Separation of achiral and chiral analytes using polymeric surfactants with ionic liquids as modifiers in micellar electrokinetic chromatography.

Abstract

In this study, we report the use of ionic liquids as modifiers in the separation of achiral and chiral analytes in micellar electrokinetic chromatography. In this investigation, polymeric surfactants and ionic liquids were added to a low-conducting buffer solution. The polymeric surfactants used in this study were poly(sodium N-undecylelinic sulfate) and poly(sodium oleyl-l-leucylvalinate). The ionic liquids used in this study were chosen because of their high conductivity, hydrophobicity, and good solvating properties. Thus, it was expected that these ionic liquids would have the ability to assist in the separation of hydrophobic mixtures while maintaining adequate background current. Three analyte mixtures were separated using various buffer combinations of polymeric surfactant and ionic liquids. The ionic liquids were shown to improve the resolution and peak efficiency of the analytes while maintaining adequate background current.

Full text

Page 1

S eparation of Ac hiral and Chiral Analytes Using
Polymeric S urfac tants with Ionic Liquids as
Modifiers in Mic ellar Elec trokinetic
Chromatography
Simon M. Mwongela, Abdulqawi Numan, Nicole L. Gill, Rezik A. Agbaria, and Isiah M. Warner*
Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803
In this study, we report the use of ionic liquids as
modifiers in the separation of achiral and chiral analytes
in micellar electrokinetic chromatography. In this inves-
tigation, polymeric surfactants and ionic liquids were
added to a low-conducting buffer solution. The polymeric
surfactants used in this study were poly(sodium N-
undecylelinic sulfate) and poly(sodium oleyl-L-leucylvali-
nate). The ionic liquids used in this study were chosen
because of their high conductivity, hydrophobicity, and
good solvating properties. Thus, it was expected that these
ionic liquids would have the ability to assist in the
separation of hydrophobic mixtures while maintaining
adequate background current. Three analyte mixtures
were separated using various buffer combinations of
polymeric surfactant and ionic liquids. The ionic liquids
were shown to improve the resolution and peak efficiency
of the analytes while maintaining adequate background
current.
Ionic liquids (ILs), sometimes called molten salts, are liquids
at ambient temperatures and consist entirely of ionic species. In
the past, ILs were mainly ofinterest toelectrochemists. However,
recently it has become apparent that a wide range of chemical
reactions can be conducted using this class of solvents.1For
example, many recent scientific investigations have focused on
ILs2-29as a new class of solvents for various chemical reactions,
such as liquid-liquid extraction,5,6organic synthesis,7-12electro-
chemistry,13-16catalysis for cleantechnology,17-21ultralowvolatility
liquid matrixes for matrix-assisted laser desorption/ionization
(MALDI) mass spectrometry,22and separations.23-27
The typical ionic liquid consists of a bulky pyridinium or
imidazolium cation paired with a variety of anions.3,13,14Ionic
liquids have many properties of conventional organic solvents,
such as excellent solvation qualities, a low viscosity, and a wide
temperaturerange.3,16-18Twocharacteristics that ILs donot share
with conventional organic solvents arevolatility andgoodelectrical
conductivities.3,4,7,16Ionic liquids are environmentally benign,
