Micellar electrokinetic chromatography

Micellar electrokinetic chromatography (MEKC) can be used to separate neutral or ionic species by their differential distribution between an aqueous mobile phase and a pseudostationary micellar phase. Capillary electrophoresis (CE) is an older, but related technique that predates MEKC.

Overview

Micellar electrokinetic chromatography (MEKC) is a relatively new technique that has been developed to expand the utility of capillary electrophoresis (CE). In traditional electrophoresis, electrically charged analytes move in a conductive liquid medium under the influence of an electric field. Introduced in the 1960’s, the technique of capillary electrophoresis (CE) was designed to separate species based on their size to charge ratio in the interior of a small capillary filled with an electrolyte. Limited by its relatively poor ability to separate neutral or poorly dissolved analytes, capillary electrophoresis can be augmented by the introduction of a surfactant to the background buffer solution. This technique, termed micellar electrokinetic chromatography, separates neutral and ionic analytes between the hydrophobic interior of a micelle and the hydrophilic buffer solution.¹ The advent of MEKC has substantially increased the amount and type of compounds that can be separated by capillary electrophoresis.

Instrumentation

The instrumentation needed to perform capillary electrophoresis, as well as MEKC, is relatively simple. A basic schematic of a capillary electrophoresis system is shown in figure 1. The system’s main components are a sample vial, source and destination vials, a capillary, electrodes, a high-voltage power supply, a detector, and a data output and handling device. The source vial, destination vial and capillary are filled with an electrolyte such as an aqueous buffer solution. To introduce the sample, the capillary inlet is placed into a vial containing the sample and then returned to the source vial (sample is introduced into the capillary via capillary action. The migration of the analytes is then initiated by an electric field that is applied between the source and destination vials and is supplied to the electrodes by the high-voltage power supply. The analytes separate as they migrate and are detected near the outlet end of the capillary. The output of the detector is sent to a data output and handling device such as an integrator or computer. The data is then displayed as an electropherogram, which reports detector response as a function of time. Separated chemical compounds appear as peaks with different retention times in an electropherogram.²

Figure 1: Diagram of capillary electrophoresis system

Detection

Separation by capillary electrophoresis can be detected by several detection devices. The majority of commercial systems use UV or UV-Vis absorbance as their primary mode of detection. In these systems, a section of the capillary itself is used as the detection cell. The use of on-tube detection enables detection of separated analytes with no loss of resolution. In general, capillaries used in capillary electrophoresis are coated with a polymer for increased stability. The portion of the capillary used for UV detection, however, must be optically transparent. Bare capillaries can break relatively easily and, as a result, capillaries with transparent coatings are available to increase the stability of the cell window. The path length of the detection cell in capillary electrophoresis (~ 50 micrometer) is far less than that of a traditional UV cell (~ 1 cm). According to the Beer-Lambert law, the sensitivity of the detector is proportional to the path length of the cell. To improve the sensitivity, the path length can be increased, though this results in a loss of resolution. The capillary tube itself can be expanded at the detection point, creating a “bubble cell” with a longer path length or additional tubing can be added at the detection point as shown in figure 2. Both of these methods, however, will decrease the resolution of the separation.¹

Figure 2: Techniques for increasing the pathlength of the capillary: a.) a bubble cell and b.) a z-cell (additional tubing).²

Fluorescence detection can also be used in capillary electrophoresis for samples that naturally fluoresce or are chemically modified to contain fluorescent tags. This mode of detection offers high sensitivity and improved selectivity for these samples, but cannot be utilized for samples that do not fluoresce. The set-up for fluorescence detection in a capillary electrophoresis system can be complicated. The method requires that the light source be concentrated at the capillary lumen, which can be difficult for many light sources.¹ Laser-induced fluorescence has been used in CE systems with detection limits as low as 10^-18 to 10^-21 mol. The sensitivity of the technique is attributed to the high intensity of the incident light and the ability to accurately focus the light on the capillary diameter. ²

In order to obtain the identity of sample components, capillary electrophoresis can be directly coupled with mass spectrometers. In most systems, the capillary outlet is introduced into an ion source that utilizes electrospray ionization (ESI), which is itself coupled to a mass spectrometer. This set-up, however, requires volatile buffers, which will affect the range of separation modes that can be employed and the degree of resolution that can be achieved. ¹

Modes of separation

The separation of compounds by capillary electrophoresis is dependent on the differential migration of analytes in an applied electric field. The electrophoretic migration velocity (u_p) of an analyte toward the electrode of opposite charge is given by

