- Email: [email protected]
Ion chromatographic determination of organic acids in food samples using a permanent coating graphite carbon column
Talanta 72 (2007) 305–309
Short communication
Ion chromatographic determination of organic acids in food samples using a permanent coating graphite carbon column Kenji Yoshikawa ∗ , Miho Okamura, Miki Inokuchi, Akio Sakuragawa Department of Materials and Applied Chemistry, College of Science and Technology, Nihon University, 1-8-14 Kanda-Surugadai, Chiyoda-ku, Tokyo 101-8308, Japan Received 1 September 2006; received in revised form 14 October 2006; accepted 14 October 2006 Available online 22 November 2006
Abstract From the viewpoint of a graphite carbon column with excellent durability, it was applied to the ion chromatography (IC) of several organic acids. The carbon column was permanently coated with the cetyltrimethylammonium (CTMA) ion, and the elution behaviors of several organic acids (acetic acid, lactic acid, succinic acid, malic acid, tartaric acid, citric acid) and inorganic anions (Cl− , NO2 − , NO3 − , SO4 2− ) were examined according to a non-suppressed IC coupled with conductivity detector, when an ion-exchange ability was given to the graphite carbon column. When salicylic acid and sodium salicylate were used as a mobile phase, each organic acid are analyzed approximately 10 min. But the separation of malic acid, chloride and nitrite was difficult. When benzoic acid and 2-amino-2-hydroxymethyl-1,3-puropanediol (tris aminomethane) were used as a mobile phase, tartaric acid and citric acid, etc. with large valency showed tendency to which the width of each peaks extended and retention time increased. However, it was possible to separate excellently for the analytes detected within 10 min. The developed method was then applied to the determination of organic acids in several food samples. © 2006 Elsevier B.V. All rights reserved. Keywords: Graphite carbon column; Organic acid; Cetyltrimethylammonium; Benzoic acid; Ion chromatography
1. Introduction Among the wide variety of methods available for determination of ions, ion chromatography (IC) based on ion-exchangers and conductivity detection has been extensively studied [1–9]. In general, organic polymer and silica-based porous ionexchangers have been used as column packing for liquid chromatography [10–12]. Organic polymer-based ion-exchangers can be used a mobile phase of wide-ranging pH. But organic solvent and pressure resistance present any problems. On the other hand, silica-based ion-exchangers such as an ODS (C18 ) column are generally used with an ion-interaction reagent to achieve separation of ions [13,14]. However, because the silica base is dissolved under the alkaline conditions, it may be affected ion-interaction reagent, i.e. alkaline surfactant. Moreover, It is difficult to use an ODS column for a long time.
∗
Corresponding author. Tel.: +81 3 3259 0801; fax: +81 3 3293 7572. E-mail address: [email protected] (K. Yoshikawa).
0039-9140/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2006.10.020
A graphite carbon column is an alternative non-polar and inactive as stationary phase for liquid chromatography, and it has excellent chemical and physical resistance [15]. Therefore, the buffer solution in a wide-ranging pH (1–14) can be used as a mobile phase. Graphite carbon is used not only an absorbent of gas and organic matter but also a packing material for liquid chromatography, and used to analyze aromatic compounds [16–18], enantiomer of amino acids [19], element of food [20] and drug [21,22]. Though a graphite carbon column has some superior characteristics other than general-purpose column such as a ODS column, etc., it is not so widespread [23–25]. In recent years, applications of the IC that used the ion interaction chemical reagent for the graphitie carbon column that doesn’t have ion exchange ability are reported. These techniques are mainly based on two different methods. One is dynamic coating method [17,26] that added ion interaction reagent such as tetrabutylammonium hydroxide (TBA-OH), polyethylene imine (PEI), etc. in a mobile phase. In this method, it is not necessary to coat an ion interaction reagent on the column surface. However, in the case of non-suppressed IC that we applied this time, because a conductivity of the ion interaction reagent
306
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
Table 1 Operating conditions for some organic acids Column Mobile phase
Carbon IC BI-01 (100 mm × 4.6 mm i.d.) (1) 0.35 mM salicilic acid–0.l mM sodium salicylate (pH 3.5) (2) 2.0 mM benzoic acid–1.2 mM tris aminomethane (pH 4.4)
Flow rate
(1) 0.8 ml/min (2) 1.0 ml/min
Column temperature Injection volume Detection
40 ◦ C 100 l Conductivity
above-mentioned included in a mobile phase is high, more sensitive analysis is difficult, and a baseline is unstable. The other is permanent coating method [27] that coated the quaternary ammonium salt as an ion exchange site on the graphite carbon surface. In this method, it takes the time to coat an ion interaction reagent on the column surface. However, because it is not necessary to add an ion interaction reagent in the mobile phase, it is possible to apply the suppressed and non-suppressed IC. Nagashima et al. [27] have also reported a similar method for inorganic anions by suppressed IC. But, there were only a few reports on simultaneous determination of organic acids by using a permanent coating graphite carbon column. In this work, cetyltrimethylammonium (CTMA)-coated graphite carbon column was used to separate six common organic acids (acetic acid, lactic acid, succinic acid, malic acid, tartaric acid, citric acid). These organic acids were directly detected by non-suppressed IC coupled with conductivity detector. Moreover, the separation behaviors of four inorganic anions (Cl− , NO2 − , NO3 − , SO4 2− ) that included in food sample were examined. The objective of this work was to establish optimal conditions of simultaneous determination of organic acids and apply to the determination of organic acids in several food samples. 2. Experimental 2.1. Apparatus The ion chromatographic equipment consisted of a Model ICA-5120 pump, a Model ICA-5220 conductivity detector, a Model ICA-5410 column oven (Toa DKK, Tokyo, Japan), and a Model 21 chromatocorder (System Instruments, Tokyo, Japan). The separation column was a Carbon IC BI-01 (100 mm × 4.6 mm i.d., average particle size 3.0 m, carbon content larger than 99.5%, Bio-Tech Research, Saitama, Japan).
