Anal. Chem. 1998, 70, 1085-1091
Separation of Chemical Warfare Agent Degradation Products by the Reversal of Electroosmotic Flow in Capillary Electrophoresis Alaa-Eldin F. Nassar* and Samuel V. Lucas
Battelle Memorial Institute, 2012 Tollgate Road, Bel Air, Maryland 21015 William R. Jones
Thermo Bioanalysis Corporation, 8 East Forge Parkway, Franklin, Massachusetts 02038 Lynn D. Hoffland
U.S. Army Material Command Treaty Laboratory, Aberdeen Proving Ground, Maryland 21010
We report the development of analyses for nerve agent degradation products or related species by the reversal of electroosmotic flow in capillary electrophoresis (CE). The developed methods were used in this laboratory for analysis of samples in the second and third official proficiency tests (International Round-Robins) for the Provisional Technical Secretariat/Preparatory Commission for the Organization for the Prohibition of Chemical Weapons, and those results are reported here. Analytes studied include methylphosphonic acid (a dibasic acid), the monoisopropyl ester of ethylphosphonic acid, and the monoalkyl esters of methylphosphonic acid (R ) ethyl, isopropyl, isobutyl, pinacolyl (3,3-dimethyl-2-butyl), cyclohexyl, and 2-ethylhexyl). The cationic surfactants used here for the reversal of electroosmotic flow are didodecyldimethylammonium hydroxide and cetyltrimethylammonium hydroxide. CE methods using conductivity or indirect UV detection provide a good separation efficiency and very high sensitivity for the analysis of such compounds. The detection limits for these species were about 75 µg/L when using conductivity detection and about 100 µg/L when using indirect UV detection. Because pH plays an important role in the CE separation of the alkylphosphonic acids and their monoesters, the influence of pH on these separation systems was investigated. Electrolytes were stable for at least 3 months. Excellent separation efficiency and freedom from interference due to common anions were obtained in the developed methods which typically achieved complete separations in less than 3 min. The method was applied to aqueous leachates of soil, wipes of surfaces, and vegetation sampled from a field site known to have been exposed to nerve agents and subsequently cleaned up. The data from these environmental samples indicated that the method can be expected to be useful for environmental monitoring. Capillary electrophoresis (CE) has generated considerable interest in recent years because of its capability for using simple S0003-2700(97)00971-2 CCC: $15.00 Published on Web 02/14/1998
© 1998 American Chemical Society
instrumentation to achieve fast and highly efficient separations while consuming minimal sample.1,2 It requires limited sample preparation and can separate analytes at relatively low concentration. CE is versatile as a separation technique for ionic and nonionic forms of organic and inorganic molecules.3-8 Alkylphosphonic acids and their monoesters are important hydrolysis products of nerve agents and related species. Determination of these compounds in aqueous and/or environmental matrixes is challenging because they have no chromophore for UV or fluorescence detection and are highly polar compounds. CE, with conductivity detection or indirect UV detection, can facilitate this analysis. Organophosphorus cholinesterase-inhibiting compounds include many familiar insecticides and the nerve agents shown in Table 1. While these nerve agents are more potent neurotoxins in man than the structurally similar pesticides, exposed humans respond much better to timely treatment and are much less likely to have residual effects or relapses in recovery from nerve agent exposure versus organophosphorus pesticide exposure.9 Scheme 1 shows the hydrolysis pathways for nerve agents. In an aqueous medium, displacement of the fluoro group on phosphorus by the hydroxide ion is the major pathway. However, other nucleophiles can displace fluoride to form stable phosphonylated products. For example, the first-order reaction of Sarin (GB) with monoethanolamine/water has been reported.10,11 The G-type organophosphonates are hydrolytically unstable, especially at high pH, (1) Landers, J. P., Ed. Handbook of Capillary Electrophoresis, 2nd ed.; CRC Press: Boca Raton, FL, 1996. (2) (a) Jorgenson, J. W.; Lukacs, K. D. Anal. Chem. 1981, 53, 1298-1302. (b) Jorgenson, J. W.; Lukacs, K. D. Science, 1983, 222, 181-189. (3) Dabey-Zlotorzynska, E.; Dlouhy, J. F. J. Chromatogr. A 1994, 671, 389395. (4) Huang, X.; Luckey, J. A.; Gordon, M. J.; Zare, R. N. Anal. Chem. 1989, 61, 766-770. (5) Oehrle, S. A. J. Chromatogr. A 1994, 671, 383-387. (6) Romano, J.; Jandik, P.; Jones, W. R.; Jackson, E. J. Chromatogr. 1991, 546, 411-421. (7) Nassar, A.-E. F.; Guarco, F. J.; Gran D. E.; Stuart, J. D.; Reuter, W. M. J. Chromatogr. Sci. 1998, 36, 19-22. (8) Lucy, C. A.; McDonald, T. L. Anal. Chem. 1995, 67, 1074-1078. (9) Sidell, F. R. J. Appl. Toxicol. 1994, 14 (2), 111-113.
