Microemulsion Electrokinetic Chromatography: A Review of Chiral Separation Capability and Optimization

Erin J. Ennis
Drexel University, Department of Chemistry
Submitted December 3rd, 2011

Abstract
Microemulsions are transparent, thermodynamically stable droplets consisting of a surfactant, co-surfactant, and oil. Derived from the need to separate neutral compounds quickly and efficiently, microemulsion electrokinetic chromatography (a subset of electrokinetic chromatography) provides opportunity for typical and chiral based separations. The aim of this paper is to review the history of microemulsions in separation science, the optimization necessities and applications in typical microemulsion electrokinetic chromatography, and capabilities of chiral separations. Focus will be on chiral microemulsion electrokinetic chromatography and the optimization needed to produce substantially quick, efficient, and reproducible separations.

Introduction
Capillary electrophoresis was developed in the late 1960’s as an alternative separation method to high performance liquid chromatography (HPLC). Capillary electrophoresis, sometimes called capillary zone electrophoresis or CZE, gained momentum in 1981 with the first literature publication by Jim Jorgenson [1]. CZE uses the mobility and mass to charge ratio of analytes to separate them quickly and efficiently. Jorgenson proposed the change from columns (typically in HPLC techniques) to narrow bore tubes or capillaries, which would utilize high voltages to move analytes [1]. At a pH higher than three, silanol groups adhered to the capillary walls ionize and interact with cations in the buffer system or “mobile phase” [1]. These interactions create a diffuse double layer of ions which flow from the postive (anodic) end to the negative (cathodic) end. This movement, termed electroosmotic flow (EOF), explains why anionic analytes migrated to the cathode despite their attraction to the anode [1]. Anions, while attracted to the anode, are swept up by the larger EOF towards the cathode. The EOF also explains why heavier analytes, which had lower mobilities and were unable to “swim” against the flow, were effectively eluted in a similar matter. This general “swimming” system provides high efficiency, low waste, and fast separations.

However, CZE systems have one main drawback. While cationic analytes swim with the EOF, eluting first, and anionic analytes attempt to swim against the EOF, eluting last, neutral analytes have no interaction with the EOF and elute at exactly in a singular band [2]. It is impossible to separate neutral compounds from each other in the CZE system. In 1989, Terabe and coworkers suggested a method they termed electrokinetic chromatography (EKC) as an alternative that would be able to separate neutrals [2].

Electrokinetic Chromatography
While CZE can be considered an electrophoretic separation method, EKC is considered primarily a chromatographic method [2]. A quasi phase, or an additive in the buffer mixture used to cause interactions with the analytes and enable separations, is added to a typical CZE system [2]. Quasi phases have come to be called “pseudostationary phases” or PSP’s. EOF and the electrophoretic mobilities of the analytes are still used as separation and transportation mechanisms, but the PSP’s allow for additional interactions that cause cationic, neutral, or anionic species to be separated utilizing the same experiment [2]. Analytes interact with the additives through partitioning mechanisms and their individual hydrophobicities [2]. Some analytes will interact longer than others, allowing these uncharged molecules to separate efficienctly. Figure 1 below shows this process [2].
ErinFile.png

There are many different types of PSP that can be added to a buffer to facilitate the separation of neutral compounds. While focus will be primarily on one type of EKC, it is important to mention other additives that have been used substantially for these separations.

Micellar EKC
Micellar EKC (MEKC) experiments involve the addition of micelles. In 1984, Terabe published on the use of sodium dodecyl sulfate (SDS) for EKC based separations [2]. SDS, a negatively charged surfactant, arranges itself in an aggregate form so the hydrophilic head group faces outward towards the buffer system and the hydrophobic tail groups gather in a central region. This particular anionic micelle travels in the opposite direction of traditional EOF (towards the positive anode) while a cationic surfactant micelle travels in the same direction as EOF [2]. Increased concentration of surfactant would cause longer interactions and therefore a longer elution time, but in general MEKC has relatively high resolution, short retention times, and the Joule heating can be controlled.

Cyclodextrins
Cyclodextrin (CD) modified EKC can also be used to separate neutral compounds. These PSP’s form inclusion complexes, which allow interactions for appropriately sized analytes [2]. CD’s need to be modified in order to add a charged group if they are to be electrophoretically active, but overall CD modified EKC is effective in the separation of isomers and analytes normally insoluble in water based systems [2].

