ARTICLE pubs.acs.org/ac
Extraction Chromatographic Methods in the Sample Preparation Sequence for Thermal Ionization Mass Spectrometric Analysis of Plutonium Isotopes Jay W. Grate,* Matthew J. O’Hara, Anne F. Farawila, Matthew Douglas, Morgan M. Haney, Steven L. Petersen, Tapas C. Maiti, and Christopher L. Aardahl Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington 99352, United States
bS Supporting Information ABSTRACT: A sample preparation sequence for actinide isotopic analysis by thermal ionization mass spectrometry (TIMS) is described that includes column-based extraction chromatography as the first separation step, followed by anion-exchange column separations. The sequence is designed to include a wet ashing step after the extraction chromatography to prevent any leached extractant or oxalic acid eluent reagents from interfering with subsequent separations, source preparation, or TIMS ionization. TEVA resin and DGA resin materials, containing extractants that consist only of C, N, O, and H atoms, were investigated for isolation of plutonium. Radiotracer level studies confirmed expected high yields from column-based separation procedures. Femtogram-level studies were carried out with TIMS detection, using multiple monoisotopic spikes applied sequentially throughout the separation sequence. Pu recoveries were 87% and 86% for TEVA and DGA resin separations, respectively. The Pu recoveries from 400 μL anion-exchange column separation sequences were 89% and 93% for trial sequences incorporating TEVA and DGA resin. Thus, a prior extraction chromatography step in the sequence did not interfere with the subsequent anion-exchange separation when a simple wet ash step was carried out in between these column separations. The average measurement efficiency for Pu, encompassing the chemical separation recoveries and the TIMS ionization efficiency, was 2.73% ( 0.77% (2σ) for the DGA resin trials and 2.67% ( 0.54% for the TEVA resin trials, compared to 3.41% and 2.37% (average 2.89%) for two control trials. These compare with an average measurement efficiency of 2.78% ( 1.70%, n = 33 from process benchmark analyses using Pu spikes processed through a sequence of oxalate precipitation, wet ash, iron hydroxide precipitation, and anion-exchange column separations. We conclude that extraction chromatography can be a viable separation procedure as part of a multistep sequence for TIMS sample preparation.
n the field of radiochemical analysis, and trace actinide analysis in particular, sample preparation and chemical separations precede detection by radiometric or mass spectrometric means.1 These processes are particularly rigorous for detection by thermal ionization mass spectrometry (TIMS). TIMS is a widely used mass spectrometric method for obtaining the highest quality isotopic information, providing excellent isotopic resolution, high precision, and extremely low detection limits.18 It is the analytical reference method for certifying various standard materials. However, TIMS requires essentially pure sources in order to achieve the performance for which it is noted. In addition to removing or separating ions that will cause the isobaric, molecular ion, or spectral interferences common to mass spectrometric approaches,1,9,10 it is necessary to take great care to remove anything that may interfere with the ionization process or contribute to backgrounds. The analysis of plutonium isotopes in environmental samples is a typical TIMS application, which is of interest in the fields of environmental contamination, safeguards, nuclear forensics and archeology, and the study of historical samples of interest. More generally, isotopic analyses by TIMS are important in geochemistry, planetary sciences, and chemical oceanography. Sample preparation for actinide analysis by TIMS involves steps to bring the sample into solution, separate actinides from
I
r 2011 American Chemical Society
major matrix components, separate actinides from one another, and prepare the source on the filament for TIMS.3,6,1115 Typical separation approaches to achieve these objectives are precipitations16 as a matrix separation method followed by anionexchange column-based separations to isolate actinide elements. Precipitation methods include oxalate (e.g., with a rare earth carrier such as Nd) followed by wet ashing the oxalate, and iron hydroxide precipitation.3,6,11,1620 Multiple anion-exchange separations on decreasing size scales may be carried out, sometimes including a microscale separation on a single large anion-exchange bead.3,1315,21 Separations are performed on decreasing volume scales because background interferences in TIMS are often limiting factors that determine the achievable detection limits, and the eluent reagents used to perform the separations introduce new impurities into the sample (even using high-purity reagents). The only way to minimize these impurities is to scale down the separation procedure. Finally, sources are prepared by either depositing a solution on the filament, or by adsorbing the sample on a single bead, which is then loaded on the filament and Received: August 17, 2011 Accepted: October 17, 2011 Published: October 17, 2011 9086
