Vacancy ion-exclusion chromatography of inorganic acids on a weakly acidic cation-exchange resin column

Journal of Chromatography A, 1118 (2006) 41–45

Vacancy ion-exclusion chromatography of inorganic acids on a weakly acidic cation-exchange resin column Masanobu Mori a,∗ , Hideyuki Itabashi a , Murad I.H. Helaleh b , Krzysztof Kaczmarski c , Bronisław Gł´od d , Teresa Kowalska e , Qun Xu f , Mikaru Ikedo g , Wenzhi Hu h , Kazuhiko Tanaka i a Faculty of Engineering, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma, Japan Faculty of Chemistry, Technical University of Rzesz´ow, Al Powsta´nc´ow Warszawy 6, 35-959 Rzesz´ow, Poland c Kuwait Institute for Scientific Research (KISR), Central Analytical Laboratory (CAL), P.O. Box 24885, Safat 13109, Kuwait d Meat and Fat Research Institute, Warsaw, Poland e Institute of Chemistry, Silesian University, 9 Szkolna Street, 40-006 Katowice, Poland f Dionex China Ltd., Room 2311, Huaihai Zhong Road, Shanghai 200021, China g Graduate School of Engineering, Chubu University, Kasugai, Aichi 487-8501, Japan h Graduate School of Environmental Science, Hokkaido University, Sapporo 060-0810, Japan i Graduate School for International Development and Cooperation, Hiroshima University 1-5-1 Kagamiyama, Higashi-Hiroshima 739-8529, Japan b

Available online 2 February 2006

Abstract Vacancy ion-exclusion chromatography (VIEC) for inorganic acids such as H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF is tested on a polymethacrylate-based weakly acidic cation-exchange resin column in the H+ -form. That is, mixture of inorganic acids in the mobile phase is adsorbed to the resin phase passing through the separation column, and each vacant peak induced by injecting water is determined. Retention times are dependent on the degrees of retention for each analyte in the resin phase. In VIEC, well-shaped peaks of inorganic acids are produced, leading to efficient separations. However, retention behaviors of inorganic acids were strongly affected by the concentrations of the acids in the mobile phase. Sulfosalicylic acid was mixed with inorganic acids in the mobile phase prior to the introduction of a separation column in order to obtain the well-resolutions in the lower concentrations of the acids. By using this method, the separations of inorganic acids could be achieved in the range of 0.01–1 mM, and the linear ranges could be extended over two-orders of magnitude. This is considered since the protonated carboxylic groups fixed on the resin phase were increased with increasing the acid concentrations in the mobile phase, and the penetration effects for the acids to the resin phase were thus enhanced. The detection limits (S/N = 3) were below 1.0 ␮M for all analyte acids. Precision values for retention times were below 0.32% and for peak area were below 0.91%. © 2006 Elsevier B.V. All rights reserved. Keywords: Vacant peak; Ion-exclusion chromatography; A weakly acidic cation-exchange resin; Inorganic acids

1. Introduction A vacant peak (i.e., system peak or elution dip) in ion chromatography with conductivity detection is induced when a diluted acid, base or salt is added in the mobile phase in order to improve the resolutions of analytes injected to a separation column. The retention time, peak shape and/or area of vacant peak are dependent on the kind or the concentration of a compound added in the mobile phase.

∗

Corresponding author. Tel.: +81 277 30 1271; fax: +81 277 30 1271. E-mail address: m [email protected] (M. Mori).

0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.01.017

The retention behavior of vacant peak has been often studied in the field of chromatography [1–5]. In ion chromatography, these studies are mainly aimed to establish the removal method of vacant peak in order to solve several analytical problems, e.g., the decrease of conductivity response of analyte, and/or the overlapping of analyte peaks and a vacant peak [4,5]. Recently, a novel basic concept of ion-exclusion chromatography for determining the vacant peak as analytes has been reported. This is called vacancy ion-exclusion chromatography (VIEC) [6–11]. In VIEC, a mixture of analytes is used as a mobile phase. The direction of peak with conductivity is negative when pure water is injected to separation column. Negative

