Preparation and Application of Methacrylate-Based Cation-Exchange

This GMA-based cation-exchange monolithic column could withstand high linear velocities of at ... Methacrylate Polymer Monoliths for Separation Applic...
0 downloads 0 Views 115KB Size
Download PDF
Anal. Chem. 2004, 76, 7007-7012

Preparation and Application of Methacrylate-Based Cation-Exchange Monolithic Columns for Capillary Ion Chromatography Yuji Ueki, Tomonari Umemura,*,† Jinxiang Li, Tamao Odake, and Kin-ichi Tsunoda

Department of Chemistry, Faculty of Engineering Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515, Japan

Polymer-based strong cation-exchange monolithic capillary columns with different capacities were constructed for ion chromatography by radical polymerization of glycidyl methacrylate (GMA) and ethylene dimethacrylate in a 250-µm-i.d. fused-silica capillary and its subsequent sulfonation based on ring opening of epoxides with 1 M Na2SO3. The cation-exchange capacities can easily and reproducibly be controlled in the range of up to 300 µequiv/mL by changing the immersion time of the epoxycontaining polymer in the Na2SO3 solution. The chromatographic performance of the produced monolithic capillary columns was evaluated through the separation of a model mixture of common cations such as Na+, NH4+, K+, Mg2+, and Ca2+. As an example, these cations could be well separated from one another on a 15-cm-long cation-exchange monolithic column (column volume, 7.4 µL) with a capacity of 150 µequiv/mL by elution with 10 mM CuSO4. The pressure drop of this 15-cm column was ∼1 MPa at a normal linear velocity of 1 mm/s (a flow rate of 3 µL/min), and the numbers of theoretical plates for the cations were above 3000 plates/15 cm. This GMAbased cation-exchange monolithic column could withstand high linear velocities of at least 10 mm/s. Over a period of at least two weeks of continuous use, no significant changes in the selectivity and resolution were observed. The applicability of a flow rate gradient elution and the feasibility of direct injection determination of major cations in human saliva sample were also presented. Capillary liquid chromatography (LC) has gained momentum driven by the availability of sophisticated instruments,1-3 and especially, its applications in the fields of pharmaceutical and biochemical research have grown tremendously because of its attractive features such as low sample and reagent consumptions, good compatibility with mass spectrometry, and the possibility of a high degree of automation.4-6 * To whom correspondence should be addressed. E-mail: umemura@ apchem.nagoya-u.ac.jp. Tel.: +81-52-789-5288. Fax: +81-52-789-4665. † Present address: Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan. (1) Gusev, I.; Huang, X.; Horva´th, C. J. Chromatogr., A 1999, 855, 273-290. (2) Tsuda, T.; Novotny, M. Anal. Chem. 1978, 50, 271-275. (3) Takeuchi, T.; Ishii, D. J. Chromatogr. 1980, 190, 150-155. 10.1021/ac040079g CCC: $27.50 Published on Web 11/03/2004