Analytical Chemistry, Vol. 70, No. 6, March 15, 1998 1085
Table 1. Description of the Chemical Warfare Agents (CWAs) (Organophosphorus) compound name 2-propyl methylphosphonofluoridate
CWA abbreviation
CAS no.a
structure
Sarin (GB)
O CH3
P
O
MW
107-44-8
140
96-64-0
182
329-99-7
180
50782-69-9
267
159939-87-4
267
CH(CH3)2
F
pinacolyl methylphosphonofluoridate
Soman (GD)
O CH3
P
CH3 O
CH
F
cyclohexyl methylphosphonofluoridate
GF
C
CH3
CH3 CH3
O CH3
P O F
O-ethyl S-[2-(diisopropylamino)ethyl]methylphosphonothiolate
VX
O CH3
CH(CH3)2
P S CH2CH2
CH(CH3)2
CH3CH2O
O-isobutyl S-[2-(diethylamino)ethyl]methylphosphonothiolate
V-type
O CH3
CH2CH3
P S CH2CH2N
(CH3)2CHCH2O a
N
CH2CH3
Chemical Abstracts Service Registry Number, provided by the author.
Scheme 1. Hydrolysis Pathways for Nerve Agents
resulting in their rapid decomposition to the corresponding monoester methylphosphonic acid. Further hydrolysis leading to methylphosphonic acid (MPA) is substantially slower, so that (10) (a) Vasil’ev, I. A.; Shvyryaev, B. V.; Liberman, B. M.; Sheluchenko, V. V.; Petrunin, V. A.; Gorskii, V. G. Mendeleev Chem. J. 1996, 39 (4), 3-10. (b) Greenhalgh, R.; Heggie, R. M.; Weinberger, M. A. Can. J. Chem. 1970, 48, 1351-1357. (11) Nassar, A.-E. F.; Lucas, S. V.; Myler, C. A.; Campisano, M.; Hoffland, L. D., submitted, 1997.