nonvolatile, andnonflammablewith ahigh thermal stability.4Their
miscibility in water is heavily dependent on the type of anion
forming the ionic liquid. In addition, the length of the alkyl chain
on the pyridiniumor imidazoliumcation has aconsiderable effect
on the properties of the ionic liquid.19Wilkes and Zaworotko7
concludedthatthel-ethyl-3-methylimidazolium(EMIM) cationwas
chemically and electrochemically robust and was generally useful
for synthesis of ionic liquids. Fuller and co-workers28fully
characterized the low-melting salt, l-ethyl-3-methylimidazolium
hexafluorophosphate (EMIMPF6, mp 58-60 °C) and concluded
* Corresponding author. Phone:
E-mail: iwarner@lsu.edu.
(1) Freemantle, M. Chem. Eng. News 2000, (May 15), 37.
(2) Osteryoung, R. A. Molten Salt Chemistry: An Introduction and Selected
Application; Mamantov, G., Marassi, R., Eds.; NATO ASI Series, Ser. C.
1987, Vol. 202, p 329.
(3) Fannin, A. A.; Floreani, A. D.; King, A. L.; Landers, S. J.; Piersma, J. B.;
Stech, J. D.; Vaughn, L. R.; Wilkes, S. J.; William, L. J. J. Phys. Chem. 1984,
88, 2614.
(4) Anderson, J. L.; Ding, J.; Welton, T.; Armstrong, D. W. J. Am. Chem. Soc.
2002, 124 (47), 14247
(5) Selvan, M. S.; McKinley, M. D.; Dubois, R. H.; Atwood, J. L. J. Chem. Eng.
Data 2000, 45 (5), 841.
(6) Dai, S.; Ju, Y. H.; Barnes, C. E. J. Chem. Soc., Dalton Trans. 1999, 1201.
(7) Wilkes, J. S.; Zaworotko, M. J. J. Chem. Soc., Chem. Commun. 1992, 965.
(8) Adams, C. J.; Earle, M. J.; Roberts, G.; Seddon, K. R. Chem. Commun. 1998,
2097.
(9) Earle, M. J.; McCormac, P. B.; Seddon, K. R. Chem. Commun. 1998, 2245.
(10) Leadbeater, N. E.; Torenius, H. M. J. Org. Chem. 2002, 67, 3145.
(11) Welton, T. Chem. Rev. 1999, 99, 2071-2083.
(12) Nara, S. J.; Harjani, J. R.; Salunkhe, M. M. Tetrahedron Lett. 2002, 43, 2979.
(225) 578-3945. Fax:(225) 578-3971.
(13) Wilkes, J. S.; Levisky, J. A.; Wilson, R. A.; Hussey, C. L. Inorg. Chem. 1982,
21, 1263.
(14) Sanders, J. R.; Ward, E. H.; Hussey, C. L. J. Electrochem. Soc. 1986, 133,
325.
(15) Dickinson, V. E.; Willaims, M. E.; Hendrickson, S. M.; Masui, H.; Murray,
R. W. J. Am. Chem. Soc. 1999, 121, 613.
(16) Hussey, C. L. Pure Appl. Chem. 1988, 60, 1763.
(17) Seddon, K. R. J. Chem. Technol. Biotechnol. 1997, 68, 351.
(18) Adams, C. J.; Earle, M. J.; Seddon, K. R. Green Chem. 2000, 21.
(19) Bowlas, C. J.; Bruce, D. W.; Seddon, K. R. Chem. Commun. 1996, 1625.
(20) Erbeldinger, M.; Mesiano, A. J.; Russell, A. J. Biotechnol. Prog. 2000, 16
(6), 1129.
(21) Mehnert, C. P.; Cook, R. A.; Dispenziere, N. C.; Afeworki, M. J. Am. Chem.
Soc. 2002, 124 (44), 12932.
(22) Armstrong, D. W.; Zhang, L.-K.; He, L.; Gross, M. L. Anal. Chem. 2001,
73 (15), 3679.
(23) Huddleston, J. G.; Willauer, H. D.; Swatloski, R. P.; Visser, A. E.; Rogers, R.
D. Chem. Commun. 1998, 1765.
(24) Dullius, J. E. L.; Suarez, P. A. Z.; Einloft, S.; de Souza, R. F.; Dupont, J.;
Fischer, J.; DeCian, A. Organometallics 1998, 17, 815.
(25) Yanes, E. G.; Gratz, S. R.; Baldwin, M. J.; Robison, S. E.; Stalcup, A. M.
Analyst 2000, 125, 1919.
(26) Yanes, E. G.; Gratz, S. R.; Baldwin, M. J.; Robison, S. E.; Stalcup, A. M.
Anal. Chem. 2001, 73, 3838.
(27) Armstrong, D. W.; He, L.; Liu, Y.-S. Anal. Chem. 1999, 71, 3873.
(28) Fuller, J.; Carlin, H. C.; DeLong, D.; Harwoth, Chem. Commun. 1994, 299.
(29) Holbrey, J. D.; Seddon, K. R. J. Chem. Soc., Dalton Trans. 1999, 2133.
Anal. Chem. 2003, 75, 6089-6096
10.1021/ac034386i CCC: $25.00
Published on Web 09/20/2003
© 2003 American Chemical Society
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003
6089

