CHINESE JOURNAL OF CHROMATOGRAPHY Volume 24, Issue 6, November 2006 Online English edition of the Chinese language journal Cite this article as: Chin J Chromatogr, 2006, 24(6): 597–600.
RESEARCH PAPER
Separation of Phthalates in Nonaqueous Micelle Using Capillary Electrokinetic Chromatography HUANG Rui1, MU Xiaojing1, YIN Yongguang2, WEI Weili1, CHEN Zhitao3, XIA Zhining1,* 1
College of Chemistry and Chemical Engineering, Chongqing University, Chongqing 400044, China Research Center for Eco-Environmental Sciences, the Chinese Academy of Sciences, Beijing 100085, China 3 College of Bioengineering, Chongqing University, Chongqing 400044, China 2
Abstract: On the basis of nonaqueous capillary electrophoresis (NACE) and micellar electrokinetic chromatography (MEKC), a novel technique, nonaqueous micellar electrokinetic chromatography (NAMEKC), has been established. NAMEKC has the advantages of NACE and uses the separation mechanism of MEKC, showing special advantages for separation of hydrophobic compounds. Separation of three of the priority pollutants named by the US Environmental Protection Agency (EPA), i.e. dimethyl phthalate (DMP), diethyl phthalate (DEP), and dibutyl phthalate (DBP), was realized in 15 min by NAKEKC. Important factors on separation, such as the amount of water added in the electrophoretic running buffer, the acidity of water phase, organic additive, and the concentration of SDS, were investigated. The proportion of water in the electrophoretic running buffer could affect the current and the stability of SDS micelle. Organic additives and the acidity of water phase showed no effect on resolution. The concentration of SDS was a dominant factor, strongly affecting the partition of analytes in micelle. DMP, DEP, and DBP were separated in a short time under the optimized operation conditions using 20 mmol/L NaH2PO4 and 120 mmol/L SDS in formamide/water (9/1, v/v). The application of NAMEKC leads to successful separation of the three typical hydrophobic compounds, which provides a novel means to separate and analyze hydrophobic compounds. Key Words: nonaqueous micellar electrokinetic chromatography; nonaqueous capillary electrophoresis; micellar electrokinetic chromatography; phthalate; formamide
Micellar electrokinetic chromatography (MEKC), established by Terabe, has expanded the scope of detection of analytes by capillary electrophoresis (CE) [1–3], especially the neutral molecules. However, MEKC is incapable of analyzing some organic samples because of their low solubility in micelle. Ever since Walbroehl and Jorgenson [4] first introduced nonaqueous medium into CE, nonaqueous capillary electrophoresis (NACE) has been rapidly developed because it not only can dissolve many nonionic organic compounds but also can broaden the range of electroosmotic flow (EOF) [5–7]. However, it fails to dissolve some extremely hydrophobic compounds. Furthermore, the weak solvophobic interaction results in low resolution of these highly hydrophobic organic compounds. Lin et al. [8–10] discovered that surfactants could form micelle in nonaqueous medium of relatively higher cohesive energy density. For instance, sodium dodecyl sulfate (SDS)
can form stable micelle in formamide (FA) medium with critical micelle concentration (CMC) of 20–30 mmol/L. On the basis of the mechanism of MEKC, the separation and analysis of hydrophobic compounds in such FA medium possess both advantages of MEKC and NACE, thereby resulting in higher resolution. According to published reports [8–10], polycyclic aromatic hydrocarbons (PAHs), steroids, and alkylphenones were separated when organic additives were added to FA medium, such as SDS, sodium 2-ethylhexyl sulfosuccinite (AOT), or sodium taurodeoxycholate (STDC). Because of the better solubility of analytes compared with MEKC, this novel method was called nonaqueous micellar electrokinetic chromatography (NAMEKC). NAMEKC has great potential in increasing the dissolution and widening the application range of MEKC since its establishment. Phthalates have been widely used as plasticizers in plastic industry. However, they are also one of the most common
Received February 27, 2006; revised May 10, 2006 *Corresponding author. E-mail:
[email protected] This work was supported by the National Natural Science Foundation of China (Grant No. 20375051), the Trans-Century Training Program Foundation for the Talents by Ministry of Education of China (Educational Department [20013]) and the Innovation Foundation for the Graduate Education of Chongqing University, China (Grant No. 2004A007). Copyright © 2006, Chinese Chemical Society and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier BV. All rights reserved.