1086 Analytical Chemistry, Vol. 70, No. 6, March 15, 1998
the half ester/monoacid would be the prevalent species in lightly weathered environmental samples. These hydrolysis products are different from those of organophosphorus pesticides.12,13 In nerve agents; the methyl group is bonded to a phosphorus atom, leading to alkyl methylphosphonates. Pesticides are mainly thiophosphates, with hydrolysis leading to dialkyl thiophosphoric acid derivatives.12 Sensitive and fast analytical methods with high resolution can be a significant asset in developing or studying chemical-based systems for the destruction of chemical warfare nerve agents. The method reported here analyzes for the degradation products, which are less toxic than the nerve agents and have structures similar to their corresponding parent agent. Thus, this method offers a convenient and safe route to monitoring the progress of destruction. Also, sensitivity, speed, and high resolution are valuable assets for analysis of these degradation products in agent destruction trails. Gas chromatography/mass spectrometry (GC/MS) and HPLC with fluorescence detection are the most widely used analytical techniques for analysis of polar molecules after derivatization.14-16 However, preparation of the sample can be elaborate and timeconsuming. Microcolumn liquid chromatography and CE with flame photometric detection for the determination of a series of organophosphoric and organophosphonic acids in environmental samples have been reported.17-20 Recently, it has been reported that ion chromatography (IC) can be used for the analysis of the acids shown in Table 2.21-23 Through the use of cleanup steps to (12) Wills, E. R. J.; Hulst, A. G. J. Chromatogr. 1988, 454, 261-272. (13) Kingery, A. F.; Allen, H. E. Toxicol. Environ. Chem. 1995, 47, 155-184. (14) Tornes, J. A.; Johnsen. B. A. J. Chromatogr. 1989, 467, 129-138. (15) Shih, M. L.; Smith, J. R.; McMonagle, J. D.; Dolzine, T. W.; Gresham, V. C. Biol. Mass Spectrom. 1991, 20, 717-723. (16) Roach, M. C.; Ungar, L. W.; Zare, R. N.; Reimer, L. M.; Pompliano D. L.; Frost, J. W. Anal. Chem. 1987, 59, 1056-1059. (17) Purdon, J. G.; Pagotto, J. P.; Miller, R. K. J. Chromatogr. 1989, 475, 261272. (18) Kientz, C. E.; Verweij, A.; de Jong, G. J.; Brinkman, U. A. T. J. Microcolumn Sep. 1992, 4, 465-475.
Table 2. Hydrolysis Products of Nerve Agents and Related Compounds compound
a
parent nerve agent
MW
(a) Compounds from Known Nerve Agents methylphosphonic acid, MPA (2)a Sarin, Soman, VX, GF ethyl methylphosphonate, EMPA (3)a VX isopropyl methylphosphonate, IMPA (4)a Sarin pinacolyl methylphosphonate, PMPA (8)a Soman a isobutyl methylphosphonate, i-BuMPA (5) V-type cyclohexyl methylphosonate, CHMPA (7)a GF
96 124 138 180 152 178
(b) Related Compounds O-isopropylethylphosphonic acid, i-PEPA (6)a O-(2-ethylhexyl)methylphosonic acid, EHMPA (1)a ethylphosphonic acid, EPA isopropylphosphonic acid, i-PPA
153 208 110 124
The number corresponds to the peak in the electropherograms in Figures 1-3.
reduce chloride and a preconcentration anion-exchange injection setup, IC detection limits in the lower microgram-per-liter or submicrogram-per-liter range and detectability in field samples (groundwater/soil leachate) in the single-digit to low two-digit microgram-per-liter range have been reported.21 However, this approach involves a rather demanding procedure and appears vulnerable to sample matrix issues when pushed to these limits. The advantages of CE include short analysis time, high selectivity, less sample consumption, and less preparation times. Recent studies using CE analysis of alkylphosphonic acids and their monoesters have been successful;24-29 however, problems encountered include common ion interference, such as carbonate, chloride, and fluoride;28 buffer instability; derivitization for direct UV;27 and analysis times of up to 15 min.29a Our specific interest in this work derives