Determination of the lanthanides, uranium and plutonium by means of on-line high-pressure ion chromatography coupled with sector field inductively coupled plasma-mass spectrometry to characterize nuclear samples

https://doi.org/10.1016/j.chroma.2019.460839Get rights and content

Highlights

  • Plutonium, uranium and the lanthanides are separated in a single run using 1 M nitric acid followed by a gradient of oxalic acid concentrations (0.1–0.15 M) at pH 4.5 on CG5A and CS5A mixed bed ion exchange columns in less than 60 min.

  • The HPIC-SF-ICP-MS method offers time savings and reduces the radiation dose to the analyst compared to gravitational ion exchange chromatography (requires weeks) followed by TIMS for the analysis of spent nuclear fuel.

  • The HPIC-SF-ICP-MS method was found to result in linearly increasing pulse count signals over the range 1–10 µg.L−1 for Nd, 10–100 µg.L−1 for Gd, 1–10 µg.L−1 for Pu and 0.5–24 µg.L−1 for U isotopes.

  • Nuclide-specific mass fractions, derived from nuclide-specific concentration measurements (external calibration), for Nd (e.g. m(142Nd)/m(Nd)) and Gd isotopes (e.g. m(152Gd)/m(Gd)) in a Gd spent nuclear fuel were 97–103% and 90–110%, respectively, of those obtained using TIMS

  • Concentrations of Pu and U determined using HPIC-SF-ICP-MS were 105% and 97%, respectively, compared to their theoretical concentration in a 242Pu spiked soil sample.

Abstract

High-pressure ion chromatography (HPIC) was coupled with sector field inductively coupled plasma-mass spectrometry (SF-ICP-MS) to separate plutonium (Pu), uranium (U), neodymium (Nd) and gadolinium (Gd) nuclides from isobaric nuclides and to quantify them with high sensitivity. In this study, mixed bed ion exchange columns CG5A and CS5A were used, from which Pu and U were eluted first using 1 M nitric acid. The lanthanides were then separated using a gradient of 0.1–0.15 M oxalic acid with the pH adjusted to 4.5. The HPIC-SF-ICP-MS method was validated using different sample matrices, i.e. spent nuclear fuel and soil. The method was found to be repeatable and gave rise to transient signals suitable for quantification of nuclide-specific concentrations using external calibration. In terms of accuracy, the HPIC-SF-ICP-MS measurement results were in good agreement with those obtained using thermal ionization mass spectrometry (TIMS). Finally, the method provides an improvement in sample throughput (≤60 minutes per sample) and reduces exposure of the operator to radiation compared to off-line gravitational chromatography followed by TIMS.

Introduction

Mass spectrometry (MS) is a widely recognized powerful characterization method for a variety of samples [1], [2], [3]. Different mass spectrometric techniques exist, such as secondary ion MS (SIMS), laser ionization MS (LIMS), glow discharge MS (GDMS), thermal ionization MS (TIMS) and inductively coupled plasma-mass spectrometry (ICP-MS). In the nuclear field, TIMS and ICP-MS are the most frequently applied methods and they are also benchmark techniques for routine measurements, such as post-irradiation examination of spent nuclear fuel (SNF) [1]. A particular advantage of ICP-MS is its capability of determining long-lived radionuclides with a half-life exceeding 106 years [4]. Additionally, it provides a broad elemental coverage [2], very low detection limits [3], measurement of isotope ratios and precise determination of the nuclide-specific concentration of fission products and actinides for the determination of the SNF burnup. The burnup is the number of fissions undergone by the fuel [5] and is usually expressed as the percentage of fissile metal atoms that underwent fission (% FIMA) [6]. The SNF burnup can be determined by destructive radiochemical analysis using ICP-MS or TIMS to measure the nuclide-specific concentration of a fission product monitor, usually 148Nd [7], and that of the residual heavy metal elements of the fuel: U and Pu [8], [9], [10]. In the case of experimental fuels such as gadolinium nuclear fuel, where Gd is included in the fuel to serve as a burnable neutron absorber in the fuel assembly thus improving rector performance [11], also the determination of the nuclide-specific concentrations of the Gd isotopes would be of interest. However, both techniques (ICP-MS and TIMS) suffer from isobaric spectral overlap hampering the accurate determination of nuclide-specific concentrations. More specifically, the signals of different nuclides having the same mass number, such as 150Nd and 150Sm or 142Ce and 142Nd, can even not be resolved by a sector field mass spectrometer operated at the highest possible mass resolution setting of 10000. Thus, such overlap has to be overcome by chemical separation of the interfering nuclides prior to introduction into the mass spectrometer. The chemical separation can be performed using gravitational ion chromatography or high-pressure ion chromatography (HPIC). To avoid a tedious and time-consuming method, HPIC is widely used [2], [6], [12]. Isolation of the analytes of interest by means of HPIC can be combined with MS detection, either off-line or on-line. On-line coupling is not compatible with TIMS, but with ICP-MS it offers a safe, time-efficient method for the characterization of SNF and other sample types and is the main focus of the current study.