© 2004 American Chemical Society

The growing demand for high-throughput analysis in such genome-related research has promoted the development of highspeed LC techniques and tools.7-9 The straightforward approach for rapid and highly efficient separations is to use short columns packed with smaller particles at the highest flow rate possible. An increase in column temperature is also beneficial to such efficient separations due to the reduced eluent viscosity and the enhanced mass transfer, although its use is limited by thermal stability of analytes and packing materials. To fulfill these requirements for high-throughput analysis, especially designed HPLC systems that can be operated at high pressure and high temperature have been constructed and ultrafast separations on the time scale of seconds are being achieved.10-14 But, this approach required an especially designed setup and is inherently restricted by pressure drop problem. To alleviate the pressure drop problem and to achieve highly efficient separation, considerable efforts have also been dedicated to the preparation and application of monolithic columns having high porosity.1,15-21 Due to the characteristic low flow resistance and the resultant applicability of extremely high flow rate, highspeed separations can be achieved within the pressure limitations of commercially available HPLC equipment. In addition, monolithic (4) Nakayama, H.; Uchida, K.; Shinkai, F.; Shinoda, T.; Okuyama, T.; Seta, K.; Isobe, T. J. Chromatogr., A 1996, 730, 279-287. (5) Oberacher, H.; Huber, C. G. Trends Anal. Chem. 2002, 21, 166-174. (6) Huang, X.; Zhang, S.; Schultz, G. A.; Henion, J. Anal. Chem. 2002, 74, 23362344. (7) Xie, S.; Allington, R. W.; Svec, F.; Fre´chet, J. M. J. J. Chromatogr., A 1999, 865, 169-174. (8) Shen, Y.; Zhao, R.; Belov, M. E.; Conrads, T. P.; Anderson, G. A.; Tang, K.; Pasˇa-Tolic´, L.; Veenstra, T. D.; Lipton, M. S.; Udseth, H. R.; Smith, R. D. Anal. Chem. 2001, 73, 1766-1775. (9) Walcher, W.; Oberacher, H.; Troiani, S.; Ho¨lzl, G.; Oefner, P.; Zolla, L.; Huber, C. G. J. Chromatogr., B 2002, 782, 111-125. (10) Miller, D. J.; Hawthorne, S. B. Anal. Chem. 1997, 69, 623-627. (11) Kephart, T. S.; Dasgupta, P. K. Anal. Chim. Acta 2000, 414, 71-78. (12) Yan, B.; Zhao, J.; Brown, J. S.; Blackwell, J.; Carr, P. W. Anal. Chem. 2000, 72, 1253-1262. (13) MacNair, J. E.; Lewis, K. C.; Jorgenson, J. W. Anal. Chem. 1997, 69, 983989. (14) Chong, J.; Hatsis, P.; Lucy, C. A. J. Chromatogr., A 2003, 997, 161-169. (15) Hjerte´n, S.; Liao, J.-L.; Zhang, R. J. Chromatogr., A 1989, 473, 273-275. (16) Svec, F.; Fre´chet, J. M. J. Anal. Chem. 1992, 64, 820-822. (17) Palm, A.; Novotny, M. V. Anal. Chem. 1997, 69, 4499-4507. (18) Premstaller, A.; Oberacher, H.; Huber, C. G. Anal. Chem. 2000, 72, 43864393. (19) Fujimoto, C. Anal. Chem. 1995, 67, 2050-2053. (20) Fields, S. M. Anal. Chem. 1996, 68, 2709-2712. (21) Minakuchi, H.; Nakanishi, K.; Soga, N.; Ishizuka, N.; Tanaka, N. Anal. Chem. 1996, 68, 3498-3501.

Analytical Chemistry, Vol. 76, No. 23, December 1, 2004 7007

columns have an important advantage over conventional particlepacked columns in terms of simplicity of the construction; monolithic stationary phases are directly produced inside the column tubing by an in situ polymerization technique and then have no need for particle-size classification, the packing, and frit making. Further, conveniently, this in situ technique is favorable for the fabrication of a narrower bore column; the smaller the inner diameter of the column, the more uniform monoliths with sufficient mechanical strength can be produced due to the small reaction volume and the high surface-to-volume ratio in the capillary.22,23 Thus, much attention has been focused on the monolithic technology, and the research and application of monolithic technology have been extended from the original continuous separation media to solid-phase extractors and catalytic microreactors.24 Monolithic columns can be classified into two general categories: organic polymer-based and silica-based columns. The primary practical monoliths were compressed soft polyacrylamide gels developed by Hjerte´n et al. in 1989,15 although the research on monolithic or continuous media had already emerged at the end of the 1960s.25,26 Subsequently, Svec and Fre´chet introduced an entirely new class of rigid macroporous polymer monoliths in the early 1990s.16 Since the extensive work, organic polymer-based monoliths based on polymethacrylates, polystyrenes, or polyacrylamides have been extensively studied and successfully used for chromatographic separations of mainly macromolecules such as proteins, nucleic acids, and polysaccharides.1,17,18 On the other hand, useful inorganic silica-based monoliths for chromatographic applications were later reported in 1996.20,21 Minakuchi et al. introduced a new sol-gel process based on hydrolysis and polycondensation of alkoxysilanes to produce a porous silica monolith with a bimodal pore structure of macropores (throughpores) and mesopores. The sophisticated silica monolithic columns with a uniform skeletal structure provided high permeability and good separation efficiency and have been in widespread use. Both monoliths have been shown to have strengths and weaknesses. Silica-based monoliths cannot generally be used with alkaline eluents due to the chemical instability. In contrast, organic polymer-based monoliths are stable under wide pH conditions but provide poor mechanical stability due to swelling and shrinking in contact with some organic solvents. In addition, the existence of micropores at the polymer surface is another major drawback of polymer-based monoliths, which have an adverse effect on the separation efficiency of small molecules. However, organic polymerbased monoliths have considerable advantages exceeding the drawbacks, including wide pH stability, inertness to biomolecules, absence of deleterious effects from silanol, and facility for modification. Also, organic polymer monoliths can easily be prepared under mild and facile conditions by using readily available inexpensive components such as an oven and a water aspirator. Thus, considerable effort has been dedicated to the