from the desire to develop CE methods for the Treaty Laboratory samples, to eliminate the undesirable derivatization procedure necessary for analysis by GC/MS. Also, the complexity of the sample matrix has proven to be problematic for analysis by LC/MS. Herein, we report that the alkylphosphonic acids and their monoesters shown in Table 2 can be separated within 3 min by reversing the electroosmotic flow. This methodology has been successfully applied to two Provisional Technical Secretariat/ Preparatory Commission for the Organization for the Prohibition of Chemical Weapons (PTS/OPCW) Round-Robin samples. Com(19) (a) Kientz, C. E.; Verweij, A.; de Jong, G. J.; Brinkman, U. A. T. J. Microcolumn Sep. 1992, 4, 477-483. (b) Kientz, C. E.; Hooijschuur, E. W. J.; Brinkman, U. A. T. J. Microcolumn Sep. 1997, 9, 253-259. (20) de Griend C. E. S.; Kientz, C. E.; Brinkman, U. A. T. J. Chromatogr. A 1994, 673, 299-302. (21) Kingery, A. F.; Allen, H. E. Anal. Chem. 1994, 66, 155-159. (22) Schiff, L. J.; Pleva, S. G.; Sarver, E. W. In Ion Chromatographic Analysis of Environmental Pollutants; Mulick, J. D., Sawicki, E., Ed.; Ann Arbor Science: Ann Arbor, MI, 1979; Vol. 2, pp 329-344. (23) Bossle, P. C.; Reutter, D. J.; Sarver, E. W. J. Chromatogr. 1987, 407, 400404. (24) Kostiainen, R.; Bruins, A. P.; Hakkinen, V. M. A. J. Chromatogr. A 1993, 634, 113-118. (25) Cheicante, R. L.; Stuff, J. R.; Durst, D. H. J. Cap. Electrophor. 1995, 2, 157163. (26) Cheicante, R. L.; Stuff, J. R.; Durst, D. H. J. Chromatogr. A 1995, 711, 347352. (27) Robin, W. H.; Wright, B. W. J. Chromatogr. A 1994, 680, 667-673. (28) Oehrle, S. A.; Bossle, P. C. J. Chromatogr. A 1995, 692, 247-252. (29) (a) Mercier, J.-P.; Morin, P.; Dreux, M.; Tambute, A. J. Chromatogr. A 1996, 741, 279-285. (b) Pianetti, G. A.; Taverna, M.; Baillet, Aa.; Mahuzier, G.; Baylocq-Ferrier, D. J. Chromatogr. 1993, 630, 371-377.
pounds identified were ethylphosphonic acid, O-isopropylethylphosphonic acid, isopropylphosphonic acid, and O-(2-ethylhexyl)methylphosphonic acid. Electrolytes were stable for at least 3 months, and no interferences from other anions were found. EXPERIMENTAL SECTION Apparatus. All experiments were performed on either a Thermo CE (now Thermo Bioanalysis) Crystal 310 capillary electrophoresis system with a Crystal 1000 conductivity detector using a 60-cm × 50-µm (i.d.) ConCapI fused-silica capillary (Santa Fe, NM) or a Hewlett-Packard 3D capillary electrophoresis system with UV detector using a 75-µm (i.d.) × 56-cm (effective length) fused-silica capillary. Both instruments have the capability to control the temperature of the fused-silica capillary. The electropherogram signals were recorded using 4880 software for the conductivity detector and HP Chem-Station software for indirect UV. Reagents. Alkylphosphonic acids and their monoesters were synthesized in our laboratory (U.S. Army Material Command Treaty Laboratory, Aberdeen Proving Ground, MD). All reagent solutions were prepared using deionized water (Winokur Water System Corp., Barnstead, Dubuque, IA): didodecyldimethylammonium bromide (DDAB) (>99%) was from Eastman Kodak (Rochester, NY); cetyltrimethylammonium bromide (CTAB) was from Aldrich (Milwaukee, WI); isooctylphenoxypolyethoxyethanol (Triton X-100) is a registered tradename of Union Carbide; and L-histidine (L-R-amino-β-imidazolylpropionic acid, His), 2-[N-morpholino]ethanesulfonic acid, (MES), phenylphosphonic acid, and boric acid were from Sigma Chemical Co. (St. Louis, MO). Solutions were stored in Nalgene plastic bottles. All buffers were degassed and filtered through a 0.45-µm cellulose nitrate membrane filter prior to use. All other chemicals were reagent grade. Didodecyldimethylammonium hydroxide (DDAOH) and cetyltrimethylammonium hydroxide (CTAOH) were obtained from their respective bromide salts; 25 mM concentrations of each bromide salt were passed through a hydroxide form Dionex OnGuard A cartridge, (Dionex Corp., Sunnyvale, CA). This conversion step was necessary to eliminate a large system peak found in the opposite direction from where bromide (the analyte) is visualized in the electropherogram.1 Procedures. New capillaries were pretreated with a 10-min rinse of deionized water, followed by 10 min of 0.5 M sodium hydroxide, and then 10 min of deionized water, and finally a 10Analytical Chemistry, Vol. 70, No. 6, March 15, 1998