used as a standard solution of organic acids. Sodium chloride, sodium nitrite, sodium nitrate and sodium sulfide (Wako Pure Chemical, Osaka, Japan) were used as a standard solution of inorganic anions. To coat the graphite carbon column, acetonitrile (Kanto Chemical, Tokyo, Japan) and cetyltrimethylammonium chloride (CTMA-Cl: Wako Pure Chemical, Osaka, Japan) was used. Deionized water, further purified with a Millipore Milli-Q system (Bedford, MA, USA) with a specific resistance of 18.2 M. 2.3. Coating method The column was at first washed with 50 and 100% (v/v) acetonitrile each for 1 h at 0.8 ml/min, and then washed with deionized water for 1 h at 0.8 ml/min. To coat the column, 2.0 mM CTMA-Cl was passed through the column at 0.8 ml/min for 2 h. After dynamically coating, the column was washed with deionized water for 2 h, and used as a permanent coating graphite carbon column. The coating carbon column is stable and available for 3 months. In the permanent coating method, normal chain alkyl group that has hydrophobic part of the quaternary ammonium salt has been adsorbed by the surface of carbon that is hydrophobic particle. Moreover, because the polarity part of quaternary ammonium salt faces the mobile phase, ion exchange and separates ions were performed. Although main separation mechanism is ion exchange, hydrophobic interaction and distribution interaction of the stationary phase also take part. 2.4. Operating conditions The operating conditions for ion chromatography are shown in Table 1. 0.35 mM salicylic acid–0.1 mM sodium salicylate (flow rate, 0.8 ml/min) was used as a mobile phase with salicylate. 2.0 mM benzoic acid–1.2 mM tris aminomethane (flow rate, 1.0 ml/min) was used as a mobile phase with benzoate.
2.2. Reagents 3. Results and discussion Salicylic acid and sodium salicylate (Wako Pure Chemical, Osaka, Japan) were used as a mobile phase with salicylate. Benzoic acid and 2-amino-2-hydroxymethyl-1,3-puropanediol (tris aminomethane: Wako Pure Chemical, Osaka, Japan) were used as a mobile phase with benzoate. Acetic acid, lactic acid, succinic acid, malic acid, tartaric acid and citric acid (Wako Pure Chemical, Osaka, Japan) were
3.1. Mobile phase with salicylate When analyzed organic acids and inorganic anions by nonsuppressed IC, aromatic compounds are usually used as a mobile phase. At first, salicylic acid and sodium salicylate were used. A typical chromatogram of six organic acids and four inorganic
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
307
Fig. 2. Effect of tris aminomethane concentration on retention time (benzoic acid concentration fixed 2.0 mM). Ions (mg l−1 ): acetic (50); 䊉 lactic (20); Cl− (5); NO2 − (5); succinic (50); NO3 − (5); malic (50); ♦ tartaric (50); citric (50); × SO4 2− (20); column: Carbon IC BI-01; mobile phase: 2.0 mM benzoic acid–tris aminomethane; column temperature: 40 ◦ C.
Fig. 1. Chromatogram of standard mixture. Ions (mg l−1 ): acetic (50); lactic (20); succinic (50); NO2 − (5); malic (50); Cl− (5); NO3 − (5); tartaric (50); citric (50); SO4 2− (20); column: Carbon IC BI-01; mobile phase: (a) 0.35 mM salicylic acid–0.1 mM sodium salicylate (pH 3.5, flow rate 0.8 ml/min); (b) 2.0 mM benzoic acid–1.2 mM tris aminomethane (pH 4.4, flow rate 1.0 ml/min); column temperature: 40 ◦ C.
