Anal. Chem. 2009, 81, 485–489
Reductive Amination of Aldehyde 2,4-Dinitorophenylhydrazones Using 2-Picoline Borane and High-Performance Liquid Chromatographic Analysis Shigehisa Uchiyama,* Yohei Inaba, Mariko Matsumoto, and Gen Suzuki Department of Environmental Health, National Institute of Public Health, 2-3-6, Minami, Wako City, Saitama 351-0197, Japan A new method for the determination of carbonyls in air using 2,4-dinitrophenylhydrazine (DNPH) has been developed. The traditional method for the measurement of carbonyl compounds, using DNPH to form the corresponding 2,4-dinitrophenylhydrazone (DNPhydrazone) derivatives, is subject to analytical errors because DNPhydrazones form both E- and Z-geometrical isomers as a result of the CdN double bond. To overcome this issue, a method for transforming the CdN double bond into a CsN single bond, using reductive amination of DNPhydrazone derivatives, has been developed. Reductive amination of aldehyde DNPhydrazones was carried out by adding 2-picoline borane acetonitrile solution in eluate through the DNPH-cartridge. The amination reactions of C1-C10 aldehyde DNPhydrazones were completely converted into the reduced forms within 40 min in the presence of 1 mmol/L 2-picoline borane and 20 mmol/L of phosphoric acid. These reduced forms were very stable and did not change when stored for 2 weeks at room temperature. The absorption maximum wavelengths of the reduced forms from C1-C10 aldehyde DNPhydrazones were 351-352 nm and shifted 6-7 nm toward shorter wavelengths when compared to the corresponding DNPhydrazones, and the molar absorption coefficients were 1.5 × 104 (C1) to 2.2 × 104 L/mol/cm (C10). Complete separation between C1-C10 aldehyde DNPhydrazones and the corresponding reduced forms can be achieved by operating the HPLC in gradient mode using an Ascentis RP-Amide column (150 mm × 4.6 mm i.d.). The RSDs of DNPhydrazone (Z + E) peak areas ranged from 0.40-0.66 and those of the corresponding reduced forms ranged from 0.26-0.41. It was shown that the reductive amination method gave improved HPLC analytical precision because of the absence of isomers. Carbonyl compounds are ubiquitous air pollutants in the global environment. They are formed naturally in the troposphere by oxidation of hydrocarbons1-3 and are also produced by the * Corresponding author. National Institute of Public Health, 2-3-6, Minami, Wako City, Saitama 351-0197, Japan. E-mail:
[email protected]. (1) Grosjean, D.; Williams, E. L.; Grosjean, E.; Andino, J. M.; Seinfeld, J. H. Environ. Sci. Technol. 1993, 27, 2754–2758. 10.1021/ac802163y CCC: $40.75 2009 American Chemical Society Published on Web 12/04/2008
reaction between ozone and terpenoid from woody materials and cleaning products in the indoor environment.4,5 The major anthropogenic emission sources affecting humans in the indoor environment are products containing carbonyl compounds such as resins, glues, insulating materials, oriented strand board, plywood, and fabrics. Long-term exposure to relatively high levels of carbonyl compounds such as formaldehyde and acetaldehyde is known to increase the risk of asthma6 and cancer.7 Accurate aldehyde measurements are therefore important both for determining the formation mechanism of aldehydes and for evaluating the implications for human health. The most widely used method for qualitative and quantitative analysis of carbonyl compounds is the 2,4-dinitrophenylhydrazine (DNPH) derivatization followed by high-performance liquid chromatography (HPLC). The specific reaction of carbonyl compounds with DNPH forming the corresponding 2,4-dinitrophenylhydrazone (DNPhydrazone) is one of the most important qualitative and quantitative methods in organic analysis. It was