1087
min rinse of the analysis buffer. Rinses were performed at 2000 mbar for the conductivity detection method and at 900 mbar for the indirect UV detection method. Samples and standards were introduced by keeping the pressure times time product at 300 mbar-s and were the same for both detection methods. Constant voltage (negative polarity, detector side anodic) was used throughout to drive the separations, -25 and -30 kV for the conductivity and UV methods, respectively. The wavelength used for indirect UV detection was 210 nm. The capillary temperature was at ambient for the conductivity method, while the indirect UV method was optimal at 40 °C. The two electrolytes used for conductivity detection both contained 30 mM His, 30 mM MES, and 0.03 wt % Triton X-100 and differed only in the electroosmotic flow (EOF) modifier. For buffer 1, the EOF modifier was 0.35 mM DDAOH, and for buffer 2, it was 0.35 mM CTAOH. Two electrolytes used for indirect UV detection both contained 200 mM boric acid, 10 mM phenylphosphonic acid, and 0.03 wt % Triton X-100 and differed again by the EOF modifier [0.35 mM DDAOH (buffer 3) and 0.35 mM CTAOH (buffer 4)]. A third indirect UV electrolyte (buffer 5) contained the same amount of boric and phenylphosphonic acids as buffers 3 and 4 but did not contain Triton X-100 or any EOF modifiers. Sodium hydroxide at 0.2 N was used to adjust the pH of the indirect UV electrolytes. Field Sample Preparation. Environmental samples consisting of soil, cotton wipes of painted surfaces, and vegetation were taken from a military area known to have been exposed to nerve agents and subsequently cleaned up. Soil (5 g) was sonicated for 5 min with 10 mL of deionized water and then centrifuged for 15 min, and the supernatant was removed. This process was repeated twice more, and the combined aqueous leachate was filtered through 0.45-µm membrane filters. The surface wipes and vegetation samples were similarly prepared, but centrifugation was omitted. RESULTS AND DISCUSSION Cationic Surfactants and Control of Electroosmotic Flow. Interactions between surfactants and the surface of the fused silica CE capillary are essential for control of electroosmotic flow (EOF). Surfactants can play important roles in separations at concentrations below the critical micelle concentration (cmc) by acting as solubilizing agents for hydrophobic solutes or by functioning as ion-pairing reagents or as capillary surface modifiers. The interaction of a surfactant with the solutes can occur by two mechanisms: ionic interactions with the charged end of the surfactant or hydrophobic interactions between the alkyl chain and hydrophobic moieties of the solute. Chart 1 shows two amphiphilic lipid molecules with cationic headgroups: DDAOH with a double chain and CTAOH with a single chain.30 Cationic surfactants such as DDAOH or CTAOH can form a double layer on the surface of the fused silica and reverse the capillary wall charge, thereby reversing the direction of the EOF. (30) For some recent studies, please see: (a) Nassar, A.-E. F.; Bobbitt, J. M.; Stuart, J. D.; Rusling, J. F. J. Am. Chem. Soc. 1995, 117, 10986-10993. (b) Nassar, A.-E. F.; Narikiyo, Y.; Sagara, T.; Nakashima, N.; Rusling, J. F. J. Chem. Soc., Faraday Trans. 1995, 91 (21), 1775-1782. (c) Nassar, A.-E. F.; Zhang, Z.; Hu, N.; Rusling, J. F.; Kumosinski, T. F. J. Phys. Chem. B 1997, 101, 2224-2231. (d) Nassar, A.-E. F.; Willis, W.; Rusling, J. F. Anal. Chem., 1995, 67, 2386-2392.