HUANG Rui et al. / Chinese Journal of Chromatography, 2006, 24(6): 597–600
kinds of pollutants, which may interfere in the secretion of hormone and induce cancer after a long-term accumulation in the body. Therefore, many countries including USA have considered phthalates as the priority pollutants. At present, phthalates are mainly separated and analyzed by gas chromatography (GC) and high performance liquid chromatography (HPLC) [11–13]. Baker [14] found that the retention times of phthalates in reversed phase HPLC (RP-HPLC) were rather long because of high hydrophobicity. Therefore, it is of interest to find an efficient and convenient separation method. In this study, phthalates were analyzed using NAMEKC. As important factors that determine separation, the effect of the composition of non-aqueous medium was investigated. SDS micelle was used as pseudo-stationary phase in FA medium. To separate these hydrophobic components, concentration of SDS, and acidity and ionic strength of water phase were optimized, and dimethyl phthalate (DMP), diethyl phthalate (DEP), and dibutyl phthalate (DBP) were successfully separated.
1 Experimental section 1.1 Instruments Capillary electrophoresis system was set up with 0–30 kV adjustable DC power supply (Tianjin Dongwen DC Power Supply Plant, China), 190–900 nm UV-Vis spectrometer (Beijing Zolix Instrument Co., Ltd, China), 71D101-CR131 photomultiplier tube detector (Beijing Saifan Optical Instrument Co., Ltd, China) and HW-2000 Chromatographic Workstation (Nanjing Qianpu Software Co., Ltd, China) for data collection and operation. Uncoated silica capillary (inner diameter 75 μm, total length 49.5 cm, and effective length 40 cm, Hebei Yongnian Ruifeng Chromatographic Implements Co., Ltd, China). 1.2 Reagents FA (Chengdu Kelong Chemical Reagent Plant, China), methanol (Chongqing Chuandong Chemical Co., Ltd, China), DMP (Chongqing Dongfang Reagent Plant, China) and DBP (Chengdu Jinshan Chemical Reagent Plant, China) were of analytical grade. DEP (Guangdong Shantou Xinning Chemical Plant, China) and SDS (Tianjin Tiantai Chemical Co., Ltd, China) were of chemically pure. 1.3 Operations Bare fused silica capillaries were pretreated by sequentially washing with 0.1 mol/L NaOH for 20 min, H2O for 10 min, and pure FA for 20 min. Between runs, they were rinsed with running medium for 5 min. Electrokinetic injections at 20 kV for 5 s were used to introduce samples into the capillary. Separations were performed at 20 kV. Detection wavelength was 254 nm. All runs were performed at ambient temperature. The running buffer was FA/H2O (9/1, v/v) containing 20 mmol/L NaH2PO4 and 120 mmol/L SDS. The concentrations of DMP, DEP, and DBP
were 20 mmol/L.
2 Results and discussion 2.1 Electrophoresis of dialkyl phthalates in pure FA medium DMP, DEP, and DBP were separated in pure FA medium, and the electropherogram obtained is shown in Fig. 1. The migration times of all the three dialkyl phthalates were 14.7 min and the apparent mobility was 11.2 × 10-5 cm2/ s·V. There was no evidence of separation. The slight tailing of the peak might be induced by the adsorption of dialkyl phthalates to the capillary wall. Since dialkyl phthalates are neutral in FA medium, they do not exhibit efficient mobility. Otherwise, in aqueous capillary zone electrophoresis (CZE), the migration time of EOF marker acetone was 8–9 min. It indicated that EOF in pure FA medium was lower than that in water, provided all other conditions are the same. The migration time of the marker acetone was also 14 min, which implied that the efficient mobility of the three compounds was zero. Although the resolution could be improved 1.5 times when FA was used as electrophoresis medium [15], these three analytes could not be separated at all. It was obvious that there was tremendous difficulty in separating the three dialkyl phthalates.
Fig. 1 Electropherogram of dialkyl phthalates in formamide medium Conditions: applied voltage, 20 kV; detection at 254 nm; electrophoretic injection, 20 kV × 5 s; without SDS and aqueous solution. 1. DMP; 2. DEP; 3. DBP.