anions is shown in Fig. 1(a)). Each organic acid and inorganic anion except a sulfite is detected nearly 10 min. But the separation of malic acid, chloride and nitrite was difficult. Although the mobile phase concentration changed, simultaneous analysis of the organic acids and inorganic anions were difficult in a mobile phase with salicylate. 3.2. Mobile phase with benzoate 3.2.1. Effect of benzoic acid and tris aminomethane concentration In order to achieve a separation of three kinds of ions (malic acid, chloride, nitrite), benzoic acid and tris aminomethane were used as a mobile phase with benzoate. Effect of tris aminomethane concentration on the retention behavior was examined with the concentration of benzoic acid fixed at 2.0 mM. The relationship between the retention time of analyte and the concentration of tris aminomethane is shown in Fig. 2. Although the retention time of most analytes steady or decrease with increasing tris aminomethane concentration, citric acid showed opposite tendency. As the tris aminomethane concentration increases (pH of a mobile phase increased), dissociation of citric acid advances. As a result, the optimum tris aminomethane concentration was determined to be 1.2 mM, based on the resolution and retention time for each analytes. Furthermore, effect of benzoic acid concentration on the retention behavior was examined with the concentration of tris aminomethane fixed at 1.2 mM. As the concentration of benzoic acid increased, retention time decreases so that dissociation of
each organic acid retrogresses. As a result, the optimum benzoic acid concentration was determined to be 2.0 mM for the same reasons as for tris aminomethane. Moreover, mobile phase concentration rate fixed (benzoic acid:tris aminomethane = 2.0:1.2), the relationships between the logarithm of the total concentration of mobile phase [E] and the capacity factor [k ] are shown in Fig. 3. In the ion-exchange mode, the following relational formula (1) is obtained between the concentration of mobile phase [E] and capacity factor [k ]: y Log k = − log [E] + const. (1) x In this formula, x shows the valency of mobile phase ion, y shows the valency of analyte ion and −(y/x) shows the slope of the straight line [28]. As shown in Fig. 3, in this condition (pH 4.4), the slope of the straight line is negative. The slope of monova-
Fig. 3. Relationship between capacity factor (k ) and total concentration of mobile phase. Ions (mg l−1 ): acetic (50); 䊉 lactic (20); Cl− (5); NO2 − (5); succinic (50); NO3 − (5); malic (50); ♦ tartaric (50); citric (50); column: Carbon IC BI-01; mobile phase concentration rate: benzoic acid:tris aminomethane = 2.0:1.2 (pH 4.4); log k = (−y/x)log[E] + const., where k is the capacity factor, x the valency of mobile phase ion, y the valency of analyte ion and E is the total concentration of mobile phase.
308
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
Table 2 Analytical precision of organic acids Organic acid
Acetic Lactic Succinic Malic Tartaric Citric
Linearity
Detection limit (mg l−1 ) (S/N = 3)
Reproducibility
Range (mg l−1 )
r2 (n = 6)
mg l−1
%R.S.D. (n = 6)
10–200 10–100 10–200 20–120 30–200 40–200
0.9992 0.9997 0.9999 0.9990 0.9991 0.9998
50 50 50 50 50 50
0.46 0.40 0.40 0.73 0.84 0.75
1.2 2.3 2.9 4.5 2.8 3.8
R.S.D.: relative standard deviation.
lent inorganic anions (Cl− , NO2 − and NO3 − ) is approximately −1. However, because the slope of organic acids (especially malic acid and tartaric acid) is between −2 and −1, the separation mode could not be decided that the ion-exchange mode was predominant indiscriminately. As a result, the separation mechanism of the organic acid by the permanent coating carbon column is considered that other modes like the ion exclusion, polarity interaction etc. also take part besides the ion exchange mode. So that the dissociation constant of each organic acid may greatly influence the separation, it is important to control the separation behavior of the analysis, and know the separation mode of the column to understand the dissociation of the organic acid. 3.2.2. Effect of column temperature and flow rate Under the composition of the mobile phase, effect of column temperature on the retention behavior was examined from 30 to 50 ◦ C. Although the retention time of monovalent ions decreased with increasing temperature, divalent ions showed opposite tendency. The optimum column temperature was determined to be
Fig. 4. Chromatogram of organic acids and inorganic anions in nutritions drink sample. Ions (mg l−1 ): acetic; lactic; Cl− ; succinic; NO3 − ; malic; citric; column: Carbon IC BI-01; mobile phase: 2.0 mM benzoic acid–1.2 mM tris aminomethane (pH 4.4); column temperature: 40 ◦ C.
Table 3 Analytical results for organic acids in food sample Sample
Organic acid
Dil. found (mg l−1 )
Added (mg l−1 )
Found (mg l−1 )
Recovery (%)
Concentration (mg l−1 )
Nutoritions drink (1)
Acetica Lactica Succinica Malica Citricb
27.5 23.2 – 18.8 158.9
50.0 30.0 – 50.0 30.0
70.8 50.6 – 68.9 188.4
91.4 91.2 – 100.2 99.7
110.0 92.8
Nutoritions drink (2)
Aceticc Lacticc Succinicd Citricc
6.2 73.8 104.2 82.0
50.0 30.0 30.0 50.0
49.4 95.8 134.3 139.8
88.8 92.3 100.1 105.9
12.5 147.7 2083.8 163.9
Moromi vinegar
Acetice Succinice Malice Citrice
64.5 42.4 27.7 170.6
50.0 50.0 50.0 30.0
111.2 91.8 72.1 189.6
97.1 99.3 92.8 94.5
2065.4 1358.4 886.2 5460.0
Yogurt
Lacticb Succinica Citrica
42.7 – 90.8
50.0 – 50.0
79.6 – 145.0
85.9 – 103.0
2134.5
DL: determination limit. a 1/4 dil. b 1/50 dil. c 1/2 dil. d 1/20 dil. e 1/32 dil.