1088 Analytical Chemistry, Vol. 70, No. 6, March 15, 1998
Figure 1. Separation of the alkylphosphonic acids. Experimental conditions: buffer, 200 mM borate, 10 mM phenylphosphonic acid; 75-µm (i.d.) × 56-cm fused-silica capillary; peaks were detected by indirect UV detection at 210 nm; injection, 6 s at 50 mbar; and temperature, 40 °C. (A) Without cationic surfactant, pH 6.0; voltage, 30 kV (buffer 5). (B) With cationic surfactant (0.35 mM DDAOH) and 0.03 wt % Triton X-100, pH 4.0; voltage, 30 kV (negative polarity) (buffer 3). Anions: 1 ) EHMPA, 2 ) MPA, 3 ) EMPA, 4 ) IMPA, 5 ) i-BuMPA, 6 ) i-PEPA, 7 ) CHMPA, and 8 ) PMPA.
Chart 1. Structure of Cationic Surfactants Used to Control EOF in Buffer
There are many ways to control the EOF, such as varying the polarity of the voltage applied to the capillary, derivatizing the inner surface of the capillary, changing the buffer pH or concentration, adding organic modifiers to the buffer, or adding cationic surfactants.31-33 Figure 1 shows a comparison between separations of the alkylphosphonic acids with and without cationic surfactant. The electropherogram in Figure 1A shows that the analysis time using buffer 5, without surfactants, was 8 min. We determined that pH 6.0 produced the optimal separation efficiency (31) Hayes, M. A.; Kheterpal, I.; Ewing, A. G. Anal. Chem. 1993, 65, 26-31. (32) Lucy, C. A.; Underhill, R. S. Anal. Chem. 1996, 68, 300-305. (33) Schwer, C.; Kenndler, E. Anal. Chem. 1991, 63, 1801-1807.
Figure 2. Comparison of the electropherograms of the alkylphosphonic acids. Experimental conditions: buffer, 200 mM borate, 10 mM phenylphosphonic acid, 0.03 wt % Triton X-100, and 75-µm (i.d.) × 56-cm fused-silica capillary; peaks were detected by indirect UV detection at 210 nm; injection, 6 s at 50 mbar; voltage, 30 kV (negative polarity); and temperature, 40 °C. (A) 0.35 mM DDAOH at pH 4.0 (buffer 3). (B) 0.35 mM CTAOH at pH 4.5 (buffer 4). Anions: 1 ) EHMPA, 2 ) MPA, 3 ) EMPA, 4 ) IMPA, 5 ) i-BuMPA, 6 ) i-PEPA, 7 ) CHMPA, and 8 ) PMPA.
for buffer 5. Comparison of parts A and B of Figure 1 also show the improved resolution when the cationic surfactant DDAOH is used, as well as the reduction in analysis time. As shown in Figure 2, while the migration time is shorter using CTAOH, the resolution is better with DDAOH. These results show that, with respect to resolution and signal-to-noise ratio, the use of DDAOH is preferable. DDAOH is particularly suited for such applications since its bilayers are in the liquid crystal phase at room temperature.30 Effect of pH on Separation. One of the major aims of this work was to study the use of various electrolytes for separation of the alkylphosphonic acids while optimizing each for pH. The five indirect UV and conductivity electrolytes described earlier in the Experimental Section were investigated at a pH range from 3.0 to 7.5. In CE separation of the alkylphosphonic acids, the change in pH can affect both solute charge and EOF, thus influencing resolution. The negatively charged wall causes adsorption of cationic solutes through Coulombic interactions. The double-layer effect has been established theoretically34,35 and experimentally31 from electrochemistry and colloid and surface chemistry. Over the pH range from 3.0 to 7.5, strong influences on the separation of the alkylphosphonic acids were observed when using conductivity or indirect UV detection. Figure 3 compares the separation of the alkylphosphonic acids using DDAOH at pH 4.0, 4.5, and 5.0, with pH 4.0 providing the greatest resolution. The optimally efficient pHs for the five buffers used were 5.5, buffer 1; 6.0, buffer 2; 4.0, buffer 3; 4.5, buffer 4; and 6.0, buffer 5. Detector Response for Alkylphosphonic Acids and Their Monoesters. Studies of the detection sensitivity were performed by conductivity and indirect UV detection. The conductivity detection showed no significant difference in the response from that of indirect UV. The average detection limit is 75 µg/L for the conductivity detection and 100 µg/L for indirect UV detection. The most striking observation is that analyte response reproduc(34) Bard, A. J.; Faulkner, L. R. Electrochemistry Methods: Fundamentals and Applications; John Wiley & Sons: New York, 1980. (35) Hiemenz, P. C., Rajagopalan, R., Ed. In Principles of Colloid and Surface Chemistry, 3rd ed.; Marcel Dekker: New York, 1997.