42

M. Mori et al. / J. Chromatogr. A 1118 (2006) 41–45

vacant peak appears corresponding to each acid in the mobile phase, leading to well-shaped peaks. VIEC has been applied mainly to partially ionized species such as aliphatic carboxylic acids [6–8,11], and aromatic carboxylic acids [9]. The partially ionized species are converted to molecular form with increasing the concentrations of analyte acids in the mobile phase, and thus, the molecular formed acids are easily penetrated on the resin phase, depending on their hydrophobic adsorptions. Consequently, the peak shape and the resolutions of weak acids in VIEC are better than those of conventional ion-exclusion chromatography. In this study, VIEC for inorganic acids such as H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF on an acidic cation-exchange resin column in the H+ -form was tested. As a separation column used in this study, a polymethacrylate-based weakly acidic cation-exchange resin column (e.g., TSKgel-OApak A) is recommended rather than a polystyrene–divinylbenzene-based strongly acidic cation-exchange resin column (e.g., TSKgelSCX). This is because the retention behaviors of analyte acids can be manipulated by the degree of the dissociation/protonation of ionic functional groups (–COOH) fixed on the resin phase with the concentration of the acids in the mobile phase. Also, in conventional ion-exclusion chromatography, inorganic acids such as H2 SO4 , HCl and HNO3 can be separated on a polymethacrylate-based weakly acidic cation-exchange resin with the mobile phase including tartaric acid or sulfosalicylic acid [12–14]. Unfortunately, the separation of H3 PO4 from strong acids in conventional ion-exclusion chromatography could not be achieved due to their tailing peaks and their poor resolutions. We report on the features of VIEC of inorganic acids in following terms: (1) comparison of VIEC and conventional ionexclusion chromatography on the retention behaviors of inorganic acids; (2) comparison of a weakly acidic and a strongly acidic cation-exchange resin columns used in VIEC of inorganic acids; and (3) the improvement of the peak resolutions of inorganic acids by introducing the online pre-column reaction.

diluted with water as necessary. The composition of the mobile phase in this report was the mixture of H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF. 2.3. Procedures The experimental conditions for ion chromatography were the following: column temperature 40 ◦ C; eluent flow-rate 0.6 ml/min; and injection volume 30 ␮l. The injected sample was water. A mobile phase including a mixture of inorganic acids was equilibrated thoroughly with the separation column until a stable eluent background conductivity level was obtained in the range of ca. 30–60 min, depending on the kinds of mobile phases. 3. Results and discussion 3.1. Ion-exclusion chromatography of inorganic acids In conventional ion-exclusion chromatography of inorganic acids on a polymethacrylate-based weakly acidic cationexchange resin, an overlapped peak for each acid except for HF was obtained using water as a mobile phase, as shown in Fig. 1A. This was because analyte acids were repelled by ionized functional groups (–COO− ) fixed on the resin phase at pH 5.6 of water and were passed through quickly. When a hydrophobic strong acid such as sulfosalicylic acid was used as a mobile phase, an improved separation was obtained as shown in Fig. 1B. The vacant peak caused by the adsorption of the sulfosalicylic acid in the mobile phase was clearly evident at a retention time of ca. 12 min. However, the complete separation of HCl, H3 PO4 and HNO3 could not be obtained even after increasing concentrations of sulfosalicylic acid in a mobile phase. This would be mainly due to insufficient adsorption and desorption of the acids to cation-exchanger (–COOH) on the resin phase [11]. 3.2. Vacancy ion-exclusion chromatography of inorganic acids

2. Experimental 2.1. Instrumentations Ion chromatographic separation was performed on a Tosoh Model IC-2001 equipped with the following devices: vacuum degasser, pump, column oven, auto-sampler, and conductivity detector (Tosoh, Tokyo, Japan). A polymethacrylatebased weakly acidic cation-exchange resin column in the H+ -form, Tosoh TSKgel Super-IC-A/C (3 ␮m particle size, 0.1 mequiv./ml-capacity, 30 cm × 6 mm I.D.), was used as the separation column. 2.2. Reagents All reagents were of analytical reagent-grade, purchased from Wako (Osaka, Japan). The preparations of the standard solutions were dissolved with distilled and deionized water. Appropriate amounts of analyte samples at the concentration of 0.1 M were