Polymers
Polymers can also be used for neutral separations in EKC modes. These compounds work as carriers for ion pairs, in which the formation constants cause different velocities amongst the analytes due to an overwhelming “change” in charge [2]. However, polymeric PSP’s tend to produce broad peaks due to extensive interactions [2].

Microemulsion Electrokinetic Chromatography
This paper will focus on microemulsion electrokinetic chromatography (MEEKC). In particular, emphasis will be on methods, optimization, and applications, with considerable evaluation of the chiral capabilities of microemulsion components.

Watarai first published about MEEKC as a technique comparable if not greater than MEKC in 1991, making MEEKC a relatively new separation mechanism [3]. Microemulsions are considered microheterogeneous liquids that typically form as oil in water droplets [3]. These droplets are optically transparent, thermodynamically stable, and are highly soluble [3]. The first microemulsion, composed of water, SDS, 1-butanol, and heptane, was used to separate both ionic and nonionic analytes [3]. Microemulsions can be created over a wide range of pH’s [3].

Microemulsions are formed by the combination of a surfactant, co-surfactant, and oil. The surfactant chain resides in the oil droplet while the head group interacts with the buffer system [4]. The co-surfactant stabilizes the microemulsion and lowers the surface tension of the droplet [4]. While SDS is the most commonly used surfactant and a short chain alcohol is usually employed as the co-surfactant, combinations of these compounds are endless [5].

While high pH buffers are common in microemulsions (phosphate, for example) in order to create high EOF, they are not always necessary [4]. Charged compounds will separate by their electrophoretic mobilities and interactions with the PSP while neutral compounds will separate solely by their interactions [4]. In order to make separations run as quickly as possible, a high EOF is the most valuable [4].

MEEKC has multiple advantages over MEKC. MEEKC has a wider range of migration times and the microemulsion, which is larger in size than the micelle aggregate, does not lead to zone broadening [3]. Microemulsions also possess higher solubility (3) and are less rigid than typical micelles [4]. However, MEEKC is not nearly as well established as MEKC. Additionally, the multiple PSP additives in MEEKC make optimization difficult and formulation much more complex [4]. It is essential to consider optimization parameters for microemulsion formulations

Optimization
Surfactant
While SDS is the most common surfactant choice, anionic bile salts have been used in negatively charged droplets as well [4]. Cationic surfactants, like CTAB, create positive droplets that require polarity changes in order to create elution [4]. This is due to the reversal of the EOF under cationic conditions [4]. Neutral surfactants include triton X-100 are useful, but would be unable to separate neutral compounds [4].

Increasing the chain length of the surfactant produces more stabile microemulsions and decreases the polydispersity (creates more spherical microemulsions) [4]. Increasing the concentration increases the ionic strength of the buffer system, which can cause longer elution times [4]. Stability of microemulsions tends to be lost at low concentrations of surfactant [4]. The current is also affected by the surfactant’s salt identity [5]. Lithium dodecyl sulfate creates less current than sodium dodecyl sulfate [5]. Overall, surfactant concentration and choice is essential for the creation of stable and useful microemulsions.

pH
pH is also essential in optimization. A pH of 7-9 creates a high EOF, although low pH will suppress the EOF for more reproducible experiments [4]. Too high of a pH will eliminate the ability of analytes to ionize and interact with the droplets. Altria uses the example of parahydroxybenzoate preserves, which at high pH elute with very poor peak shape due to changed interactions [5].

Oil
While octane and heptane are the most commonly used oils in microemulsion formulation, oil choice seems to have only minor effects on the microemulsion capabilities in terms of chromatographic figures of merit [5]. Diethyl ether, cyclohexane, chloroform, and ethyl acetate can also be used as oils, while chiral oils can be employed to facilitate chiral separations [5]. While reproducibility tends to be higher with octane, migration times and selectivity tend to remain the same in separations [4].

Organic Solvents
Organic solvents can be added to microemulsion systems to aid in separating highly retained species [4]. This helps to improve resolution and minimize broadening [5]. However, some organic solvents (like 2-propanol) can act as a co-surfactant at high concentrations [5]. Additionally, there are limitations to the amount of organic solvents that can be added without disrupting the microemulsion formation, forcing compounds back into their immiscible phases [4].

Co-surfactant
Co-surfactants, as discussed briefly, can also be optimized. 1-butanol tends to be the most favored co-surfactant [5]. However, since co-surfactants have a direct effect on the viscosity and therefore EOF of the system, co-surfactants should be chosen carefully [5]. Increasing co-surfactant concentration can reduce migration times and increase the size of the droplet formed [4].