Figure 3. Effect of pH on the resolution of the alkylphosphonic acids using 0.35 mM DDAOH. (a) pH 5.0; (b) pH 4.5; and (c) pH 4.0. Anions: 1 ) EHMPA, 2 ) MPA, 3 ) EMPA, 4 ) IMPA, 5 ) i-BuMPA, 6 ) i-PEPA, 7 ) CHMPA, and 8 ) PMPA. Experimental conditions were the same as in Figure 2.
ibility over a 3-month period was very satisfactory when using a surfactant in the buffer. A report which systematically determined the accuracy, precision, linearity, specificity/selectivity, limit of quantitation, limit of detection, ruggedness, recovery, and solution stability for this method has been submitted for publication.11 Second Official PTS/OPCW Proficiency Test Analysis (April 1997). The water sample, along with its corresponding blank and standard addition spikes, was analyzed by CE, as shown in Figure 4. The water sample gave two major peaks and one minor peak that were not found in the corresponding blank. The first major peak had CE mobility similar to that of ethylphosphonic acid (EPA), the second major peak had mobility similar to that of O-isopropylethylphosphonic acid (i-PEPA), and the third (minor) peak in the water sample between EPA and i-PEPA was not positively identified but had CE mobility similar to that of isopropylphosphonic acid (i-PPA). A possible explanation for the minor peak i-PPA is that it is an impurity in the i-PEPA used to prepare the Round-Robin samples, since it was also found in the i-PEPA standard used in our laboratory (Figure 4). These CE analysis results were confirmed by derivatization followed by flow injection API/MS/MS, GC/MS/MS, and GC/AED.36 Third Official PTS/OPCW Proficiency Test Analysis (October 1997). Figure 5 shows a water sample along with the blank as analyzed using CE. The upper trace is the water blank, where no peak was detected between 4 and 6 min. The lower trace is the unknown water sample, where two major peaks were found at 4.89 and 5.60 min. Standard addition experiments showed that the first peak had the same mobility as isopropylphosphonic acid (i-PPA), and the second peak had the same mobility as O-(2-ethylhexyl)methylphosphonic acid (EHMPA). The standard addition data also provided quantitation results for i-PPA and EHMPA, 8.8 and 6.7 µg/mL, respectively. It is important to note that this sample contained carbonate and an amine, but when using buffer 5 no interference was observed. Also, we observed that O-(2-ethylhexyl)methylphosphonic acid forms an ion pair with cationic surfactants in this matrix, which contains an amine, so it was necessary to use a surfactant-free buffer (buffer 5). These CE identifications were confirmed by derivatization followed by flow injection API/MS/MS, GC/MS/ (36) U.S. Army Material Command Treaty Laboratory, Aberdeen Proving Ground, MD, 1997, unpublished results.
Analytical Chemistry, Vol. 70, No. 6, March 15, 1998
1089
Figure 4. Analysis of aqueous sample from second official PTS/ OPCW proficiency test. Conductivity detection conditions: sample was introduced by pressure into the capillary for 0.2 min at 25 mbar and a constant potential of 25 kV (negative polarity), detector side anodic. Electrolytes were 30 mM His, 30 mM MES, 0.03 wt % Triton X-100, and 0.35 mM CTAOH; and temperature was 25 °C (buffer 2).
Figure 6. Analysis of phosphonic compounds extracted from soil and surface wipe by CE. (Experimental conditions as Figure 1B.)