$u_{p}=\mu _{p}E$ (1)

where μ_p is the electrophoretic mobility and E is the electric field strength. The electrophoretic mobility is proportional to the ionic charge of a sample and inversely proportional to any frictional forces present in the buffer. When two species in a sample have different charges or experience different frictional forces, they will separate from one another as they migrate through a buffer solution. The frictional forces experienced by an analyte ion depend on the viscosity (η) of the medium and the size and shape of the ion.¹ Accordingly, the electrophoretic mobility of an analyte in a given pH is given by

$\mu _{p}={\frac {z}{6\pi \eta r}}$ (2)

where z is the net charge of the analyte and r is the Stokes radius of the analyte The Stokes radius is given by

$r={\frac {k_{b}T}{6\pi \eta }}$ (3)

where k_b is the Boltzmann constant, and T is temperature. As a result of equations 2 and 3, electrophoretic mobility of the analyte is proportional to the charge of the analyte and inversely proportional to its molecular weight. The electrophoretic mobility can be determined experimentally from the migration time and the field strength:

$\mu _{p}=\left({\frac {L}{t_{r}}}\right)\left({\frac {L_{t}}{V}}\right)$ (4)

where L is the distance from the inlet to the detection point, t_r is the time required for the analyte to reach the detection point (migration time), V is the applied voltage (field strength), and L _t is the total length of the capillary.¹ Since only charged ions are affected by the electric field, neutral analytes are poorly separated by capillary electrophoresis.

The velocity of migration of an analyte in capillary electrophoresis will also depend upon the rate of electroosmotic flow (EOF) of the buffer solution. In a typical system, the electroosmotic flow is directed toward the negatively charged cathode so that the buffer flows through the capillary from the source vial to the destination vial. Separated by differing electrophoretic mobilities, analytes migrate toward the electrode of opposite charge. ² As a result, negatively charged analytes are attracted to the positively charged anode, counter to the EOF, while positively charged analytes are attracted to the cathode, in agreement with the EOF as depicted in figure 3.

Figure 3: Diagram of the separation of charged and neutral analytes (A) according to their respective electrophoretic and electroosmotic flow mobilities

The velocity of the electroosmotic flow, u_o can be written as:

$u_{o}=\mu _{o}E$ (5)

where μ_o is the electroosmotic mobility, which is defined as:

$\mu _{o}={\frac {\epsilon _{r}\zeta }{\eta }}$ (6)

where ζ is the zeta potential of the capillary wall, and ε_r is the dielectric constant of the buffer solution. Experimentally, the electroosmotic mobility can be determined by measuring the retention time of a neutral analyte using equation 3.¹ The velocity (u) of an analyte in an electric field can then be defined as:

$u_{p}+u_{o}=(\mu _{p}+\mu _{o})E$ (7)

Since the electroosmotic flow of the buffer solution is generally greater than that of the electrophoretic flow of the analytes, all analytes are carried along with the buffer solution toward the cathode. Even small, triply charged anions can be redirected to the cathode by the relatively powerful EOF of the buffer solution. Negatively charged analytes are retained longer in the capilliary due to their conflicting electrophoretic mobilities.² The order of migration seen by the detector is shown in ‘’figure 3’’ where small multiply charged cations migrate quickly and small multiply charged anions are retained strongly.¹

Electroosmotic flow is observed when an electric field is applied to a solution in a capillary that has fixed charges on its interior wall. Charge is accumulated on the inner surface of a capillary when a buffer solution is placed inside the capillary. In a fused-silica capillary, silanol (Si-OH) groups attached to the interior wall of the capillary are ionized to negatively charged silanoate (Si-O^-) groups at pH values greater than three. The ionization of the capillary wall can be enhanced by first running a basic solution, such as NaOH or KOH through the capillary prior to introducing the buffer solution. Attracted to the negatively charged silanoate groups, the positively charged cations of the buffer solution will form two inner layers of cations (called the diffuse double layer) on the capillary wall as shown in figure 4. The first layer is referred to as the fixed layer because it is held tightly to the silanoate groups. The outer layer, called the mobile layer, is farther from the silanoate groups. The mobile cation layer is pulled in the direction of the negatively charged cathode when an electric field is applied. Since these cations are solvated, the bulk buffer solution migrates with the mobile layer, causing the electroosmotic flow of the buffer solution. Other capillaries including Teflon capillaries also exhibit electroosmotic flow. The EOF of these capillaries is probably the result of adsorption of the electrically charged ions of the buffer onto the capillary walls.² The rate of EOF is dependent on the field strength and the charge density of the capillary wall. The wall’s charge density is proportional to the pH of the buffer solution. The electroosmotic flow will increase with pH until all of the available silanols lining the wall of the capillary are fully ionized.¹

Figure 4: Depiction of the interior of a fused-silica gel capillary in the presence of a buffer solution.

