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 pubs.acs.org/ac
Characterization of Actinides Complexed to Nuclear Fuel Constituents Using ESI-MS Luther W. McDonald, IV,*,† James A. Campbell,‡ Thomas Vercouter,§ and Sue B. Clark∥ †
Department of Civil and Environmental Engineering, University of Utah, Salt Lake City, Utah 84112, United States Chemical and Biological Signature Sciences Group, Pacific Northwest National Laboratory, Richland, Washington 99352, United States § CEA, DEN, DANS, Department of Physico-Chemistry, F-91191 Gif-sur-Yvette, France ∥ Department of Chemistry, Washington State University, Pullman, Washington 99164, United States ‡
S Supporting Information *
ABSTRACT: Electrospray ionization-mass spectrometry (ESI-MS) was tested for its use in monitoring spent nuclear fuel (SNF) constituents including U, Pu, dibutyl phosphate (DBP), and tributyl phosphate (TBP). Both positive and negative ion modes were used to evaluate the speciation of U and Pu with TBP and DBP. Furthermore, apparent stability constants were determined for U complexed to TBP and DBP. In positive ion mode, TBP produced a strong signal with and without complexation to U or Pu, but, in negative ion mode, no TBP, U-TBP, or Pu-TBP complexes were observed. Apparent stability constants were determined for [UO2(NO3)2(TBP)2], [UO2(NO3)2(H2O)(TBP)2], and [UO2(NO3)2(TBP)3]. In contrast DBP, U-DBP, and Pu-DBP complexes were observed in both positive and negative ion modes. Apparent stability constants were determined for the species [UO2(DBP)], [UO2(DBP)3], and [UO2(DBP)4]. Analyzing mixtures of U or Pu with TBP and DBP yielded the formation of ternary complexes whose stoichiometry was directly related to the ratio of TBP to DBP. The ESI-MS protocols used in this study will further demonstrate the utility of ESI-MS and its applicability to process control monitoring in SNF reprocessing facilities.
C
monobutyl phosphate (MBP).1,5−10 The primary degradation product, DBP, has adverse effects on the efficiency of U and Pu extraction including poor phase separation,11,12 decreased masstransfer coefficients,13 and poor fission-product/actinide separation factors.1,13 In addition, many studies have reported that the complexation of DBP to ZrIV, PuVI, and other metals results in the formation of a third layer made up of emulsions between the aqueous and organic interface.1,11,14−19 This third phase and/or precipitate, which is formed, creates a large uncertainty in the given Pu and U concentration at a given stage of the PUREX process. Development of a method to quantitatively characterize actinide speciation in SNF will help elucidate the species leading to the formation of this third phase while helping maintain proliferation resistance and enhanced safety during reprocessing. Many reports have demonstrated the potential of ESI-MS for characterizing the speciation of lanthanides and naturally occurring actinides (i.e., 238U and 232Th), and two reports have provided studies of Ce and Zr complexed to TBP and/or DBP using ESI-MS.20,21 However, no reports have developed methods to characterize U and Pu complexed to TBP and DBP using ESI-MS, and the analysis of any transuranic species by
hemical reprocessing of irradiated fuel provides an opportunity to recycle U and Pu after removal of the fission products that are neutron poisons, which are generated during irradiation and energy production.1 However, a disadvantage of reprocessing is the opportunity to divert special nuclear material for other purposes, such as nuclear weapons production. Consequently, international treaties such as the Nuclear Nonproliferation Treaty (NPT) establish monitoring protocols for safeguards and nuclear materials accountability. Typically, only isotopic information and quantities on U and/or Pu are reported,2 with no additional molecular detail. While accuracy and sensitivity for the isotopic information is maximized, the speed of reporting results is often slower than desired, providing no opportunity to use the information to support process control or rapidly detect an abrupt diversion. The application of ESI-MS to in-line or near real time monitoring would enable improved process control monitoring and could serve as an additional tool to support materials accountability. In this study, the complexation of U and Pu with complexants found in the most common reprocessing process (i.e., plutonium uranium refining by extraction (PUREX)) was studied.1,3,4 In this process, tributyl phosphate (TBP) in an immiscible organic solvent such as kerosene extracts PuIV and UVI from aqueous solutions derived of dissolved SNF. The TBP can be degraded by both radiolysis and acid hydrolysis, resulting in a production of dibutyl phosphate (DBP) and © XXXX American Chemical Society
Received: September 2, 2015 Accepted: January 29, 2016
A
DOI: 10.1021/acs.analchem.5b03352 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 1. Positive ion mass spectra of 0.1 mM uranyl with 2.85 mM HNO3 and 0.5 mM DBP: (A) cone voltage = 20 V and (B) cone voltage = 80 V. The 1:4 uranyl to DBP complex is most prominent at the lower cone voltage. Increasing the cone voltage causes collision induced dissociation of the larger complexes to produce primarily the 1:3 uranyl to DBP complex. Aggregation of uranyl species is also observed at the higher cone voltage, and these species are expected to be artifacts of the electrospray. Complete identification of all species observed is provided in Supporting Information, Tables S-2 and S-3.