dx.doi.org/10.1021/ac202150v | Anal. Chem. 2011, 83, 9086–9091
Analytical Chemistry pyrolyzed. For the latter approach, the sample is typically captured on a small anion-exchange bead.3,11,14,2228 In recent decades, extraction chromatographic procedures have been developed to simplify and streamline radiochemical sample preparation procedures.1,9,10,2938 In this approach, a polymer resin that is impregnated with an extractant is used for column-based separations. Typically, the resin is used to capture actinides of interest to separate them from gross matrix interferences, and sometimes to individually separate actinides from one another, using a sequence of eluents to release actinides individually or in small groups. Extraction chromatographic methods are advantageous because they are better at performing separations on complex sample matrixes; they have better kinetics than other column-based methods (e.g., solid-phase extraction with covalently linked ligands); extraction chromatographic materials have excellent capacity, they do not require a manual solidliquid separation (e.g., in contrast to precipitations) because the solid is stationary, and they are amenable to automation. There have been limited reports of using extraction chromatography in sample preparation sequences for isotopic analysis using TIMS detection. TOPO-impregnated resin or UTEVA resin have been applied to purification of spent fuel or mixed oxide fuels.3944 In trace elemental analysis of environmental samples using TIMS, there has been limited use of extraction chromatography,6,7,28,4547 and some papers have noted that leached extractant can interfere with TIMS ionization.7,46 Generally, the extraction chromatography eluates of interest have been further processed before TIMS source preparation. By and large, the use of extraction chromatography has not been embraced by those carrying out trace actinide isotopic analysis via TIMS. Caution with regard to the use of extraction chromatography for actinide separations prior to using TIMS detection arises from the fact that these resins typically leach extractant that can contaminate downstream solutions and surfaces. The major concerns are that these extractants may interfere with subsequent chemical and microscale separations, may carry interferences forward that would otherwise be removed, raising backgrounds and detection limits, may interfere with source preparation by capturing the ions of interest in condensed extractant droplets and thus prevent the ions from being captured on anionexchange beads used for source preparation, and may ultimately lead to interference with the ionization process on the filament (as some authors cited above have noticed). In this paper, we demonstrate that extraction chromatography can be successfully used in the trace analysis of plutonium with TIMS as the final detection step. We are not proposing the direct use of extraction chromatography fractions for TIMS source preparation, for the reasons just noted. Rather, our objective is to use extraction chromatography as a matrix separation step that can replace complex, tedious, and difficult-to-automate precipitation steps. Matrix separation is the stage of purification that most benefits from the noted advantages of extraction chromatography (see above). In this concept, sample purification using anion-exchange steps after matrix separation, and prior to source preparation, are retained as part of the overall sample preparation sequence. It is a bonus that, in comparison to precipitations for matrix separation, extraction chromatography can also isolate actinides from one another individually or as groups. Accordingly, we have performed two series of studies: in the first, we perform extraction chromatographic separations and anion-exchange separations using radiotracer detection methods
ARTICLE
Figure 1. (A) Radiotracer studies for chemical recoveries from extraction chromatography (EC) and anion-exchange (AnIX) column separations. In addition, an experiment was done linking the two AnIX separations. (B) A schematic diagram of the sample preparation sequence incorporating EC and multiple AnIX column separations prior to TIMS source preparation and analysis, showing the use of multiple monoisotopic tracers. In this approach, the extraction chromatography replaces one or more manual precipitation steps that would otherwise be used to separate the actinides from the bulk sample matrix. The rows listed above the B scheme indicate rows in Table 3, relating to which portions of the sequence are covered by the listed yields using isotope ratio methods.