improvement of mechanical strength of polymer-based monoliths, and various promising results of the applications to capillary LC as well as capillary electrochromatography have been demonstrated.24,27 However, there are few applications to the separation of small analytes, especially to ion chromatography (IC),28,29 probably because micropores existing at the polymer surface are strongly recognized to negatively affect column efficiency and peak symmetry for the separation of small molecules. In the present paper, the applicability of organic polymer-based monoliths to capillary IC has been experimentally investigated. Organic polymerbased monolithic capillary columns were prepared by a two-step procedure consisting of polymerization and sulfonation of glycidyl methacrylate (GMA),30,31 and the chromatographic performance was evaluated through the separation of a model mixture of common cations. In addition, some promising results of flow rate gradient separation and direct injection determination are presented.

(22) Liao, J.-L.; Zeng, C.-M.; Palm, A.; Hjerte´n, S. Anal. Biochem. 1996, 241, 195-198. (23) La¨mmerhofer, M.; Svec, F.; Fre´chet, J. M. J.; Lindner, W. Anal. Chem. 2000, 72, 4623-4628. (24) Yu, C.; Svec, F.; Fre´chet, J. M. J. Electrophoresis 2000, 21, 120-127. (25) Prˇistoupil, T. I.; Kramlova´, M.; Kubı´n, M.; Sˇ pacˇek, P. J. Chromatogr. 1972, 67, 362-365. (26) Hileman, F. D.; Sievers, R. E.; Hess, G. G.; Ross W. D. Anal. Chem. 1973, 45, 1126-1130.

(27) Ericson, C.; Hjerte´n, S. Anal. Chem. 1999, 71, 1621-1627. (28) Hatsis, P.; Lucy, C. A. Analyst 2002, 127, 451-454. (29) Xu, Q.; Mori, M.; Tanaka, K.; Ikedo, M.; Hu, W. J. Chromatogr. 2004, 1026, 191-194. (30) Paul, S.; Ra˘ nby, B. Macromolecules 1976, 9, 337-340. (31) Hradil, J.; Svec, F.; Aratskova, A. A.; Beljakova, L. D.; Orlov, V. I.; Yashin, Ya. I. J. Chromatogr. 1989, 475, 209-217. (32) Li, Y.-M.; Liao, J.-L.; Nakazato, K.; Mohammad, J.; Terenius, L.; Hjerte´n, S. Anal. Biochem. 1994, 223, 153-158.

7008

Analytical Chemistry, Vol. 76, No. 23, December 1, 2004

EXPERIMENTAL SECTION Apparatus. The capillary IC system was composed of a pump (model LC-10ADVP, Shimadzu, Kyoto, Japan), a sample injector (model 7520, Rheodyne, Cotati, CA) with a 0.2-µL sample reservoir, and an on-capillary type UV absorption detector (model CE-1575, Jasco, Tokyo, Japan). A polyimide-coated fused-silica capillary tubing (250-µm i.d., 350-µm o.d., GL Science, Tokyo, Japan) filled with in situ polymerized and functionalized cationexchange monolithic polymer was used as a separation column. The separation of cations was performed by an isocratic elution with 10 mM copper sulfate, while the detection was carried out by indirect UV absorption at 210 nm. Reagents. (3-Methacryloxypropyl)trimethoxysilane was obtained from Shin-etsu Chemicals. (Tokyo, Japan). GMA and ethylene dimethacrylate (EDMA) were purchased from Wako Pure Chemicals (Osaka, Japan). Organic solvents (propan-1-ol, butane-1.4-diol, ethanol, and acetone), 2,2′-azoisobutyronitrile, inorganic salts, and proteins were also obtained from Wako Pure Chemicals. These reagents were of highest commercially available purity and used as received. The water used for sample and eluent preparations was purified with a Milli-Q deionization system (Nihon Millipore Kogyo, Tokyo). Preparation of Cation-Exchange Monolithic Columns. The capillary inner wall was pretreated with (3-methacryloxypropyl)trimethoxysilane to ensure the anchoring of monolithic polymer matrix, according to the procedure described by Li et al.32 In a 1-m piece of the pretreated capillary, GMA-based continuous polymer matrix was prepared by in situ polymerization. The preparation conditions were as follows, if not otherwise stated. Methacrylic monomers (0.9 mL of GMA and 0.3 mL of EDMA) were dissolved into a ternary porogenic solvent (1.05 mL of propan-1-ol, 0.6 mL of butane-1.4-diol, and 0.15 mL of water). After purging with nitrogen for 3 min, 12 mg (corresponding to 1 wt %