Buffer
The buffer that the microemulsion is incorporated in is one of the most important optimization parameters in an MEEKC system. Low ionic strength buffers are typical choices, mainly for their ability to produce high EOF and allow high voltages [4]. The suppression of EOF is possible over a wide range of buffer ionic strength to facilitate better separations, while zwitterionic buffers (ACES and TRIS) can be employed to reduce current even farther [5]. Buffering capacity over the pH range of choice also needs to be taken into account.

Temperature
Finally, the temperature of the system can be optimized [5]. MEEKC separations can be temperature controlled by instrument software. Since the electrophoretic mobility of the analytes and the viscosity of the buffer are dictated by temperature, finding an optimum will allow for enhancement of resolution, migration time, and peak shape [5]. Temperature can also aid in the solubility of analytes in microemulsions.

Applications
In recent years, MEEKC has been used for a wide variety of applications. Neutral, basic, and acidic analytes have been separated in one quick, efficient experiment [5]. Microemulsions have been used in the separations of proteins, hops, and steroids [4]. Pesticides and suntan lotion derivatives have been separated and qualitatively identified [5]. One particularly interesting application involved the separation of heroin using both MEEKC and modified MEKC at pH 9.5 [6]. Both a standard heroin sample and 17 real samples of drug mixtures were separated within 10 minutes with good reproducibility [6]. This particular application shows the MEEKC applicability to the pharmaceutical and forensic industries.

Chiral Separations with MEEKC
In recent years, MEEKC has gained popularity in chiral separations. Chiral compounds, or those that have a non superimposable mirror image, are very prominent in pharmaceutical drugs. Typically, only one of the chiral enantiomers are active or necessary in an industrial formulation; which drives the need for rapid and reproducible separations. This can be done in two ways: indirect and direct detection [7]. Indirect detection involves reactions of diastereomers [7]. Chiral MEEKC involves direct chiral separations, or temporary interactions, by using chiral selector components [7]. The first chiral separation using MEEKC, by Aiken and Huie, presented 2R,3R di-n-butyl tartrate as a chiral selecting oil [8]. Other additives or changes can be made to create chiral microemulsions. Tartaric acid esters exhibit a wide range of enantioselectivity [9]. Chiral polymers, like Poly-D-SUV, have been added to microemulsions to promote the separation of barbituate and binaphthyl derivatives [8]. CD’s and crown ethers have been used for their interactive abilities [7]. The microemulsion surfactant, co-surfactant, and oil can also exchange for chiral selective compounds. Zheng and coworkers presented a chiral co-surfactant, (2-alkanol) in 2004 [10]. Zheng found that too low of a concentration of co-surfactant (1%) and too high of a concentration (6%) affected the enantiomeric separation capability of the microemulsion [10].

Dodecoxycarbonyl valine (DDCV) was found to be a useful chiral surfactant in the early 2000’s [11]. It is available in R and S forms [12]. Mertzman published a comparison for DDCV between MEEKC, MEKC, and modified MEKC to show the benefits of chiral MEEKC [11]. Frictional drag effects play a large part in the efficiency of the systems; micelles are the most mobile due to size [11]. MEEKC, however, has an offset effect, with their larger size being negligible when compared to the change in charge on the droplet [11]. Overall, the study by Mertzman shows that, while efficiency and concentration ranges in chiral MEEKC make it ultimately more useful, the long preparation times can hinder study and use [11].

Regardless of the choice, the aim of chiral MEEKC is to promote enantiomeric interactions with the microemulsion, with one enantiomer interacting more than the other and promoting separation [9]. Optimization of the chiral MEEKC system is needed in order to compete with the simpler MEKC systems.

Optimization
Buffer
Buffers choice needs to be optimized in chiral separations. ACES, a typically used buffer, tends to need a long equilibration time in order to establish reproducibility [13]. Phosphate buffer improves ruggedness and robustness, increases efficiency, decreases migration time, and slightly decreases resolution [13]. It also increases EOF, although in this particular paper it was more due to a change in capillary conditioning than ionic strength [13]. Tetrapropyl ammonium hydroxide (TPAH), an additive used in washing, can decrease electromigration dispersion by adding non-competitive salts to the buffer [13].

