Aqueous in Situ Derivatization of Carboxylic Acids by an Ionic

Journal of Chromatography A, 1145 (2007) 241–245

Short communication Aqueous in situ derivatization of carboxylic acids by an ionic carbodiimide and 2,2,2-triﬂuoroethylamine for electron-capture detection Quincy LaRon Ford, Justina Marie Burns, John Lee Ferry ∗ Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC 29208, USA Received 28 November 2006; received in revised form 18 January 2007; accepted 23 January 2007 Available online 1 February 2007

Abstract We report a technique for the rapid, room temperature derivatization of aqueous carboxylic acids to the corresponding 2,2,2-trifluoroethylamide derivative. 3-Ethyl-1-[3-(dimethylamino)propyl]carbodiimide hydrochloride (EDC) and 2,2,2-trifluoroethylamine hydrochloride (TFEA) were added to aqueous samples of several acids of interest in environmental analytical chemistry, including benzoic acid, ibuprofen, clofibric acid, monochloroacetic acid, dicholoroacetic acid, bromoacetic acid, monochlorophenoxy acetic acid, dichlorophenoxy acetic acid, trichlorophenoxy acetic acid, acetylsalicylic acid, and cholorobenzoic acid. Amidization was essentially complete within ten minutes, and subsequent liquid–liquid extraction of the amides with methyl tert-butyl ether (MTBE) demonstrated recoveries of over 85%. The starting materials, both quaternary ammonium salts, were not co-extracted with the derivative, yielding much cleaner samples than historically obtained from carbodiimide based techniques. The fluorinated amides produced had excellent chromatographic characteristics for gas chromatography and were easily detected by electron-capture detection (ECD) or electron impact mass spectrometry. This method is suggested as a sensitive alternative to more traditional acidification, extraction, and ex situ derivatization techniques. © 2007 Elsevier B.V. All rights reserved.

Keywords: Aqueous carboxylic acids; 3-Ethyl-1-[3-(dimethylamino)propyl]carbodiimide hydrochloride (EDC); 2,2,2-Triﬂuoroethylamine (TFEA); Aqueous in situ derivatization; Fluorinated amide

1. Introduction of the analyte. For environmental samples, the most common derivatization technique is methylation, using highly reactive Trace organic analysis has always been challenged by polar reagents such as diazomethane, BF3/methanol, or more recently carboxylic acids in aqueous matrices. Typical extraction/pre- methanol/HCl or H2SO4 [12–23]. These methods have been concentration procedures rely on similar values of the applied to simple carboxylic acids as well as polyfunctional octanol–water partitioning coefficient (Kow) between the aque- acids (including aldo and keto acids, haloacids, aromatic acids, ous analyte and the extracting solvent [1–3]. Carboxylic acids etc.) generated by a wide manifold of oxidative processes, typically have much lower Kow values than most other organ- both natural and technological [24–33]. Silylation reagents ics [4–8]. It is occasionally possible to increase the Kow to a are also applied to the analysis of acids in environmental level where extraction is possible by protonating the carboxy- samples, although they sometimes yield complicated mass late group through acidification [5–7,9–11]. However, the polar spectra and are also reactive toward alcohols and phenols nature of the carboxylate substituent often makes chromato- [34,35]. Several extensive reviews on carboxylic acid analysis graphic resolution in a complex matrix problematic, especially have been published that address derivatizations suitable for when dealing with small carboxylic acids that may co-elute with gas chromatographic (GC) and LC-based analytical techniques salts during liquid chromatography (LC). [36–40]. These difficulties are typically addressed post-extraction by In recent years, carbodiimides have seen considerable use as employing derivatization strategies to increase the volatility dehydrating agents for coupling carboxylic acids with amines to produce amide derivatives [41–46]. The technique is a traditional synthetic approach for generating amides, and its use ∗ Corresponding author. Tel.: +1 803 777 2646; fax: +1 803 777 9521. in the environmental field has been somewhat limited due to E-mail address: [email protected] (J.L. Ferry). the exacting conditions required for its application (typically the