to determine the chemical recoveries for these column-based separation steps, as indicated in Figure 1A. In the second series of tests, conducted at femtogram levels of plutonium, we perform a complete sequence of sample preparation steps followed by TIMS detection. This sequence includes an extraction chromatographic separation, a set of intermediate wet ash and redox adjustment steps, followed by anion-exchange column separations on decreasing volume scales as shown in Figure 1B. This approach was quantified using multiple monoisotopic tracers added at different points in the sequence (as shown in Figure 1B), in order to track chemical yields throughout the TIMS sample preparation process. The final sample measurement efficiencies, which include recovery and TIMS ionization efficiency, were also determined. None of the prior studies using extraction chromatography and TIMS (cited above) have traced yields through a multistep separation sequence in this level of detail. Our studies show that extraction chromatography used in this manner does not compromise subsequent anion-exchange separations, nor does it interfere with TIMS ionization. Hence, extraction chromatography has the potential to replace precipitation steps as a matrix separation approach when preparing samples for TIMS. As has been amply demonstrated in the past, column-based extraction chromatographic separations are readily automated.3438,48
’ EXPERIMENTAL SECTION Radiotracer Materials and Separations. TEVA resin extraction chromatographic resin columns (2 mL disposable dry-pack cartridges) and AG 1-X4 anion-exchange columns (2 mL and 100 μL column bed in, respectively, 5 and 1 mL columns) were purchased from Eichrom Technologies, Inc. Tracers of 239Pu, 230 Th, 241Am, and 237Np were high-purity standards purchased from Eckert & Ziegler Isotope Products. The 233U was from NIST SRM 995. Column-based separations were conducted using a fluidic system with a digital syringe pump as the fluid 9087
dx.doi.org/10.1021/ac202150v |Anal. Chem. 2011, 83, 9086–9091
Analytical Chemistry drive to deliver reagents at controlled flow rates. The 239Pu, 230 Th, and 233U were analyzed using a liquid scintillation analyzer (Perkin-Elmer Tri-Carb 3100TR), whereas the 241Am and 237Np were analyzed using a γ counter (Perkin-Elmer Wizard 1470). The percent recovery threshold for each collected fraction for both the liquid scintillation and γ counting was 0.5%. Detailed procedures for actinide separations at radiotracer levels are provided in the text and tables in the Supporting Information. Caution: radioactive isotopes represent radiological hazards. TIMS Recoveries and Analysis. All acid reagents in the separations for TIMS analysis were Optima grade. Triplicate samples were prepared for both DGA resin and TEVA resin extraction chromatography by spiking 2.4 1015 g of 244Pu into 0.5 mL of 1.2 M HCl. Solutions were evaporated to neardryness in Teflon vials, then brought up in 6 mL of the respective condition, load, and wash solutions: 7.5 M HNO3/0.02 M Fe(II)/0.1 M sulfamic acid/0.01 M NaNO2 for the DGA resin trials and 7.5 M HNO3/0.02 M Fe(II)/0.1 M sulfamic acid for the TEVA resin trials. A solution of 0.2 M NaNO2 was added to the sample just before the load onto the TEVA column. A vacuum box system (Eichrom Technologies, Inc., Darien, IL) was fitted with prepackaged 2 mL cartridges (50100 μm particle size) of each extraction resin and conditioned with 10 mL of the condition solution. Subsequent load, rinse, and elution steps were conducted at approximately 1 mL/min flow rates. Samples were loaded onto the respective cartridges and sample vials rinsed with 10 mL of the wash solution in 23 mL increments prior to addition to the cartridges to ensure quantitative transfer. DGA resin: The DGA resin cartridges were rinsed with 5 mL of 1 M HCl/0.2 M NaNO2. A 15 mL rinse with 1 M HCl followed, performed to remove any trivalent actinides or lanthanides present. The plutonium was eluted from the cartridges using 20 mL of 0.5 M HNO3/0.03 M oxalic acid. TEVA resin: The TEVA resin cartridges were rinsed with 5 mL of 6 M HCl/0.2 M NaNO2. A 7.5 mL rinse with 9 M HCl followed to elute any Th present. The plutonium was eluted from the column with 10 mL of 0.2 M HCl. In all cases, plutonium fractions were eluted into glass beakers which contained 3 mL of 1.2 M HCl and approximately 7.5 mg of Fe(III) to minimize sorption of the trace quantity of plutonium to the vessel. A perchloric acid wet ash was performed after extraction chromatography by evaporating to near-dryness a mixture of 1 mL each of concentrated nitric and perchloric acids (1), and 1 mL of concentrated perchloric acid only (3). Samples were then triply transposed with 1 mL of concentrated nitric acid, and a consistent quantity of 242Pu tracer, approximately 2.5 1015 g, was added to each sample. To restore the plutonium tetravalent oxidation state, the species which has high retention on anionexchange resin in nitric acid, a freshly prepared reducing reagent solution49 composed of 7.5 M HNO3/0.02 M Fe(II)/0.1 M sulfamic acid/0.1 M NaNO2 was added to each sample prior to loading onto a 400 μL AG1-X4 (100200 mesh) column. Following conditioning of the column with 2 mL of the load solution, the sample was quantitatively transferred to the column and rinsed with 2.5 mL of 7.5 M HNO3 and 2 mL of 9 M HCl, then the plutonium was eluted with 2 mL of 1.2 M HCl into a Teflon vial. All samples were spiked with a consistent quantity of 239Pu tracer, approximately 2.5 1015 g, doubly transposed with concentrated nitric acid, then brought up in 550 μL of 7.5 M HNO3 and quantitatively loaded onto a 100 μL AG1-X4 column. In similar fashion as above, the column was rinsed with
ARTICLE
Figure 2. Organic extractants utilized in TEVA resin and DGA resin extraction chromatographic resins.