of the amount of total monomers) of 2,2′-azoisobutyronitrile was added to the monomer solution. This monomer solution was immediately aspirated into the capillary, and both ends of the capillary were sealed with silicon septa. The polymerization was left to proceed for 24 h at 60°C. The monolithic polymer gel thus synthesized was washed with ethanol and with water to remove the porogenic solvent and remaining monomers present in the column. Subsequently, epoxy groups in the polymer matrix were reacted with a Na2SO3 solution to introduce sulfonate groups, referring to the method of Paul and Ra˚nby.30 A 1 M concentration of Na2SO3 was passed through the monolithic column at a flow rate of 3 µL/min. This sulfonation was allowed to proceed at 75 °C in a column oven (model CTO-2A, Shimadzu). The cationexchange columns thus produced were rinsed with 10 mM HNO3 and then with a plenty of water for at least 24 h. Estimation of Cation-Exchange Capacity by a Breakthrough Method. The cation-exchange capacity was estimated through a breakthrough method, where 10 mM CuSO4 was passed through the column, and UV absorption at 210 nm was monitored. RESULTS AND DISCUSSIONS Preparation of Cation-Exchange Monolithic Stationary Phases with Different Capacities. It is conceivable that sulfonate groups can be directly incorporated into the polymer matrix by using 2-acrylamido-2-methylpropanesulfonic acid (AMPS) as a comonomer, but the monolithic polymer incorporating AMPS was prone to swell and then exhibited poor rigidity due to the softness. To build up a rigid polymer network, in the present paper, a twostep procedure consisting of the synthesis of a rigid polymer matrix (step 1) and the subsequent sulfonation (step 2) was attempted. As an approach for affording sulfonate groups to the polymer matrix, the following two ways are considered: (1) aromatic substitution with chlorosulfonic acid or fuming sulfuric acid; (2) ring opening of epoxy groups with sodium sulfite. This ring-opening reaction proceeds under mild conditions desirable for in situ preparation. Thus, the synthesis of epoxy-containing polymer matrix and the sulfonation based on ring opening were adopted and examined. To form the epoxy-containing polymer matrix, GMA was chosen and polymerized with EDMA in the presence of a porogenic solvent. The contents of the porogenic solvent and EDMA in the polymerization mixture have significant influences on the hydrodynamic permeability and mechanical strength. In terms of the porogenic solvent, a mixture of propan-1-ol, butane1.4-diol, and water (7:4:1, v/v) was used, referring to the appropriate preparation conditions described by Peters et al.33 and Coufal et al.,34 while the total concentration of monomers (%T) and the proportion of cross-linker (%C) were fixed at 40 (w/v) and 25% (w/w), respectively, as a good compromise among several requirements (high incorporation of functional monomer, high permeability, and mechanical strength). The permeability of the GMA-based monolithic columns thus constructed was examined through the measurements of back pressure. The pressure drops of 15-cm columns ranged from 0.8 (33) Peters, E. C.; Petro, M.; Svec, F.; Fre´chet, J. M. J. Anal. Chem. 1997, 69, 3646-3649. (34) Coufal, P.; C ˇ iha´k, M.; Sucha´nkova´, J.; Tesarˇova´, E.; Bosa´kova´, Z.; Sˇ tulı´k, K. J. Chromatogr., A 2002, 946, 99-106.

