Electrochemically Modulated Separation, Concentration, and

Anal. Chem. 2006, 78, 8535-8542

Electrochemically Modulated Separation, Concentration, and Detection of Plutonium Using an Anodized Glassy Carbon Electrode and Inductively Coupled Plasma Mass Spectrometry William J. Clark, Jr.,† Sea H. Park, Debra A. Bostick, Douglas C. Duckworth,*,‡ and Gary J. Van Berkel

Chemical Science Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6375

Plutonium is shown to be retained on anodized glassy carbon (GC) electrodes at potentials positive of +0.7 V (vs Ag/AgCl reference) and released upon potential shifts to values negative of +0.3 V. This phenomenom has been exploited for the separation, concentration, and detection of plutonium by the coupling an electrochemical flow cell on-line with an ICPMS system. The electrochemically controlled deposition and analysis of Pu improves detection limits by analyte preconcentration and by matrix and isobaric ion elimination. Information related to the parametric optimization of the technique and hypotheses regarding the mechanism of electrochemical accumulation of Pu are reported. The most likely accumulation scenario involves complexation of Pu(IV) species, produced under a controlled potential, with anions retained in the anodization film that develops during the activation of the GC electrode. The release mechanism is believed to result from the reduction of Pu(IV) in the anion complex to Pu(III), which has a lower tendency to form complexes. A rapidly expanding field of research involves the coupling of electrochemical systems on-line with mass spectrometry (MS) to control or study products of electrochemical reactions or to increase the analytical sensitivity and resolution. Several review articles have been published in the last two decades chronicling the linking of electrochemical techniques with mass spectrometry.1-7 Much has been published recently regarding electrospray ionization mass spectrometry and the underlying electrochemical * To whom correspondence should be addressed. E-mail: douglas.duckworth@ pnl.gov. † Current address: School of Science, Penn State Erie, The Behrend College, Erie, PA 16563. ‡ Current address: Pacific Northwest National Laboratory, P.O. Box 999, Richland, WA 99352. (1) Zhou, F. Trends Anal. Chem. 2005, 24, 218-227. (2) Karst, U. Angew. Chem., Int. Ed. 2004, 43, 2476-2478. (3) Diehl, G.; Karst, U. Anal. Bioanal. Chem. 2002, 373, 390-398. (4) Zhou, F. Electroanalysis 1996, 8, 855-861. (5) Volk, K. J.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1992, 64, 21A-33A. (6) Bittins-Cattaneo, B.; Cattaneo, E.; Ko¨nigshoven, P.; Vielstich, W. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1991; Vol. 17, pp 181-220. (7) Chang, H.; Johnson, D. C.; Houk, R. S. Trends Anal. Chem. 1989, 8, 328333. 10.1021/ac061538b CCC: $33.50 Published on Web 11/15/2006