1200 μL of 7.5 M HNO3 and 450 μL of 9 M HCl, then the plutonium was eluted with 500 μL of 1.2 M HCl into a Teflon vial. Samples were transposed to nitric acid, and then the Pu was incorporated into a nominally 150 μm diameter anion resin bead by equilibrating the bead with successively smaller volumes of the sample in a 7.5 M HNO3 medium. The beads were transferred to high-purity rhenium filaments, then the beads were pyrolyzed under vacuum through increasing current (0.5 A/h), and the filament carburized using benzene to improve the ionization efficiency of actinides.
’ RESULTS AND DISCUSSION Approach. Extraction chromatographic materials were selected to separate the actinides from bulk matrix interferences, with the criterion that the extractant could be completely destroyed in a wet ashing step following the column separation. TEVA resin,50 containing a quaternary ammonium salt as the extractant, and DGA resin,51 containing N,N,N0 ,N0 -tetra-n-octyldiglycolamide, were investigated. These organic extractants (Figure 2) comprise carbon, nitrogen, oxygen, and hydrogen atoms. Hence, a wet ash step of the Pu/Np elution fraction will destroy any potential leached extractant so that it cannot interfere in subsequent separation and source preparation steps. In addition, oxalic acid, a complexant used to elute plutonium from DGA resin, can be destroyed by this wet ashing procedure. We are particularly interested in the analysis of plutonium isotopes. These selected resins can separate plutonium from uranium, as well as separate plutonium from trivalent actinides (americium) and lanthanides. Depending on the resin and procedure, they may also separate thorium. It is not necessary to separate plutonium from neptunium, as neptunium does not interfere in the isotopic analysis of plutonium. In this regard, matrix separation using extraction chromatography is potentially advantageous over precipitation procedures that do not provide this level of actinide separation. For example, a ferric hydroxide precipitation does not separate these actinides, whereas an oxalate precipitation can separate uranium but not the others. Methods for the extraction chromatographic separations are given in the Experimental Section. TEVA resin is impregnated with a liquid anion exchanger, and typical actinide separations consist of loading the sample in strong nitric acid, where tetravalent actinides Pu, Np, and Th are retained, while U and Am pass through the column during sample load and column wash steps.50 The Th is released in high concentrations of hydrochloric acid. The Pu/Np fraction is eluted with low concentrations of hydrochloric acid. We found that, at hydrochloric acid concentrations below 0.1 M HCl, the extractant leached from the resin to yield a turbid eluate.52 At 0.2 M HCl, the eluate was visibly clearer (see Figure 3). As HCl concentrations rise above 1 M HCl, it is known that capacity factor values for retention of Pu increase substantially.50 9088
dx.doi.org/10.1021/ac202150v |Anal. Chem. 2011, 83, 9086–9091
Analytical Chemistry
ARTICLE
Figure 3. Eluates from TEVA columns at increasing HCl concentration, illuminated with a green laser from right to left to reveal scattering by immiscible droplets of leached extractant. All TEVA columns were pretreated with ca. 5 mL of 9 M HCl and purged with air. Then 10 mL of (left to right) DI water, 0.025 M HCl, or 0.1 M HCl was delivered at 1 mL/min through each cartridge, respectively, and collected in the vials pictured.
Table 1. Measured Chemical Yields (%) for Actinide Radiotracers in the Pu/Np Elution Fraction separation(s)\actinide
a
Am
U
Th
Np
Pu
1a. TEVA resin