ESI-MS is rare.22−32 This is likely due to the added safety precautions required for handling radioactive materials. In this study, ESI-MS was used to characterize U and Pu speciation with TBP and DBP in water/methanol solutions. Since Pu is more hazardous than U, it required additional safety precautions. All Pu measurements were performed in a glovebox using nano-ESI. The lower flow rate of the nanoESI enabled the analysis of very small volumes reducing the radioactivity of the samples and minimizing the amount of Pu waste generated. Results from this work improve the basic science understanding of actinide speciation in the gas phase, while the developed analytical techniques and the identification of U and Pu gas phase species provide the foundation for characterizing and monitoring SNF reprocessing systems.
■
source capillary at lower voltages resulted in an unstable production of droplets, and operating at higher voltages resulted in a corona discharge of ions. Fragmentation was induced on the ionized analytes by adjusting the cone voltage, while the extractor voltage was held at ground. The source capillary temperature was 100 and 80 °C for the micro- and nano-ESI source, respectively. Nitrogen was used as the desolvation gas at 500 L/h and 300 °C and as the cone gas at 50 L/h in the micro-ESI source. No desolvation or cone gas was used for the nano-ESI source. For MS/MS studies, collision induced dissociation (CID) was used to fragment the ions using argon as the collision gas with the collision energy varying from 5 V to 70 V. To obtain accurate mass measurements with the QTOF, the instrument was calibrated daily with a NaI standard solution in 50% (v/v) isopropyl alcohol (Acros Organics Spectroscopy grade) and water (milli-Q). Spectra were recorded from 100−1800 m/z at 1 s/scan and 6 s/scan for the QTOF and QQQ, respectively. Millipore deionized water was used for the preparation of all samples. U samples were prepared from a 10,000 mg/L stock solution of 238U as uranyl in 5% nitric acid from Spex CertiPrep. The samples were diluted to 0.1 mM in 50% (v/v) methanol (Acros Organics Spectroscopy grade) and water with concentrations of TBP (Marsan) and/or DBP (purified from Panchim) ranging from 0 mM to 0.5 mM. The proton concentration was maintained by the fixed nitric acid concentration at 2.85 mM, and the ionic strength was not adjusted in any samples. All U samples were prepared in triplicate. The mean and standard deviation were determined from the triplicates using the total abundance of each ion containing uranyl. The Pu solutions were prepared from a mother solution containing 617.7 ppm 242Pu, 3894 ppb 238Pu, and 5902 ppb 239Pu. An aliquot of the mother solution was dried overnight under nitrogen and then diluted in 2 mM nitric acid (Merck, Suprapur). Samples for analysis by nano-ESI-MS were diluted to 0.1 mM 242Pu in 50% (v/v) methanol (Acros
EXPERIMENTAL SECTION
Safety: Pu is a transuranic element requiring extra safety precautions. All solutions of Pu were handled and prepared in a negative pressure glovebox equipped with a high-efficiency particulate air (HEPA) filtered exhaust. In addition, the nanoESI-MS used for all analysis was enclosed within a negative pressure glovebox with a HEPA filter. Sample volumes and concentrations were kept to a minimum to reduce the radioactivity of the samples. U and Pu samples were analyzed using a Waters QTof II and Water μ-Quattro triple-quadrupole mass spectrometer, respectively. For U samples, a Harvard apparatus syringe pump was used to introduce the samples at 5 μL/min to the electrospray source. For Pu samples, a nano-ESI interface (Micromass, Manchester, UK) equipped with a 20 μL emitter tip placed 3 mm from the inlet orifice was used as the electrospray needle. The solution flow rate was estimated to be between 50−75 nL/ min. The micro-ESI source capillary was held at −2.25 kV for negative ionization and at +2.5 kV for positive ionization, while the nano-ESI source capillary was held at −2.00 kV for negative ionization and +2.25 kV for positive ionization. Operating the B
DOI: 10.1021/acs.analchem.5b03352 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry Organics Spectroscopy grade) and water. Varying concentrations of TBP and DBP were added to select Pu samples. The total volume of each sample was limited to 100 μL to minimize the volume of radioactive waste. All samples were left to equilibrate for 24 h before analysis by nano-ESI-MS.