© 2006 American Chemical Society

mechanisms and control of the ionization process.8-12 The focus of the work reported herein is to use a controlled-potential electrochemical cell interfaced on-line with inductively coupled plasma mass spectrometry (ICPMS) for elemental and isotopic analysis. An electrochemical control process is used for analyte separation and concentration. The coupling of electrochemical preconcentration cells on-line with mass spectral detection has proven particularly useful for trace metals analysis.4,13-23 Stripping voltammetry allows low detection limits as a result of analyte preconcentration onto an electrode (a mercury drop, mercury thin film, or solid electrode). However, achieving reproducible results at low analyte concentrations requires near-ideal experimental conditions. Matrix species can affect electron-transfer kinetics or coreact to shift or overlap stripping potentials. Qualitative information obtained from stripping voltammetry is less than ideal, and positive species identification is often impossible. Mass spectrometry has the benefit of positive species identification via mass-to-charge along with low detection limits. Matrix effects and nominal mass overlap can (8) Van Berkel, G. J.; Kertesz, V. Anal. Chem. 2005, 77, 8041-8049. (9) Kertesz, V.; Van Berkel, G. J.; Granger, M. C. Anal. Chem. 2005, 77, 43664373. (10) Van Berkel, G. J.; Kertesz, V.; Ford, M. J.; Granger, M. C. J. Am. Soc. Mass Spectrom. 2004, 15, 1755-1766. (11) Van Berkel, G. J.; Asano, K. G.; Granger, M. C. Anal. Chem. 2004, 76, 1493-1499. (12) Van Berkel, G. J.; Asano, K. G.; Kertesz, V. Anal. Chem. 2002, 74, 50475056. (13) Pretty, J. R.; Van Berkel, G. J. Rapid Commun. Mass Spectrom. 1998, 12, 1644-1652. (14) Pretty, J. R.; Van Berkel, G. J.; Duckworth, D. C. Int. J. Mass Spectrom. 1998, 178, 51-63. (15) Pretty, J. R.; Duckworth, D. C.; Van Berkel, G. J. Anal. Chem. 1998, 70, 1141-1148. (16) Pretty, J. R.; Duckworth, D. C.; Van Berkel, G. J. Anal. Chem. 1997, 69, 3544-3551. (17) Baca, A. J.; Ce, La, Ree, A. B.; Zhou, F.; Mason, A. Z. Anal. Chem. 2003, 75, 2501-2511. (18) Cao, G. X.; Jimenez, O.; Zhou, F. J. Am. Soc. Mass Spectrom. 2006, 17, 945-952. (19) Hwang, T. J.; Jiang, S. J. J. Anal. Atom. Spectrom. 1996, 11, 353-357. (20) Pretty, J. R.; Blubaugh, E. A.; Caruso, J. A.; Davidson, T. M. Anal. Chem. 1994, 66, 1540-1547. (21) Pretty, J. R.; Blubaugh, E. A.; Caruso, J. A. Anal. Chem. 1993, 65, 33963403. (22) Pretty, J. R.; Blubaugh, E. A.; Evans, E. H.; Caruso, J. A.; Davidson, T. M. J. Anal. At. Spectrom. 1992, 7, 1131-1137. (23) Pretty, J. R.; Evans, E. H.; Blubaugh, E. A.; Shen, W. L.; Caruso, J. A.; Davidson, T. M. J. Anal. At. Spectrom. 1990, 5, 437-443.

Analytical Chemistry, Vol. 78, No. 24, December 15, 2006 8535

reduce the sensitivity and resolution of this technique also. The use of mass spectrometry as a detection method for analytes under electrochemical control provides improvements over each individual technique. Here, a small-volume electrochemical flow cell is linked on-line with an inductively coupled plasma orthogonal acceleration time-of-flight mass spectrometer. Coupling these techniques allows matrix elimination concurrently with sample preconcentration. Mass spectral interferences are eliminated, and detection limits are improved. This is particularly important in isotope ratio mass spectrometry where minor isotopes can have quite low abundances. A novel electrochemically modulated separation (EMS) method is reported for plutonium, which is shown to be concentrated onto an anodized glassy carbon (GC) electrode at potentials more positive than +0.7 V (vs Ag/AgCl reference). A release of retained Pu from the electrode is accomplished by application of potentials negative of +0.3 V (vs Ag/AgCl). Most metals preconcentrate during excursions to more negative potential values. Pu accumulation was first observed during electrochemical cleaning cycles (positive potential steps to release analyte) employed while studying uranium preconcentration and analysis, as described by this laboratory previously.15 This difference between Pu and U (and many other metals) offers attractive opportunities. For example, trace level and isotopic analysis of 239Pu are difficult in the presence of higher concentrations of 238U since uranium forms a hydride species (238U-1H) with a nominal mass of 239. Accumulation and preconcentration of Pu, but not U, should allow lower detection limits for Pu in this matrix. Parametric optimization studies and fundamental electrochemical processes related to Pu preconcentration and analysis via EMS-ICPMS are reported. EXPERIMENTAL SECTION Instrumentation. The instrumental setup consists of an electrochemical flow cell connected in-line with an inductively coupled plasma mass spectrometer. Details of the electrochemical cell and mass spectrometer are given below. Samples are introduced via flow injection using a microprocessor-controlled gas displacement pump (Microneb 2000, CETAC Technologies, Inc., Omaha, NE). The gas displacement pump system (pressurized with Ar) controls the flow rate of the carrier solution (0.46 M HNO3, Ultrex Grade, J.T. Baker) and employs an automatic injection valve system. A 1-mL sample loop is connected to the electrochemical cell via 127-µm-i.d. PEEK tubing (Upchurch Scientific, Oak Harbor, WA). The electrochemical cell is then connected to the mass spectrometer with an 80-mm length of the same PEEK tubing. The electrochemical flow cell has been described in detail previously.14-16 It is a three-electrode planar flow-by cell utilizing a GC working electrode and platinum auxiliary electrode. The reference electrode is Ag/AgCl (model RE-30, Bioanalytical Systems, (BAS) West Lafayette, IN), and all potentials listed are measured relative to this electrode. The working electrode was custom modified (BAS) to offset the 6-mm-diameter glassy carbon electrode location (auxiliary electrode upstream of working electrode) to minimize deposition at the auxiliary electrode and to reduce instabilities produced by gas evolution. A 12.7-µm spacing gasket separates the working and auxiliary electrodes. Electrochemical control is performed using a potentiostat (model 832A, CH Instruments, Austin, TX). 8536

