- Email: [email protected]
On-line sequential preconcentration of inorganic anions by anion-selective exhaustive injection and base-stacking in capillary zone electrophoresis
Journal of Chromatography A, 1109 (2006) 285–290
On-line sequential preconcentration of inorganic anions by anion-selective exhaustive injection and base-stacking in capillary zone electrophoresis Zhao-Xiang Zhang a,b , You-Zhao He a,∗ , Yan-Yun Hu a a
b
Department of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026, China College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, China Received 28 September 2005; received in revised form 29 December 2005; accepted 10 January 2006 Available online 14 February 2006
Abstract A sequential electrostacking method based on anion-selective exhaustive injection (ASEI) and base-stacking (BS) is presented for the preconcentration and determination of inorganic anions by capillary zone electrophoresis (CZE) in this paper. Tetradecyltrimethylammonium bromide as an electroosmotic flow (EOF) modifier was added into the buffer to suppress EOF of the capillary. Firstly, a water plug was hydrodynamically injected into the capillary. During ASEI under negative high voltage, the sample anions migrated quickly towards the boundary between the water plug and buffer in the capillary. Then an alkaline zone was injected electrokinetically to concentrate the anions further. With the sequential electrostacking method, the preconcentration factor of (0.8–1.3) × 105 was obtained compared with the conventionally electrokinetic injection and the relative standard deviation of peak area was 3.3–5.3% (n = 5). The detection limits of ASEI–BS–CZE for six inorganic anions were 6–14 ng/L. The proposed method has been adopted to analyze six anions in cigarette samples successfully. © 2006 Elsevier B.V. All rights reserved. Keywords: Inorganic anions; Anion-selective exhaustive injection; Base-stacking; Capillary zone electrophoresis; Cigarette
1. Introduction Capillary electrophoresis (CE) has been employed for rapid separation of inorganic anions in biochemical and environmental studies [1]. However, its extremely short light path for ultraviolet–visible (UV–vis) detection still limits the separation technique as a sensitive one. On-column stacking is an effective means to improve the detection sensitivity of UV–vis in CE. Park and Lunte [2] presented an acid-stacking method to concentrate cationic analytes, in which fast migrating hydrogen ions from acid zone reacted with the buffer weak anions, produced a low conductivity region and made the cationic stacking. Similarly, by electrokinetically injecting an alkaline plug into a capillary, anionic stacking occurred [3,4]. Cao and co-workers [5,6] proposed a “transient moving chemical reaction boundary” method to concentrate zwitterionic analytes, in which the charge polarity of zwitterionic analytes was changed during their migration from basic
∗
Corresponding author. Tel.: +86 551 3607072; fax: +86 551 3603388. E-mail address: [email protected] (Y.-Z. He).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.01.028
sample zone into acidic buffer region and the analytes were stacked at the moving boundary. Britz-Mckibbin et al. [7] proposed a dynamic pH junction-sweeping method to concentrate flavin derivatives with the improved sensitivity of more than 1200-fold. Terabe and co-workers developed a dual preconcentration method of cation-selective exhaustive injection and sweeping to concentrate organic base [8] and metal ions [9], which improved the detection sensitivity to a million-fold for organic bases and 140,000-fold for divalent and trivalent metal ions. To obtain a high concentration factor and satisfied separation properties, a large volume of sample solution should be introduced into and its medium ought to be removed out of separation capillary. The removal of sample medium can not only enhance the analyte concentration, but also be propitious to the following separation, for it can provide an effective separation length and eliminate the non-uniform distribution of electric field strength. A simple way to remove sample medium is the change of the electrode polarity after sample injection, in which the reversed EOF can expel the sample medium out of the capillary inlet [10]. However, it is difficult to monitor the current of the polarity conversion exactly. Another way adopted for large volume injection
286
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
and sample medium removal is the EOF restriction. By using the coated capillary [11–13], low pH buffer [14], non-aqueous CE [15] or EOF modifier [16,17], EOF can be suppressed to be even lower than the analyte electrophoresis mobility. Kim et al. [12] proposed anion-selective exhaustive injection-sweep micellar electrokinetic capillary chromatography (MECC) to concentrate organic acids with a polyacrylamide-coated capillary. With low pH buffer, sample medium can be removed by an electroosmotic flow pump under negative voltage during analyte stacking [14]. However, it is only fit for some organic and inorganic anions retained in low pH solution. By using methanol as the buffer solvent, non-aqueous CE can improve the detection sensitivity of organic anions by field-enhanced sample injection and reduce sample medium introduction [15]. EOF modifiers have been used to remove sample medium from the capillary inlet under a reversed potential after the hydrodynamic sample injection. The reported methods improved the detection sensitivity by 30–40 folds because of the deficient sample injection [16,17]. In this paper, a sequential electrostacking method based on anion-selective exhaustive injection (ASEI) and base-stacking (BS) was developed with the preconcentration factor of 105 and was used to concentrate and determine six inorganic anions in cigarette samples successfully. 2. Experimentals 2.1. Apparatus A 1229 HPCE Analyzer (Institute of New Technology Application, Beijing, China) with detection at 254 nm, a N−2000 double-channel chromatography processor (Institute of Intelligence Information and Engineering, Zhejiang Univ., Zhejiang, China) and an S–2200 ultrasonic cleaner (120 W, 35 kHz, J & L Ultrasonic Science and Technology Corp., Shanghai, China) were employed in this work. A 50 m i.d. capillary with total and effective length of 48 and 33 cm, respectively, was purchased from Yongnian Chromatographic Components Ltd. (Hebei, China). 2.2. Reagents All the chemicals were of analytical reagent grade and the required solutions were prepared with tri-distilled water (SZ-3 tri-distilled water system, Huxi Anal. Instrument Factory, Shanghai, China). Tetradecyltrimethylammonium bromide (TTAB) was purchased from Sigma (St. Louis, MO, USA). Sodium chromate, Tris and glacial acetic acid (HAc), etc. were purchased from Chemical Reagent Co. (Shanghai, China). Anion stock solutions (1000 mg/L) were prepared from their sodium or potassium salts. Three mixed standard solutions in the range of 0.020–6.0 g/L were prepared by diluting the stock solutions with tri-distilled water.