This coupling has been investigated for more than twenty years with different types of ion chromatography columns and different kinds of ICP-MS instruments. For the separation of the lanthanides and actinides, columns such as mixed bed cation and anion exchange CG5A and CS5A [1], [3], [8], [13], [14], cation exchange (such as LUNA SCX [8], Shodex IC R-621 [9] and CS10 [2], [15]) and reversed phase monolith columns (which were modified into a dynamic cation-exchange support using camphor-10-sulfonic acid [6] or n-octane sulfonic acid [16]) are mainly used. With the mixed bed ion exchange columns, lanthanides are eluted using either α-hydroxyisobutyric acid (α-HIBA) [6], [8] in the order of increasing ionic radius (from Lu to La) or using oxalic acid [2], [5], [11], [12], [17], [18] in the reverse order (from La to Lu). The actinides (such as U and Pu) are eluted using nitric acid [2] or hydrochloric acid [8] with this type of column. Alternatively, 2-hydroxy-2-methylbutyric acid [8] (HMB) can be used with cation exchange columns to elute the lanthanides according to their increasing ionic radius. Finally, with the dynamically modified monolith columns, α-HIBA is used to elute the lanthanides and actinides [6], [19].

In previous studies, HPIC has been coupled to quadrupole-based and to multi-collector sector field ICP-MS instruments [1], [8], [19], [20] to determine isotope ratios based on transient signals of analytes eluting from the different types of ion exchange columns. However, no previous study has investigated the coupling of HPIC to a single-collector sector field ICP-MS (SF-ICP-MS) instrument to study lanthanides and actinides in various sample matrices. Nevertheless, the use of a single-collector sector field instrument can be advantageous for applications that require high sensitivity, but are less demanding in terms of isotope ratio precision [10]. A SF-ICP-MS unit offers a substantially higher sensitivity than does a MC-ICP-MS instrument, because it uses an electron multiplier for ion detection, while MC-ICP-MS uses Faraday collectors for this purpose. As a result, the concentrations of radionuclides in the samples to be measured by SF-ICP-MS can be much lower compared to those to be measured by MC-ICP-MS. The purchase cost of a SF-ICP-MS instrument is also considerably lower (2- to 3-fold) than that for MC-ICP-MS instrumentation [10]. Another advantage of SF-ICP-MS over MC-ICP-MS is the higher number of nuclides that can be monitored in one injection or measurement method. Although the analyte nuclides are not monitored simultaneously, the single detector of a SF-ICP-MS instrument is not as limited by a certain mass range and maximum number of isotopes that can be monitored, as is the case for a MC-ICP-MS instrument, in which the (custom) detector configuration determines the maximum number and mass range of isotopes that can be monitored simultaneously. Modern quadrupole ICP-MS instruments also offer high sensitivity in different matrices, but their compact size renders handling and maintenance inconvenient after their nuclearization in a glovebox. In addition, the flat-top peaks obtainable with SF-ICP-MS (at low mass resolution and even at medium mass resolution with an exit slit wider than the entrance slit [21]) result in better isotope ratio precisions than those attainable with quadrupole-based ICP-MS [3], [10]. These flat-top peaks are a characteristic of the mass spectrometer design, and, when measuring in low resolution mode, this holds true whether measuring off-line or on-line. Therefore, the coupling of HPIC to SF-ICP-MS to measure radionuclides offers a fit-for-purpose precision and sensitivity and reduces exposure of the operator to radiation. With this in mind, the objective of the current paper does not involve measurement of isotope ratios, but is strictly focused on the separation of lanthanides, uranium and plutonium in different sample matrices and the quantification of concentrations of neodymium, gadolinium, uranium and plutonium isotopes from transient signals using external calibration.

This paper presents an on-line HPIC-SF-ICP-MS separation method for the lanthanides, U and Pu developed and validated in our laboratory, for the characterization of SNF and environmental soil samples.

Section snippets

Reagents and standards

Chemicals used for preparing the eluent solutions included: Trace Metal Grade (TMG) oxalic acid (C2H2O4, 99.999% w/w) purchased from Sigma-Aldrich (Overijse, Belgium), TMG ammonium hydroxide (NH4OH, 20–22% w/w), OPTIMATM grade nitric acid (HNO3, 67–69% w/w) and hydrochloric acid (HCl, 32–35% w/w) for ultra-trace analysis obtained from Fisher Scientific (Merelbeke, Belgium). Spex Certiprep mono-elemental standard solutions for In, La, Ce, Pr, Nd, Sm, Eu, Gd, Lu, Tl and U at a concentration of