first published by Allen8 and Brady9 in the early 1930s. The main advantage of the DNPH method is the ability to analyze various aldehydes and ketones simultaneously in a complex mixture. Sampling can be performed using acidic solutions of DNPH in impingers10 or with acidic, DNPH-coated, solid sorbents in a cartridge. A number of cartridge devices containing solid sorbents coated with DNPH are commercially available for sampling aldehydes in air. Recently, the use of DNPH-coated silica gel has also become widespread for active11 and diffusive12 air sampling methods. Because of the (2) Grosjean, E.; de Andrade, J. B.; Grosjean, D. Environ. Sci. Technol. 1996, 30, 975–983. (3) Possanzini, M.; Di Palo, V.; Cecinato, A. Atmos. Environ. 2002, 36, 3195– 3201. (4) Singer, B. C.; Coleman, B. K.; Destaillats, H.; Hodgson, A. T.; Lunden, M. M.; Weschler, C. J.; Nazaroff, W. W. Atmos. Environ. 2006, 40, 6696– 6710. (5) Nicolas, M.; Ramalho, O.; Maupetit, F. Atmos. Environ. 2007, 41, 3129– 3138. (6) Nordman, H.; Keskinen, H.; Tuppurainen, M. J. Allergy Clin. Immunol. 1985, 75, 91–99. (7) Kerns, W. D.; Pavkov, K. L.; Donofrio, D. J.; Gralla, E. J.; Swenberg, J. A. Cancer Res. 1983, 43, 4382–4392. (8) Allen, C. F. H. J. Am. Chem. Soc. 1930, 52, 2955–2959. (9) Brady, O. L. J. Chem. Soc. 1931, 756-759. (10) Grosjean, D. Environ. Sci. Technol. 1982, 16, 254–262. (11) Determination of Formaldehyde in Ambient Air Using Adsorbent Cartridge Followed by High Performance Liquid Chromatography (HPLC). U.S. Environmental Protection Agency, Office of Research and Development: Research Triangle Park, NC, 1999; Compendium method TO-11A.
Analytical Chemistry, Vol. 81, No. 1, January 1, 2009
485
importance of the method, it has been introduced as a standard procedure by several national and international standardization bodies. However, this method is prone to analytical error because DNPhydrazone derivatives have both E- and Z-geometrical isomers. The formation of isomeric DNPhydrazones from asymmetric carbonyl compounds in the liquid phase has long since been known.13-15 Behforouz and co-workers16 and Tayyari and co-workers17 reported that a trace of acid catalyzed the E-Z isomerization which was detected via the melting point anomalies it caused. Uchiyama and co-workers18 reported that purified aldehyde-DNPhydrazone demonstrated only the E-isomer; however, under the addition of acid or UV irradiation, both E- and Z-isomers were seen. The spectra of Z-isomers were different from E-isomers, and the absorption maxima of the E-isomers were shorter (5-8 nm) than the corresponding Z-isomers. An equilibrium Z/E isomer ratio was observed in 0.02-0.2% v/v phosphoric acid solutions. In the case of saturated aldehyde DNPhydrazones, the equilibrium Z/E isomer ratios were 0.14-0.32.18 In the case of unsaturated aldehyde DNPhydrazones; the equilibrium Z/E isomer ratios were 0.018-0.035 for 2-alkenals or 0.13-0.21 for other alkenals.19 In the case of ketones, the equilibrium Z/E isomers ratios were 0.20-0.26.20 Catalytic acid such as phosphoric acid is required for the reaction of DNPH and carbonyl compounds. Therefore, the preparative solution for HPLC is acidic and isometric problems described above are inevitable. In order to resolve the isometric problem, it is necessary to transform the CdN double bond to the CsN single bond. Pyridine-borane has been used as a reductive amination reagent for aldehydes.21-23 However, this reagent is quite unstable to heat and attempted distillation of the liquid residue at reduced pressures sometimes results in violent decompositions.24-26 Thus, extreme care must be used if this reagent is handled in large quantities. Sato and co-workers27 have developed an expeditious, easy-to handle, and environmentally friendly approach to the synthesis of a variety of amines through a three-component onepot reaction of carbonyl compounds, amines, and 2-picoline borane. The later is a thermally stable transparent solid and can be stored on a shelf for months without appreciable loss of (12) Uchiyama, S.; Hasegawa, S. Atmos. Environ. 