Fig. 1. The derivatization of carboxylic acids by carbodiimides and amines is a multistep process based on the reaction of an acylisourea (generated by the reaction of the carbodiimide and carboxylic acid) with the amine (shown for EDC and TFEA for illustrative purposes) [51]. analyte, carbodiimide, and amine are reacted in organic solvents) group of environmentally relevant acids to their corresponding [43]. Two notable in situ applications are for the general deriva- 2,2,2-trifluoroethylamides, using a water-soluble carbodiimide tization of carboxylic acids by 2,2,3,3,4,4,5,5-octafluoropentyl for derivatizing carboxylic acids in the aqueous phase. The chloroformate and the derivatization of haloacetic acids in nat- derivatives are fluorinated amides with excellent chromato- ural waters using 2,4-difluroaniline [41,42,47]. However, these graphic capabilities that are easily detected by electron-capture methods require either synthesis of unique starting materials (i.e. techniques and also yield significant molecular ions during mass octafluorpentyl chloroformate) and/or application of hazardous spectrometric analysis (Fig. 2). reagents [41,42,47]. The descriptive mechanism for carbodiimide-mediated 2. Experimental derivatization is well known (Fig. 1) [48]. A given carboxylate and carbodiimide react to form an intermediate complex that is 2.1. Materials activated to attack by an electrophile such as an amine [40,49]. This attack displaces the carbodiimide to generate the corre- All reagents were used as received and made fresh. sponding amide. In this work, we report the derivatization of a 1-[3-(Dimethylamino)propyl]-3-ethylcarbodimide hydrochlo-

Fig. 2. The mass spectrum of N-2,2,2-trifluoroethylbenzamide (benzoic acid derivative) with significant peaks (GC-ITMS, unit resolution). Q.L. Ford et al. / J. Chromatogr. A 1145 (2007) 241–245 243 ride salt (EDC), 2,2,2-trifluoroethylamine hydrochloride salt 3. Results and discussion (TFEA), monochlororacetic acid (99%), dichloroacetic acid (99%), sodium hexanoate, sodium octanoate, sodium decanoate, 3.1. Method optimization sodium dodecanoate, and 1,4-dibromobenzene (DBB) were purchased from Aldrich (St. Louis, MO, USA); benzoic It was found that the extent of derivatization was pH and acid from Fisher (Waltham, MA, USA); sodium phos- time dependent. Optimal reaction parameters were determined phate dibasic from Mallinckrodt (Phillipsburg, NJ, USA); by holding the acid concentration constant while pH (1–9) and and methyl tert-butyl ether (MTBE) from EM Science reaction time (0–10 min) were varied (Fig. 3). After quenching (Lawrence, KA, USA). ␣-Methyl-4-[isobutyl]phenylacetic by liquid-liquid extraction, the organic phase was immediately acid (sodium salt) (ibuprofen-99%), 2-(p-chlorophenoxy)-2- removed. Generally under optimal conditions (pH 5 and a reac- methylpropionic acid (clofibric acid-98%), p-chlorophenoxy tion time of 10 min), the derivatization yield was approximately acetic acid (99%), 2,4-dichlorophenoxyacetic acid (98%), 2,4,5- 85% for most acids. Yields were determined by comparison trichlorophenoxyacetic acid (98%) were obtained from Sigma of derivatized, extracted samples with synthesized standards (St. Louis, MO, USA). (purity assayed by NMR and direct probe introduction HRMS). Dichloroacetic acid deviated from this profile with an optimal 2.2. Derivatization derivatization of pH 4.0. Successful derivatization required pro- tonation of the intermediate; speculatively, the strongly electron A 1.0-mL volume of aqueous pH 5.0 dibasic phosphate withdrawing character of the dichloromethyl group makes this buffer (0.5 M), 1.0 mL of aqueous EDC (0.4 M), and 1.0 mL more difficult and required a lower pH (Fig. 1). of aqueous TFEA (0.4 M) were added to 10.0 mL of a given The reproducibility of the method was determined for three aqueous sample containing the organic acid(s). The resulting analytes (monochloroacetic acid, p-chlorophenoxyacetic acid, solution was mixed for approximately 1 min and then allowed to benzoic acid) representative of the three acid types in this react for 10 min at room temperature. The mixture was extracted study. Relative standard errors ranged from ∼0.009 to 0.03 with 2 mL of MTBE containing 0.85 ␮M 1,4-dibromobenzene as an external standard and immediately analyzed.