In the experiments carried out in this study, it was discovered that there was 4 µA current in FA medium without any electrolyte, which might result from self-electrolysis of FA. Hence, the electrostatic interaction between FA and analytes should be considered to be utilized for the separation. 2.2 Effect of the amount of water added to running buffer The amount of water added to FA medium influenced the separation of the phthalates. The effect of water at various amounts of 10%, 15%, and 20% (v/v) was separately investigated. The increase of the amount of water added caused in-
HUANG Rui et al. / Chinese Journal of Chromatography, 2006, 24(6): 597–600
crease in current and decrease in migration times of DMP, DEP, and DBP, which provided the basis for separation. However, when the amount of water ranged from 10% to 20%, the resolution changed slightly. The increase in current was caused by the higher molar conductivity of the same com+ pound in water, since the ionization of water produced H and OH . In such medium, the electrostatic interaction might be the dominant interaction force that facilitated separation. However, this action force was not sufficient to obtain the baseline separation. All resolutions were less than 0.8. Furthermore, when the amount of water in the nonaqueous SDS micelle reached 20%, the current was unstable, and the baseline of electropherogram showed noticeable fluctuation. The reason probably was the increase of conductivity caused by increase in water and little amount of precipitated SDS. Therefore, 10% water was available. Owing to the addition of water, much more silica hydroxyl ionized on the inner wall of the capillary [16]. Consequently, the EOF accelerated, as well as the migration of components. 2.3 Effects of the acidity of water phase and organic additives The acidity of buffer is an important factor that determines separation in common aqueous MEKC [17]. By modifying the EOF and the charges of surfactants, acidity can be utilized to realize the separation of analytes. The effect of the acidity was investigated through varying the proportion of acid and base in the water added to the running medium. Over the investigated range, the migration mobility was hardly controlled by acidity. To obtain lower current, 20 mmol/L NaH2PO4 was adopted as the running buffer in water phase. When 10% methanol or 10% acetonitrile was added into the above-mentioned medium, the migration times of the three dialkyl phthalates were prolonged, whereas the resolution did not change. The same phenomenon was also observed in 120 mmol/L SDS nonaqueous micelle in FA medium. Therefore, the effect of organic additives on separation could be ignored. 2.4 Effect of the concentration of SDS Since the solvophobic interaction weakened in FA medium, the distribution of analytes in SDS micelle decreased as well, which altered the migration behavior and promoted the separation. The conductivity was lower and the solubility of SDS was higher in the FA medium compared with the water medium. Hence, it was possible to form micelle with higher concentration of SDS in running medium. With the increasing concentration of SDS, the electrophoresis current increased. The electrophoresis current was linear with the SDS concentration when it was below 120 mmol/L. The increase of current showed obvious deviation from the linear relation beyond 160 mmol/L of SDS because of the excessive Joule heat. The concentration of SDS in the water phase of FA medium played an important role in affecting the mobility of the three
dialkyl phthalates as seen in Fig. 2. With 10 mmol/L SDS, there was no evidence of separation. A clue that separation occurred at 20 mmol/L resulted from two possible reasons. One was the function of SDS micelle if 20 mmol/L SDS exceeded its CMC in the present medium; the other was the interaction between SDS molecules and analyte molecules if 20 mmol/L did not achieve the CMC of SDS. Eventually, baseline separation was realized at the SDS concentration of 120 mmol/L.
Fig. 2 Effect of SDS concentration in formamide medium on apparent mobilities of phthalates Buffer: 20 mmol/L NaH2PO4 in formamide /water (9/1, v/v). Other conditions as in Fig. 1. 1. DMP; 2. DEP; 3. DBP.