1999, 33, 1999–2005. (13) Ramirez, F.; Kirby, A. F. J. Am. Chem. Soc. 1954, 76, 1037–1044. (14) Uralets, V. P.; Rijks, J. A.; Leclercq, P. A. J. Chromatogr. 1980, 194, 135– 144. (15) Binding, N.; Mu ¨ ller, W.; Witting, U. Fresenius J. Anal. Chem. 1996, 356, 315–319. (16) Behforouz, M.; Bolan, J. L.; Flynt, M. S. J. Org. Chem. 1985, 50, 1186– 1189. (17) Tayyari, S. F.; Speakman, J. L.; Arnold, M. B.; Cai, W.; Behforouz, M. J. Chem. Soc., Perkin Trans. 1998, 2, 2195–2200. (18) Uchiyama, S.; Ando, M.; Aoyagi, S. J. Chromatogr., A 2003, 996, 95–102. (19) Uchiyama, S.; Matsushima, E.; Aoyagi, S.; Ando, M. Anal. Chim. Acta 2004, 523, 157–163. (20) Uchiyama, S.; Kaneko, T.; Tokunaga, H.; Ando, M.; Otsubo, Y. Anal. Chim. Acta 2007, 605, 198–204. (21) Pelter, A.; Rosser, R. M.; Mills, S. J. Chem. Soc., Perkin Trans. 1984, 1, 1717–720. (22) Bomann, M. D.; Guch, I. C.; DiMare, M. J. Org. Chem. 1995, 60, 5995– 5996. (23) Moormann, A. E. Synth. Commun. 1993, 23, 789–795. (24) Ryschkewitsch, G. E.; Birnbaum, E. R. Inorg. Chem. 1965, 4, 575–578. (25) Baldwin, R. A.; Washburn, R. M. J. Org. Chem. 1961, 26, 3549–3550. (26) Brown, H. C.; Domash, L. J. Am. Chem. Soc. 1956, 78, 5384–5386. (27) Sato, S.; Sakamoto, T.; Miyazawa, E.; Kikugawa, Y. Tetrahedron 2004, 60, 7899–7906.
486
Analytical Chemistry, Vol. 81, No. 1, January 1, 2009
reduction capability. The use of 2-picoline borane eliminates the problems encountered with the use of other less stable reducing agents such as pyridine borane. In this study, a method of reductive amination of DNPhydrazone derivatives using 2-picoline borane was developed for the high-precision analysis of aldehydes in air. EXPERIMENTAL SECTION Apparatus and Reagents. The HPLC system (Shimadzu, Kyoto, Japan) used included two LC-20AD pumps, an SIL-20AC autosampler, and an SPD M20A photodiode array detector. Two analytical columns were adopted. The first, used in gradient mode for C1-C10 aldehyde derivative analysis, was an Ascentis RPAmide, 3 µm particle size, 150 mm × 4.6 mm i.d. column (Supelco Inc., Bellefonte, PA), where solution A of the mobile phase mixture was acetonitrile/water (40/60 v/v) containing 5 mmol/L ammonium acetate and solution B was acetonitrile/ water (75/25 v/v). In this case, HPLC elution was carried out with 100% A for 5 min, followed by a linear gradient from 100% A to 100% B in 50 min and then held for 10 min. The flow rate of the mobile phase was 1.2 mL/min. The second, used in isocratic mode for effective separation of acetaldehyde derivatives, was an Ascentis Express C18, 2.7 µm particle size, 150 mm × 4.6 mm i.d. column (Supelco Inc., Bellefonte, PA). In this case, the mobile phase mixture was acetonitrile/cyclopentyl methyl ether/water (40:2:60 v/v) containing 5 mmol/L ammonium acetate. The column temperature was 40 °C, the autosampler temperature was 25 °C, and the injection volume was 10 µL. The water used in HPLC and sample preparation was deionized and purified using a Milli-Q Water System equipped with a UV lamp (Millipore, Bedford, MA). Acetonitrile was HPLC grade from Riedel-de Haen (AG, Seelze-Hannover, Germany). 