2.3. Standards

Pure standards (98%) of the above amides were prepared for high-resolution mass spectrometry (HRMS) under the following conditions: equal parts of aqueous pH 5.0 phosphate buffer, 0.4 M aqueous EDC, 0.4 M aqueous TFEA, and 0.1 M aqueous carboxylic acid were combined (25 ◦C) and extracted with of MTBE after 10 min. The MTBE was removed by rotary evaporation and the resulting dry crys- tals or oils were analyzed by a VG 70S Magnetic Sector high-resolution mass spectrometry system with direct probe introduction.

2.4. Derivative analysis

Instrumental detection limits were determined on a Hewlett-Packard 5890 gas chromatograph equipped with an electron-capture detection (ECD) system. Derivatization was qualified with a Varian 3800 gas chromatograph equipped with a Saturn 2000 ion trap mass spectrometry (ITMS; Supplementary Fig. 2, S1–S10) system. Identical column (30 m × 0.25 mm I.D., 0.25 ␮m film thickness DB-5 MS-grade) and flow parameters (He carrier gas, 1.3 mL/min) were used for both analyses. The injector port was operated in splitless mode at 250 ◦C. Haloacetic acid derivative extracts were chromatographed using the following conditions: the column was held isothermally at 50 ◦C for 2 min, heated at 5 ◦C/min to 100 ◦C and held for 1 min, followed by heating to 200 ◦Cat10◦C/min. Other, less Fig. 3. (a) The derivatization of benzoic acid is essentially complete (∼80%) volatile acids were chromatographed with an initial column within 5 min. (b) The yield of the corresponding amide was determined as a ◦ ◦ function of pH, with a maximum sample recovery at pH 5. Optimization exper- temperature of 100 C followed by heating to 290 C at a rate of iments were conducted at 0.4 M TFEA, 0.4 M EDC, 100 ␮M carboxylic acid, ◦ 15 C/min. and 25 ◦C. 244 Q.L. Ford et al. / J. Chromatogr. A 1145 (2007) 241–245

Table 1 Carboxylic acids in this study were found to have detection limits in the micromolar range IUPAC target acid name Corresponding triﬂuoroethyl amide Exact mass of GC-ECD detection GC-ECD detection limit, Standard (common name) derivativea limit, ␮M (qualitative) ␮M (quantitative) error

Benzoic acid N-2,2,2-Trifluoroethylbenzamide 203.0558 10.48 67.34 0.0045 2-(4-Isobutylphenyl) N-2,2,2-Trifluoroethyl-2-(4-(2- 287.1497 45.77 74.76 0.0115 propanoic acid methylpropyl)phenyl)propanamide hydrochloride (ibuprofen-HCl) 2-(4-Chlorophenoxy)-2- 2-(4-Chlorophenoxy)-N-2,2,2 295.0587 19.48 84.65 0.0109 methylpropanoic acid trifluoroethyl-2-methyl propanamide (clofibric acid) 2,2-Dichloroacetic acid 2,2-Dichloro-N-2,2,2- 208.9962 24.05 110.25 0.0086 (dichloroacetic acid) trifluoroethylethanamide 2-Chloroacetic acid 2-Chloro-N-2,2,2-trifluoroethylethanamide 175.0012 39.99 150.79 0.0104 (monochloracetic acid) 2-Bromoacetic acid 2-Bromo-N-2,2,2-trifluoroethylethanamide 218.9507 60.64 184.46 0.0177 (bromoacetic acid) 2-(2,4,5-Trichlorophenoxy) N-2,2,2-Trifluoroethyl-2-(2,4,5- 334.9494 21.68 76.69 0.0089 acetic acid trichlorophenoxy)ethanamide 2-(2,4-Dichlorophenoxy) 2-(2,4-Dichlorophenoxy)-N-2,2,2- 300.9884 25.89 85.24 0.0087 acetic acid trifluoroethylethanamide 2-(4-Chlorophenoxy)acetic 2-(4-Chlorophenoxy)-N-2,2,2- 267.0274 22.77 94.26 0.0092 acid trifluoroethylethanamide 2-Acetoxybenzoic acid 2-(2,2,2-Trifluoroethylcarbamoyl)phenyl 261.0613 44.56 102.26 0.0090 (acetylsalicylic acid) acetate a High-resolution mass spectrometry performed on aVG 70S Magnetic Sector high-resolution mass spectrometer (HRMS) with direct probe introduction. across the test set (n = 4 for all samples). Monochloroacetic acid Acknowledgements (334.7 ␮M) was recovered with a relative standard deviation of 0.048; p-chlorophenoxyacetic acid (202.1 ␮M) was recovered The authors are very grateful to Dr. Mike Walla, Dr. Scott with a relative standard deviation of 0.028; and benzoic acid Reese, Dr. Wayne Fai Chan, and Dr. Bill Cotham for helpful (404.1 ␮M) was recovered with a relative standard deviation of discussion and assistance with mass spectrometry. This work 0.009. Method detection limits were determined using the guide- was supported by the Carolina Venture Fund and the US Envi- lines for data analysis in US Environmental Protection Agency ronmental Protection Agency Grant RD83-1042. (EPA) method 552 (quantification limit must have a value of at least 3σ) [50] (Table 1). Appendix A. Supplementary data