(0.500 g) was weighed in a 50 mL conical flask. After adding 20 mL water, the flask was capped and the anionic analytes were extracted with the ultrasonic cleaner for 15 min. The treated solution was filtered with a 0.22 m acetic cellulose membrane, diluted to 50 mL and reserved in refrigerator. The running buffer contained 40 mmol/L sodium chromate as background electrolyte (BGE), 0.40 mmol/L TTAB and 20 mmol/L Tris, and adjusted to pH 8.9 with HAc. 2.4. ASEI−BS−CE procedure The capillary was flushed daily in the order of H2 O (1 min), 1.0 mol/L HCl (5 min), H2 O (1 min), 1.0 mol/L NaOH (15 min), H2 O (1 min) and conditioned with the running buffer for 10 min by pressure. Between two runs, the capillary was conditioned with the running buffer for 6 min. The main steps of ASEI−BS−CE are illustrated in Fig. 1. Firstly, the capillary was pre-equilibrated with the running buffer containing TTAB, which was dynamically adsorbed on the inner wall of the capillary (Fig. 1a). Before sample injection, a water plug was hydrodynamically injected into the capillary. TTAB on the capillary inner surface of the water section was dissolved into the water plug and the capillary surface restored negative charges, i.e. negative zeta potential (Fig. 1b). With the voltage of −6 kV for 210 s in ASEI, the anionic analytes migrated rapidly into the capillary through the water plug and stacked on the boundary between the water plug and running buffer region (Fig. 1c). Once the anions reached the boundary, their velocities slowed down for the electric field strength decreased in the running buffer region. Meanwhile, the water plug was expelled because of the EOF direction towards the capillary inlet (Fig. 1d). At the same time, the capillary section was filled with the running buffer and TTAB in the buffer was adsorbed on the capillary inner wall again, so the EOF of the capillary became smaller and smaller. After sample loading, an alkaline zone was injected from 0.1 mol/L NaOH solution at −6 kV for 120 s (Fig. 1e), a neutralization reaction between OH− and Tris+ from the buffer took place and reduced the conductivity of analyte zone. In this way, the injected anions were condensed further. By replacing the alkaline solution with buffer, CZE separation was carried out at −6 kV (Fig. 1f).
2.3. Cigarette sample and buffer preparation Cigarette samples were dried at 60 ◦ C for 4 h, ground into powder and sieved with a 40 mesh sifter. Cigarette sample
Fig. 1. Schematic diagrams of ASEI–BS–CE processes.
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
3. Results and discussion 3.1. Separation conditions Sodium chromate, potassium hydrogen phthalate and paminobenzoic acid were examined as BGE for indirect UV detection. The results indicate that CrO4 2− is much suitable for the anion separation and its mobility close to the average mobility of the inorganic anions. As described by Kohlrausch regulating function [18], when the analyte zone has a higher mobility than the running buffer, the leading edge of the analyte zone will be diffuse and the trailing edge sharp. Contrarily, the former will be sharp and the latter diffuse. In Fig. 2, it can be found that the mobility of CrO4 2− is located between Cl− and F− . The effect of Na2 CrO4 concentration on the anion separation is also illustrated in Fig. 2. When Na2 CrO4 concentration is 10 mmol/L, the peaks of SO4 2− and NO2 − overlap completely and those of F− and HPO4 2− cannot be separated on baseline (Fig. 2a). When Na2 CrO4 concentration increases to 20 mmol/L, SO4 2− peak shifts backward and superposes NO3 − one, but the baseline separation of F− and HPO4 2− is achieved. In Fig. 2b, six anions can be completely separated when Na2 CrO4 concentration increases to 40 mmol/L. If the Na2 CrO4 concentration increases further, the peak height of six anions decreases and the retention time of anions are prolonged. It implies that the increase of Na2 CrO4 concentration will lead to the increase of the ionic strength in BGE, the decrease of the anionic mobility [19,20], and thus the prolongation of the anionic retention time. So 40 mmol/L Na2 CrO4 was chosen in this work. For base-stacking, the BGE solution should contain a weak base ion, such as Tris+ or NH4 + [3,4]. Several buffer systems
Fig. 2. Effect of Na2 CrO4 concentration on inorganic anion separation. The capillary is 50 m i.d. with its total length of 48 cm and effective length of 33 cm; the sample solution is injected at −6 kV for 10 s and the separation voltage is −6 kV; the running buffer contains pH 8.9, 20 mmol/L Tris, 0.4 mmol/L TTAB and CrO4 2− concentration: (a) 10 and (b) 40 mmol/L. Each anion concentration is 5.0 mg/L. 1: Cl− ; 2: NO2 − ; 3: NO3 − ; 4: SO4 2− ; 5: F− and 6: HPO4 2− .
287
were investigated, including ammonium tartrate/ammonium hydroxide, ammonium oxalate/ammonium hydroxide and Tris+ /HAc. A system peak can appear behind F− or SO4 2− peak with ammonium tartrate or ammonium oxalate buffer, which can interfere on the analyte peak of F− or SO4 2− . However, the system peak of Tris+ /HAc buffer is behind the analyte peaks and its influence on the anion separation and peak shape can be ignored. When the pH value of the buffer is lower than 8.5, the buffer color changes to orange, which results from a reaction equilibrium between CrO4 2− and Cr2 O7 2− . Meanwhile, it makes the peak width and retention time increased. By changing the buffer pH from 8.9 to 11.3, no significant difference can be observed on the separation efficiency and anionic resolution. However, when the pH value is higher than 11.3, there are two phosphate peaks of HPO4 2− and PO4 3− in the electropherogram, and the PO4 3− peak can appear in front of peak F− . So pH 8.9 Tris+ /HAc buffer was used in this work. TTAB in buffer can be dynamically adsorbed on the capillary inner wall and affect EOF of the capillary. When TTAB concentration is less than 0.38 mmol/L, the mobility of EOF increases towards the cathode relatively. For the electrophoretic mobility of analyte anions is towards anode, the peaks of the anionic analytes are broadening. When the concentration of TTAB locates in the range of 0.38–0.42 mmol/L, the EOF becomes extremely low (2.6 × 10−5 cm2 /V s) and the peak width and separation efficiency of the anions are satisfied. So 0.40 mmol/L TTAB was adopted in this work.