Figure 5. Analysis of the aqueous sample and the blank from the third official PTS/OPCW proficiency test. Experimental conditions: buffer, 200 mM borate, 10 mM phenylphosphonic acid, pH 6.0; 75µm (i.d.) × 56-cm fused-silica capillary; peaks were detected by indirect UV detection at 210 nm; injection, 6 s at 50 mbar; temperature, 40 °C; and voltage, 30 kV (buffer 5).
MS, and GC/AED.36 The i-PPA result was also confirmed by LC/ MS of the free acid.36 Application of Environmental Samples. The soil, surface wipe, and vegetation samples were collected from a military area, and aqueous leachates of these samples were analyzed by CE. The aqueous samples extracted from vegetation leachates were analyzed by CE/conductivity and no significant interference due to the matrix was observed. The soil and surface wipe leachates were analyzed by CE/UV detection, and Figure 6 shows the detection of IMPA, PMPA, and MPA in them. The concentarions of IMPA, PMPA, and MPA in these leachates are 27.4, 24.2, and 2.5 µg/mL (soil) and 8.8, 9.9, and 0.3 µg/mL (surface wipe), respectively. These leachate levels equate to 0.16, 0.14, and 0.01 mg/g of soil for IMPA, PMPA, and MPA, respectively, and 2.2, 2.5, and 0.08 µg/cm2 of surface wipes, respectively, as before. The standard addition spikes were analyzed by CE, and these CE analysis results were confirmed by derivatization, followed by GC/ MS as well as GC/AED. None of the analytes were detected in the vegetation leachates (but no interferances were found in those samples, either), and the corresponding detection limits were estimated at 0.11, 0.12, and 0.10 µg/g vegetation for IMPA, PMPA, amd MPA, respectively. Although these data for environmental samples are limited in scope, we believe the results demonstrate 1090 Analytical Chemistry, Vol. 70, No. 6, March 15, 1998
the potential applicability of our CE method to environmental samples. CONCLUSIONS In summary, the reliable CE methods described in this paper for the separation of alkylphosphonic acids and their monoesters are improved with the addition of cationic surfactants for EOF reversal. The co-electroosmotic conditions reduced the analysis time and eliminated the baseline anomalies observed when the EOF modifier was absent. Optimum resolution of the key analytes was obtained using acidic electrolyte pHs, where the actual pH required was dependent upon the type of buffer used. Using acidic pH electrolytes eliminated two common interferences. The first interference, fluoride, is reactive in the acid form and is not observed in the separation due to interactions with the inner wall of the capillary. The second interference is the weakly acidic carbonate anion, which is also not observed in the separation, as it has a pKa1 value of 6.3. The acidic buffer brings all ionized forms of carbonate in the sample to a partially ionized form, which lowers both mobility and signal response. Acidic electrolytes also increase the shelf life of the buffer, as it does not absorb atmospheric CO2 (which drops the pH) as readily as alkaline pH electrolytes. The shelf life of the buffers for both detection methods is 3 months, and this level of storage stability would be a great benefit in many other CE applications. The derivatization steps required for direct UV visualization of the nonchromophoric analytes are eliminated for both indirect UV and conductivity detection, as the analytes are detected as differences between the measured properties of the electrolyte co-ion and the analyte itself. The results from the second and third official PTS/OPCW proficiency tests indicate that the
standard additions in this method give an accurate analytical framework. Although application of the method to environmental samples reported in this paper is limited, the results reported here are encouraging with regard both to usable sensitivity of the method and ability to reject environmental background matrix with no sample pretreatment requirements. Thus, the method should be useful for the environmental detection screening of species which might indicate the past use or manufacture of G- or V-type nerve agents.
ACKNOWLEDGMENT We thank the U.S. Army Material Command Treaty Laboratory, Aberdeen Proving Ground, MD, for their support. The authors express sincere appreciation to R. Calloway, A. Emery, T. Adams, L. Slivon, and G. Bowen for their technical assistance and many helpful suggestions. Received for review September 4, 1997. January 2, 1998.
Accepted
AC9709716
Analytical Chemistry, Vol. 70, No. 6, March 15, 1998
1091