3.2. Peak determination Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chroma.2007.01.096. GC-ECD peaks were identified using a ratio of the relative retention times found on the GC-ITMS (verified by comparison of the mass spectra with those from high-resolution MS References experiments) to the relative retention time found on the GC- [1] A.J. Nunez, L.F. Gonzalez, J. Janak, J. Chromatogr. 300 (1984) 127. ECD when identical chromatographic conditions were used. [2] E.C. Price, Anal. Meas. Instrum. Process Pollut. Control (1981) 119. This comparison enabled the positive identification of all deriva- [3] H.S. Hertz, W.E. May, S.A. Wise, S.N. Chesler, Anal. Chem. 50 (1978) tives in subsequent GC-ECD analyses (Supplementary Table 428A. S-1). [4] EPI Suite US Environmental Protection Agency, Syracuse Research Cor- poration, 2000. [5] W.J. Dunn, J.H. Block, R.S. Pearlman, Partition Coefficient: Determination 4. Conclusions and Estimation, Pergamon Press, New York, 1986. [6] S.W. Karickhoff, D.S. Brown, Determination of octanol/water distribution coefficients, water solubilities, and sediment/water partition coefficients This paper presents a method for derivatizing carboxylic for hydrophobic organic pollutants, Environmental Protection Agency acids in aqueous matrices with ionic reagents that are not (EPA/600/4-79/032), 1979. prone to liquid–liquid extraction. The use of carbodiimides with [7] R.P. Schwarzenbach, P.M. Gschwend, D.M. Imboden, Environmental quaternary ammonium substituents eliminates the non-polar Organic Chemistry, Wiley, New York, 1993. contaminants typically associated with the use of more tradi- [8] J.R. Seward, T.W. Schultz, SAR QSAR Environ. Res. 10 (1999) 557. [9] E.A. Dietz, K.F. Singley, J. Liq. Chromatogr. 17 (1994) 1637. tional carbodiimides, such as dicyclohexylcarbodiimide. The [10] E.W. Cook, Chem. Ind. 18 (1971) 485. resulting fluorinated amides are readily chromatographed and [11] T. Niwa, K. Meada, T. Ohki, A. Saito, I. Tsuchida, J. Chromatogr. 225 easily detectable by electron-capture techniques. (1981) 1. Q.L. Ford et al. / J. Chromatogr. A 1145 (2007) 241–245 245