2,4-Dinitrophenylhydrazine hydrochloride (>98%) was from Tokyo Kasei Co. Ltd. (Tokyo, Japan). Phosphoric acid (85% solution in water), 2-picoline borane complex (95%), cyclopentyl methyl ether (>99.90%), and ammonium acetate (99.999%) were from SigmaAldrich Inc., St. Louis, MO. Silica gel (105-210 µm particle size, 120 Å mean pore size) was from AGC Si-Tech. Co., Ltd. (Fukuoka, Japan). Preparation of the DNPH Cartridge for Collection of Carbonyls. Silica gel (50 g) was washed with acetonitrile (3 × 500 mL). To the washed silica gel was added a solution consisting of 2,4-dinitrophenylhydrazine hydrochloride (0.25 g) and phosphoric acid (0.5 mL) dissolved in acetonitrile (250 mL). The mixture was stirred and the solvent was evaporated to dryness at 40 °C under vacuum using a rotary evaporator. DNPH-coated silica particles were packed into a polyethylene cartridge (Rezorian tube, Supelco Inc., Bellefonte, PA) and stored in a refrigerator at 4 °C. RESULTS AND DISCUSSION Reductive Amination of Aldehyde DNPhydrazones with 2-Picoline Borane. The carbonyl compounds react with acidified DNPH in the solid or liquid phase (Scheme 1). As described above, aldehyde-DNPhydrazones exist as (E)- and (Z)-geometrical isomers around the CdN double bond, and isomerization during sample preparation is unpredictable and can cause analytical errors.1-6 Moreover, DNPhydrazone formation is reversible which
Scheme 1
Scheme 2
introduces even more uncertainty. In acidic aqueous solutions, the DNPhydrazone derivatives can hydrolyze back to the carbonyl compounds and DNPH until equilibrium is attained.14 However, 2-picoline borane reduces DNPhydrazones under acidic conditions to form the corresponding secondary amines as reduced forms. This mechanism is illustrated in Scheme 2. For example, acetaldehyde DNPhydrazone is transformed into N-(2,4dinitrophenyl)-N′-ethylhydrazine as it’s reduced form. This reaction is a nonreversible reaction, and the reduced form has no geometrical isomers because the CdN double bond has been eliminated. Reductive amination of aldehyde DNPhydrazones is carried out either by adding 2-picoline borane acetonitrile solution in the eluate through the DNPH-cartridge or by eluting the DNPHcartridge with 2-picoline borane acetonitrile solution. Reductive amination with 2-picoline borane is a specific reaction for the CdN double bond. Under all experimental conditions in this study, 2-picoline borane reduced only the CdN double bond and did not reduce the CdC double bond. Spectrum Profiles of the Reduced Forms of DNPhydrazones. Figure 1 shows the UV spectra of acetaldehyde DNPhydrazone and its reduced form (N-(2,4-dinitrophenyl)-N′-ethylhydrazine). Reduced forms of other DNPhydrazone derivatives (e.g., for formaldehyde, propionaldehyde, and all C1C10 aldehydes)
Figure 1. UV spectra of acetaldehyde DNPhydrazone and its reduced form.
show the same spectral profiles as the acetaldehyde derivative. The absorption maxima of the reduced forms of the C1-C10 aldehyde DNPhydrazone derivatives were 351 nm (C1,C2) and 352 nm (C3-C10) and were shifted toward shorter wavelengths by 6-7 nm relative to corresponding DNPhydrazones. The molar absorption coefficients were 1.5 × 104 (C1), 1.6 × 104 (C2, C3), 1.7 × 104 (C4-C9), and 2.2 × 104 L/mol/cm (C10). Separation of Aldehyde DNPhydrazones and Their Reduced Forms. In HPLC analysis of DNPhydrazones, good separation results were usually obtained using an ODS or Amide column and acetonitrile/H2O as the mobile phase. However, in the case of acetaldehyde DNPhydrazone derivatives, it is impossible to separate E-acetaldehyde DNPhydrazone from its reduced form using this HPLC method. An alternative HPLC method for acetaldehyde derivatives was therefore developed as follows: Formaldehyde, acetaldehyde and propionaldehyde DNPhydrazone derivatives (50 µmol/L each) were prepared in acetonitrile. A 3 mL portion of this solution was poured into 5 mL vial and added 60 µL of 0.1 mol/L 2-picoline borane acetonitrile solution. Phosphoric acid was added