Figure 1. Variation in the degree of sulfonation of GMA-based monolithic polymer with immersion time in 1 M Na2SO3 solution. The solution pH and reaction temperature were kept at pH 7 and 75 °C, respectively.

to 1.6 MPa (in most cases, 1 MPa) at a normal flow rate of 3 µL/ min (a linear velocity of 1 mm/s). Incidentally, six columns (in the case of 15-cm column from 1-m capillary) are obtained at a time. Among these six columns, a quite good column-to-column reproducibility was obtained; the relative standard deviations in pressure drop were within (7%. The mechanical strength for high flow rate operations was also evaluated through the values of back pressure at different flow rates. The back pressure linearly increased with an increase in the flow rate of at least up to 10 mm/s velocity, and in this range examined (1-10 mm/s), the pressure drop was almost constant over a period of at least two weeks. Also, this GMA-based monolith was not found to be compressed even at high pressures over 25 MPa. The morphology of GMA-based polymer monoliths formed inside the capillary was examined using scanning electron microscopy. Heterogeneous polymer globules with diameters of 2-5 µm were observed, and the globules were interconnected to form clusters. It was also found that micrometer-sized flowthrough-pores existed between the polymer globules, resulting in high permeability and thereby low flow resistance. The preparation of cation-exchange columns with different capacities is of great importance for a broad variety of applications. Thus, control of the degree of sulfonation was attempted. The degree of sulfonation is dependent on several parameters such as reaction immersion time, reaction temperature, concentration of Na2SO3, and solution pH. Among them, reaction time may be the most convenient and simple parameter for producing a variety of exchange capacities, and then the effect of the reaction time on the sulfonation was examined. Figure 1 shows the relationship of the reaction time and the amount of sulfonate groups (denoted as cation-exchange capacity). The solution pH of 1 M Na2SO3 and the reaction temperature were kept at pH 7 and 75 °C, respectively. As can be seen in Figure 1, the cation-exchange capacities of the monolithic columns immersed for 1, 4, 8, and 12 h were 19, 52, 80, and 92 µequiv/mL of the column volume, respectively. The relative standard deviations of the reproducibility of sulfonation were less than 8.2% (n ) 3). The capacity linearly increased with increase in the reaction time of up to 6 h and thereafter was constant at ∼90 µequiv/mL. It was confirmed that various levels of sulfonation could successfully be achieved simply by changing Analytical Chemistry, Vol. 76, No. 23, December 1, 2004

7009

Figure 2. Effect of solution pH of 1 M Na2SO3 on cation-exchange capacity. Reaction time and temperature were fixed at 12 h and 75°C, respectively.

the reaction time and that monolithic capillary columns with different capacities of up to 90 µequiv/mL could be produced. Incidentally, residual epoxy groups were converted to diols by acid hydrolysis so that they might not act as anchoring sites. The preparation of ion-exchange columns with higher capacities may also be an important subject. Figure 2 shows the variation in the cation-exchange capacity with the solution pH of 1 M Na2SO3, where the reaction time was set at 12 h. As the solution pH was increased from 6 to more alkaline pH values, the capacity increased and reached 300 µequiv/mL under the condition of pH 11. Here, this 30% GMA/10% EDMA-based polymer matrix included 2300 µmol/mL epoxy groups, while the amount of sulfonate groups was 300 µmol/mL. Then, the percentage of the epoxy groups reacting with Na2SO3 can be estimated to be at most 13% epoxy groups in the monolithic column. According to previous data obtained from a batch experiment, epoxy groups were largely converted to sulfonate groups under this reaction conditions (pH 11, 75°C). The low incorporation efficiency of sulfonate groups in the present experiment may be mainly due to the deficiency of available epoxy groups on the through-pore surface, while a small part of the epoxy groups may be converted to diols by hydrolysis. That is, a considerable amount of epoxy groups may exist inside the polymer gel and not be accessible to the reaction with Na2SO3. Separation of Common Monovalent and Divalent Cations. The separation performance and selectivity of the cation-exchange columns thus produced were evaluated through the separation of a model mixture of common cations such as Na+, NH4+, K+, Mg2+, and Ca2+. Figure 3 shows the separations on three cationexchange monolithic columns with different capacities, where 10 mM CuSO4 was used as an eluent and the flow rate was set at 3 µL/min. Analyte cations were indirectly detected with UV absorption at 210 nm. The retention times of the cations increased along with increase in the capacity, and monovalent and divalent cations were efficiently separated on the 15-cm column with the capacity of 150 µequiv/mL. The numbers of theoretical plates were above 3000 plates/15 cm (above 20 000 plates m-1) for these cations. This separation efficiency is comparable to that of commercially available conventional-size cation-exchange columns. It was confirmed that the performance of organic polymer-based monolithic capillary columns is sufficient for practical use in IC. 7010 Analytical Chemistry, Vol. 76, No. 23, December 1, 2004