Page 2

that EMIMPF6interionic interactions were dominated by cation-
anion Coulombic attraction with minimal hydrogen bonding. The
ionic liquidl-ethyl-3-methylimidazoliumtetrafluoroborate(EMIMBF4,
mp 15 °C) was found to be a liquid at room temperature and was
highly miscible in water.26,29
In addition to the characterization work mentioned above,
Armstrong et al. recently developed a solvation model for ionic
liquids based on multiple solvation interactions.4They concluded
that ionic liquids were complex solvent systems, as compared to
conventional organic solvents. Using the solvation model, Arm-
strong and co-workers were able to show that the anion had
greater effect on the hydrogen bond basicity of the ionic liquid,
as compared to the cation.
The application of ILs for the separation of various classes of
compounds has also been recently recognized. Yanes et al.
reported the development of a fairly robust capillary electrophore-
sis (CE) method for the separation of polyphenols found in grape
seedextracts inwhich theIL tetraethylammoniumtetrafluroborate
was used as the only background electrolyte (BGE).25More
recently, Yanes and co-workers developed a CE method for the
same analysis using 1-ethyl-3-methylimidazolium-based ionic liq-
uids as the BGE.26In another publication, Armstrong et al.
examined two ionic liquids (1-butyl-3-methylimidazolium hexafluo-
rophosphate (BMIMPF6) and 1-butyl-3-methylimidazolium-chlo-
ride (BMIMCl)) as stationary phases in gas-liquid chromatog-
raphy.27
In micellar electrokinetic chromatography (MEKC), a surfac-
tant is added into the buffer to separate either chiral or achiral
analytes. Theuseofpolymeric surfactants bearing both chiral and
achiral ionic headgroups has been reported.30-33In some cases,
organic modifiers havebeenusedintheseparationofhydrophobic
environmental pollutants, such as polycyclic aromatic hydrocar-
bons (PAHs), and alkyl aryl ketones, to help resolve such
mixtures.33,34To date, the use of ionic liquids as modifiers in
MEKC using polymeric surfactants has not been investigated.
Whenusing ionic liquids, thechoiceoftheBGE is very important,
because most ionic liquids are highly conductive. There are
several advantages for using ionic liquids over organic solvents
as modifiers. Ionic liquids are soluble in water, have a good
electrical conductivity, and act as good electrolytes in CE either
whenusedindependently or whenmixedwith other buffers. Ionic
liquids are slightly more viscous than organic solvents; therefore,
low concentrations may be required for buffer modifications to
achieve better separations. In addition, ionic liquids are less
volatile and are referred to as ªgreen solventsº, meaning they are
environmentally friendly. In contrast, organic solvents are poor
conductors of electricity, and high concentrations of organic
solvents inthebuffer causecurrentbreakdowns inCE. Inaddition,
most organic solvents are highly volatile and harmful to the
environment.
(30) Palmer, C. P. Electrophoresis 2002, 23, 3993.
(31) Shamsi, S. A.; Palmer, C. P.; Warner, I. M. Anal. Chem. 2001, 73, 140A.
(32) Maichel, B.; Kenndler, E. Electrophoresis 2000, 21, 3160.
(33) Dube, S.; Smith, R. M. Chromatographia 2003, 57, 485.
(34) Shamsi, S. A.; Akbay, C.; Warner, I. M. Anal. Chem. 1998, 70, 3078.
Figure 1. Chemical structures of the ionic liquids (ILs) used as modifiers in the MEKC separation experiments.
6090
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003

