n column (GL Science, Tokyo, Japan), a UV detector (Jasco, Japan) with a modified flow cell and Chromatopac C-R7Ae as a recorder (Shimadzu, Kyoto, Japan). The flow rate of the mobile phase was kept constant at 4 μL∙min−1. Scanning electron micrographic images for morphology observation of the monolithic column were obtained using a SEM S-4800 (Hitachi, Japan).
All chemicals used were of analytical grade. 2,2’-azo-bis (isobuthyronitrile) (AIBN), glycidyl methacrylate (GMA) and ethylene dimethacrylate (EDMA) were purchased from Wako Pure Chemical Industries (Osaka, Japan). 1-propanol and 1,4-butanediol were obtained from Nacalai Tesque (Kyoto, Japan). Trimethylamine (30 wt% in H2O) and tetrahydrofuran were from Kanto Chemical (Tokyo, Japan). The purified water was prepared using GS-590 water distillation system (Advantec, Tokyo, Japan). Other chemicals were used as received.
2.3. Preparation of Methacrylate-Based Anion-Exchange Monolithic Column
The fused silica capillary was first activated by 1 mol∙L−1 NaOH, purified water, and 1 mol∙L−1 HCl. 3-(trimethoxysilyl)-propyl methacrylate (γ-MAPS) solution (30% (v/v) in aceton) was used to fill the activated capillary. After sealing the capillary two ends, the reaction was allowed to perform 60˚C for 24 h in waterbath. Then, the capillary was washed thoroughly with aceton. The N2 was flown through the capillary to dry the inner surface before further use.
The monolithic column was prepared by in situ polymerization. A polymerization mixture containing GMA (30% (v/v)), EDMA (10% (v/v)), 1-propanol (35% (v/v)), 1,4-butanediol (20% (v/v)), water (5% (v/v)), and AIBN (1% (w/v) of the total monomer amount). This mixture solution was ultrasonically homogenized for 5 min and immediately aspirated into the pretreated capillary. After sealing both ends, the treated capillary was placed in the waterbath to proceeds the polymerization at 60˚C for 24 h. The monolith was washed with methanol to remove the unreacted monomers and remaining porogenic solvent present in the column. Subsequently, epoxy groups in the monolith were reacted with trimethylamine to get strong anion exchanger, as the following procedure: a trimethylamine solution (50% (v/v) in tetrahydrofuran) was passed through the monolithic column at flow rate 4 μL∙min−1 for 2 h. Then, the monolithic column was placed in the oven to proceeds the modification at 80˚C for 5 h. The anion exchange monolithic column produced was washed with methanol for 3 h at 2μL∙min−1 of flow rate.
3. Results and Discussion
3.1. Preparation of Methacrylate-Based Anion-Exchange Monolithic Column
The monolithic column was prepared directly in the capillary by in situ polymerization method. There are two step were adopted in this experiment to form the polymer network. First, the synthesis of a rigid polymer matrix by using GMA as monomer, EDMA as crosslinker, and ternary porogen which consists of 1-propanol, 1,4-butanediol, and water. Then, the introduction of trimethylamine as strong anion exchange moiety via ring-opening reaction of the epoxy group. The composition of the monomer, crosslinker, porogen solvent; polymerization reaction time; and modification conditions would have great effect on the monolithic structure.
The morphology of the monolithic column was an important parameter that can affect the capability and efficiency separation. The morphology of the monolithic column was examined by scanning electron microscopy (SEM). Figure 1(a) demonstrates that the morphology of monolithic column were solid and completely attached to
Figure 1. Scanning electron microphotographs of monolithic column. (a) Wide-view and (b) close-up-view.
Figure 2. Plots of the flow rate of water against the back pressure of the monolithic column. Column, 80 ´ 0.32 mm i.d.; mobile phase: water.
the inner surface of the capillary column. It indicated that the monolith was covalently bonded to the capillary. The successful attachment of monolith into the inner surface of the capillary column influenced by the pre-treatment step of the capillary column. As shown in Figure 2(b), the obtained monolith displayed porous network with globular structure. The continuous porous channels in the monolith bed which were formed by mesopores and through pores can also be seen.
The permeability of monolithic column was examined by measure the back pressure for different flow rate using water as mobile phase. The flow rate ranged from 0.5 to 4 μL∙min−1. Column permeability is affected by amount of porogenic solvent. A sufficient amount of porogen would result good permeability so as the mobile phase and sample solution would flow through the column under small back pressure. On the contrary, insufficient amount of porogen would make the mobile phase and sample solution flow through the column under large back pressure . The permeability of the monolithic column was calculated as 9.88 ´ 10−13∙m2 which indicated that the monolithic column had good permeability. The permeability (B0) was calculated by using Darcy’s Law ,
where F was the linear velocity of the mobile phase, η was the dynamic viscosity of the mobile phase (η = 0.089 Pa s for water), L was the effective column length, and ∆P was pressure drop.