3.2. Anion-selective exhaustive injection The water plug can provide high field strength to introduce more analyte anions into the capillary during ASEI. TTAB on the capillary inner wall can be dissolved into the water plug, which enhances the negative zeta potential in the water section of the capillary and increases the EOF towards the capillary inlet under high negative voltage [16,17]. During ASEI, the analyte anions are introduced into the capillary at −6 kV and the water plug is removed out of the capillary inlet by the EOF. The amount of anions injected is dependent on the sample injection time under a fixed injection voltage. The effect of water and sample injection time on peak height and separation efficiency was investigated, as shown in Fig. 3. The injection time of water plug was evaluated with 120 (Fig. 3a), 180 (Fig. 3b), 240 (Fig. 3c), 300 (Fig. 3d), 360 (Fig. 3e) and 420 s (Fig. 3f) under 10 cm height difference, respectively. The sample injection time was changed from 50 to 230 s at −6 kV. It can be found that the optimal peak height and separation efficiency are obtained with the water/sample injection time pair of 120 s/70 s, 180 s/90 s, 240 s/130 s, 300 s/160 s, 360 s/180 s and 420 s/210 s, respectively. It also shows that the peak height increases with the increase of the water/sample injection time pair up to 300 s/160 s (Fig. 3a–d), in which the corresponding length of water plug is about 5.3 cm. With the water/sample injection time pair longer than 300 s/160 s, both the peak height and separation efficiency decrease (Fig. 3e and f).
288
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
Fig. 3. Effect of water/sample injection time on peak height and separation efficiency of Cl− (other inorganic anions display the similar results). Cl− concentration is 1.0 g/L; the buffer contains 40 mmol/L Na2 CrO4 , 0.4 mmol/L TTAB and 20 mmol/L Tris at pH 8.9; other conditions are the same as in Fig. 2. Water injection time is (a) 120, (b) 180, (c) 240, (d) 300, (e) 360 and (f) 420 s.
3.3. ASEI–BS sequential preconcentration According to ASEI conditions aforementioned, the optimal stacking can be obtained with the sample injection time of 160 s. If the injection time is longer than 160 s, it can cause the decrease of separation efficiency. To introduce more analytes into the capillary, a base-stacking method was adopted and OH− from 0.1 mol/L NaOH was introduced at −6 kV after sample injection. Since the mobility of OH− is higher than the analyte ions, the fast-migrating OH− passes through the sample zone, reacts with the weak base cation Tris+ from the buffer, creates a low
conductivity region [3] and condenses the anionic analyte zone further. The effect of sample and base injection time on the peak height and separation efficiency of the anions was also investigated with both the injection voltages of −6 kV. Firstly, the injection time of alkaline zone was surveyed with the sample injection time of 210 s at −6 kV and the water injection time of 420 s under 10 cm height difference, corresponding to the water plug length of 7.4 cm, as shown in Fig. 3f. The effect of the alkaline injection time on the anion stacking is illustrated in Fig. 4. It can be found that the separation efficiency of the anions is improved by increasing the alkaline injection time from 90 to 120 s (Fig. 4a and b). However, when the injection time increases to 150 s, a buffer matrix peak and shifted baseline can appear in the electropherogram (Fig. 4c), which indicates that the OH− amount has been excessive. The optimal separation efficiency and resolution can be obtained with the alkaline injection time of 120 s. The shifted baseline may result from the excessive concentration of OH− and the decreased concentration of CrO4 2− , for the migration charges of the anions should be constant when the current equilibrium is achieved in CZE. The buffer matrix peak may be Ac− stacked in the corresponding low-conductivity region formed by TrisOH. The experiment results also indicate that the slightly excessive injection of alkaline zone will not reduce the separation efficiency and anionic resolution (Fig. 4c). However, when the injection time of alkaline zone is longer than 160 s, both the separation efficiency and resolution will decrease besides the buffer matrix peak and shifted baseline. According to the above-mentioned conditions in Section 3.2, excessive sample injection times were examined with 180, 190, 200, 210 and 220 s, respectively, and then the injection
Fig. 4. Electropherograms of inorganic anions with different alkaline injection time in ASEI–BS–CZE. The water and sample injection time is 420 s under 10 cm height difference and 210 s at −6 kV, respectively; the anion concentration is 0.2 g/L; other conditions are the same as in Fig. 3. Injection time of OH− from 0.1 mol/L NaOH at −6 kV is: (a) 90, (b) 120 and (c) 150 s.
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
289
by the buffer one, the current trace becomes plateau (after point c ) and CZE separation is carried out. In ASEI–BS–CZE, the sample vial is replaced by the alkaline one and the electric current experiences a decrease course again because OH− migrates toward the anode rapidly, reacts with Tris+ from the buffer region, produces a low conductivity zone of TrisOH and leads to stack the injected anions (210 s at −6 kV) further. Obviously, the current decrease from 41 to 10 A (c → d) results from the formation of the low conductivity zone. By comparing the current variation of ASEI with that of ASEI–BS, the mechanism of the second concentration step can be demonstrated as a basestacking. At the time d, the buffer vial replaces the alkaline solution one and the equilibrium current of CZE is recovered gradually. In addition, all the separated peaks are located after the time of e in ASEI–BS–CZE. Fig. 5. Current variation of ASEI–CZE (curve 1) and ASEI–BS–CZE (curve 2). The sample injection time is 160 s at −6 kV for ASEI–CZE, and the sample and base injection time are 210 and 120 s at −6 kV for ASEI–BS–CE.