Surfactant
Surfactant choice and concentration can also optimize chiral separations. Increasing the concentration expands the elution range, increases the mobility of the microemulsion, and increases the negative charge and hydrophobicity of the droplet [13]. This is due to a greater surface area coupled with additional negative charges. This increase also moderately increases enantioselectivity and resolution [13]. By coupling buffer optimization and concentration, efficiencies can increase up to 140,000 [13]. However, increasing concentration of all microemulsion parameters does not have the same effect and the components must be optimized individually [13].

Oil
Ethyl acetate can be used as an oil to facilitate high voltages, short analysis time, and surfactant concentration changes [12]. While it is known that the oil has only minor effects on the separation it can alter the retention factors [12]. In a comparison published by Mertzman and coworkers, the concentration seems more important than the type, although ethyl acetate still had the largest elution range and separated more compounds than similar oils [12].

Cosurfactant
Co-surfactant can also be optimized, although the size of the co-surfactant plays little role beyond appropriate penetration into the core [14]. Retention factors had no drastic change with different co-surfactants while elution order (and therefore retention of the enantiomers) was very effected [14]. Enantioselectivity is enhanced with a secondary alcohol and a short or cyclic chain length, although 1-butanol tends to be an exception [14]. It’s suggested that secondary alcohols tend to open the chiral center of the microemulsion more and allow for increased interactions [14].

Temperature
Temperature also plays an important role in separations. Resolution and retention decrease with temperature increase at low concentrations of chiral surfactant [15]. The enantioselectivity does not seem to change with temperature, although the temperature was only altered in this paper from 15 to 35 degrees Celsius [15]. Very little information is available about the true value of temperature dependency but it would be beneficial to attempt temperature optimization over a large range of degrees [15].

Purity
One of the final optimization factors of chiral MEEKC is chiral purity. Ratios of R to S were evaluated to determine the purity and separation capability of one chiral component (in this case, DDCV) [16]. Enantioselectivity was lower with R than with S, and the racemic combination (50/50) did not provide any enantiomeric separation [16]. Polarimeter and FAB-MS experiments were also conducted to show that commercially purchased R-DDCV tended to have 2.5 to 3.5 % impurities [16]. Efficiencies also increased more with increased S-DDCV percentages than R percentages [16]. This optimization is important in determining which form of the chiral selector is best for the separation at hand.

Multiple Chiral Selectors
While the discussion on single parameter chiral systems is inherently important, it is also necessary to consider multiple chiral-component microemulsions. A 2 component mix of DDCV with chiral S-2-hexanol (co-surfactant) was studied by Kahle and coworkers [17]. While the retention factors stayed the same, the elution range and enantioselectivity increased [17]. Efficiency was slightly better in the 1 component system, but the 2 component system was superior in terms of resolution [17]. Kahle and coworkers also attempted a separation using racemic hexanol, but efficiency was decreased and resolution was only moderate [18]. Combinations of chiral selectors, such as RS and SS (surfactant/co-surfactant) showed superiority in using the same chirality in both groups [18]. Diethyl tartrate was also examined as a chiral oil with chiral surfactant [19]. Separations were possible, but zone broadening was also prevalent [19].

Kahle and coworkers also examined 3 component systems, where surfactant (DDCV), co-surfactant (S-2-hexanol), and oil (di-n-butyl tartrate) were all used in combination [20]. The method was optimized so that zero, 1, 2, and 3 chiral components were all examined. Enantioselectivity increased with 3 components [20]. Resolution only marginally increased from 2 to 3, but drastically increased from 1 to 3 [20]. Efficiency was reduced slightly and there was increased zone broadening [20]. This was due to increased interactions of the enantiomers with the components [20]. Finding chiral components that interact less strongly with the enantiomers but still present the same benefits as the 3 component system would be the next step in chiral MEEKC optimization.

Conclusion
MEEKC, which grew from the need of a technique beyond CZE to separate neutral analytes, has grown immensely in the last few years. Optimization of traditional MEEKC parameters is necessary in order to achieve rapid, reproducible results with high efficiencies and good resolution for a large range of applications. The need for chiral separations, particularly in the pharmaceutical industry, drove the invention of chiral MEEKC. Similar optimization is needed to achieve comparable results to other chiral methods. Multiple chiral components also provide an opportunity for greater enhancement of MEEKC. With continued work, chiral MEEKC will grow as a primary method for enantiomeric and general separation.

