Comparison of Solution Chemistries of Orthophthalaldehyde and 2, 3

Comparison of Solution Chemistries of Orthophthalaldehyde and 2,3- ... Citation data is made available by participants in Crossref's Cited-by Linking ...
0 downloads 0 Views 130KB Size
Anal. Chem. 2006, 78, 7802-7808

Comparison of Solution Chemistries of Orthophthalaldehyde and 2,3-Naphthalenedicarboxaldehyde N. Salem and P. Zuman*

Department of Chemistry, Clarkson University, Potsdam, New York 13699-5810

Polarography was used to obtain the concentrations of the dialdehydic (10%), monohydrated acyclic (5%), and cyclic hemiacetal form (85%) of orthophthalaldehyde (OPA). For 2,3-naphthalenedicarboxaldehyde (NDA) these values were estimated to be 15, 7, and 78%. Addition of water in unbuffered solutions followed first-order kinetics with rate constants 0.0018 s-1 for OPA and 0.0012 s-1 for NDA. Dehydration to form both the dialdehyde and the monohydrate is both acid- and base-catalyzed. Both dialdehydes yield on reaction with OH- ions geminal diol anion, which is electro-oxidized to a carboxylic acid. In the most frequently used pH range for the determination of amino acids, NDA can undergo reaction with OH- ions, but OPA does not. In aqueous solutions, NDA is less strongly hydrated than OPA. Most frequently used analytical methods for determination of low concentrations of amino acids are based on formation of a fluorescent isoindole derivative. These species are formed in a reaction of aromatic vicinal dicarboxaldehydes in the presence of a nucleophile. Initially, analytical procedures were based on the reaction of amino acids with orthophthalaldehyde (OPA) in the presence of a thiolate as the nucleophile.1,2 More recently has been promoted the use of a reagent consisting of 2,3-naphthalenedicarboxaldehyde (NDA) with cyanide ions as the nucleophile.3-5 The structure of the fluorophore, a substituted isoindole derivative, was confirmed both for OPA6-8 and for NDA.3-5 The mechanism by which this fluorophore is formed is, nevertheless, not yet understood2,9,10 for reactions of both OPA and NDA. The uncertainty about the actual reaction course is reflected in a considerable variety of empirically developed compositions of reaction mixtures and reaction conditions for individual applications * Corresponding author: Telephone: 001-315-2682340. Fax 001-315-2686610. E-mail address: [email protected]. (1) Roth, M. Anal. Chem. 1971, 43, 880-882. (2) Zuman, P. Chem. Revi. 2004, 104, 3217-3238. (3) de Montigny, P.; Stobaugh, J. F.; Givens, R. S.; Carlson, R. G.; Srinivasachar, K.; Sternson, L. A.; Higuchi, T. Anal. Chem. 1987, 59, 1096-1101. (4) Sternson, L. A.; Stobaugh, J. F.; Repta, A. J. Anal. Biochem. 1985, 144, 233-246. (5) Zuman, P. Electroanalysis 2006, 18, 131-140. (6) Simons, S. S., Jr.; Johnson, D. F. J. Am. Chem. Soc. 1976, 98, 7098-7099. (7) Simons, S. S., Jr.; Johnson, D. F. Anal. Biochem. 1976, 82, 250-254. (8) Simpson, R. C.; Spriggle, J.; Veenig, H. J. J. Chromatogr. 1983, 261, 407414. (9) Wong, O. S.; Sternson, L. A.; Schowen, R. L. J. Am. Chem. Soc. 1985, 107, 6421-6422. (10) Trepman, E.; Chen, R. F. Arch. Biochem. Biophys. 1980, 204, 524-532.