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
40 ◦ C, based on the resolution and retention time for each analytes. Moreover, the optimum flow rate of mobile phase was determined to be 1.0 ml/min for the same reasons as for column temperature. 3.2.3. Chromatogram of organic acids and inorganic anions A typical chromatogram of six organic acids and four inorganic anions is shown in Fig. 1(b). Because eluting power is weaker to a mobile phase with benzoate than a mobile phase with salicylate, tartaric acid and citric acid, etc. with large valency showed tendency to which the width of each peaks extended and retention time increased. But it was possible to separate excellently for the analytes detected within 10 min. Especially, the separation of the Cl− , NO2 − and malic acid became possible. 3.2.4. Analytical precision The linearity of calibration curves, reproducibility and detection limit are shown in Table 2. The calibration curves obtained from the peak areas for the organic acids were linear, with good correlation coefficients of 0.999. The relative standard deviations (R.S.D.) of peak areas were between 0.40 and 0.84% for six repeated measurements. The detection limits, summarized in Table 2, were calculated at the S/N ratio of 3. The detection limits of organic acids were between 1.2 and 4.5 mg/l. 3.2.5. Application to food samples The proposed method was applied to analyze organic acids in food samples. Yogurt sample was prepared by adding 5 g of it in 20 ml of deionized water and then mixing in 20 min. After the sample solution had been centrifuged in 2000 rpm, 100 l was injected into IC. Chromatogram of nutritions drink sample is shown in Fig. 4. Analytical results of organic acids in the other food sample were shown in Table 3. In addition, to confirm the reliability of the method, organic acids of each concentration is added to the food sample, and recovery test was demonstrated (shown in Table 3). Because the recovery showed nearly 100%, this method can be applied to the determination of organic acids without the matrix effect. 4. Conclusion In this work, simultaneous determination of six common organic acids and four inorganic anions using a stationary phase of the CTMA-coated graphite carbon column and a mobile phase with benzoate was demonstrated. This method showed excel-
309
lent linearity, reproducibility, lower detection limits. Therefore, it was applied to the determination of organic acids in food samples such as nutritious, vinegar and yogurt without the matrix effect. The graphite carbon column has been used for six years, and the excellent durability of the column was confirmed in all our experiments. Acknowledgements We appreciate for the support of Kenji Nakamura and Kazue Ochi (Department of Materials and Applied Chemistry, College of Science and Technology, Nihon University) in this work. References [1] M.C. Bruzzoniti, E. Mentasti, C. Sarzanini, Anal. Chim. Acta 382 (1999) 291. [2] C.L. Guan, J. Ouyang, Q.L. Li, B.H. Liu, W.R.G. Baeyens, Talanta 50 (2000) 1197. [3] S.D. Kumar, B. Maiti, P.K. Mathur, Talanta 53 (2001) 701. [4] A. Alcazar, P.L. Fernandez-Caceres, M.J. Martin, F. Pablos, A.G. Gonzalez, Talanta 61 (2003) 95. [5] R. Garc´ıa-Fern´andez, J.I. Garc´ıa-Alonso, A. Sanz-Medel, J. Chromatogr. A 1033 (2004) 127. [6] M.M. McDowell, M.M. Ivey, M.E. Lee, V.V.V.D. Firpo, T.M. Salmassi, C.S. Khachikian, K.L. Foster, J. Chromatogr. A 1039 (2004) 105. [7] A.A. Krokidis, N.C. Megoulas, M.A. Koupparis, Anal. Chim. Acta 535 (2005) 57. [8] R. Lin, B. De Borba, K. Srinivasan, A. Woodruff, C. Pohl, Anal. Chim. Acta 567 (2006) 135. [9] B. Zhu, Z. Zhong, J. Yao, J. Chromatogr. A 1118 (2006) 106. [10] J. Qiu, X. Jin, J. Chromatogr. A 950 (2002) 81. [11] J.J. Kirkland, J. Chromatogr. A 1060 (2004) 9. [12] K. Horvath, P. Hajos, J. Chromatogr. A 1104 (2006) 75. [13] J.S. Fritz, Z. Yan, P.R. Haddad, J. Chromatogr. A 997 (2003) 21. [14] S. Pelletier, C.A. Lucy, J. Chromatogr. A 1125 (2006) 189. [15] T. Hanai, J. Chromatogr. A 989 (2003) 183. [16] J.H. Knox, B. Kaur, High-Performance Liquid Chromatography, J. Wiley & Sons, New York, 1989, p. 189. [17] J.H. Knox, Q.H. Wan, Chromatographia 42 (1996) 83. [18] J.L. Gundersen, J. Chromatogr. A 914 (2001) 161. [19] Q.H. Wan, P.N. Shaw, M.C. Davies, D.A. Barrett, J. Chromatogr. A 765 (1997) 187. [20] J. Rosen, K.E. Hellenas, The Analyst 127 (2002) 880. [21] E. Forgacs, T. Cserhati, J. Chromatogr. B 681 (1996) 197. [22] I.D. Wilson, Chromatographia 52 (1996) S28. [23] Y. Inamoto, S. Inamoto, T. Hanai, M. Tokuda, O. Hatase, K. Yoshi, N. Sugiyama, T. Kinoshita, Biomed. Chromatogr. 12 (1998) 239. [24] A. Karlsson, M. Berglin, C. Charron, J. Chromatogr. A 797 (1998) 75. [25] P. Chaimbault, C. Elfakir, M. Lafosse, J. Chromatogr. A 797 (1998) 83. [26] T. Okamoto, A. Isozaki, H. Nagashima, J. Chromatogr. A 800 (1998) 239. [27] H. Nagashima, T. Okamoto, J. Chromatogr. A 855 (1999) 261. [28] P. Hajos, O. Horvath, V. Denke, Anal. Chem. 67 (1995) 434.