time of alkaline solution was evaluated. The optimal results were obtained with the sample/ alkaline injection time pair of 180 s/50 s, 190 s/70 s, 200 s/100 s and 210 s/120 s, respectively. The experiment results show that the alkaline injection time should match the sample injection one in order to obtain an optimal separation efficiency and resolution. However, when the sample injection time is longer than 220 s, the baseline separation of NO2 − and NO3 − cannot be achieved no matter how to adjust the injection time of alkaline zone. So 210 and 120 s were adopted for the sample and alkaline injection time in this work, respectively, in which the water injection time was 420 s. 3.4. Mechanism of sequential preconcentration For elucidating the mechanism of the sequential preconcentration, the electrophoretic currents of ASEI–CZE and ASEI–BS–CZE were investigated. The sample injection time was 160 s at −6 kV for ASEI–CZE, while the sample and base injection time was 210 and 120 s, respectively, at −6 kV for ASEI–BS–CE. As shown in Fig. 5, curve 1 (a → b → c → e ) and 2 (a → b → c → d → e) represent the current variation in ASEI–CZE and ASEI–BS–CZE, respectively. The periods of a → c (a → c ) and c → d correspond to the sample and base injection, respectively. The periods after c in ASEI–CZE and after d in ASEI–BS–CZE are relative to the electrophoresis separation. During ASEI, the electric current increases from 3 to 9 A in 150 s (a → b ) and from 2 to 8 A in 200 s (a → b) gradually, then enhances rapidly to the top value of 42 A only in 10 s (b → c and b → c ). It manifests that the initially low electric current results mainly from the existence of the water plug, i.e. 5.3 cm in ASEI and 7.4 cm in ASEI–BS. During the sample injection, the water plug is expelled out of the capillary inlet by EOF and the electric current increases gradually. When the water plug is nearly eliminated, the electric current increases rapidly (b → c and b → c). In ASEI–CZE, the sample vial is replaced
3.5. Analytical characteristics Comparing with the conventional electrokinetic injection, the preconcentration factor (PF) with ASEI–BS has been calculated. The PF of (0.8–1.3) × 105 by ASEI–BS for six inorganic anions can be achieved under similar separation efficiencies. The limit of detection (LOD) is obtained with three-time standard deviation of 11 noise peaks of the baseline near the analyte peaks. For the proposed method of ASEI–BS–CZE, the LODs of six inorganic anions are in the range of 6–14 ng/L. The relative standard deviations (RSDs) of peak height, peak area and migration time are 3.7–5.8, 3.3–5.3 and 4.3–6.9% for five individual determinations, respectively. 3.6. Analysis of cigarette sample Inorganic anions are the important components and can influence the quality of cigarettes. Nitrate and nitrite can affect the formation of other nitrogen compounds in cigarette; meanwhile, nitrite is the precursor of tobacco-specific nitrosamines [21]. In addition, superabundant fluoride can affect the immune system and the hemoglobin amount of human body, etc. So it is necessary to determine the inorganic anions, especially the trace anions of NO2 − and F− in cigarette samples. In the sample preparation, the extraction time and water volume were investigated with the ultrasonic cleaner. The water volume and extraction time were examined from 5 to 30 mL and 5 to 20 min, respectively. In accordance with the experiment results, 20 mL water and 15 min are selected for the extraction of 0.500 g cigarette sample. With the selected conditions aforementioned, six inorganic anions in real cigarette sample were determined and the electropherogram is shown in Fig. 6. The results show that the proposed method can be adopted for the analysis of the inorganic anions in real cigarette sample, especially for the trace anions of NO2 − and F− . The quantitative determination is based on the calibration curves of peak area, as shown in Table 1. The mixed calibration solutions with three concentration levels and the samples were detected in one day. The RSDs of peak area are less than 5.3% and the recovery of the analyte anions ranges from 93.5 to 98.7%.
290
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
Table 1 Calibration data, RSD and concentration of anions in cigarette sample Anion
Cl− NO2 − NO3 − SO4 2− F− HPO4 2−
Calibration data Slope (mAU/g/L)
Intercept (mAU)
Linear correlation coefficient (r)
2.6 × 102 4.2 × 102 2.0 × 102 1.7 × 102 1.9 × 102 1.2 × 102
3.7 −5.6 8.9 6.3 −3.3 1.4
0.998 0.994 0.997 0.990 0.996 0.992
Relative standard deviation of peak area (RSD, %, n = 5)
Concentration (g/g)
3.3 4.2 4.7 3.4 5.3 4.4
(5.5 (1.2 0.23 0.25 (8.5 (1.6
± ± ± ± ± ±
0.4) × 10−2 0.3) × 10−3 0.05 0.02 0.1) × 10−4 0.3) × 10−2
tion factor. TTAB added into the buffer can suppress EOF in the capillary. The introduced water plug has the functions of dissolving TTAB from the capillary inner wall, enhancing EOF towards the capillary inlet and introducing large numbers of analyte anions into the capillary. With ASEI–BS, more analytes can be injected into the capillary than with ASEI alone and the injected analyte zone can be condensed further. The preconcentration factor of (0.8–1.3) × 105 was obtained compared with the electrokinetic injection. The proposed method has successfully determined six anions in real cigarette samples, especially for the analysis of trace anions of NO2 − and F− . Acknowledgement The authors thank the Natural Science Foundation of China (Nos. 20275035 and 29975026) for financial supports to this work. References [1] [2] [3] [4] [5]
Fig. 6. Electropherogram of six anions in cigarette sample. Extraction conditions: 0.500 g cigarette sample is extracted in 20 mL water for 15 min by the ultrasonic cleaner at room temperature; other conditions are the same as in Fig. 4. 1: Cl− ; 2: NO2 − ; 3: NO3 − ; 4: SO4 2− ; 5: F− and 6: HPO4 2− .
[6] [7] [8] [9] [10] [11] [12] [13] [14] [15]
4. Conclusions
[16] [17]
ASEI–BS–CZE is a simple, convenient and high sensitivity method to analyze trace amounts of anions. The enrichment method based on ASEI and BS can provide high preconcentra-
[18] [19] [20] [21]
E. Santoyo, R. Garc´ıa, R. Abella, J. Chromatogr. A 920 (2001) 325. S. Park, C.E. Lunte, J. Microcol. Sep. 10 (1998) 511. Y. Xiong, S.R. Park, H. Swerdlow, Anal. Chem. 70 (1998) 3605. S.D. Arnett, C.E. Lunte, Electrophoresis 24 (2003) 1745. C.X. Cao, Y.Z. He, M. Li, Y.T. Qian, L. Yang, J. Chromatogr. A 952 (2002) 39. C.X. Cao, Y.Z. He, M. Li, Y.T. Qian, Anal. Chem. 74 (2002) 4167. P. Britiz-McKibbin, K. Otsuka, S. Terabe, Anal. Chem. 74 (2002) 3736. J.P. Quirino, S. Terabe, Anal. Chem. 72 (2000) 1023. K. Isoo, S. Terabe, Anal. Chem. 75 (2003) 6789. E.S. Burgi, R.L. Chien, Anal. Biochem. 202 (1992) 306. J.H. Lee, O.K. Chio, H.S. Jung, Electrophoresis 21 (2000) 930. J.B. Kim, K. Otsuka, S. Terabe, J. Chromatogr. A 932 (2001) 129. L.Y. Zhu, H.K. Lee, Anal. Chem. 73 (2001) 3065. L.Y. Zhu, C.H. Tu, H.K. Lee, Anal. Chem. 74 (2002) 5820. A. Maci`a, F. Borrull, C. Aguilar, M. Calull, Electrophoresis 24 (2003) 2779. D.S. Burgi, Anal. Chem. 65 (1993) 3726. M. Albert, L. Debusschere, C. Demesmay, J. Chromatogr. A 757 (1997) 291. F. Kohlrausch, Ann. Phys. (Leipzig) 62 (1897) 209. J.C. Reijenga, E. Kenndler, J. Chromatogr. A 659 (1994) 403. W. Friedl, J.C. Reijenga, E. Kenndler, J. Chromatogr. A 709 (1995) 163. H.R. Burton, N.K. Dye, J. Agric. Food Chem. 42 (1994) 2007.