7802 Analytical Chemistry, Vol. 78, No. 22, November 15, 2006

(Table 1). Only in some reported procedures has the importance of sequence of the addition of components of the reaction mixture been stressed. The ratios of concentrations of individual components, reaction time, and recommended temperature vary widely. The roles of pH, kind of buffer, and concentration have not been investigated in sufficient detail. Borate buffers have been used in the majority of proposed procedures, but the possibility of formation of complexes with the geminal diol form of the used dialdehydes has not been ruled out. Even more serious is the use of primary alcohols in some reported procedures to increase the solubility of some compounds. These alcohols are stronger nucleophiles than water, and formation of hemiacetals and acetals in their presence might further complicate the investigated system. Acetonitrile, present in other reaction mixtures, can affect hydration-dehydration equilibria, especially at higher concentrations. In these investigations, only the stability of the isoindole derivative has been tested. A better understanding of the sequence of reactions resulting in the formation of the isoindole derivative would enable a more rational approach to the finding of optimum conditions for preparation of the reaction mixture and to establishing optimum reaction conditions for determination of amino acids and other primary amines. The first step in this direction is the present attempt to contribute to the elucidation of the solution chemistry of OPA and NDA. A better understanding of the equilibria and kinetics involving species present in aqueous solutions of the two dialdehydes may offer a indicator for the reason for the differences in their reactivities. MATERIALS AND METHODS Materials. The studied compounds, orthophthalaldehyde and 2,3-naphthalenedicarboxaldehyde, were supplied by Ralph N. Emanuel LTD and Aldrich, respectively. Acetonitrile used to prepare stock solutions of the studied compounds was obtained from J. T. Baker. Chemicals used for preparation of buffers (primary and secondary sodium phosphate, sodium acetate, boric and hydrochloric acids, and sodium hydroxide) were of analytical grade. Solutions of sodium hydroxide were carbonate-free. Measurements. The polarographic current-voltage curves were recorded using Sargent polarograph model XVI from E. H. Sargent and Company. The capillary electrode used had the following characteristics: m ) 2.2 mg·s-1, t1 ) 5.0 s, and h ) 64 cm. A Kalousek cell31 with a reference electrode separated by a liquid (11) Oates, M. D.; Jorgenson, J. W. Anal. Chem. 1989, 61, 432-435. 10.1021/ac061185g CCC: $33.50

© 2006 American Chemical Society Published on Web 10/20/2006

Table 1. Reported Conditions for the Reaction of 2,3-Naphthalenedicarboxaldehyde with Amino Acids, Peptides, or Other Amines buffer borate

phosphate borate phosphate borate

a

pH

[NDA], mM

[CN-], mM

9.1 9.5 9.5 9.5 9.5 8.0 9.5 10.0 9.1-9.5 9.5 6.0 9.5 9.0 9.0 6.8 9.5 9.2 8.5 9.5 9.0 9.1

0.4 20 20 30 1.0 0.2 10 0.1 1.0 0.1 0.6 0.1 0.2 0.18 5.0 450 0.55 0.6 0.5 4.3 0.3

1.0 20 20 130 0.5 1.0 100 0.1 0.5 0.2 1.0 1.0 2.0 20 5.0 440 0.4 0.6 0.5 4.3 3.0

[amine], mM

teaction,time, min

temp, °C

solvent

25

20% MeOHa 28% MeOHa 10% MeCNb b

10-3 30 2 0.06 10-2-10-5 10-3 10-3-10-5 0.25 0.06 0.0025 10-2-10-6 10-2-10-4

3 120 5 20 3 >60 8 30 60 60 0.5 25

75 10-3-10-4 10-2 10-2-10-3

20 15 0.3-5-30 15

c 75% MeCN 25

25 25 4 60

MeCNb d c 5% MeCN

ref 3 11 12 13 14 15 16 17 14 19 20 21 22 22 24 25 26 27 28 29 30

NDA stock solution in MeOH. b NDA stock solution in MeCN. c NDA stock solution in 80% MeCN, 20% i-PrOH. d Lipid dissolved in 33% EtOH.