Short communication
Ion chromatographic determination of organic acids in food samples using a permanent coating graphite carbon column Kenji Yoshikawa ∗ , Miho Okamura, Miki Inokuchi, Akio Sakuragawa Department of Materials and Applied Chemistry, College of Science and Technology, Nihon University, 1-8-14 Kanda-Surugadai, Chiyoda-ku, Tokyo 101-8308, Japan Received 1 September 2006; received in revised form 14 October 2006; accepted 14 October 2006 Available online 22 November 2006
Abstract From the viewpoint of a graphite carbon column with excellent durability, it was applied to the ion chromatography (IC) of several organic acids. The carbon column was permanently coated with the cetyltrimethylammonium (CTMA) ion, and the elution behaviors of several organic acids (acetic acid, lactic acid, succinic acid, malic acid, tartaric acid, citric acid) and inorganic anions (Cl− , NO2 − , NO3 − , SO4 2− ) were examined according to a non-suppressed IC coupled with conductivity detector, when an ion-exchange ability was given to the graphite carbon column. When salicylic acid and sodium salicylate were used as a mobile phase, each organic acid are analyzed approximately 10 min. But the separation of malic acid, chloride and nitrite was difficult. When benzoic acid and 2-amino-2-hydroxymethyl-1,3-puropanediol (tris aminomethane) were used as a mobile phase, tartaric acid and citric acid, etc. with large valency showed tendency to which the width of each peaks extended and retention time increased. However, it was possible to separate excellently for the analytes detected within 10 min. The developed method was then applied to the determination of organic acids in several food samples. © 2006 Elsevier B.V. All rights reserved. Keywords: Graphite carbon column; Organic acid; Cetyltrimethylammonium; Benzoic acid; Ion chromatography
1. Introduction Among the wide variety of methods available for determination of ions, ion chromatography (IC) based on ion-exchangers and conductivity detection has been extensively studied [1–9]. In general, organic polymer and silica-based porous ionexchangers have been used as column packing for liquid chromatography [10–12]. Organic polymer-based ion-exchangers can be used a mobile phase of wide-ranging pH. But organic solvent and pressure resistance present any problems. On the other hand, silica-based ion-exchangers such as an ODS (C18 ) column are generally used with an ion-interaction reagent to achieve separation of ions [13,14]. However, because the silica base is dissolved under the alkaline conditions, it may be affected ion-interaction reagent, i.e. alkaline surfactant. Moreover, It is difficult to use an ODS column for a long time.
∗
Corresponding author. Tel.: +81 3 3259 0801; fax: +81 3 3293 7572. E-mail address: [email protected] (K. Yoshikawa).
0039-9140/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2006.10.020
A graphite carbon column is an alternative non-polar and inactive as stationary phase for liquid chromatography, and it has excellent chemical and physical resistance [15]. Therefore, the buffer solution in a wide-ranging pH (1–14) can be used as a mobile phase. Graphite carbon is used not only an absorbent of gas and organic matter but also a packing material for liquid chromatography, and used to analyze aromatic compounds [16–18], enantiomer of amino acids [19], element of food [20] and drug [21,22]. Though a graphite carbon column has some superior characteristics other than general-purpose column such as a ODS column, etc., it is not so widespread [23–25]. In recent years, applications of the IC that used the ion interaction chemical reagent for the graphitie carbon column that doesn’t have ion exchange ability are reported. These techniques are mainly based on two different methods. One is dynamic coating method [17,26] that added ion interaction reagent such as tetrabutylammonium hydroxide (TBA-OH), polyethylene imine (PEI), etc. in a mobile phase. In this method, it is not necessary to coat an ion interaction reagent on the column surface. However, in the case of non-suppressed IC that we applied this time, because a conductivity of the ion interaction reagent
306
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
Table 1 Operating conditions for some organic acids Column Mobile phase
Carbon IC BI-01 (100 mm × 4.6 mm i.d.) (1) 0.35 mM salicilic acid–0.l mM sodium salicylate (pH 3.5) (2) 2.0 mM benzoic acid–1.2 mM tris aminomethane (pH 4.4)
Flow rate
(1) 0.8 ml/min (2) 1.0 ml/min
Column temperature Injection volume Detection
40 ◦ C 100 l Conductivity
above-mentioned included in a mobile phase is high, more sensitive analysis is difficult, and a baseline is unstable. The other is permanent coating method [27] that coated the quaternary ammonium salt as an ion exchange site on the graphite carbon surface. In this method, it takes the time to coat an ion interaction reagent on the column surface. However, because it is not necessary to add an ion interaction reagent in the mobile phase, it is possible to apply the suppressed and non-suppressed IC. Nagashima et al. [27] have also reported a similar method for inorganic anions by suppressed IC. But, there were only a few reports on simultaneous determination of organic acids by using a permanent coating graphite carbon column. In this work, cetyltrimethylammonium (CTMA)-coated graphite carbon column was used to separate six common organic acids (acetic acid, lactic acid, succinic acid, malic acid, tartaric acid, citric acid). These organic acids were directly detected by non-suppressed IC coupled with conductivity detector. Moreover, the separation behaviors of four inorganic anions (Cl− , NO2 − , NO3 − , SO4 2− ) that included in food sample were examined. The objective of this work was to establish optimal conditions of simultaneous determination of organic acids and apply to the determination of organic acids in several food samples. 2. Experimental 2.1. Apparatus The ion chromatographic equipment consisted of a Model ICA-5120 pump, a Model ICA-5220 conductivity detector, a Model ICA-5410 column oven (Toa DKK, Tokyo, Japan), and a Model 21 chromatocorder (System Instruments, Tokyo, Japan). The separation column was a Carbon IC BI-01 (100 mm × 4.6 mm i.d., average particle size 3.0 m, carbon content larger than 99.5%, Bio-Tech Research, Saitama, Japan).