On-line sequential preconcentration of inorganic anions by anion-selective exhaustive injection and base-stacking in capillary zone electrophoresis Zhao-Xiang Zhang a,b , You-Zhao He a,∗ , Yan-Yun Hu a a
b
Department of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026, China College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, China Received 28 September 2005; received in revised form 29 December 2005; accepted 10 January 2006 Available online 14 February 2006
Abstract A sequential electrostacking method based on anion-selective exhaustive injection (ASEI) and base-stacking (BS) is presented for the preconcentration and determination of inorganic anions by capillary zone electrophoresis (CZE) in this paper. Tetradecyltrimethylammonium bromide as an electroosmotic flow (EOF) modifier was added into the buffer to suppress EOF of the capillary. Firstly, a water plug was hydrodynamically injected into the capillary. During ASEI under negative high voltage, the sample anions migrated quickly towards the boundary between the water plug and buffer in the capillary. Then an alkaline zone was injected electrokinetically to concentrate the anions further. With the sequential electrostacking method, the preconcentration factor of (0.8–1.3) × 105 was obtained compared with the conventionally electrokinetic injection and the relative standard deviation of peak area was 3.3–5.3% (n = 5). The detection limits of ASEI–BS–CZE for six inorganic anions were 6–14 ng/L. The proposed method has been adopted to analyze six anions in cigarette samples successfully. © 2006 Elsevier B.V. All rights reserved. Keywords: Inorganic anions; Anion-selective exhaustive injection; Base-stacking; Capillary zone electrophoresis; Cigarette
1. Introduction Capillary electrophoresis (CE) has been employed for rapid separation of inorganic anions in biochemical and environmental studies [1]. However, its extremely short light path for ultraviolet–visible (UV–vis) detection still limits the separation technique as a sensitive one. On-column stacking is an effective means to improve the detection sensitivity of UV–vis in CE. Park and Lunte [2] presented an acid-stacking method to concentrate cationic analytes, in which fast migrating hydrogen ions from acid zone reacted with the buffer weak anions, produced a low conductivity region and made the cationic stacking. Similarly, by electrokinetically injecting an alkaline plug into a capillary, anionic stacking occurred [3,4]. Cao and co-workers [5,6] proposed a “transient moving chemical reaction boundary” method to concentrate zwitterionic analytes, in which the charge polarity of zwitterionic analytes was changed during their migration from basic
∗
Corresponding author. Tel.: +86 551 3607072; fax: +86 551 3603388. E-mail address: [email protected] (Y.-Z. He).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.01.028
sample zone into acidic buffer region and the analytes were stacked at the moving boundary. Britz-Mckibbin et al. [7] proposed a dynamic pH junction-sweeping method to concentrate flavin derivatives with the improved sensitivity of more than 1200-fold. Terabe and co-workers developed a dual preconcentration method of cation-selective exhaustive injection and sweeping to concentrate organic base [8] and metal ions [9], which improved the detection sensitivity to a million-fold for organic bases and 140,000-fold for divalent and trivalent metal ions. To obtain a high concentration factor and satisfied separation properties, a large volume of sample solution should be introduced into and its medium ought to be removed out of separation capillary. The removal of sample medium can not only enhance the analyte concentration, but also be propitious to the following separation, for it can provide an effective separation length and eliminate the non-uniform distribution of electric field strength. A simple way to remove sample medium is the change of the electrode polarity after sample injection, in which the reversed EOF can expel the sample medium out of the capillary inlet [10]. However, it is difficult to monitor the current of the polarity conversion exactly. Another way adopted for large volume injection
286
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
and sample medium removal is the EOF restriction. By using the coated capillary [11–13], low pH buffer [14], non-aqueous CE [15] or EOF modifier [16,17], EOF can be suppressed to be even lower than the analyte electrophoresis mobility. Kim et al. [12] proposed anion-selective exhaustive injection-sweep micellar electrokinetic capillary chromatography (MECC) to concentrate organic acids with a polyacrylamide-coated capillary. With low pH buffer, sample medium can be removed by an electroosmotic flow pump under negative voltage during analyte stacking [14]. However, it is only fit for some organic and inorganic anions retained in low pH solution. By using methanol as the buffer solvent, non-aqueous CE can improve the detection sensitivity of organic anions by field-enhanced sample injection and reduce sample medium introduction [15]. EOF modifiers have been used to remove sample medium from the capillary inlet under a reversed potential after the hydrodynamic sample injection. The reported methods improved the detection sensitivity by 30–40 folds because of the deficient sample injection [16,17]. In this paper, a sequential electrostacking method based on anion-selective exhaustive injection (ASEI) and base-stacking (BS) was developed with the preconcentration factor of 105 and was used to concentrate and determine six inorganic anions in cigarette samples successfully. 2. Experimentals 2.1. Apparatus A 1229 HPCE Analyzer (Institute of New Technology Application, Beijing, China) with detection at 254 nm, a N−2000 double-channel chromatography processor (Institute of Intelligence Information and Engineering, Zhejiang Univ., Zhejiang, China) and an S–2200 ultrasonic cleaner (120 W, 35 kHz, J & L Ultrasonic Science and Technology Corp., Shanghai, China) were employed in this work. A 50 m i.d. capillary with total and effective length of 48 and 33 cm, respectively, was purchased from Yongnian Chromatographic Components Ltd. (Hebei, China). 2.2. Reagents All the chemicals were of analytical reagent grade and the required solutions were prepared with tri-distilled water (SZ-3 tri-distilled water system, Huxi Anal. Instrument Factory, Shanghai, China). Tetradecyltrimethylammonium bromide (TTAB) was purchased from Sigma (St. Louis, MO, USA). Sodium chromate, Tris and glacial acetic acid (HAc), etc. were purchased from Chemical Reagent Co. (Shanghai, China). Anion stock solutions (1000 mg/L) were prepared from their sodium or potassium salts. Three mixed standard solutions in the range of 0.020–6.0 g/L were prepared by diluting the stock solutions with tri-distilled water.