References

[1]. Jorgenson, J., et al. (1981). "High-Resolution Separations Based on Electrophoresis and Electroosmosis." Journal of Chromatography A. 218, (1-3): 209-216.Link

[2]. Terabe, S. (1989). "Electrokinetic chromatography: an interface between electrophoresis and chromatography." TrAC Trends in Analytical Chemistry. 8, (4):129-134.Link

[3]. Watarai, H. (1991). "Microemulsion Capillary Electrophoresis." Chemistry Letters. (3): 391-394.Link

[4]. Altria, K. (2000). "Background theory and applications of microemulsion electrokinetic chromatography." Journal of Chromatography A. 892, (1-2): 171-186.Link

[5]. Altria, K., et al. (2003) "Background and operating parameters in microemulsion electrokinetic chromatography." Electrophoresis. 24, (3): 315-324 Link

[6]. Wen, T., et al. (2006). "Comparison of microemulsion electrokinetic chromatography and solvent modified micellar electrokinetic chromatograph on rapid separation of heroin, amphetamine and their basic impurities." Talanta. 71, (2): 854-860. Link

[7]. Kahle, K., et al. (2007). "Review of aqueous chiral electrokinetic chromatography (EKC) with an emphasis on chiral microemulsion EKC." Electrophoresis. 28, (15): 2503-2526. Link

[8]. Iqbal, R., et al. (2004). "Chiral separations in microemulsion electrokinetic chromatography - Use of micelle polymers and microemulsion polymers." Journal of Chromatography A.1043, (2): 291-302. Link

[9]. Aiken, J., et al. (1993). "Use of a Microemulsion System to Incorporate a Lipophilic Chiral Selector in Electrokinetic Capillary Chromatography." Chromatographia. 35, (7-8): 448-450. Link

[10]. Zheng, Z., et al. (2004). "Separation of enantiomers in microemulsion electrokinetic chromatography using chiral alcohols as cosurfactants." Electrophoresis. 25, (18-19): 3263-3269. Link

[11]. Mertzman, M., et al. (2005). "Comparison of dodecoxycarbonylvaline microemulsion, solvent-modified micellar and micellar pseudostationary phases for the chiral analysis of pharmaceutical compounds." Electrophoresis. 26, (21): 4153-4163. Link

[12]. Mertzman, M., et al. (2004). "Effect of oil substitution in chiral microemulsion electrokinetic chromatography." Electrophoresis. 25, (4-5): 723-732.Link

[13]. Mertzman, M., et al. (2004). "Effect of surfactant concentration and buffer selection on chromatographic figures of merit in chiral microemulsion electrokinetic chromatography." Electrophoresis. 25, (18-19): 3247-3256. Link

[14]. Kahle, K., et al. (2004). "Chiral Microemulsion electrokinetic chromatography: Effect of cosurfactant identity on enantioselectivity, methylene selectivity, resolution, and other chromatographic figures of merit." Electrophoresis. 27, (21): 4321-4333.Link

[15]. Mertzman, M., et al. (2005). "Temperature effects on chiral microemulsion electrokinetic chromatography employing the chiral surfactant dodecoxycarbonylvaline." Journal of Chromatography A. 1073, (1-2): 181-189. Link

[16]. Kojtari, A., et al. (2009). "Effect of microemulsion component purity on the chromatographic figures of merit in chiral microemulsion electrokinetic chromatography." Journal of Chromatography A. 1216, (16): 3488-3491. Link

[17]. Kahle, K., et al. (2006). "Chiral microemulsion electrokinetic chromatography with two chiral components: Improved separations via synergies between a chiral surfactant and a chiral cosurfactant." Electrophoresis. 27, (4): 896-904.Link

[18]. Kahle, K., et al. (2007). "Two-chiral-component microemulsion electrokinetic chromatography-chiral surfactant and chiral oil: Part 1. Dibutyl tartrate." Electrophoresis. 28, (11): 1723-1734. Link

[19]. Kahle, K., et al (2007). "Two-chiral component microemulsion EKC - chiral surfactant and chiral oil. Part 2: Diethyl tartrate." Electrophoresis. 28, (15): 2644-2657. Link

[20]. Kahle, K., et al. (2007). "Influence of microemulsion chirality on chromatographic figures of merit in EKC: Results with novel three-chiral-component microemulsions and comparison with one- and two-chiral-component microemulsions." Electrophoresis. 28, (17): 3024-3040. Link