Analytical Chemistry, Vol. 78, No. 24, December 15, 2006

Mass spectra were acquired with an ICP-orthogonal acceleration time-of-flight mass spectrometer (Optimass 8000, GBC Scientific, Dandenong, Australia). Samples were introduced into the mass spectrometer from the electrochemical flow cell via a glass concentric nebulizer (0.4 µL/min MicroMist, Glass Expansion, Inc.) and 50-mL cyclonic spray chamber (Jacketed Tracey, Glass Expansion, Inc.). Data acquisition software (Optimass version 1.2, GBS Scientific) utilized time-scan mode to collect temporal data (1-s integrations) and integrated responses for selected masses (0.8 amu mass integration width). For graphical display, the raw data (time and signal) were exported to spreadsheet software (Excel 2003, Microsoft). Reagents. A 244Pu standard (IRMM-042a, Isotopic Reference Materials and Measurements) was diluted with 0.46 M HNO3 (Ultrex, J. T. Baker) and had a concentration of 1 ng/g Pu (1 ppb Pu) unless otherwise noted. A lutetium (High Purity Standards, Inc., Charleston, SC) internal standard was used (1 ng/g Lu). Lu was not retained during EMS; thus, its temporal mass spectral profile was constant. Ultrapure water (Milli-Q purification system, Millipore Corp., Bedford, MA) was used for all dilutions. Procedure. Specific details for each experiment are included in the accompanying text. The GC working electrode was polished using successively finer particle sized abrasives; SiC paper (600, 800, 1200 grit; Beuhler) was followed by alumina slurries (1.0, 0.3, 0.05 µm; Beuhler). After each step, the electrode was rinsed and subjected to a short sonication cycle (300 s) have shown a steady increase in Pu signal during accumulation, indicating a reduction in the accumulation or retention efficiency. The cause is unclear, but it is likely that electroactive surface sites are occupied by retained nonelectroactive Pu species saturating the electrode. It is also possible that increased local concentration of retained Pu (a) alters the complexation equilibria lowering retention efficiency, (b) changes reaction kinetics via concentration effects, illustrated by the Butler-Volmer equation,34 or (c) affects mass transport into and out of the anodization film. The leveling of the EF at longer tacc could also indicate the effect of electrode loading on reaction kinetics or mass transport during the release cycle. (31) Nagaoka, T.; Fukunaga, T.; Yoshino, T.; Watanabe, I.; Nakayama, T.; Okazaki, S. Anal. Chem. 1988, 60, 2766-2769. (32) Bowers, M. L.; Hefter, J.; Dugger, D. L.; Wilson, R. Anal. Chim. Acta 1991, 248, 127-142. (33) Kepley, L. J.; Bard, A. J. Anal. Chem. 1988, 60, 1459-1467. (34) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, 2nd ed.; Wiley: New York, 2001; p 96.

Analytical Chemistry, Vol. 78, No. 24, December 15, 2006

8539

Figure 5. Effects of sample matrix acidity on the enhancement factor and deposition efficiency. All other experimental parameters held constant as described in the text.