(0.500 g) was weighed in a 50 mL conical flask. After adding 20 mL water, the flask was capped and the anionic analytes were extracted with the ultrasonic cleaner for 15 min. The treated solution was filtered with a 0.22 m acetic cellulose membrane, diluted to 50 mL and reserved in refrigerator. The running buffer contained 40 mmol/L sodium chromate as background electrolyte (BGE), 0.40 mmol/L TTAB and 20 mmol/L Tris, and adjusted to pH 8.9 with HAc. 2.4. ASEI−BS−CE procedure The capillary was flushed daily in the order of H2 O (1 min), 1.0 mol/L HCl (5 min), H2 O (1 min), 1.0 mol/L NaOH (15 min), H2 O (1 min) and conditioned with the running buffer for 10 min by pressure. Between two runs, the capillary was conditioned with the running buffer for 6 min. The main steps of ASEI−BS−CE are illustrated in Fig. 1. Firstly, the capillary was pre-equilibrated with the running buffer containing TTAB, which was dynamically adsorbed on the inner wall of the capillary (Fig. 1a). Before sample injection, a water plug was hydrodynamically injected into the capillary. TTAB on the capillary inner surface of the water section was dissolved into the water plug and the capillary surface restored negative charges, i.e. negative zeta potential (Fig. 1b). With the voltage of −6 kV for 210 s in ASEI, the anionic analytes migrated rapidly into the capillary through the water plug and stacked on the boundary between the water plug and running buffer region (Fig. 1c). Once the anions reached the boundary, their velocities slowed down for the electric field strength decreased in the running buffer region. Meanwhile, the water plug was expelled because of the EOF direction towards the capillary inlet (Fig. 1d). At the same time, the capillary section was filled with the running buffer and TTAB in the buffer was adsorbed on the capillary inner wall again, so the EOF of the capillary became smaller and smaller. After sample loading, an alkaline zone was injected from 0.1 mol/L NaOH solution at −6 kV for 120 s (Fig. 1e), a neutralization reaction between OH− and Tris+ from the buffer took place and reduced the conductivity of analyte zone. In this way, the injected anions were condensed further. By replacing the alkaline solution with buffer, CZE separation was carried out at −6 kV (Fig. 1f).
2.3. Cigarette sample and buffer preparation Cigarette samples were dried at 60 ◦ C for 4 h, ground into powder and sieved with a 40 mesh sifter. Cigarette sample
Fig. 1. Schematic diagrams of ASEI–BS–CE processes.
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
3. Results and discussion 3.1. Separation conditions Sodium chromate, potassium hydrogen phthalate and paminobenzoic acid were examined as BGE for indirect UV detection. The results indicate that CrO4 2− is much suitable for the anion separation and its mobility close to the average mobility of the inorganic anions. As described by Kohlrausch regulating function [18], when the analyte zone has a higher mobility than the running buffer, the leading edge of the analyte zone will be diffuse and the trailing edge sharp. Contrarily, the former will be sharp and the latter diffuse. In Fig. 2, it can be found that the mobility of CrO4 2− is located between Cl− and F− . The effect of Na2 CrO4 concentration on the anion separation is also illustrated in Fig. 2. When Na2 CrO4 concentration is 10 mmol/L, the peaks of SO4 2− and NO2 − overlap completely and those of F− and HPO4 2− cannot be separated on baseline (Fig. 2a). When Na2 CrO4 concentration increases to 20 mmol/L, SO4 2− peak shifts backward and superposes NO3 − one, but the baseline separation of F− and HPO4 2− is achieved. In Fig. 2b, six anions can be completely separated when Na2 CrO4 concentration increases to 40 mmol/L. If the Na2 CrO4 concentration increases further, the peak height of six anions decreases and the retention time of anions are prolonged. It implies that the increase of Na2 CrO4 concentration will lead to the increase of the ionic strength in BGE, the decrease of the anionic mobility [19,20], and thus the prolongation of the anionic retention time. So 40 mmol/L Na2 CrO4 was chosen in this work. For base-stacking, the BGE solution should contain a weak base ion, such as Tris+ or NH4 + [3,4]. Several buffer systems
Fig. 2. Effect of Na2 CrO4 concentration on inorganic anion separation. The capillary is 50 m i.d. with its total length of 48 cm and effective length of 33 cm; the sample solution is injected at −6 kV for 10 s and the separation voltage is −6 kV; the running buffer contains pH 8.9, 20 mmol/L Tris, 0.4 mmol/L TTAB and CrO4 2− concentration: (a) 10 and (b) 40 mmol/L. Each anion concentration is 5.0 mg/L. 1: Cl− ; 2: NO2 − ; 3: NO3 − ; 4: SO4 2− ; 5: F− and 6: HPO4 2− .
287
were investigated, including ammonium tartrate/ammonium hydroxide, ammonium oxalate/ammonium hydroxide and Tris+ /HAc. A system peak can appear behind F− or SO4 2− peak with ammonium tartrate or ammonium oxalate buffer, which can interfere on the analyte peak of F− or SO4 2− . However, the system peak of Tris+ /HAc buffer is behind the analyte peaks and its influence on the anion separation and peak shape can be ignored. When the pH value of the buffer is lower than 8.5, the buffer color changes to orange, which results from a reaction equilibrium between CrO4 2− and Cr2 O7 2− . Meanwhile, it makes the peak width and retention time increased. By changing the buffer pH from 8.9 to 11.3, no significant difference can be observed on the separation efficiency and anionic resolution. However, when the pH value is higher than 11.3, there are two phosphate peaks of HPO4 2− and PO4 3− in the electropherogram, and the PO4 3− peak can appear in front of peak F− . So pH 8.9 Tris+ /HAc buffer was used in this work. TTAB in buffer can be dynamically adsorbed on the capillary inner wall and affect EOF of the capillary. When TTAB concentration is less than 0.38 mmol/L, the mobility of EOF increases towards the cathode relatively. For the electrophoretic mobility of analyte anions is towards anode, the peaks of the anionic analytes are broadening. When the concentration of TTAB locates in the range of 0.38–0.42 mmol/L, the EOF becomes extremely low (2.6 × 10−5 cm2 /V s) and the peak width and separation efficiency of the anions are satisfied. So 0.40 mmol/L TTAB was adopted in this work.
3.2. Anion-selective exhaustive injection The water plug can provide high field strength to introduce more analyte anions into the capillary during ASEI. TTAB on the capillary inner wall can be dissolved into the water plug, which enhances the negative zeta potential in the water section of the capillary and increases the EOF towards the capillary inlet under high negative voltage [16,17]. During ASEI, the analyte anions are introduced into the capillary at −6 kV and the water plug is removed out of the capillary inlet by the EOF. The amount of anions injected is dependent on the sample injection time under a fixed injection voltage. The effect of water and sample injection time on peak height and separation efficiency was investigated, as shown in Fig. 3. The injection time of water plug was evaluated with 120 (Fig. 3a), 180 (Fig. 3b), 240 (Fig. 3c), 300 (Fig. 3d), 360 (Fig. 3e) and 420 s (Fig. 3f) under 10 cm height difference, respectively. The sample injection time was changed from 50 to 230 s at −6 kV. It can be found that the optimal peak height and separation efficiency are obtained with the water/sample injection time pair of 120 s/70 s, 180 s/90 s, 240 s/130 s, 300 s/160 s, 360 s/180 s and 420 s/210 s, respectively. It also shows that the peak height increases with the increase of the water/sample injection time pair up to 300 s/160 s (Fig. 3a–d), in which the corresponding length of water plug is about 5.3 cm. With the water/sample injection time pair longer than 300 s/160 s, both the peak height and separation efficiency decrease (Fig. 3e and f).