Separation of lanthanides

Using a gradient of oxalic acid concentrations from 0.1 to 0.15 M at a pH of 4.5 resulted in the lanthanides eluting from the column in the order La to Lu (Fig. 2), with a resolution >1.0 and a separation factor (α) > 1.1, as shown in Table 3. Lanthanides exist in solution as trivalent ions, with almost identical chemical properties, which makes their separation based solely on the selectivity of the ion exchanger challenging [3], [13]. However, the use of a complexing agent, such as oxalate [1]

Conclusions

In conclusion, a HPIC-SF-ICP-MS method for the separation of Pu, U and the lanthanides was developed with the aid of simulated speciation diagrams from Hydra/Medusa. In a single run, Pu and U were eluted separately with 1 M nitric acid as neutral plutonyl and uranyl nitrate complexes, respectively, before the lanthanides, which were eluted as anionic oxalate complexes by using a gradient of 0.1–0.15 M oxalic acid at pH 4.5. All analytes eluted as single peaks, suitable for quantitative analysis

CRediT authorship contribution statement

Nancy Nazem Wanna: Data curation, Writing - original draft. Karen Van Hoecke: Visualization, Validation, Investigation, Supervision. Andrew Dobney: Supervision, Validation, Writing - review & editing. Mirela Vasile: Supervision, Validation, Writing - review & editing. Thomas Cardinaels: Project administration, Writing - review & editing. Frank Vanhaecke: Supervision, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to thank Lesley Adriaensen, Peter Van Bree, Marc Verwerft, Els Verheyen, Prisca Verheyen and Arnaud Campsteyn for providing the samples and standards and sharing their expertise on HPIC and SF-ICP-MS operation.

References (32)

  • F. Gueguen et al.

    Neodymium isotope ratio measurements by LC-MC-ICPMS for nuclear applications: investigation of isotopic fractionation and mass bias correction

    J. Anal. Atom. Spectrom.

    (2015)
  • K. Van Hoecke et al.

    Isolation of lanthanides from spent nuclear fuel by means of high performance ion chromatography (HPIC) prior to mass spectrometric analysis

    J. Radioanal. Nucl. Chem.

    (2017)
  • M. Moldovan et al.

    High resolution sector field ICP-MS and multicollector ICP-MS as tools for trace metal speciation in environmental studies: a review

    Hour. Anal. Atom. Spectrom.

    (2004)
  • Characteristics and use of Urania-Gadolinia fuels

    (1995)
  • L. Ettre

    Nomenclature for chromatography (IUPAC recommendations 1993)

    Pure Appl. Chem.

    (1993)
  • S. Cotton

    Lanthanide and Actinide Chemistry

    (2006)
  • Cited by (13)

    • Novel biocompatible and sensitive visual sensor based on aggregation-induced emission for on-site detection of radioactive uranium in water and live cell imaging

      2023, Science of the Total Environment
      Citation Excerpt :

      In the traditional method the environmental samples were usually collected on site and then sent back to the laboratory for analysis, which is very disadvantageous for the timely warning and accurate tracing of the uranium contamination in the emergency case. Therefore, instead of the laboratory testing methods, such as surface-enhanced Raman spectroscopy (SERS) (Jiang et al., 2022; Ruan et al., 2007), atomic absorption spectroscopy (AAS) (Yildiz et al., 2016), atomic emission spectroscopy (AES) (Jamali et al., 2006), inductively coupled plasma-mass spectrometry (ICP-MS) (Wanna et al., 2020; Oshita et al., 2003), the real-time and on-site detection of uranium in the field are highly desirable. In addition, developing the convenient visual imaging method for the in-situ detection of trace uranium in living organisms is also of practical significance in consideration of its severe biological toxicity.

    • Prompt digestion and HPIC separation of rare earth elements in surrogate post-detonation debris material with detection by ICP-MS and gamma spectroscopy

      2022, Talanta
      Citation Excerpt :

      Digested samples from the microwave and ABF procedure were also diluted in 2% HNO3 and injected onto the column. This resulted in poor separation with all REEs eluting without baseline separation This was a surprising result because it has previously been reported that REEs are loaded onto mixed bed ion exchange columns as cations and eluted as anionic complexes [24]. In principle, this is advantageous because it allows the REE cations to remain at the top of the column until the complexing agent is added.

    • Quantification of uranium, plutonium, neodymium and gadolinium for the characterization of spent nuclear fuel using isotope dilution HPIC-SF-ICP-MS

      2021, Talanta
      Citation Excerpt :

      The spikes and 3.2 μg of KMnO4 were added to the samples at least 12 h before the analysis, to ensure sufficient time for isotopic equilibration and to convert all Pu to the desired oxidation state (+VI) [19], respectively. The elution method for the separation of Pu, U and the lanthanides in one run as described in Ref. [18] was used. The uranium fraction was collected automatically between 4.25 and 21.67 min by diverting the flow of the column effluent away from the ICP-MS nebulizer and towards the fraction collection system, using a 6-port 2-way valve.

    View all citing articles on Scopus
    View full text