It is important to note that the GC electrode returns to a usable state following an EMS experiment. Even during extended accumulation times exhibiting reduced % DE, the % DE returns to initial values in subsequent EMS experiments. Performance is reestablished upon release of Pu from the anodized GC electrode using a negative potential step to 100 mV. Multiple EMS experiments may be performed without further electrode conditioning. The electrode lifetime lasts well beyond a single-day usage. Three flow rates (2.3, 8.2, and 14.2 µL/min) were examined with Vacc ) 1.0 V, tacc ) 300 s, trinse ) 300 s, and Vstrip ) -0.2 V. DE decreased from 90 to 60% as the flow rate increased. The fwhm improved from 22 to 8 s as the flow rate increased. Consequently, the EF was highest for the midrange flow rate in this study, changing from 13 to 39 to 25 for the three increasing flow rates. A maximum EF for the midrange flow rate reported here can be explained by considering the effects of flow rate on accumulation and release processes. Increased flow rate results in less efficient accumulation of Pu, but increases analyte throughput at constant tacc. Release transient narrowing (smaller fwhm) with increased flow rates results from less time for band broadening between electrode and detector. During release, narrower fwhm at higher flow rates result in higher EF values but the transient peak area remains relatively constant with respect to the total mass retained. Acid Concentration. The effect of HNO3 concentration on Pu accumulation and release was investigated. The concentration of HNO3 for sample dilution varied from 0.5 to 5% (v/v), ∼0.11.1 M with VCS ) 0.2 V, Vacc ) 1.0 V, tacc ) 120 s, trinse ) 100 s, and Vstrip ) -0.2 V. Figure 5 shows the strong dependence of both % DE and EF on HNO3 concentration. The correlation between % DE and EF is not surprising, given that all electrochemical parameters were held constant. As shown, the % DE and EF both decreased with increasing HNO3 concentration. EF 8540

Analytical Chemistry, Vol. 78, No. 24, December 15, 2006

ranged from ∼20 to ∼1 (no enhancement) over the narrow range of HNO3 concentrations studied. The working hypothesis for Pu accumulation is that Pu(IV) is complexed with nitrate anions intercalated within the anodized glassy carbon matrix. As discussed previously, the oxidation of Pu(III) to Pu(IV) is required prior to accumulation. Formation of anion complexes with anions located within the anodized glassy carbon matrix is thought to be the second requirement for EMS accumulation. Glassy carbon contains short-range graphitic characteristics that allow electrical conductivity. Anodized GC electrodes have a layer of graphitic oxide on the surface, with a oxygen to carbon (O/C) ratio in the range of 0.25.26 Graphite forms intercalation compounds with anions reversibly injected between graphitic layers when subjected to certain positive potentials. Positive of a particular limit, irreversible lattice damage and the formation of graphite oxide occurs. Similar injection of ions into anodized GC has been observed.26,31 Retained anions can be released from the graphite or GC by exposure to negative potentials. Depth profiling by secondary ion mass spectrometry of GC electrodes anodized in various media showed that the ions migrate well into the electrode.32 S and P were detected at depths near 300 nm for GC electrodes anodized for 10 min in H2SO4 and H3PO4, respectively. These electrodes were cycled between +0.6 and +2.3 V (or +3.0 V). Presumably, the lower potential limit was not negative enough to release the anions that had been adsorbed. In the present study, the anodizing procedure cycled between +1.85 and +0.85 V. NO3or SO42- (or HSO4-) ions are likely retained GC electrode during anodization. Pu species differ in their tendency to form complex ions, decreasing as Pu4+ . Pu3+ ≈> PuO22+ > PuO2+. Pu(IV) forms complexes much more readily than the other Pu species. The only species (besides water) present for complexation is NO3- in the sample and carrier solutions. NO3- forms strong complexes with

Pu(IV), forming Pu(NO3)3+, Pu(NO3)22+, ..., Pu(NO3)62-. The predominant solution species in 3000 times [Pu], 239Pu has been analyzed in the presence of 238U via selective retention. Investigations that are

8542

Analytical Chemistry, Vol. 78, No. 24, December 15, 2006

currently underway to determine species that codeposit with Pu show relatively few examples. ACKNOWLEDGMENT This research was sponsored by the U.S. Department of Energy’s Office of Nuclear Nonproliferation Research and Development and Office of Basic Energy Sciences, under contract DEAC05-00OR22725 with Oak Ridge National Laboratory, managed and operated by UTsBattelle, LLC. The authors thank Dr. Richard Haire for constructive comments and Dr. Shane M. Peper for assistance during manuscript preparation. Received for review August 17, 2006. Accepted October 10, 2006. AC061538B