288
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
Fig. 3. Effect of water/sample injection time on peak height and separation efficiency of Cl− (other inorganic anions display the similar results). Cl− concentration is 1.0 g/L; the buffer contains 40 mmol/L Na2 CrO4 , 0.4 mmol/L TTAB and 20 mmol/L Tris at pH 8.9; other conditions are the same as in Fig. 2. Water injection time is (a) 120, (b) 180, (c) 240, (d) 300, (e) 360 and (f) 420 s.
3.3. ASEI–BS sequential preconcentration According to ASEI conditions aforementioned, the optimal stacking can be obtained with the sample injection time of 160 s. If the injection time is longer than 160 s, it can cause the decrease of separation efficiency. To introduce more analytes into the capillary, a base-stacking method was adopted and OH− from 0.1 mol/L NaOH was introduced at −6 kV after sample injection. Since the mobility of OH− is higher than the analyte ions, the fast-migrating OH− passes through the sample zone, reacts with the weak base cation Tris+ from the buffer, creates a low
conductivity region [3] and condenses the anionic analyte zone further. The effect of sample and base injection time on the peak height and separation efficiency of the anions was also investigated with both the injection voltages of −6 kV. Firstly, the injection time of alkaline zone was surveyed with the sample injection time of 210 s at −6 kV and the water injection time of 420 s under 10 cm height difference, corresponding to the water plug length of 7.4 cm, as shown in Fig. 3f. The effect of the alkaline injection time on the anion stacking is illustrated in Fig. 4. It can be found that the separation efficiency of the anions is improved by increasing the alkaline injection time from 90 to 120 s (Fig. 4a and b). However, when the injection time increases to 150 s, a buffer matrix peak and shifted baseline can appear in the electropherogram (Fig. 4c), which indicates that the OH− amount has been excessive. The optimal separation efficiency and resolution can be obtained with the alkaline injection time of 120 s. The shifted baseline may result from the excessive concentration of OH− and the decreased concentration of CrO4 2− , for the migration charges of the anions should be constant when the current equilibrium is achieved in CZE. The buffer matrix peak may be Ac− stacked in the corresponding low-conductivity region formed by TrisOH. The experiment results also indicate that the slightly excessive injection of alkaline zone will not reduce the separation efficiency and anionic resolution (Fig. 4c). However, when the injection time of alkaline zone is longer than 160 s, both the separation efficiency and resolution will decrease besides the buffer matrix peak and shifted baseline. According to the above-mentioned conditions in Section 3.2, excessive sample injection times were examined with 180, 190, 200, 210 and 220 s, respectively, and then the injection
Fig. 4. Electropherograms of inorganic anions with different alkaline injection time in ASEI–BS–CZE. The water and sample injection time is 420 s under 10 cm height difference and 210 s at −6 kV, respectively; the anion concentration is 0.2 g/L; other conditions are the same as in Fig. 3. Injection time of OH− from 0.1 mol/L NaOH at −6 kV is: (a) 90, (b) 120 and (c) 150 s.
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
289
by the buffer one, the current trace becomes plateau (after point c ) and CZE separation is carried out. In ASEI–BS–CZE, the sample vial is replaced by the alkaline one and the electric current experiences a decrease course again because OH− migrates toward the anode rapidly, reacts with Tris+ from the buffer region, produces a low conductivity zone of TrisOH and leads to stack the injected anions (210 s at −6 kV) further. Obviously, the current decrease from 41 to 10 A (c → d) results from the formation of the low conductivity zone. By comparing the current variation of ASEI with that of ASEI–BS, the mechanism of the second concentration step can be demonstrated as a basestacking. At the time d, the buffer vial replaces the alkaline solution one and the equilibrium current of CZE is recovered gradually. In addition, all the separated peaks are located after the time of e in ASEI–BS–CZE. Fig. 5. Current variation of ASEI–CZE (curve 1) and ASEI–BS–CZE (curve 2). The sample injection time is 160 s at −6 kV for ASEI–CZE, and the sample and base injection time are 210 and 120 s at −6 kV for ASEI–BS–CE.