On the other hand, the back pressures dependence on flow rate was a straight line with a correlation coefficient R2 0.992 (Figure 2), which indicated the good mechanical stability of the prepared monolithic column.
3.2. Separation of Inorganic Anions
Three anions were first used to evaluate the performances of the monolithic column. Figure 3 demonstrates the separation of three inorganic anions on monolithic column using various mobile phases with the same concentration. The analytes are detected at 210 nm. All the analytes can well separated. Potassium chloride provided better resolution of the anions in a shorter retention time and more reproducible signals than the others.
In order to increase the retention time of analytes, the concentration of the mobile phase should be higher. Potassium chloride was examined as the mobile phase in the 50 - 200 mM concentration range. Separations of five inorganic anions were shown in Figure 4. The elution order was iodate, bromate, nitrite, bromide and nitrate. The present system is more sensitive to determination of iodate, nitrite and nitrate compared with bromate and bromide. It’s seen that, the retention time of analytes could be increase with the increasing of the mobile phase
Figure 3. Effect of the mobile phase on the separation of inorganic anions. Column, 80 ´ 0.32 mm i.d.; mobile phase: 50 mM LiCl 50 (a), 50 mM NaCl (b), 50 mM KCl (c), 50 mM RbCl (d), 50 mM CsCl (e), 50 mM NH4Cl (f); flow rate: 4 μL∙min−1; injection volume, 0.2 µL; wavelength of UV detection: 210 nm; analytes: 1 = iodate, 2 = nitrite, 3 = nitrate, 1.0 mM each.
Figure 4. Effect of concentration of potassium chloride mobile phase on the separation of inorganic anions. Column dimension: 80 ´ 0.32 mm i.d.; concentration of mobile phase: (a). 50 mM, (b). 100 mM KCl, (c). 150 mM, (d). 200 mM; flow rate: 4 μL∙min−1; injection volume: 0.2 µL; wavelength of UV detection, 210 nm; analytes, 1 = iodate, 2 = bromate, 3 = nitrite, 4 = bromide, 5 = nitrate, 1.0 mM each.
concentration. On the other hand, if the concentration of the mobile phase was too high, the analytes could not be separated completely. Considering the experimental results, 100 mM of potassium chloride was selected as a mobile phase for the following experiments.
3.3. Analytical Figures of Merit
The RSD of the retention time for the six succescive chromatographic run under the optimum condition were in the 1.09% - 1.75% range. The RSD for the retention time were less than 2%. It showed that this method had good repeatability.
The calibration curves of the five inorganic anions are shown in Figure 5. The calibration graphs showed linear relationships between the peak area and the concentration. It can be seen from the good R-square values obtained.
The limits of detection (LOD) were 0.15, 0.18, 0.14, 0.15, 0.08 mM for iodate, bromate, nitrite, bromide, nitrate, respectively. On the other hand, the limits of quantitation (LOQ) of iodate, bromate, nitrite, bromide, nitrate were 0.50, 0.61, 0.46, 0.49, 0.26 mM, respectively. The values of LOD and LOQ were low enough. It showed that this method had good sensitivity.
3.4. Practical Application
The monolithic column was applied to the determination of inorganic anions present in tap water and ground water water samples. The results are shown in Figure 6. There were no inorganic anions identified in the tap water sam
Figure 5. Calibration curves for five inorganic anions. Column, 80 ´ 0.32 mm i.d.; mobile phase, 100 mM KCl; flow rate, 4 µL min-1; injection volume: 0.2 µL; wavelength of UV detection: 210 nm; ▲ = iodate, ○ = bromate, ■ = nitrite, □ = bromide, ● = nitrate.
Figure 6. Separation of inorganic anions in water samples. Column dimension: 80 ´ 0.32 mm i.d.; mobile phase: 100 mM KCl; flow rate: 4 μL∙min−1; injection volume: 0.2 µL; wavelength of UV detection, 210 nm; sample: tap water (a) and ground water samples (b).
ple (Figure 6(a)). On the other hand, nitrate was identified in the ground water sample (Figure 6(b)), with concentration of nitrate is 0.08 mM. The concentration of nitrate as calculated according to the peak area was 0.08 mM.
A poly(glycidyl methacrylate-co-ethylene dimethacrylate) anion exchange monolithic column was successfully produced by in situ polymerization further modified with trimethylamine via ring-opening reaction of epoxy group. Morphology of the monolithic column was studied by scanning electron microscopy. Mechanical stability and permeability of the column were both good. In general, this method provides good precision of retention time, acceptable linearity and good sensitivity. The present method could be applied to the determination of inorganic anions contained in tap water and ground water with the good values for recovery.
The authors gratefully acknowledge the Japan Student Services Organization (JASSO) for the scholarships over. The authors would like also to thank to the Dean of Faculty of Engineering, Gifu University, Japan and the Dean of Faculty of Mathematics and Natural Sciences, Andalas University, Indonesia, for their generous support.