[12] G.A. Boorman, V. Dellarco, J.K. Dunnick, R.E. Chapin, S. Hunter, F. [30] Chlorinated Acids by GC/ECD, US Environmental Protection Agency, Hauchman, H. Gardner, M. Cox, R.C. Sills, Environ. Health Perspect. 107 1995 (600/R-95-131, 515.1). (1999) 207. [31] Chlorinated Acids (Herbicides) by LSE & GC/ECD, US Environmental [13] B. Cancho, F. Ventura, M.T. Galceran, J. Chromatogr. A 841 (1999) 197. Protection Agency, 1995 (600/R-95-131, 0515.2). [14] B. Davis, Anal. Chem. 49 (1977) 832. [32] Chlorinated Acids (Herbicides) in DW/LLE GC/ECD, US Environmental [15] H. Weinberg, Anal. Chem. 71 (1999) 801a. Protection Agency, 1996 (815/R-00-014, 0515.3). [16] V. Lopez-Avila, Crit. Rev. Anal. Chem. 29 (1999) 195. [33] Haloacetic Acids by LLE & GC/ECD, US Environmental Protection [17] J.H. Brill, B.A. Narayanan, J.P. McCormick, Appl. Spectrosc. 45 (1991) Agency, 1990 (600/4-90-020, 0552). 1617. [34] F. Bartolozzi, G. Bertazza, D. Bassi, G. Cristoferi, J. Chromatogr. A 758 [18] J.W. Hodgeson, J. Collins, R.E. Barth, Method 552 Determination of (1997) 99. Haloacetic Acids and Dalapon in Drinking Water by Liquid-Liquid Extrac- [35] M.N. Sarrion, F.J. Santos, M.T. Galceran, Anal. Chem. 72 (2000) 4865. tion, Derivatization, and Gas Chromatography with Electron Capture [36] D.F. Detar, R. Silverstein, J. Am. Chem. Soc. 88 (1966) 1020. Detection, Rev. 1.0., U.S. Environmental Protection Agency, 1990. [37] D.F. Detar, R. Silverstein, J. Am. Chem. Soc. 88 (1966) 1013. [19] J. Horner, S.S.Q. Hee, R.G. Sutherland, Anal. Chem. 46 (1974) 110. [38] J.F. Poduslo, G.L. Curran, J. Neurochem. 66 (1996) 1599. [20] E.T. Urbansky, J. Environ. Monitor. 2 (2000) 285. [39] J.F. Poduslo, G.L. Curran, J. Neurochem. 67 (1996) 734. [21] K. Blau, G.S. King, Handbook of Derivatives Chromatography, Heydon & [40] F. Kurzer, K. Douraghi, Chem. Rev. 67 (1967) 107. Sons, London, 1979. [41] V.Maurino, C. Minero, E. Pelizzetti, S. Angelino, M. Vincenti, J. Am. Soc. [22] J.W. Hodgeson, J.W. Munch, D.J. Munch, A.M. Pawlecki, Method 552.2 Mass Spectrom. 10 (1999) 1328. Determination of Haloacetic Acids and Dalapon in Drinking Water by [42] B.F. Scott, M. Alaee, Water Qual. Res. J. Can. 33 (1998) 279. Liquid–Liquid Extraction, Derivatization, and Gas Chromatography with [43] A. Vairavamurthy, K. Mopper, Anal. Chim. Acta 237 (1990) 215. Electron Capture Detection, US Environmental Protection Agency, 1995. [44] R.D. Mortimer, J. Chromatogr. 516 (1990) 427. [23] M. Tedetti, K. Kawamura, B. Charriere, N. Chevalier, R. Sempere, Anal. [45] M. Mikolajczyk, P. Kielbasinski, Tetrahedron 37 (1981) 233. Chem. (2006). [46] H. Miwa, C. Miyama, M. Yamamoto, J. Chromatogr. 321 (1985) 165. [24] Chlorinated Acids in Water–HPLC, US Environmental Protection Agency, [47] M. Vincenti, S. Biazzi, N. Ghiglione, M.C. Valsania, S.D. Richardson, J. 1992 (821/R-93-010-B, 0555). Am. Mass Spectrom. 16 (2005) 803. [25] B. Aikawa, R.C. Burk, Int. J. Environ. Anal. Chem. 66 (1997) 215. [48] J. March, Advanced Organic Chemistry, third ed., Wiley, New York, [26] J.Z. Yu, H.E. Jeffries, R.M. Lelacheur, Environ. Sci. Technol. 29 (1995) 1985. 1923. [49] H.G. Khorana, Chem. Rev. 53 (1953) 145. [27] Chlorinated Acids in Water by GC/ECD, US Environmental Protection [50] J.W. Hodgeson, J. Collins, R.E. Barth, Method 552 Determination of Agency, 1989 (821/R-93-010-B, 0515.1). Haloacetic Acids and Dalapon in Drinking Water by Liquid–Liquid [28] Chlorinated Acids—Liquid-Solid Extraction/GC/ECD, US Environmental Extraction, Derviatization, and Gas Chromatography with Electron Protection Agency, 1992 (0515.2). Capture Detection, Rev. 1.0, US Environmental Protection Agency, [29] Chlorinated Acids in Water by LSE/GC/ECD, US Environmental Protec- 1990. tion Agency, 1992 (821/R-93-010-B, 0515.2). [51] D.G. Hoares, D.E. Koshland Jr., J. Biol. Chem. 242 (1967) 2447.