The peak resolutions of the inorganic acids were largely improved by using VIEC on a weakly acidic cation-exchange resin, which was comprised of a mixture of six inorganic acids used as a mobile phase and water as an injection sample, as shown in Fig. 2. Negative vacant peaks corresponding to each acid presented in the mobile phase appeared, leading to wellshaped and well-resolved peaks. The identities of these peaks were confirmed by using the mobile phase containing each acid separately. The separation of H3 PO4 from other strong acids was also achieved due to the improvement of the peak shapes. In contrast, when a polystyrene–divinylbenzene-based strongly acidic cation-exchange resin column (TSKgel SCX: 30 cm × 7.8 mm I.D.; 5 ␮m-particle and 1.5 eq./l-capacity) was used under the same conditions as shown in Fig. 2, a single negative peak was obtained (not shown data). These results imply that VIEC of inorganic acids is dependent on the negatively charged functional groups fixed on resin phase. In a weakly acidic cation-exchange resin column, the pen-

M. Mori et al. / J. Chromatogr. A 1118 (2006) 41–45

43

Fig. 1. Ion-exclusion chromatograms of six inorganic acids on a weakly acidic cation-exchange column with mobile phase of (A) water and (B) 2 mM sulfosalicylic acid. Column: TSKgel Super-IC-A/C (30 cm × 6 mm I.D.); injected sample: a mixture of H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF (0.5 mM for each); injection volume: 30 ␮l; column oven: 40 ◦ C; flow rate: 0.6 ml/min. Peaks: (1) H2 SO4 , (2) HCl, (3) H3 PO4 , (4) HNO3 , (5) HI, (6) HF and (7) vacant peak of sulfosalicylic acid.

etrations of analyte acids to the resin phase are accelerated with increasing the concentration of the protonated functional groups (e.g., –COOH) on the resin surface below pH 3–4. The differences of retention volumes between analyte acids are consequently enhanced. The symmetric peaks of analyte acids would be induced because of the balance between adsorption and desorption of acids to the resin phase with sufficient equilibrium of

the mobile phase. In a strongly acidic cation-exchange resin column, the differences of penetrations for each acid are minimized by ionic repulsion with dissociated functional groups (SO3 − ) on the resin phase in the wide range of pH. 3.3. Improvement of retention behaviors of inorganic acids in VIEC using sulfosalicylic acid as a pre-column reaction reagent The retention volumes of inorganic acids (H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF) in VIEC on a weakly acidic cation-exchange resin column were investigated in different acid concentrations in the mobile phases as well as their pH values. The retention volumes of inorganic acids were strongly affected with increases in acid concentrations in the mobile phases, as shown in Fig. 3. The acids were not almost

Fig. 2. Vacancy ion-exclusion chromatograms of six inorganic acids on a weakly acidic cation-exchange column. Conditions: column: TSKgel Super-IC-A/C (30 cm × 6 mm I.D.); mobile phase: a mixture of H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF (0.5 mM for each); injected sample: water; injection volume: 30 ␮l; column oven: 40 ◦ C; flow rate: 0.6 ml/min. Other conditions are same as in Fig. 1. Peaks: (1) H2 SO4 , (2) HCl, (3) H3 PO4 , (4) HNO3 , (5) HI and (6) HF.

Fig. 3. Retention volumes of inorganic acids against different concentrations of acids in the mobile phase on VIEC. Conditions: a composition of mobile phase: H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF; injected sample: water; range of each acid concentration: 0.04–1 mM. Plot identities: (
) H2 SO4 , () HCl, () H3 PO4 , () HNO3 , () HI, (♦) HF and (×) pH. Other conditions are same as in Fig. 2.

