fractions in highly variable proportions, including iodine attached in multiple species. For 15 this reason, several amendments to method A were introduced to maximize the extraction of all the 16 iodine species. Four versions were denoted successively as A (described above), B, C and D, different 17 changes were added from method A through to method D in subsequent additions and/or 18 modifications. Each method developed the improvements already made in the previous version, 19 building on the improved methodology each time. The aim was to ensure organic matter 20 decomposition and chemical equilibrium between iodine species, as well as avoiding the loss of 21 iodine by volatilization after acid digestion. The scheme of the modifications implemented are shown 22 in Figure 2 and can be described as follows: 23 8 Method A. As described, the extracting solution NaHSO3 of the former method is substituted with 1 hydrazine (NH2-NH2). 2 Method B. Includes a redox adjustment after microwave digestion by the addition of NH2OH·HCl 3 and Na2S2O5, in order reduce iodate (IO3-) to iodide (I-), to homogenize iodine chemical forms prior 4 to iodine purification. This way ensures that the iodine extracted from the sample and the iodine from 5 the carrier are in chemical equilibrium. 6 Method C. NaOCl is added after the microwave digestion to decompose any remaining organic 7 matter. 8 Method D. NaOH was added after microwave digestion to facilitate reduction of iodine in solution, 9 enhancing the NaOCl effect. 10 In summary, the most complex method (D) includes the following steps: 11 1) digestion of approximately 250 mg of sample and 3 mg of carrier with 10 mL of concentrated 12 HNO3 for 20 min total time at 200oC and 1800 MW potency (5 min at 100 oC, 10 min at 160 oC, 20 13 min at 200oC) using a microwave system (Milestone ETHOS™ EASY with SK-15 easy TEMP high-14 pressure rotor featuring 15 TFM vessels with a volume of 10 mL). 15 2) Recovery of the digested solution with H2O-MQ water to a final volume of 100-150 mL. 16 3) Addition of 0.5 g NaOH lentils and 1 mL NaOCl 14%. 17 4) Redox adjustment using 1 mL of 1M NH2OH(HCl) and 5 ml of 1M Na2S2O5. 18 5) Purification of iodine in a double liquid-liquid extraction. First, iodine is oxidized with 2 ml 0.45 19 g/l NaNO2 in an acidic environment to extract it in 100 mL of CH3Cl. Second, a back extraction in a 20 reducing aqueous solution (containing 0.5 mL of NH2-NH2 80% + 30 mL H2O-MQ) is used. 21 15 rates in two different cores. This approach could be used to estimate sedimentation rates in shelf seas 1 sediments. 2 5. ACKNOWLEDGEMENTS 3 This work has been partially financed by the project PGC2018-094546-B-I00 by the Spanish 4 Ministerio de Ciencia, Innovación y Universidades and the project US-1263369 by the Junta de 5 Andalucía (Consejería de Economía y Conocimiento). The authors thank Prof. X. Hou for his 6 collaboration in sending the material for the validation of the methods. 7 6. REFERENCES 8 MENDELEY BIBLIOGRAPHY PLACEHOLDER #1 9 Table I. Summary of measured isotopic ratios 129I/127I of instrumental blanks and standards, 10 samples, internal standard and chemical blanks. 11 129I/127I Instrumental blanks 1-2 × 10−13 Instrumental standard nominal value (Q2) (1.117 ± 0.046) × 10−11 Core sediment samples 2-20 × 10−12 Internal standard (XASTD) 0.6-2 × 10−12 Chemical blanks for sediment samples (after blank subtraction) Chemical blanks for internal standard (after blank subtraction) 1-2 × 10−13 0.2-0.4 × 10−13 12 16 Table II. Average values and calculated uncertainty of triplicates prepared by each method of depth 1 section 19-21 cm for core I (sandy), depth section 15-17 cm for core A (muddy) and Xi’an Center 2 Internal Standard (XAST). 3 Average 129I concentration (x1011 at/kg) Method Core I Core A XAST A 2.770 ± 1.139 12.100 ± 1.393 0.969 ± 0.612 B 1.590 ± 0.071 13.000 ± 1.490 0.592 ± 0.348 C 1.727 ± 0.543 13.433 ± 2.255 0.391 ± 0.017 D 1.520 ± 0.121 11.367 ± 0.216 0.432 ±0.108 (0.390 ± 0.046)1 Reported value [21] 0.328 ± 0.030 1corrected value discarding outlier 4 5 6 17 7. FIGURES 1 2 3 Figure 1. Sampling location and detail of the general circulation of water masses in the Celtic Sea 4 (based on [33]) 5 6 18 1 Figure 2. Scheme of the original method [13] with proposed modifications. A: Substitution of 2 NaHSO3 by NH2-NH2 as a reducing agent in back extraction process. B: Addition of NH2OH·HCl 3 and Na2S2O5 to homogenize iodine chemical forms through a redox adjustment. C: Addition of 4 NaOCl to decompose remaining organic matter. D: Addition of NaOH to facilitate iodine reduction 5 in solution after digestion and enhance NaOCl effect. 6 7 19 1 2 3 Figure 3. 129I concentration results with error (σ=2) and average of the triplicates (Av.) with 4 calculated uncertainty prepared by four performed methods for two core samples (I and A) from the 5 Celtic Sea (51o N; 6o 30’ W) and an internal standard sediment. A) Depth section 19-21 cm for core 6 I (sandy). B) Depth section 15-17 cm for core A (muddy). C) Xi’an Center Internal Standard 7 (XAST). 8 9 20 1 Figure 4. Results of 129I concentration depth profiles for Core-A and Core-I obtained using method 2 D (in 1012 at/kg) compared to known Sellafield and La Hague liquid discharges to the ocean (in kg). 3 4