junction was used for DC polarography,. The reference electrode was a saturated calomel electrode (SCE), Hg2Cl2/KCl (saturated). A 100-µL portion of the 0.01 M stock solution of a dialdehyde in acetonitrile was added to 9.90 mL of a buffer solution (phosphate, acetate, or borate) of the desired pH value. Current-voltage curves were recorded after purging with N2 for oxygen removal for ∼2 min. The final concentration of the studied compound was 0.1 mM. The studied solutions also contained 1% v/v acetonitrile. Time dependence of UV-vis absorption spectra of unbuffered aqueous solutions containing 0.05 mM OPA and 0.5% v/v aceto(12) Oates, M. D.; Jorgenson, J. W. Anal. Chem. 1989, 61, 1977-1980. (13) Oates, M. D.; Cooper, B. R.; Jorgenson, J. W. Anal. Chem. 1990, 62, 15731577. (14) de Montigny, P.; Riley, C. M.; Sternson, L. A.; Stobaugh, J. F. J. Pharm. Biomed. Anal. 1990, 8, 419-429. (15) Hayakawa, K.; Schlipp, T.; Imai, K.; Higuchi, T.; Wong, O. S. J. Agric. Food Chem. 1990, 38, 1256-1260. (16) Oates, M. D.; Jorgenson, J. W. Anal. Chem. 1990, 62, 1577-1580. (17) Nicholson, I. M.; Patel, H. B.; Kristiansson, F.; Crowle, S. C.; Dave, K.; Stobaugh, J. F.; Riley, C. M. J. Pharm. Biomed. Anal. 1990, 8, 805-816. (18) Nussbaum, M. A.; Przedwiecki, J. E.; Staerk, D. V. Anal. Chem. 1992, 64, 1259-1263. (19) Dave, K. J.; Stobaugh, J. F.; Rossi, T. M.; Riley, C. M. J. Pharm. Biomed. Analysis 1992, 10, 965-977. (20) Melanson, J. E.; Lucy, Ch. A. Analyst 2000, 125, 1049-1052. (21) Panta-Polak, T.; Kassai, M.; Grant, K. B. Anal. Biochem. 2001, 297, 128136. (22) Yang, J. Z.; Basnan, Ch.; Moore, R. D.; Stobaugh, J. F.; Borchardt, R. T. J. Chromatogr., B 2002, 780, 269-281. (23) Cho, Y.-H.; Yoo, H.-S.; Min, J.-K.; Lee, E.-Y.; Hong, S.-P.; Chung, Y.-B.; Lee, Y.-M J. Chromatogr., A 2002, 977, 69-76. (24) Manica, D. P.; Lapos, J. A.; Jones, A. D.; Ewing, A. G. Anal. Biochem. 2003, 322, 68-78. (25) Marra, M.; Bonfigli, A. R.; Testa, R.; Testa, I.; Gambini, A.; Coppa, G. Anal. Biochem. 2003, 318, 13-17. (26) Weng, Q.; Jin, W. Anal. Chim. Acta 2003, 478, 199-207. (27) Zhang, L.-Y.; Liu, Y.-M.; Wang, Z.-L.; Cheng, J.-K Anal. Chim. Acta 2004, 508, 141-145. (28) Zhang, L.-Y.; Sun, M.-X J. Chromatogr., A 2004, 1040, 133-140. (29) Furman, N. H.; Norton, D. R. Anal. Chem. 1954, 26, 1111-1115. (30) Norton, D. R.; Furman, N. H. Anal. Chem. 1954, 26, 1116-1119. (31) Heyrovsky, J.; Kalousek, M. Collect. Czech. Chem. Commun. 1939, 11, 464473.

nitrile or 0.025 mM NDA and 0.25% v/v acetonitrile in a 10-mm quartz cell were recorded using a Hewlett-Packard Agilent 8453 UV-vis spectrophotometer. RESULTS AND DISCUSSION To decide if there are any differences between equilibria and reactivity involving OPA and NDA in aqueous solutions, the rates and equilibria involving the covalent addition of water and hydroxide ions to those dialdehydes were compared using a recording of polarographic current-voltage curves and UV spectra. In all attempts to propose the reaction scheme for the reaction of those dialdehydes with nucleophiles,2 it is generally assumed that these reagents exist in aqueous solutions practically completely as a cyclic hemiacetal (III). Early polarographic studies29,30 indicated qualitatively the presence of three forms, a reducible unhydrated form (I), a reducible monohydrated acyclic form (II), and the electroinactive cyclic hemiacetal form (III).

In the reduction of the majority of organic compounds in aqueous and other protic solvents, the transfer of electrons is accom(32) Bover, W. J.; Zuman, P. J. Chem. Soc., Perkin Transactions 2 1973, 6, 786790.

Analytical Chemistry, Vol. 78, No. 22, November 15, 2006

7803

Figure 1. Dependence of polarographic reduction current-voltage curves of OPA on the concentration of OPA: (1) 5 × 10-5, (2) 1 × 10-4, (3) 1.5 × 10-4, and (4) 2 × 10-4 M in phosphate buffer, pH 5.8. Starting potential was -0.2 V.