used as a standard solution of organic acids. Sodium chloride, sodium nitrite, sodium nitrate and sodium sulfide (Wako Pure Chemical, Osaka, Japan) were used as a standard solution of inorganic anions. To coat the graphite carbon column, acetonitrile (Kanto Chemical, Tokyo, Japan) and cetyltrimethylammonium chloride (CTMA-Cl: Wako Pure Chemical, Osaka, Japan) was used. Deionized water, further purified with a Millipore Milli-Q system (Bedford, MA, USA) with a specific resistance of 18.2 M. 2.3. Coating method The column was at first washed with 50 and 100% (v/v) acetonitrile each for 1 h at 0.8 ml/min, and then washed with deionized water for 1 h at 0.8 ml/min. To coat the column, 2.0 mM CTMA-Cl was passed through the column at 0.8 ml/min for 2 h. After dynamically coating, the column was washed with deionized water for 2 h, and used as a permanent coating graphite carbon column. The coating carbon column is stable and available for 3 months. In the permanent coating method, normal chain alkyl group that has hydrophobic part of the quaternary ammonium salt has been adsorbed by the surface of carbon that is hydrophobic particle. Moreover, because the polarity part of quaternary ammonium salt faces the mobile phase, ion exchange and separates ions were performed. Although main separation mechanism is ion exchange, hydrophobic interaction and distribution interaction of the stationary phase also take part. 2.4. Operating conditions The operating conditions for ion chromatography are shown in Table 1. 0.35 mM salicylic acid–0.1 mM sodium salicylate (flow rate, 0.8 ml/min) was used as a mobile phase with salicylate. 2.0 mM benzoic acid–1.2 mM tris aminomethane (flow rate, 1.0 ml/min) was used as a mobile phase with benzoate.
2.2. Reagents 3. Results and discussion Salicylic acid and sodium salicylate (Wako Pure Chemical, Osaka, Japan) were used as a mobile phase with salicylate. Benzoic acid and 2-amino-2-hydroxymethyl-1,3-puropanediol (tris aminomethane: Wako Pure Chemical, Osaka, Japan) were used as a mobile phase with benzoate. Acetic acid, lactic acid, succinic acid, malic acid, tartaric acid and citric acid (Wako Pure Chemical, Osaka, Japan) were
3.1. Mobile phase with salicylate When analyzed organic acids and inorganic anions by nonsuppressed IC, aromatic compounds are usually used as a mobile phase. At first, salicylic acid and sodium salicylate were used. A typical chromatogram of six organic acids and four inorganic
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
307
Fig. 2. Effect of tris aminomethane concentration on retention time (benzoic acid concentration fixed 2.0 mM). Ions (mg l−1 ): acetic (50); 䊉 lactic (20); Cl− (5); NO2 − (5); succinic (50); NO3 − (5); malic (50); ♦ tartaric (50); citric (50); × SO4 2− (20); column: Carbon IC BI-01; mobile phase: 2.0 mM benzoic acid–tris aminomethane; column temperature: 40 ◦ C.
Fig. 1. Chromatogram of standard mixture. Ions (mg l−1 ): acetic (50); lactic (20); succinic (50); NO2 − (5); malic (50); Cl− (5); NO3 − (5); tartaric (50); citric (50); SO4 2− (20); column: Carbon IC BI-01; mobile phase: (a) 0.35 mM salicylic acid–0.1 mM sodium salicylate (pH 3.5, flow rate 0.8 ml/min); (b) 2.0 mM benzoic acid–1.2 mM tris aminomethane (pH 4.4, flow rate 1.0 ml/min); column temperature: 40 ◦ C.