time of alkaline solution was evaluated. The optimal results were obtained with the sample/ alkaline injection time pair of 180 s/50 s, 190 s/70 s, 200 s/100 s and 210 s/120 s, respectively. The experiment results show that the alkaline injection time should match the sample injection one in order to obtain an optimal separation efficiency and resolution. However, when the sample injection time is longer than 220 s, the baseline separation of NO2 − and NO3 − cannot be achieved no matter how to adjust the injection time of alkaline zone. So 210 and 120 s were adopted for the sample and alkaline injection time in this work, respectively, in which the water injection time was 420 s. 3.4. Mechanism of sequential preconcentration For elucidating the mechanism of the sequential preconcentration, the electrophoretic currents of ASEI–CZE and ASEI–BS–CZE were investigated. The sample injection time was 160 s at −6 kV for ASEI–CZE, while the sample and base injection time was 210 and 120 s, respectively, at −6 kV for ASEI–BS–CE. As shown in Fig. 5, curve 1 (a → b → c → e ) and 2 (a → b → c → d → e) represent the current variation in ASEI–CZE and ASEI–BS–CZE, respectively. The periods of a → c (a → c ) and c → d correspond to the sample and base injection, respectively. The periods after c in ASEI–CZE and after d in ASEI–BS–CZE are relative to the electrophoresis separation. During ASEI, the electric current increases from 3 to 9 A in 150 s (a → b ) and from 2 to 8 A in 200 s (a → b) gradually, then enhances rapidly to the top value of 42 A only in 10 s (b → c and b → c ). It manifests that the initially low electric current results mainly from the existence of the water plug, i.e. 5.3 cm in ASEI and 7.4 cm in ASEI–BS. During the sample injection, the water plug is expelled out of the capillary inlet by EOF and the electric current increases gradually. When the water plug is nearly eliminated, the electric current increases rapidly (b → c and b → c). In ASEI–CZE, the sample vial is replaced
3.5. Analytical characteristics Comparing with the conventional electrokinetic injection, the preconcentration factor (PF) with ASEI–BS has been calculated. The PF of (0.8–1.3) × 105 by ASEI–BS for six inorganic anions can be achieved under similar separation efficiencies. The limit of detection (LOD) is obtained with three-time standard deviation of 11 noise peaks of the baseline near the analyte peaks. For the proposed method of ASEI–BS–CZE, the LODs of six inorganic anions are in the range of 6–14 ng/L. The relative standard deviations (RSDs) of peak height, peak area and migration time are 3.7–5.8, 3.3–5.3 and 4.3–6.9% for five individual determinations, respectively. 3.6. Analysis of cigarette sample Inorganic anions are the important components and can influence the quality of cigarettes. Nitrate and nitrite can affect the formation of other nitrogen compounds in cigarette; meanwhile, nitrite is the precursor of tobacco-specific nitrosamines [21]. In addition, superabundant fluoride can affect the immune system and the hemoglobin amount of human body, etc. So it is necessary to determine the inorganic anions, especially the trace anions of NO2 − and F− in cigarette samples. In the sample preparation, the extraction time and water volume were investigated with the ultrasonic cleaner. The water volume and extraction time were examined from 5 to 30 mL and 5 to 20 min, respectively. In accordance with the experiment results, 20 mL water and 15 min are selected for the extraction of 0.500 g cigarette sample. With the selected conditions aforementioned, six inorganic anions in real cigarette sample were determined and the electropherogram is shown in Fig. 6. The results show that the proposed method can be adopted for the analysis of the inorganic anions in real cigarette sample, especially for the trace anions of NO2 − and F− . The quantitative determination is based on the calibration curves of peak area, as shown in Table 1. The mixed calibration solutions with three concentration levels and the samples were detected in one day. The RSDs of peak area are less than 5.3% and the recovery of the analyte anions ranges from 93.5 to 98.7%.
290
Z.-X. Zhang et al. / J. Chromatogr. A 1109 (2006) 285–290
Table 1 Calibration data, RSD and concentration of anions in cigarette sample Anion
Cl− NO2 − NO3 − SO4 2− F− HPO4 2−
Calibration data Slope (mAU/g/L)
Intercept (mAU)
Linear correlation coefficient (r)
2.6 × 102 4.2 × 102 2.0 × 102 1.7 × 102 1.9 × 102 1.2 × 102
3.7 −5.6 8.9 6.3 −3.3 1.4
0.998 0.994 0.997 0.990 0.996 0.992
Relative standard deviation of peak area (RSD, %, n = 5)
Concentration (g/g)
3.3 4.2 4.7 3.4 5.3 4.4
(5.5 (1.2 0.23 0.25 (8.5 (1.6
± ± ± ± ± ±
0.4) × 10−2 0.3) × 10−3 0.05 0.02 0.1) × 10−4 0.3) × 10−2
tion factor. TTAB added into the buffer can suppress EOF in the capillary. The introduced water plug has the functions of dissolving TTAB from the capillary inner wall, enhancing EOF towards the capillary inlet and introducing large numbers of analyte anions into the capillary. With ASEI–BS, more analytes can be injected into the capillary than with ASEI alone and the injected analyte zone can be condensed further. The preconcentration factor of (0.8–1.3) × 105 was obtained compared with the electrokinetic injection. The proposed method has successfully determined six anions in real cigarette samples, especially for the analysis of trace anions of NO2 − and F− . Acknowledgement The authors thank the Natural Science Foundation of China (Nos. 20275035 and 29975026) for financial supports to this work. References [1] [2] [3] [4] [5]
Fig. 6. Electropherogram of six anions in cigarette sample. Extraction conditions: 0.500 g cigarette sample is extracted in 20 mL water for 15 min by the ultrasonic cleaner at room temperature; other conditions are the same as in Fig. 4. 1: Cl− ; 2: NO2 − ; 3: NO3 − ; 4: SO4 2− ; 5: F− and 6: HPO4 2− .
[6] [7] [8] [9] [10] [11] [12] [13] [14] [15]
4. Conclusions
[16] [17]
ASEI–BS–CZE is a simple, convenient and high sensitivity method to analyze trace amounts of anions. The enrichment method based on ASEI and BS can provide high preconcentra-
[18] [19] [20] [21]
E. Santoyo, R. Garc´ıa, R. Abella, J. Chromatogr. A 920 (2001) 325. S. Park, C.E. Lunte, J. Microcol. Sep. 10 (1998) 511. Y. Xiong, S.R. Park, H. Swerdlow, Anal. Chem. 70 (1998) 3605. S.D. Arnett, C.E. Lunte, Electrophoresis 24 (2003) 1745. C.X. Cao, Y.Z. He, M. Li, Y.T. Qian, L. Yang, J. Chromatogr. A 952 (2002) 39. C.X. Cao, Y.Z. He, M. Li, Y.T. Qian, Anal. Chem. 74 (2002) 4167. P. Britiz-McKibbin, K. Otsuka, S. Terabe, Anal. Chem. 74 (2002) 3736. J.P. Quirino, S. Terabe, Anal. Chem. 72 (2000) 1023. K. Isoo, S. Terabe, Anal. Chem. 75 (2003) 6789. E.S. Burgi, R.L. Chien, Anal. Biochem. 202 (1992) 306. J.H. Lee, O.K. Chio, H.S. Jung, Electrophoresis 21 (2000) 930. J.B. Kim, K. Otsuka, S. Terabe, J. Chromatogr. A 932 (2001) 129. L.Y. Zhu, H.K. Lee, Anal. Chem. 73 (2001) 3065. L.Y. Zhu, C.H. Tu, H.K. Lee, Anal. Chem. 74 (2002) 5820. A. Maci`a, F. Borrull, C. Aguilar, M. Calull, Electrophoresis 24 (2003) 2779. D.S. Burgi, Anal. Chem. 65 (1993) 3726. M. Albert, L. Debusschere, C. Demesmay, J. Chromatogr. A 757 (1997) 291. F. Kohlrausch, Ann. Phys. (Leipzig) 62 (1897) 209. J.C. Reijenga, E. Kenndler, J. Chromatogr. A 659 (1994) 403. W. Friedl, J.C. Reijenga, E. Kenndler, J. Chromatogr. A 709 (1995) 163. H.R. Burton, N.K. Dye, J. Agric. Food Chem. 42 (1994) 2007.