where {ML} is the activity of the chemical species ML. In this study, the activity of each species was not determined, rather [ML] was used for the calculations. Prior studies by Hlushak et al. modeled the extraction of U by TBP in dodecane using the activities of each species.40 Working in a similar concentration scale with similar ratios of M and L should enable speciation calculations based on concentration to yield similar results. The [MLn]:[M] ratios were obtained from the speciation diagrams, and the initial concentration of the ligand in solution was used for [L].40 For uranyl-DBP complexes, previous experimental values are only available for the 1:1 and 1:4 complexes. An apparent stability constant for the 1:1 complex was calculated from the distribution of species at cone voltage = 20 V (Figure S-1A) and 80 V (Figure S-1B). Statistically, the values are the same and are within error of the previously reported values (Table 1). The 1:4 complex was only present at a low cone
■
RESULTS AND DISCUSSION Characterization of U Complexed to TBP and DBP. Uranyl nitrate solutions with either TBP or DBP were evaluated to determine the stoichiometry of U with the SNF complexants. The stoichiometry of the complexes was dependent on the applied cone voltage. For example, with the cone voltage = 20 V, the 1:3 uranyl to TBP complex and the 1:4 uranyl to DBP complex were most abundant. Increasing the cone voltage to 80 V caused fragmentation of these large complexes so that the 1:2 complex was most prevalent for uranyl with TBP and DBP. As prior reports have noted the challenges of correlating solution and gas phase species,33,34 apparent stability constants were determined from data in this study to quantitatively compare the gas phase complexes produced by ESI to known solution complexes. The apparent gas phase speciation of uranyl with TBP and DBP was modeled according to methods reported by Di Marco et al.35 The relative ion intensity (RI) was computed as the sum of all ions produced by a given complex divided by the sum of the intensities of all ions. Since TBP and DBP are reported to undergo a McLafferty-like rearrangement, the ions of a given complex were grouped according to the number of phosphorus atoms attached to the U.36,37 For example, a 1:1 complex was any U species with one phosphorus atom and a 1:2 complex was any U species with two phosphorus atoms. A detailed description of the stoichiometry of all complexes is reported in Supporting Information Table S-1. Uranyl complexation with DBP yielded the formation of 1:1, 1:2, 1:3, and 1:4 complexes (Figure 1). With the cone voltage = 20 V, an increase in DBP yielded an increase in the ion [UO2(HDBP)3(DBP)]+. This 1:4 complex was not observed with the cone voltage = 80 V (Figure 1B). Furthermore, at the higher cone voltage many dimerized uranyl-DBP complexes were observed. In solution, uranyl and DBP can polymerize in a high nitric acid media.38 While the initial solution conditions did not contain a high nitric acid content, following the ESI process and evaporation of the solvent on the formed droplets, a high nitric acid media would be present possibly favoring the formation of these polymeric species. Table S-2 shows all polymeric species observed and their possible corresponding solution species. All other U-DBP complexes can be found in Table S-3. Overall formation constants were calculated using the following equations39 β1
M + L ⇔ ML → β1 =
{ML} {M}{L}
β2
M + 2L ⇔ ML 2 → β2 = β3
M + 3L ⇔ ML3 → β3 = β4
M + 4L ⇔ ML4 → β4 =
Table 1. Published Equilibrium Constants Obtained Using Tradition Experimental Methods and Apparent Equilibrium Constants Estimated from This Work for the Formation of Uranyl-DBP Complexes equilibrium expression
[UO2 (DBP)] [UO22 +][DBP−] [UO2 (DBP)4 ] [UO22 +][DBP−]4
8.40 water47,48
8.1 ± 0.7 (dimer) (cone = 20 V)
+ UO22(aq) + 2H 2DBP2(org) ⇋ UO2 (HDBP)2 DBP2(solv) + 2H+
From this reaction, the log β4 (dimer) is found to be 8.1 ± 0.7, which is in agreement with the previously reported values (Table 1). The apparent speciation of uranyl with TBP at a low cone voltage observed the formation of 1:1, 1:2, and 1:3 complexes (Figure 2). The 1:1 complex was less than 10% of the total U species observed. Collision induced dissociation (CID) experiments were performed on m/z = 1130 Da to confirm the identification of the 1:3 complex as [UO22+(NO3−)(TBP)3]+. Soft collision energies (