panied by a transfer of one or more protons. In the elucidation of such organic electrode processes and in search of information about chemical reactions accompanying the electron transfer, it proved advantageous to follow dependences of limiting currents and half-wave potentials on pH. Polarographic current-voltage curves of OPA and NDA show two reduction waves, i1 and i2 (Figure 1). Wave i1 is attributed to the reduction of the free aldehydic form I; wave i2, to the reduction of the unhydrated formyl group in form II and of the second formyl group in the unhydrated form I. Limiting currents of both waves i1 and i2 are a linear function of the concentration of the dialdehydes. This was confirmed within the pH range where the equilibria between forms I and II, as well as those between forms II and III, are established within the time window of the duration of the electrochemical experiment, and the corresponding limiting currents i1 and i2 are pH-independent. The pH-range, where the current i1 (corresponding to the reduction of the unhydrated form I) is pH-independent, is between pH 3 and 8 for OPA and between pH 4 and 9 for NDA. The current i2, due to the reduction of the second aldehydic group in form I and the formyl group in form II remains pH-independent between pH 1 and ∼4.5 for OPA and between pH 1 and ∼3.5 for NDA (Figure 2). Outside the pH range, where currents i1 and i2 are pHindependent, both limiting currents i1 and i2 show changes with pH (Figure 2). The increases in current i1 of both OPA and NDA with increasing acidity at pH lower than about 3 (Figure 2a) correspond to an acid-catalyzed dehydration of forms II and III to yield the reducible unhydrated form I. The rate of the acidcatalyzed dehydration of forms II and III is somewhat faster for NDA than for OPA. No acid-catalyzed ring opening that would be indicated by an increase in i2 was observed for the conversion of form III into II for either OPA or NDA within the acidity range studied (Figure 2b). In the pH range between 3 and 8 for OPA and between 4 and 9 for NDA, the rates of dehydration of form II to I and of form III to II are relatively slow and catalyzed only by water. The limiting current i1 of OPA is a linear function of h1/2 (where h is the height of the mercury column above the orifice of the capillary electrode) and the plot of i ) f(h1/2) almost passes through the origin (Figure 3a). This shows that the current i1 of OPA is controlled solely by diffusion. Under such conditions, the electrolysis at the dropping 7804 Analytical Chemistry, Vol. 78, No. 22, November 15, 2006

Figure 2. Dependence on pH of limiting currents OPA (9) and NDA (() in 1 × 10-4 M solutions. (a) Limiting currents of the more positive wave, i1, and (b) limiting currents of the more negative wave, i2.

mercury electrode does not perturb the slowly established equilibrium between forms I and II and the limiting current i1 is proportional to the equilibrium concentration of form I of OPA. Comparison of current i1 of OPA with the limiting current of isophthalaldehyde (IPA) (which is in aqueous solution present in equilibrium in 8 is due to establishment of an equilibrium and to exclude the possibility of any cleavage or consecutive reaction. To test the reversibility of processes involved, a stock solution of the dialdehydes was added to a supporting electrolyte consisting of a buffer, pH 8 or 10.3, or 1.0 M NaOH. The solution was left for a short period of time to allow establishment of an equilibrium, and a polarographic i9E curve was recorded. Then acetic acid was added to adjust the pH to 5.0. After establishment Analytical Chemistry, Vol. 78, No. 22, November 15, 2006

7807

of the new equilibrium, another i9E curve was recorded and compared with an i9E curve obtained by addition of the dialdehydes directly to the acetate buffer pH 5.0 (Table 2). Under all conditions, the currents without and with exposure to higher pH were practically identical, thus proving chemical reversibility of the system. Whereas polarography proved useful for investigation of equilibria, spectrophotometry is for this purpose less suitable, because the absorption bands of individual species, which correspond to the π f π* electron transfer in the region between 250 and 280 nm. On the other hand, UV-visible spectrophotometry proved suitable to follow kinetics of hydration. Measuring the absorbance at 260 nm as a function of time after addition of the dialdehyde to distilled water (resulting in 0.05 mM OPA and 0.025 mM NDA solutions) and measuring the absorbance after an establishment of the equilibrium (Figure 7) proved that the hydration catalyzed only by the solvent follows first-order kinetics

7808

Analytical Chemistry, Vol. 78, No. 22, November 15, 2006

(Figure 8). The first-order rate constant of the addition of water was found to be 2.8 × 10-3 s-1 for NDA and 4.1 × 10-3 s-1 for OPA. CONCLUSIONS Qualitatively, the solution chemistry of the two dialdehydes, OPA and NDA, show a similarity, but both the equilibria among the three forms present and the rates of their establishment differ. Perhaps the most important difference in the pH range used in most analytical methods, pH 9-10, is that NDA already shows an acid-base reaction, either dissociation of C(OH)2 or addition of OH- to formyl groups, whereas such reactions are in this pH range absent in solutions of OPA. Received for review June 29, 2006. Accepted August 18, 2006. AC061185G