Page 3

In this paper, ionic liquids were used as modifiers in micellar
electrokinetic chromatography (MEKC) using polymeric surfac-
tants to separate both achiral and chiral compounds. The ionic
liquids used in this study were 1-butyl-3-methylimidazolium
tetrafluoroborate (BMIMBF4), l-ethyl-3-methylimidazolium tet-
rafluoroborate (EMIMBF4), l-ethyl-3-methylimidazolium hexafluo-
rophosphate (EMIMPF6), 1-ethyl-3-methylimidazolium trifluoro-
methanesulfonate(EMIMSO3F3), and1-ethyl-3-methylimidazolium
chloride(EMIMCl); their chemical structures areshowninFigure
1. Three different analyte mixtures were separated using these
ionic liquids. Those mixtures included a mixture of eight alkyl
aryl ketones, a mixture of seven phenols, and a mixture of three
chiral binaphthyl derivatives. Theperformanceofsodiumdodecyl
sulfate (SDS) with the ionic liquids was compared to that of poly-
SUS with the ionic liquids for the separation of alkyl aryl ketone
mixture.
EX PERIMENTAL SECTION
Materials. Thealkyl aryl ketones (acetophenone, propiophen-
one, butyrophenone, valerophenone, hexanophenone, heptanophen-
one, octanophenone, decanophenone, and dodecanophenone) the
phenols (4-chlorophenol, 3-chlorophenol, 4-chloro-3-methylphenol,
2-chlorophenol, 2,4,6-trichlorophenol, 2,4-dichlorophenol, and pen-
tachlorophenol), andthechiral compounds ((()-1,1′-bi-2-naphthol
(BOH), (()-1,1′-bi-2-naphthyl-2,2′-diyl hydrogenphosphate(BNP),
and 1,1′-bi-2-naphthyl-2,2′-diamine (BNA)) were purchased from
Sigma-Aldrich (Milwaukee, WI) (Figure 2). The undecylenyl
alcohol, sodium borate, chlorosulfonic acid, pyridine, oleic acid
N-hydroxysuccinimideester, sodiumhydrogencarbonate, sodium
borate, disodium phosphate, tris(hydroxymethyl)aminomethane
(TRIS), tetrahydrofuran(THF), sodiumdodecyl sulfate(SDS), and
theionic liquids EMIMBF4, EMIMPF6, EMIMSO3F3, andEMIMCl
werealsoobtainedfromSigma-Aldrich. Theionic liquidBMIMBF4
was purchased from Chemada Fine Chemicals Ltd., (Nir Itzhak,
D.N.HaNegev 85455, Israel). The dipeptide leucine-valine was
purchased from Bachem Bioscience Inc (King of Prussia, PA).
All reagents were of analytical grade and were used as received
without further purification.
Synthesis of Poly(sodium N-undecylelinic sulfate) [Poly-
(SUS)]. The sodium undecylinic sulfate (SUS) monomer was
preparedaccording toBregstrom's procedure,35andmodifications
were made according to the procedure described by Shamsi et
al.34to obtain the polymerized surfactant poly-SUS.
Synthesis of Poly(sodium oleyl-L-leucylvalinate) [Poly-(L-
SOLV)]. The L-SOLV monomer and polymer poly-L-SOLV were
synthesized using the guidelines of the procedure reported by
Wang and Warner,36with some modifications as described in a
previous study.37All polymers used in this study were found to
be >99%pure as estimated by elemental analysis.
Preparation of MEKC Buffer Solutions. The background
electrolyte (BGE) used for separation of the alkyl aryl ketones
was 100 mM TRIS at pH 10, and the BGE used for the phenols
was 20 mM sodium borate and disodium phosphate at pH 9.2.
The BGE used for the binaphthyl derivatives was 100 mM TRIS
mixed with 10 mM sodium borate at pH 10. The pH of the BGE
was adjusted with 1 M NaOH or 1 M HCl, whenever was
necessary. Anappropriateamount ofthepolymeric surfactant was
then added to the BGE. The ionic liquid was then added as a
modifier to examine the separation efficiency and resolution of
the analytes.
Capillary Electrophoresis. The MEKC experiments were
performed by useofaHewlett-Packard 3D CE instrument (Foster
city, CA) equipped with a UV diode array detector. Bare fused-
silica capillaries (effective length 40 cm, 50-µm i.d. for the
separation of the ketones and phenols compounds and effective
length 52 cm, 50-µm i.d. for the separation of the binaphthyl
compounds), were purchased from Polymicro Technologies
(35) Bergstrom, S. Z. Physiol. Chem. 1936, 236, 163.
(36) Wang, J.; Warner, I. M. Anal. Chem. 1994, 66, 3773
(37) Mwongela, S. M.; Akbay, C.; Zhu, X.; Collins, S.; Warner, I. M. Electrophoresis,
2003, submitted.
(38) Kelly, K. A.; Burns, S. T.; Khaledi, M. G. Anal. Chem. 2001, 73, 6057.
Figure 2. (a) Chemical structures of the eight alkyl aryl ketones.
The elution order is 1-8, except where peaks are coeluting. (b)
Chemical structures of the seven phenols. The elution order is 1-7,
except where peaks are coeluting. (c) Chemical structures of the three
chiral binaphthyl derivatives.
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003
6091