The limit of detection (LOD) with UV absorption at 210 nm was calculated based on a signal-to-noise ratio of 3. The LODs for the cations were less than 50 µM in 0.2-µL samples. These LODs were the results from an on-capillary detection (light pass length, 0.25 mm), and would be still enhanced by using a now commercially available UV detector equipped with microflow cell (a small volume and long path length). To ensure run-to-run and day-to-day precision in retention time, 1 mM Mg2+ was injected at least 10 times/day during a period of two weeks. The relative standard deviation (RSD) of the run-torun reproducibility was found to be below 0.4%, while the RSD value of the day-to-day reproducibility was maintained to within 1.5%. The somewhat higher RSD value in the day-to-day reproducibility may be due to the poor performance of the HPLC pump. Indeed, the monolithic columns did not show any significant changes in the selectivity and resolution during this period. Comparison of Monolithic and Particle-Packed Capillary Columns. To compare the performance of a monolithic capillary column with that of a particle-packed capillary column, a packed capillary column was prepared in the laboratory, following the procedure described by Takeuchi and Ishii.3 Develosil ODS-5 (5 µm, from Nomura Chemicals, Seto, Japan) modified with an anionic surfactant, sodium dodecyl sulfate (SDS) was used as a packing material. The SDS-modified ODS packed column showed good chromatographic performance under the eluent condition in the present work (elution with 10 mM CuSO4), and the chromatogram of a model mixture of cations is presented in Figure 4. The cations were well separated from one another on this packed capillary column, and the numbers of theoretical plates for the cations were ∼6000 per 15-cm column. This separation efficiency was comparable to or higher than that of a monolithic capillary column, but the back pressure of this packed capillary column reached 7 MPa, which was 7 times higher than that of monolithic capillary column under the same flow rate condition. Separations at High Flow Rate and Application of Flow Rate Gradient Elution. The low flow resistance and good mechanical strength of the constructed monolithic column make it possible to use high flow rate, allowing high-speed separations within the pressure drop constraints. Thus, the durability for highpressure operation and the separation efficiency at higher flow rate were elaborately examined in the flow rate range of 3-30 µL/min. Figure 5 shows some chromatograms of a model mixture obtained under the conditions of different flow rates. As expected, the analysis time was reduced along with an increase in the flow rate, but the separation efficiency deteriorated to a certain extent under the conditions of high flow rate. The van Deemter plot for Mg is presented in Figure 6, together with the data of pressure drop. When the flow rate was raised from 3 to 30 µL/min, the height equivalent to a theoretical plate was increased ∼5 times higher. Eluent flow rate can be an important operating parameter to accelerate the elution of strongly retained analytes. As an example, the chromatogram obtained by a flow rate gradient elution is shown in Figure 7. The gradient elution program was as follows; the flow rate was kept at 3 µL/min until 8 min and then increased up to 15 µL/min at a rate of 2 µL/min. As compared with the chromatogram in normal elution mode, the long retention interval between the peaks of mono- and divalent cations was saved, and

Figure 3. Separations of common cations on three cation-exchange monolithic stationary phases with different capacities of (A) 5.5, (B) 25.4, and (C) 151 µequiv/mL. Stationary phase, continuous GMA-EDMA modified with 1 M Na2SO3 inside a capillary of 0.25 mm i.d. × 150 mm long; mobile phase, 10 mM CuSO4; flow rate, 3 µL/min; injection volume, 0.2 µL; detection, indirect UV absorption at 210 nm (on-column type UV detector); sample, 2 mM each of Na+, K+, NH4+, Mg2+, and Ca2+.

Figure 4. Separation of common cations on a cation-exchange packed capillary column. Stationary phase, SDS-modified Develosil ODS-5 packed into capillary of 0.25 mm i.d. × 150 mm long. Experimental conditions are the same as in Figure 3.