44

M. Mori et al. / J. Chromatogr. A 1118 (2006) 41–45

Fig. 6. Retention volumes of analyte acids against different concentrations of inorganic acid on VIEC with sulfosalicylic acid. Flow rate: 0.3 ml/min for both pumps 1 and 2. Plot identities: (
) H2 SO4 , () HCl, () H3 PO4 , () HNO3 , () HI, (♦) HF and (×) pH. Other conditions as in Fig. 2. Fig. 4. Schematic illustration of VIEC with pre-column reaction system.

separated when the concentration of each acid was below 0.2 mM. If this VIEC system is applied to real water samples, wellresolved peaks of inorganic acids must be required in the wide concentration ranges. As a solution to this problem, the precolumn reaction system was established as shown in Fig. 4. This was assembled so that the mobile phase eluted from pump 1 could be mixed with sulfosalicylic acid from pump 2. The acid concentration of the solution mixed just before the separation column was increased. Accordingly, the ionic repulsions were weakened with the increase of the protonated cation-exchangers on the resin, and the degrees of penetration for each acid to resin phase were enhanced. The flow rates of both elution pumps of the mobile phase and sulfosalicylic acid were 0.3 ml/min, considering the retention volumes of analyte acids and the pressure to the separation column. The concentration of sul-

fosalicylic acid mixed with the mobile phase including analyte acids was prepared to be below pH 2.5 before introducing to the column. Fig. 5 shows the chromatogram for 0.2 mM of each inorganic acid in the mobile phase with 3 mM sulfosalicylic acid as the pre-column reaction. The low concentration of inorganic acids was obviously improved by VIEC with the pre-column reaction, compared with that without it. Moreover, the constant retention volumes could be obtained in range of 0.04–1 mM for each analyte acid as shown in Fig. 6. 3.4. Analytical performances The calibration graphs of peak areas of inorganic acids in VIEC with pre-column reaction were plotted against their different concentrations. The calibration graphs of all analyte acids were linear in the range of 0.01–1 mM for each. The correlation

Fig. 5. Chromatograms of inorganic acids on VIEC at (A) absence and (B) presence of the pre-column reaction. Condition: pre-column reagents: 3 mM sulfosalicylic acid; inorganic acid in the mobile phases: 0.8 mM (0.2 mM for each); flow rate: (A) 0.6 ml/min of single pump and (B) 0.3 ml/min for both pumps 1 and 2. Peak identities: (1) H2 SO4 , (2) HCl, (3) H3 PO4 , (4) HNO3 , (5) HI, (6) HF and (7) sulfosalicylic acid. Other conditions are same as in Fig. 2.

M. Mori et al. / J. Chromatogr. A 1118 (2006) 41–45 Table 1 Detection limits of analyte acids in the proposed VIECa and conventional ionexclusion chromatographyb Analyte acids

Proposed VIECa (␮M)

Conventional ion-exclusion chromatographyb (␮M)

H2 SO4 HCl H3 PO4 HNO3 HI HF

0.40 0.34 0.41 0.20 0.15 0.56

0.34 0.28 0.93 0.14 0.56 2.13

a The concentration of each acid in the mobile phase was 0.01 mM. The concentration of sulfosalicylic acid as a pre-column reaction reagent was 3.5 mM. Other chromatographic conditions were same as in Fig. 2. b The mobile phase on conventional ion-exclusion chromatography was 2 mM sulfosalicylic acid. Other conditions were same as in Fig. 1.

coefficients (r2 ) for H2 SO4 , HCl, H3 PO4 , HNO3 , HI and HF were in the range of 0.9987–0.9999. The detection limits of all analyte acids calculated at S/N = 3 were below 1.0 ␮M as summarized in Table 1. The detection limits of fully ionized acids such as H2 SO4 , HCl and HNO3 in VIEC were slightly higher than those in conventional ionexclusion chromatography with 2 mM sulfosalicylic acid. The detection limits of HI, HF and H3 PO4 in VIEC were lower than those in conventional ion-exclusion chromatography. This could be due to the enrichment effects in resin phase with the higher adsorptivity for HI, which is a strong acid having lower hydration, and for H3 PO4 and HF, which are easily converted to the molecular form below pH 2.5. The relative standard deviations (RSDs) of the retention times and peak areas of inorganic acids in the proposed VIEC were calculated by running the mobile phases containing 0.01 mM of each acid. The RSDs in the repeated chromatographic runs (n = 11) were 0.12–0.32% for the retention times and 0.26–0.91% for the peak areas. The RSDs of acids (H2 SO4 , HCl and HNO3 ) in ion-exclusion chromatography under conditions as shown in Fig. 1B were 0.2–0.5% for the retention time and 0.9–1.7% for the peak area. Therefore, the RSDs of retention behaviors of analyte acids obtained in VIEC could give better reproducible results than those in ion-exclusion chromatography. 4. Conclusions VIEC of inorganic acids was observed on a weakly acidic cation-exchange column. The retention behaviors of inorganic