anions is shown in Fig. 1(a)). Each organic acid and inorganic anion except a sulfite is detected nearly 10 min. But the separation of malic acid, chloride and nitrite was difficult. Although the mobile phase concentration changed, simultaneous analysis of the organic acids and inorganic anions were difficult in a mobile phase with salicylate. 3.2. Mobile phase with benzoate 3.2.1. Effect of benzoic acid and tris aminomethane concentration In order to achieve a separation of three kinds of ions (malic acid, chloride, nitrite), benzoic acid and tris aminomethane were used as a mobile phase with benzoate. Effect of tris aminomethane concentration on the retention behavior was examined with the concentration of benzoic acid fixed at 2.0 mM. The relationship between the retention time of analyte and the concentration of tris aminomethane is shown in Fig. 2. Although the retention time of most analytes steady or decrease with increasing tris aminomethane concentration, citric acid showed opposite tendency. As the tris aminomethane concentration increases (pH of a mobile phase increased), dissociation of citric acid advances. As a result, the optimum tris aminomethane concentration was determined to be 1.2 mM, based on the resolution and retention time for each analytes. Furthermore, effect of benzoic acid concentration on the retention behavior was examined with the concentration of tris aminomethane fixed at 1.2 mM. As the concentration of benzoic acid increased, retention time decreases so that dissociation of
each organic acid retrogresses. As a result, the optimum benzoic acid concentration was determined to be 2.0 mM for the same reasons as for tris aminomethane. Moreover, mobile phase concentration rate fixed (benzoic acid:tris aminomethane = 2.0:1.2), the relationships between the logarithm of the total concentration of mobile phase [E] and the capacity factor [k ] are shown in Fig. 3. In the ion-exchange mode, the following relational formula (1) is obtained between the concentration of mobile phase [E] and capacity factor [k ]: y Log k = − log [E] + const. (1) x In this formula, x shows the valency of mobile phase ion, y shows the valency of analyte ion and −(y/x) shows the slope of the straight line [28]. As shown in Fig. 3, in this condition (pH 4.4), the slope of the straight line is negative. The slope of monova-
Fig. 3. Relationship between capacity factor (k ) and total concentration of mobile phase. Ions (mg l−1 ): acetic (50); 䊉 lactic (20); Cl− (5); NO2 − (5); succinic (50); NO3 − (5); malic (50); ♦ tartaric (50); citric (50); column: Carbon IC BI-01; mobile phase concentration rate: benzoic acid:tris aminomethane = 2.0:1.2 (pH 4.4); log k = (−y/x)log[E] + const., where k is the capacity factor, x the valency of mobile phase ion, y the valency of analyte ion and E is the total concentration of mobile phase.
308
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
Table 2 Analytical precision of organic acids Organic acid
Acetic Lactic Succinic Malic Tartaric Citric
Linearity
Detection limit (mg l−1 ) (S/N = 3)
Reproducibility
Range (mg l−1 )
r2 (n = 6)
mg l−1
%R.S.D. (n = 6)
10–200 10–100 10–200 20–120 30–200 40–200
0.9992 0.9997 0.9999 0.9990 0.9991 0.9998
50 50 50 50 50 50
0.46 0.40 0.40 0.73 0.84 0.75
1.2 2.3 2.9 4.5 2.8 3.8
R.S.D.: relative standard deviation.
lent inorganic anions (Cl− , NO2 − and NO3 − ) is approximately −1. However, because the slope of organic acids (especially malic acid and tartaric acid) is between −2 and −1, the separation mode could not be decided that the ion-exchange mode was predominant indiscriminately. As a result, the separation mechanism of the organic acid by the permanent coating carbon column is considered that other modes like the ion exclusion, polarity interaction etc. also take part besides the ion exchange mode. So that the dissociation constant of each organic acid may greatly influence the separation, it is important to control the separation behavior of the analysis, and know the separation mode of the column to understand the dissociation of the organic acid. 3.2.2. Effect of column temperature and flow rate Under the composition of the mobile phase, effect of column temperature on the retention behavior was examined from 30 to 50 ◦ C. Although the retention time of monovalent ions decreased with increasing temperature, divalent ions showed opposite tendency. The optimum column temperature was determined to be
Fig. 4. Chromatogram of organic acids and inorganic anions in nutritions drink sample. Ions (mg l−1 ): acetic; lactic; Cl− ; succinic; NO3 − ; malic; citric; column: Carbon IC BI-01; mobile phase: 2.0 mM benzoic acid–1.2 mM tris aminomethane (pH 4.4); column temperature: 40 ◦ C.
Table 3 Analytical results for organic acids in food sample Sample
Organic acid
Dil. found (mg l−1 )
Added (mg l−1 )
Found (mg l−1 )
Recovery (%)
Concentration (mg l−1 )
Nutoritions drink (1)
Acetica Lactica Succinica Malica Citricb
27.5 23.2 – 18.8 158.9
50.0 30.0 – 50.0 30.0
70.8 50.6 – 68.9 188.4
91.4 91.2 – 100.2 99.7
110.0 92.8
Nutoritions drink (2)
Aceticc Lacticc Succinicd Citricc
6.2 73.8 104.2 82.0
50.0 30.0 30.0 50.0
49.4 95.8 134.3 139.8
88.8 92.3 100.1 105.9
12.5 147.7 2083.8 163.9
Moromi vinegar
Acetice Succinice Malice Citrice
64.5 42.4 27.7 170.6
50.0 50.0 50.0 30.0
111.2 91.8 72.1 189.6
97.1 99.3 92.8 94.5
2065.4 1358.4 886.2 5460.0
Yogurt
Lacticb Succinica Citrica
42.7 – 90.8
50.0 – 50.0
79.6 – 145.0
85.9 – 103.0
2134.5
DL: determination limit. a 1/4 dil. b 1/50 dil. c 1/2 dil. d 1/20 dil. e 1/32 dil.