to a level of 20 mmol/L and the solution was immediately analyzed by HPLC. In this case, the isocratic mode was used with the addition of cyclopentyl methyl ether (CPME) to mobile phase which facilitates separation of the acetaldehyde derivatives. Figure 2 shows the chromatograms of formaldehyde, acetaldehyde and propanal DNPhydrazone derivatives, including E/Z isomers and their corresponding reduced forms, obtained using this method. Before the addition of 2-picoline borane only E- and ZDNPhydrazone isomers were detected (upper panel). After the addition of 2-picoline borane, peaks of the reduced DNPhydrazones began to appear in front of the Z- and E-isomer peaks of the corresponding DNPhydrazone (middle panel). After 1 h, all DNPhydrazone derivatives had been reduced (lower panel). As formaldehyde is a symmetrical aldehyde, its DNPhydrazone derivative has no isomer and the peak of its reduced form appears in front of the DNPhydrazone peak. In a separate experiment, C1-C10 aldehyde DNPhydrazone derivatives were reduced using 2-picoline borane as follows. C1-C10 aldehyde DNPhydrazone derivatives (50 µmol/L each) were prepared in acetoAnalytical Chemistry, Vol. 81, No. 1, January 1, 2009
487
Figure 2. Chromatographic profiles of DNPhydrazones and their reduced forms changing with time. FA-H, formaldehyde DNPhydrazone; FA-R, reduced form from FA-H; AA-H, acetaldehyde DNPhydrazone; AA-R, reduced form from AA-H; PA-H, proponaldehyde DNPhydrazone; PA-R, reduced form from PA-H. Table 1. Reproducibility with a Relative Standard Deviation (%) of Saturated Straight-Chain C1-C10 Aldehyde DNPhydrazones and Their Reduced Formsa DNPhydrazone derivative aldehyde C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 a
Z 0.41 0.45 0.63 0.66 0.67 0.60 0.62 0.72 0.87
E
Z+E
reduced form
0.42 0.55 0.58 0.58 0.60 0.63 0.64 0.64 0.64
0.57 0.40 0.53 0.58 0.59 0.61 0.62 0.63 0.64 0.66
0.33 0.26 0.32 0.34 0.36 0.38 0.41 0.40 0.37 0.38
Reproducibility was obtained from 10 repeated runs.
nitrile. A 3 mL portion of this solution was taken in a 5 mL vial, and 60 µL of 0.1 mol/L 2-picoline borane acetonitrile solution was added. Immediately after adding phosphoric acid to contain 20 mmol/L, the solution was analyzed by HPLC in the gradient mode. Figure 3 shows the chromatograms at the state of coexistent aldehyde DNPhydrazones and their reduced forms. Complete separation between C1-C10 aldehyde DNPhydrazones and their reduced forms can be achieved for all derivatives except acetaldehyde using the gradient mode. The peak of E-acetaldehyde DNPhydrazone was overlapped by its reduced form. Before the addition of 2-picoline borane, only E- and Z-DNPhydrazone isomers were detected (upper panel). After the addition of 2-picoline borane, peaks of the reduced forms began to appear in front of the Z- and E-isomer peaks of the corresponding DNPhydrazone, except for acetaldehyde derivatives. At a time of 20 min after addition of the 2-picoline borane solution, reductive amination proceeded to 46-50% (middle panel). At a time of 80 min after addition of the 2-picoline borane solution, all DNPhydrazone derivatives, including Z- and E-isomers, were completely converted to their 488
Analytical Chemistry, Vol. 81, No. 1, January 1, 2009
Figure 3. Chromatographic profiles of DNPhydrazones and their reduced forms changing with reaction time. The number of the peak name indicates carbon number of the precursor aldehyde (1, formaldehyde; 2, acetaldehyde; 3, propanal; 4, butanal; 5, pentanal; 6, hexanal; 7, heptanal; 8, octanal; 9, nonanal; 10, decanal). “H” indicates the DNPhydrazone derivative and “R” indicates reduced form of the DNPhydrazone derivative. The prime sign indicates the Z-isomer of the DNPhydrazone derivative.