45

acids were dependent strongly on total acid concentration in the mobile phase. The well-resolutions of the analyte acids in the wide concentration ranges could be obtained by mixing with sulfosalicylic acid before the introduction of the separation column and by decreasing pH value of the solutions. This method seems to be effective for hydrophobic strong acids such as HI and partially ionized acids such as H3 PO4 and HF, in terms of the improvement of the detection limits and the peak resolutions. Further interest is whether the VIEC system can apply to real water sample. In the case of VIEC, a large amount of real water must be required, and the solutions to be analyzed must be passed through the column continuously for 30–60 min. Therefore, this method might be suitable for routine water-quality monitoring for filtration plant water or a commercialized drinking water containing the known ionic species, rather than the determination of the unknown ionic species in natural water. Further investigations for the application of VIEC to actual samples will be the subject of future work. References [1] C.N. Reilley, G.P. Hildcbrand, J.W. Ashley, Anal. Chem. 34 (1962) 1198. ˇ [2] K. Slais, M. Krejˇc´ı, J. Chromatogr. 24 (1974) 161. [3] K. Veggeland, T. Austad, Colloids Surf. A: Physicochem., Eng. Aspects 76 (1993) 73. [4] A. Caliamanis, M.J. McCormick, P.D. Carpenter, Anal. Chem. 69 (1997) 3272. [5] H.H. Streckert, B.D. Epstein, Anal. Chem. 56 (1984) 21. [6] K. Tanaka, M.-Y. Ding, M.I.H. Helaleh, H. Taoda, H. Takahashi, W. Hu, K. Hasebe, P.R. Haddad, J.S. Fritz, C. Sarzanini, J. Chromatogr. A 956 (2002) 209. [7] K. Tanaka, M.-Y. Ding, H. Takahashi, M.I.H. Helaleh, H. Taoda, W. Hu, K. Hasebe, P.R. Haddad, M. Mori, J.S. Fritz, C. Sarzanini, Anal. Chim. Acta 474 (2002) 31. [8] M.I.H. Helaleh, K. Tanaka, M. Mori, Q. Xu, H. Taoda, M.-Y. Ding, W. Hu, K. Hasebe, P.R. Haddad, J. Chromatogr. A 997 (2003) 133. [9] M.I.H. Helaleh, K. Tanaka, M. Mori, Q. Xu, H. Taoda, M.-Y. Ding, W. Hu, K. Hasebe, P.R. Haddad, J. Chromatogr. A 997 (2003) 139. [10] M. Mori, M.I.H. Helaleh, Q. Xu, W. Hu, M. Ikedo, M.-Y. Ding, H. Taoda, K. Tanaka, J. Chromatogr. A 1039 (2004) 129. [11] K. Kaczmarski,. Mori, B. Glod, T. Kowalska, K. Tanaka, Acta Chromatogr. 15 (2005) 66. [12] J. O’Reilly, P. Doble, K. Tanaka, P.R. Haddad, J. Chromatogr. A 884 (2000) 61. [13] M. Mori, K. Tanaka, M.I.H. Helaleh, Q. Xu, M. Ikedo, Y. Ogura, S. Sato, W. Hu, K. Hasebe, P.R. Haddad, J. Chromatogr. A 997 (2003) 219. [14] K. Tanaka, K. Ohta, P.R. Haddad, J.S. Fritz, A. Miyanaga, W. Hu, K. Hasebe, K.-P. Lee, C. Sarzanini, J. Chromatogr. A 920 (2001) 239.