K. Yoshikawa et al. / Talanta 72 (2007) 305–309
40 ◦ C, based on the resolution and retention time for each analytes. Moreover, the optimum flow rate of mobile phase was determined to be 1.0 ml/min for the same reasons as for column temperature. 3.2.3. Chromatogram of organic acids and inorganic anions A typical chromatogram of six organic acids and four inorganic anions is shown in Fig. 1(b). Because eluting power is weaker to a mobile phase with benzoate than a mobile phase with salicylate, tartaric acid and citric acid, etc. with large valency showed tendency to which the width of each peaks extended and retention time increased. But it was possible to separate excellently for the analytes detected within 10 min. Especially, the separation of the Cl− , NO2 − and malic acid became possible. 3.2.4. Analytical precision The linearity of calibration curves, reproducibility and detection limit are shown in Table 2. The calibration curves obtained from the peak areas for the organic acids were linear, with good correlation coefficients of 0.999. The relative standard deviations (R.S.D.) of peak areas were between 0.40 and 0.84% for six repeated measurements. The detection limits, summarized in Table 2, were calculated at the S/N ratio of 3. The detection limits of organic acids were between 1.2 and 4.5 mg/l. 3.2.5. Application to food samples The proposed method was applied to analyze organic acids in food samples. Yogurt sample was prepared by adding 5 g of it in 20 ml of deionized water and then mixing in 20 min. After the sample solution had been centrifuged in 2000 rpm, 100 l was injected into IC. Chromatogram of nutritions drink sample is shown in Fig. 4. Analytical results of organic acids in the other food sample were shown in Table 3. In addition, to confirm the reliability of the method, organic acids of each concentration is added to the food sample, and recovery test was demonstrated (shown in Table 3). Because the recovery showed nearly 100%, this method can be applied to the determination of organic acids without the matrix effect. 4. Conclusion In this work, simultaneous determination of six common organic acids and four inorganic anions using a stationary phase of the CTMA-coated graphite carbon column and a mobile phase with benzoate was demonstrated. This method showed excel-
309
lent linearity, reproducibility, lower detection limits. Therefore, it was applied to the determination of organic acids in food samples such as nutritious, vinegar and yogurt without the matrix effect. The graphite carbon column has been used for six years, and the excellent durability of the column was confirmed in all our experiments. Acknowledgements We appreciate for the support of Kenji Nakamura and Kazue Ochi (Department of Materials and Applied Chemistry, College of Science and Technology, Nihon University) in this work. References [1] M.C. Bruzzoniti, E. Mentasti, C. Sarzanini, Anal. Chim. Acta 382 (1999) 291. [2] C.L. Guan, J. Ouyang, Q.L. Li, B.H. Liu, W.R.G. Baeyens, Talanta 50 (2000) 1197. [3] S.D. Kumar, B. Maiti, P.K. Mathur, Talanta 53 (2001) 701. [4] A. Alcazar, P.L. Fernandez-Caceres, M.J. Martin, F. Pablos, A.G. Gonzalez, Talanta 61 (2003) 95. [5] R. Garc´ıa-Fern´andez, J.I. Garc´ıa-Alonso, A. Sanz-Medel, J. Chromatogr. A 1033 (2004) 127. [6] M.M. McDowell, M.M. Ivey, M.E. Lee, V.V.V.D. Firpo, T.M. Salmassi, C.S. Khachikian, K.L. Foster, J. Chromatogr. A 1039 (2004) 105. [7] A.A. Krokidis, N.C. Megoulas, M.A. Koupparis, Anal. Chim. Acta 535 (2005) 57. [8] R. Lin, B. De Borba, K. Srinivasan, A. Woodruff, C. Pohl, Anal. Chim. Acta 567 (2006) 135. [9] B. Zhu, Z. Zhong, J. Yao, J. Chromatogr. A 1118 (2006) 106. [10] J. Qiu, X. Jin, J. Chromatogr. A 950 (2002) 81. [11] J.J. Kirkland, J. Chromatogr. A 1060 (2004) 9. [12] K. Horvath, P. Hajos, J. Chromatogr. A 1104 (2006) 75. [13] J.S. Fritz, Z. Yan, P.R. Haddad, J. Chromatogr. A 997 (2003) 21. [14] S. Pelletier, C.A. Lucy, J. Chromatogr. A 1125 (2006) 189. [15] T. Hanai, J. Chromatogr. A 989 (2003) 183. [16] J.H. Knox, B. Kaur, High-Performance Liquid Chromatography, J. Wiley & Sons, New York, 1989, p. 189. [17] J.H. Knox, Q.H. Wan, Chromatographia 42 (1996) 83. [18] J.L. Gundersen, J. Chromatogr. A 914 (2001) 161. [19] Q.H. Wan, P.N. Shaw, M.C. Davies, D.A. Barrett, J. Chromatogr. A 765 (1997) 187. [20] J. Rosen, K.E. Hellenas, The Analyst 127 (2002) 880. [21] E. Forgacs, T. Cserhati, J. Chromatogr. B 681 (1996) 197. [22] I.D. Wilson, Chromatographia 52 (1996) S28. [23] Y. Inamoto, S. Inamoto, T. Hanai, M. Tokuda, O. Hatase, K. Yoshi, N. Sugiyama, T. Kinoshita, Biomed. Chromatogr. 12 (1998) 239. [24] A. Karlsson, M. Berglin, C. Charron, J. Chromatogr. A 797 (1998) 75. [25] P. Chaimbault, C. Elfakir, M. Lafosse, J. Chromatogr. A 797 (1998) 83. [26] T. Okamoto, A. Isozaki, H. Nagashima, J. Chromatogr. A 800 (1998) 239. [27] H. Nagashima, T. Okamoto, J. Chromatogr. A 855 (1999) 261. [28] P. Hajos, O. Horvath, V. Denke, Anal. Chem. 67 (1995) 434.