When the SDS concentration ranged from 10 to 80 mmol/L, the apparent mobility of the three compounds tended to reduce. At 80 mmol/L, the apparent mobility was approximately 10 × 10-5 cm2/ (s·V) because the effective mobility of phthalates was zero and that of SDS micelle negative. Along with the increase of the SDS concentration, the action force of the micelle reinforced. Consequently, the effective mobility of phthalates gradually approached the negative mobility of SDS. Therefore, the apparent mobility of DMP, DEP, and DBP declined, and the migration time increased. However, although the three compounds showed high distribution in SDS micelle when the concentration of SDS was beyond 120 mmol/L, the migration time was shortened on the contrary because the EOF showed an abrupt increase. The mechanism of this abnormal phenomenon is not clear till now. 2.5 Separation of dialkyl phthalates using NAMEKC Under the optimized running buffer of FA/H2O (9/1, v/v) containing 20 mmol/L NaH2PO4 and 120 mmol/L SDS, the baseline separation of the three dialkyl phthalates was realized with the lowest Rs of 1.5. As can be seen in Fig. 3, the distorted peak of DBP implied that the mobility of DBP was close to that of micelle. It can be concluded that the migration time window was narrow, which acted against the separation of the other compounds with higher hydrophobicity. To adjust the distribution of highly
HUANG Rui et al. / Chinese Journal of Chromatography, 2006, 24(6): 597–600
hydrophobic compounds in micellar phase, some organic reagents can be added into the buffer, such as methanol, acetonitrile, or urea. Moreover, it is feasible to add supramolecules, such as β-cyclodextrin (the solubility of β-cyclodextrin in FA is 40 times higher than that in water), into the buffer. The peak sequence of the three phthalates in NAMEKC was in accordance with their hydrophobicity. Their migration time escalated with the increase in hydrophobicity. The detection limits of DMP, DEP, and DBP were 0.36, 1.08, and 3.04 mmol/L (S/N=3), respectively.
ple components with considerable variation in their hydrophobicities. To widen the migration time window, methanol, acetonitrile, urea, or β-cyclodextrin can be added to the buffer to adjust the distribution of analytes in micellar phase. In addition, either using other surfactants (e.g. fluorocarbon surfactant or polymeric surfactant) or mixed surfactants is a valid alternate.
References [1] Terabe S, Otsuka K, Ichikawa K, Tsuchiya A, Ando T. Anal Chem, 1984, 56(1): 111 [2] Terabe S, Ozaki H, Otsuka K, Ando T. J Chromatogr, 1985, 332: 211 [3] Terabe S, Miyashita Y, Ishihama Y, Shikata O. J Chromatogr, 1993, 636: 47 [4] Walbroehl Y, Jorgenson J W. J Chromatogr, 1984, 315: 135 [5] Okada T. J Chromatogr A, 1997, 771: 275 [6] Miller J L, Khaledi M G, Shea D. J Microcolumn Sep, 1998, 10(8): 681 [7] Miller J L, Khaledi M G, Shea D. Anal Chem, 1997, 69: 1223 [8] Lin J M, Nakagawa M, Uchiyama K, Hobo T. Chromatographia, 1999, 50: 739
Fig. 3
Electropherogram of phthalates at optimized operation conditions
Buffer: 20 mmol/L NaH2PO4 and 120 mmol/L SDS in formamide/ water (9/1, v/v). Other conditions as in Fig. 1.
3 Conclusions Three dialkyl phthalates DMP, DEP, and DBP were rapidly separated with SDS as micellar phase in FA medium using NAMEKC. The lowest resolution was 1.5, and the detection limit was higher than 3.04 mmol/L for each analyte. It is obvious that SDS not only can be applied in aqueous MEKC but also displays advantages in NAMEKC. Due to the narrow migration time window, the systems introduced in this study are disadvantageous for separating the compounds with higher hydrophobicity or the complex multi-
[9] Lin J M, Nakagawa M, Uchiyama K, Hobo T. Electrophoresis, 2002, 23: 421 [10] Nakagawa M, Lin J M, Nakagama T, Uchiyama K, Hobo T. Bunseki Kagaku, 1999, 48: 239 [11] Luks-Betlej K, Popp P, Janoszka B, Paschke H. J Chromatogr A, 2001, 938: 93 [12] Yang Z J, Zhang W Y, Wang C Y, Xu D Z, Pan K Y, Wu C Y. Chinese Journal of Chromatography, 2003, 21(6): 617 [13] Chen H M, Wang C, Wang X. Chinese Journal of Chromatography, 2004, 22(3): 224 [14] Baker J K. J Pharm Biomed Anal, 1996, 15(1): 145 [15] Ward V L, Khaledi M G. J Chromatogr A, 1999, 859: 203 [16] Chen Y. Study on Capillary Electrophoresis Theory. Beijing, China: Huawen Press, 2001: 16 [17] Oliver B, Ruth F. J Chromatogr A, 1995, 717: 309