respective reduced forms (lower panel). These reduced forms were very stable and could be stored at room temperature for at least 2 weeks. Acid Catalyzed Amination of Aldehyde DNPhydrazones. Before the addition of 2-picoline borane, the peaks of the Z- and E-isomers of acetaldehyde and propionaldehyde DNPhydrazones were observed as an inevitable result of the unsymmetrical carbonyl compounds in the acetonitrile solution. When 2-picoline borane was added to the solution, new peaks for the reduced form of each derivative appeared gradually in front of the corresponding DNPhydrazone peaks, and then finally all DNPhydrazone peaks were converted to their respective reduced forms. Figure 4 shows the amination reaction of acetaldehyde DNPhydrazone with 2-picoline borane under various concentrations of phosphoric acid. Other C1-C10 aldehyde DNPhydrazones produced similar curves to those shown in Figure 3. Increasing the concentration of phosphoric acid dramatically speeded up the reaction rate. The amination reaction of aldehyde DNPhydrazones with 2-picoline borane was completed in 40 min in the presence of 20 mmol/L
Figure 5. Chromatographic profiles of DNPhydrazones (upper panel) and their reduced forms (lower panel). The number of the peak name indicates the carbon number of the precursor aldehyde. “H” indicates the DNPhydrazone derivative and “R” indicates the reduced form of the DNPhydrazone derivative. The prime sign indicates the Z-isomer of the DNPhydrazone derivative. AH and AR indicate acetone DNPhydrazone and its reduced form (N-(2,4-dinitrophenyl)-N′-isopropylhydrazine). Figure 4. Changes in acetaldehyde DNPhydrazone (upper) and its reduced form (lower) concentrations with time under various concentrations of phosphoric acid.
phosphoric acid. The facts that the derivatization reaction requires a catalytic amount of acid and that amine formation is irreversible explain the behavior observed in Figure 4. Chromatographic resolution deteriorated as the acidity of the aqueous solution increased and, above 50 mmol/L phosphoric acid, quantitation of peaks was impossible. Applications to Real Sample Analysis. An indoor air sample was drawn through a prepared DNPH cartridge at a flow rate of 100 mL/min for 24 h. The DNPH cartridge was subsequently extracted with 5 mL of acetonitrile at a flow rate of 1 mL/min. A 2 mL portion of the eluate was taken in a 5 mL vial and 200 µL of 0.1 mol/L 2-picoline borane acetonitrile solution and 20 µL of 1 mol/L phosphoric acid acetonitrile solution were added. After 24 h, the resulting derivative mixtures were analyzed by HPLC using the gradient mode. Figure 5 shows the chromatographic profiles of the DNPhydrazones and corresponding reduced forms obtained from indoor air. As can be clearly seen, adding 2-picoline borane solution to the eluate caused all DNPhydrazone derivatives, including both Z- and E-isomers, to be completely converted to their respective reduced forms (lower panel) producing a simpler chromatographic profile. Precision of the Reductive Amination Method. Aliquots (1 mL) of the saturated straight-chain C1-C10 aldehyde DNPhydrazone acetonitrile solution (100 µmol/L) were placed in 20 vials. A volume of 20 µL of 0.1 mol/L 2-picoline borane acetonitrile solution and 10 µL of 1 mol/L phosphoric acid acetonitrile solution were then added to 10 of these vials, and
10 µL of 1 mol/L phosphoric acid acetonitrile solution alone was added to the other 10 vials. After 2 h, the resulting derivative mixtures were analyzed by HPLC using the gradient mode. Reproducibility data, with relative standard deviations (RSD, %) calculated from the peak areas of saturated straightchain C1-C10 aldehyde DNPhydrazones and their respective reduced forms, are shown in Table 1. The peak area RSDs of DNPhydrazone (Z + E) derivatives ranged from 0.40-0.66 and those for the corresponding reduced forms ranged from 0.26-0.41. It was therefore shown that reductive amination improved HPLC analytical precision by eliminating the complication of isomers. The calibration curves for the reduced forms of C1-C10 aldehyde DNPhydrazone derivatives were linear for the concentrations studied (2-100 µmol/L) with correlation coefficients greater than 0.999. CONCLUSIONS The traditional method for the measurement of carbonyl compounds, using DNPH to form the corresponding DNPhydrazone derivatives, is subject to analytical errors because DNPhydrazones form both E- and Z-geometrical isomers as a result of the CdN double bond. To overcome this issue, a method for transforming the CdN double bond into a CsN single bond, using reductive amination of DNPhydrazone derivatives, has been developed. The reductive amination method developed in this study overcomes analytical errors caused by E- and Z-geometrical isomers and gives good HPLC analytical precision. Received for review November 19, 2008.
October
13,
2008.
Accepted
AC802163Y Analytical Chemistry, Vol. 81, No. 1, January 1, 2009
489