Page 4

(Phoenix, AZ). The applied voltage ranged from +20 to +30 kV.
The detection wavelength was 254 nm. The temperature of the
capillary was maintained at 20 °C for the alkyl aryl ketones and
phenols and at 15 °C for the binaphthyl derivatives by the
instrument thermostating system, which consisted of a peltier
element for forced-air cooling and temperature control. Different
capillary temperatures were chosen for the achiral and chiral
analyte separation in order to strictly optimize the separations.
The binaphthyl derivatives were prepared in 50:50 methanol/
water at concentrations of 0.1 mg/mL. The ketones and the
phenols were prepared in methanol at aconcentration of 0.1mg/
mL. The samples injection size varied from 50 mbar for 1 s to 30
mbar for 3 s. Prior to use, each new capillary was conditioned for
30 min with 1 M NaOH and then for 30 min with 0.1 M NaOH.
Finally, the capillary was rinsed for 15 min with triply distilled
deionized water. Prior to each run, the capillary was flushed with
MEKC buffer for 3 min to condition and fill the capillary. To
change from one buffer to another (i.e., buffer containing poly-
SUS and anionic liquid toanother buffer containing poly-SUS and
adifferentionic liquid), thecapillary was flushedwith 0.1M NaOH
for 10 min and rinsed with water for 5 min, then filled with the
new MEKC buffer containing the ionic liquid for 5 min. This was
done to minimize any influence of the previous buffer.
Calculations. The retention factors for two of the analytes
(i.e., acetophenone and decanophenone were determined using
eq 1,38
where teois the migration time of an unretained solute, tris the
retention of a solute, and tmcis the migration of the micelle or
polymeric surfactant. Methanol was used as the electroosmotic
flow (teo) marker and was measured from the time of injection to
thefirstdeviationfrombaseline. Themigrationtimeofthemicelle
or polymeric surfactant (tmc) was determinedusing n-dodecaphen-
one as the marker. The resolution of the chiral analytes was
calculated using eq 2,39
where tr1 and tr2 are the respective migration times of each
enantiomer, and w1and w2are the peak widths at the baseline of
each enantiomer.
RESULTS AND DISCUSSION
Separation of Alkyl Aryl Ketones. Experiments were con-
ductedtoseparateamixtureofeight alkyl aryl ketones, first using
SDS, and later poly-SUS in the buffer. The separation results
obtained at the optimized buffer pH, voltage, temperature, and
SDS concentration are illustrated in Figure 3 I. Only five out of
the eight alky aryl ketones in the mixture were resolved. The
elution order for these ketones was from the least hydrophobic
analyte acetophenone (n ) 1) to the most hydrophobic analyte
decanophenone (n ) 9). The peak numbers are based on the
labeling usedfor Figure2a. TheoptimizedSDS concentrationwas
50 mM. Higher concentrations of SDS led to poor resolution and
longer migration times of the analytes, but lower concentrations
of SDS did not enhance the separation.
To enhance the resolution of the analytes, ILs were used as
modifiers using the SDS optimized concentration (50 mM). The
UV studies showed that the ionic liquid had a strong absorbance
in the region between 215and 235nm(datanot shown). An ionic
liquidconcentrationstudy was doneusing 1-10mM ofeach ionic
liquid in order to determine the optimum amount to be added to
the MEKC buffer composed of 50 mM SDS and 100 mM TRIS at
pH 10. Figure 3 electropherograms II and III are representative
of the data obtained when ionic liquids were used as modifiers
using SDS as the pseudostationary phase. The ionic liquids
increased the migration time of the analytes but did not enhance
either theresolutionor theefficiency for SDS separations ofthese
compounds.
Theseparationoftheeight alkyl aryl ketones mixturewas also
done using poly-SUS. The poly-SUS concentration and separation
conditions were optimized. The MEKC buffer consisted of 0.5%
poly-SUS and 100 mM TRIS at pH 10. Figure 4a I shows that a
better separation was achieved using poly-SUS than SDS for the
separation of the alkyl aryl ketones mixture. Six out of the eight
alkyl aryl ketones were well-resolved and eluted in <10 minutes.
However, the last three analytes coeluted. Ofthe five ionic liquids
used as modifiers in the MEKC buffer, BMIMBF4gave the best
results. Figure 4a II-VII illustrates the effect of BMIMBF4
concentrationontheseparationoftheketonemixture. Theelution
order was determined by spiking a small amount of each analyte
into the mixture, and the peaks are numbered according to the
analyte notation in Figure 2a. Slight changes in the EOF as well
as the migration times of the first four analytes when the
concentration of BMIMBF4was increased from 1 mM to 5 mM
were observed. However, the peak efficiencies and resolution of
these analytes were almost maintained. Shorter migration times
of the last four analytes were observed as the concentration of
BMIMBF4 was increased from 0 to 2 mM. However, longer
migration times were observed for BMIMBF4concentrations >2
mM. This trend is observed by viewing the plot of capacity factor
(k′) of the least hydrophobic analyte acetophenone (n ) 1) and
the most hydrophobic analyte decanophenone (n ) 9) versus the
concentration of BMIMBF4, as shown in Figure 4b. The other
(39) Terabe, S.; Ostuka, K.; Ichikawa, K.; Ando, T. Anal. Chem. 1984, 56, 111.
Figure 3. Electropherograms showing the separation of eight alkyl
aryl ketones: (I) using 50 mM SDS, (II) using 50 mM SDS + 3 mM
BMIMBF4, and (III) using 50 µM SDS + 10 mM BMIMBF4. Condi-
tions: 100 mM TRIS buffer, pH 10.0, press. Injection, 50 mbar for 1
s; temp, 20 °C; voltage, 30 kV; detection λ, 254 nm.
k )
(tr- teo)
teo(1 - tr/tmc)
(1)
Rs) 2(tr2- tr1)
w1+ w2
(2)
6092
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003