Figure 6. Effect of flow rate on height equivalent to a theoretical plate and pressure drop. 9, height equivalent to a theoretical plate of Mg2+; 4, pressure drop.

Figure 5. Separations of common cations at different flow rates of (A) 3, (B), (C) 9, (D) 12, and (E) 15 µL/min. Sample, 2 mM each of Na+, K+, Mg2+, and Ca2+; cation exchange capacity, 80 µequiv/mL. Other experimental conditions are the same as in Figure 3.

the analysis time was reduced to the half without sacrificing the quality of chromatographic resolution. It was confirmed that flow rate gradient was effective to control the peak positions of various analytes with remarkably different retention times. Application to Direct Injection Determination of Major Cations in Human Saliva. This methacrylic-based cationexchange polymer is relatively hydrophilic and may have the potential for direct injection determination of cations due to the physiological and chemical inertness. To evaluate the inertness of this polymer to column fouling substances (in this case, proteins), an aqueous solution of each protein (bovine serum albumin (pI 4.7, MW 66 000), hemoglobin (pI 7.2, MW 64 500), and cytochrome c (pI 9.6, MW 12 400)) at 100 µM was prepared

Figure 7. Chromatograms obtained by (A) constant flow rate elution and (B) flow rate gradient elution. The gradient elution program was as follows; the flow rate was kept at 3 µL/min until 8 min and then increased up to 15 µL/min at a rate of 2 µL/min. Other experimental conditions are the same as in Figure 5.

and injected into this GMA-based cation-exchange monolithic column. As a result, each protein tested was eluted at nearly the void volume and almost completely recovered from the column under the elution condition used in this work. The present capillary IC system was applied to the determination of major cations in a human saliva sample. Human saliva samples were collected from 10 healthy males in our laboratory. A typical chromatogram of a human saliva sample without any Analytical Chemistry, Vol. 76, No. 23, December 1, 2004

7011

columns were tolerant to at least 20 times injections of the saliva samples without a significant decrease in separation efficiency.

Figure 8. Chromatogram of a human saliva sample without any treatment except filtration. Cation exchange capacity, 130 µequiv/ mL. Other experimental conditions are the same as in Figure 3.

treatment except filtration is shown in Figure 8. Four major cations attributed to Na+, K+, Mg2+, and Ca2+ were detected without interference from the matrix eluted as broad peaks within 5 min. Calibration curves of peak area versus concentration were constructed by using seven nonzero standards. Linear relationships were obtained up to 50 mM with correlation coefficients of >0.997. The relative standard deviations for 10 replicate measurements were less than 5% in the linear range, and the detection limits for Na+, K+, Mg2+, and Ca2+ in saliva samples, at a signalto-noise ratio of 3, were 83, 75, 57, and 48 µM, respectively. The concentrations in human saliva samples ranged from 13 to 20 mM for Na+, 17 to 43 mM for K+, 0.10 to 0.55 mM for Mg2+, and 0.89 to 3.1 mM for Ca2+. These values were in good agreement with those obtained by inductively coupled plasma-atomic emission spectrometry or emission flame photometry. The monolithic

7012

Analytical Chemistry, Vol. 76, No. 23, December 1, 2004

CONCLUSION Polymer-based strong cation-exchange monolithic stationary phases with different capacities for practical applications in capillary IC were produced by a two-step procedure involving the synthesis of GMA-based polymer matrix and its subsequent sulfonation by reaction with Na2SO3 to different extents. The produced monolithic columns had low flow resistance and mechanical stability good enough to operate at high linear velocities (up to at least 10 mm/s) and exhibited sufficient separation performance similar to that observed for commercially available conventional-size cation-exchange packed column. The ring-opening method allows gentle coupling of different functional groups and ligands to the polymer matrix and its finetuning. Therefore, tailor-made mixed stationary phases with variable densities of different ligands, which are congruous with each analytical purpose, would be produced. It should also be emphasized that these polymer-based monolithic columns can be produced almost without failure at a low cost of $1/column. Due to the simplicity and cost-effectiveness, a one-day or one-week disposable column could be realized in the near feature, which may be suitable for direct injection analysis of biological samples containing high levels of matrix.

Received for review April 26, 2004. Accepted September 18, 2004. AC040079G