Efficiency and resolution

The number of theoretical plates, or separation efficiency, in capillary electrophoresis is given by

$N={\frac {\mu V}{2D_{M}}}$ (8)

where N is the number of theoretical plates, μ is the apparent mobility in the separation medium and D_m is the diffusion coefficient of the analyte. According to this equation, the efficiency of separation is only limited by diffusion and is proportional to the strength of the electric field. The efficiency of capillary electrophoresis separations is typically much higher than the efficiency of other separation techniques like HPLC. Unlike HPLC, in capillary electrophoresis there is no mass transfer between phases.¹ In addition, the flow profile in EOF-driven systems is flat, rather than the rounded laminar flow profile characteristic of the pressure-driven flow in chromatography columns as shown in figure 5. As a result, EOF does not significantly contribute to band broadening as in pressure-driven chromatography and capillary electrophoresis separations can have several hundred thousand theoretical plates.³

Figure 5: Flow profiles of laminar and electroosmotic flow.

The resolution (R_s) of capillary electrophoresis separations can be written as:

$R_{s}={\frac {1}{4}}\left({\frac {\triangle \mu _{p}{\sqrt {N}}}{\mu _{p}+\mu _{o}}}\right)$(9)

According to this equation, maximum resolution is reached when the electrophoretic and electroosmotic mobilities are similar in magnitude and opposite in sign. In addition, it can be seen that high resolution requires lower velocity and, correspondingly, increased analysis time.¹

Applications of capillary electrophoresis

Due to its high efficiency and capacity for relatively high resolution, capillary electrophoresis can be used to separate and determine a wide variety of compounds including simple inorganic ions, metal ions, oligosaccharides, nucleic acids, and proteins. CE is most commonly used, however, to analyze larger, water-soluble biomolecules.² CE is a particularly attractive technique because it requires small amounts of buffer and sample to perform several analyses. In most applications, CE is used in conjunction with other techniques, offering different separation selectivity or improved quantization.¹ In general, capillary electrophoresis is capable of separating compounds with different size to charge ratios. Accordingly, the major limitation of capillary electrophoresis is that it is inefficient at separating neutral compounds.⁴

Separation of neutral analytes: MEKC

In 1984, the Terabe group reported a technique that enabled capillary electrophoresis instrumentation to be used in the separation of neutral (as well as ionic) species. In micellar electrokinetic chromatography (MEKC), samples are separated by differential partitioning between a pseudo-stationary micellar phase and an aqueous mobile phase.⁵ In most applications, MEKC is performed in open capillaries under alkaline conditions to generate a strong electroosmotic flow. The basic set-up and detection methods used for MEKC are the same as those used in CE, but the solution contains a surfactant at a concentration that is greater than the critical micelle concentration (CMC). Above this concentration, surfactant monomers are in equilibrium with micelles. Sodium dodecylsulfate (SDS) is the most commonly used surfactant in MEKC applications. The anionic character of the sulfate groups of SDS cause the surfactant and micelles to have electrophoretic mobility that is counter to the direction of the strong electroosmotic flow. As a result, the surfactant monomers and micelles migrate quite slowly, though they still move with the buffer solution toward the cathode.² During a MEKC separation, analytes distribute themselves between the hydrophobic interior of the micelle and hydrophilic buffer solution as shown in figure 6.

Figure 6: Depiction of the distribution of analytes (A) based on their hydrophobicity.

Analytes that are insoluble in the interior of micelles should migrate at the electroosmotic flow velocity, u_o, and be detected at the retention time of the buffer, t_M. Analytes that solubolize completely within the micelles (analytes that are highly hydrophobic) should migrate at the micelle velocity, u_c, and elute at the final elution time, t_c.⁶ The micelle velocity is defined by

$u_{c}=u_{p}+u_{o}$ (10)

where u_p is the electrophoretic velocity of a micelle.⁶

The retention time of a given sample should depend on the capacity factor,k¹

$k^{1}={\frac {n_{c}}{n_{w}}}$(11)

where n_c is the total number of moles of solute in the micelle and n_w is the total moles in the aqueous phase.⁶ The retention time of a solute should then be within the range:

$t_{M}\leq t_{r}\leq t_{c}$ (12)

It should be noted that charged analytes have a more complex interaction in the capillary because they exhibit eletrophoretic mobility, engage in electrostatic interactions with the micelle and participate in hydrophobic partitioning.¹

The fraction of the sample in the aqueous phase, R, is given by

$R={\frac {u_{s}-u_{c}}{u_{o}-u_{c}}}$ (13)

where u_s is the migration velocity of the solute.⁶ The value R can also be expressed in terms of the capacity factor:

$R={\frac {1}{1+k^{1}}}$ (14)

Using the relationship between velocity, tube length from the injection end to the detector cell (L), and retention time, u_eo= L/t_m, u_mc= L/t_mc and u_s= L/t_r, a relationship between the capacity factor and retention times can be formulated¹:

$k^{1}={\frac {t_{r}-t_{M}}{t_{M}(1-(t_{r}/t_{M}))}}$ (15)

This equation resembles an expression derived for k¹ in conventional packed-bed chromatography:

$k={\frac {t_{r}-t_{M}}{t_{M}}}$ (16)

The extra term enclosed in parenthesis in the denominator of eq. 15 accounts for the partial mobility of the hydrophobic phase in MEKC.¹

A rearrangement of equation 15, can be used to write an expression for the retention factor ⁷ :

$t_{r}=\left({\frac {1+k^{1}}{1+(t_{M}/t_{c})k^{1}}}\right)t_{M}$ (17)

From this equation it can be seen that all analytes that partition strongly into the micellar phase (where k¹ is essentially ∞) migrate at the same time, t_c. In conventional chromatography, separation of similar compounds can be improved by gradient elution. In MEKC, however, techniques must be used to extend the elution range to separate strongly retained analytes.¹

Elution ranges can be extended by several techniques including the use of organic modifiers, cyclodextrins, and mixed micelle systems. Short-chain alcohols or acetonitrile can be used as organic modifiers that decrease t_M and k¹ to improve the resolution of analytes that co-elute with the micellar phase. These agents, however, may alter the level of the EOF. Cyclodextrins are cyclic polysaccharides that form inclusion complexes that can cause competitive hydrophobic partitioning of the analyte. Since analyte-cyclodextrin complexes are neutral, they will migrate toward the cathode at a higher velocity than that of the negatively charged micelles. Mixed micelle systems, such as the one formed by combining SDS with the neutral surfactant Brij-35^®, can also be used to alter the selectivity of MEKC.¹

Applications of MEKC

The simplicity and efficiency of MEKC have made it an attractive technique for a variety of applications. Further improvements can be made to the selectivity of MEKC by adding chiral selectors or chiral surfactants to the system. Unfortunately, this technique is not suitable for protein analysis because proteins are generally too large to partition into a surfactant micelle and tend to bind to surfactant monomers to form SDS-protein complexes.³ Recent applications of MEKC include the analysis of uncharged pesticides⁸, essential and branched-chain amino acids in nutraceutical products⁹, hydrocarbon and alcohol contents of marjoram herb¹⁰ . MEKC has also been targeted for its potential to be used in combinatorial chemical analysis. The advent of combinatorial chemistry has enabled medicinal chemists to synthesize and identify large numbers of potential drugs in relatively short periods of time. Small sample and solvent requirements and the high resolving power of MEKC have enabled this technique to be used to quickly analyze a large number of compounds with good resolution. Traditional methods of analysis, like high-performance liquid chromatography (HPLC), can be used to identify the purity of a combinatorial library, but assays need to be rapid with good resolution for all components to provide useful information for the chemist.¹¹ The introduction of surfactant to traditional capillary electrophoresis instrumentation has dramatically expanded the scope of analytes that can be separated by capillary electrophoresis.

References

1. Cunico,R.L.; Goodin, K.M.; Wehr,T. “Basic HPLC and CE of Biomolecules” Bay Bioanalytical Laboratory: Richmond, CA, 1998.
2. Baker, D.R. “Capillary Electrophoresis” John Wiley & Sons, Inc.: New York, 1995.
3. Skoog, D.A.; Holler, F.J.; Nieman, T.A. “Principles of Instrumental Analysis, 5th ed.” Saunders college Publishing: Philadelphia, 1998.
4. Simms, P.J.; Jeffries, C.T.; Huang, Y; Zhang, L.; Arrhenius, T.; Nadzan, A.M. J. Comb. Chem. 2001, 3, 427.
5. Terabe, S.; Otsuka, K.; Ichikawa, K.; Tsuchiya, A.; Ando, T. Anal. Chem. 1984, 56, 111.
6. Terabe, S.; Otsuka, K.; Ichikawa, K.; Tsuchiya, A.; Ando, T. Anal. Chem. 1984, 56, 113.
7. Foley, J.P. Anal. Chem. 1990, 62, 1302.
8. Carretero, A.S.; Cruces-Blanco, C.; Ramirez, S.C.; Pancorbo, A.C.; Gutierrez, A.F. J. Agric. Food. Chem. 2004, 52, 5791.
9. Cavazza, A.; Corradini, C.; Lauria, A.; Nicoletti, I. J. Agric. Food Chem. 2000, 48, 3324.
10. Rodrigues, M.R.A.; Caramao, E.B.; Arce, L.; Rios, A.; Valcarcel, M. J. Agric. Food Chem. 2002, 50, 425.
11. Simms, P.J.; Jeffries, C.T.; Huang, Y.; Zhang, L.; Arrhenius, T.; Nadzan, A.M. J. Comb. Chem. 2001, 3, 427.