Page 5

ILs, EMIMSO3F3and EMIMCl, were used as modifiers in the
separation of alkyl aryl ketones, but the separation was not
enhanced.
Hydrophobic solutes are known to partition between the
aqueous phase and the polymeric pseudostationary phase in
MEKC. Because of their hydrophobicity, the alkyl aryl ketones
are retained in the polymeric pseudostationary phase more than
the aqueous phase. Figure 4a I shows that decanophenone (n )
9) interacts more with the polymeric pseudostationary phase. ILs
are conductive; thus, an addition of an IL toabuffer may enhance
its ionic strength under a controlled capillary temperature. This
fact was confirmed by an increment in the current as the
concentration of BMIMBF4was increased in the buffer, as shown
in Figure 5. Other parameters held constant. An enhancement in
the ionic strength or concentration of the background electrolyte
is expected to cause the ªvalueº of EOF to decrease and results
in an increase in the migration time of the analytes being
separated. However, in looking at Figure 5, an abnormality in the
EOF versus BMIMBF4 concentration plot is observed. Between
1 and 2 mM, the value of EOF increases with an increase in the
Figure 4. (a) Electropherograms showing the separation of eight alkyl aryl ketones: (I) no modifier; (II-VII) using 1, 2, 3, 5, 10, and 20 mM
BMIMBF4as modifiers. Conditions: 0.5% poly-SUS, 100 mM TRIS buffer, pH 10.0. Pressure injection, 50 mbar for 1 s; temp, 20 °C; voltage,
30 kV; detection λ, 254 nm. (b) Plot of capacity factor (k′) of the least hydrophobic and first analyte to elute (acetophenone) and the most
hydrophobic and last analyte to elute (decanophenone) versus the concentration of BMIMBF4.
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003
6093

Page 6

concentration of the ionic liquid. A plausible explanation given
for this trend is that, when 1-2 mM of the IL were added to the
buffer, we observed an increase in the value of EOF as aresult of
an enhancement in the current, as shown in Figure 5. It was
assumed that low concentrations of the ionic liquid caused a
compression of the mobile layer, making the value of EOF
increase. However, the addition of 3 mM or more of the IL in the
buffer ledtoadecreaseinthevalueofEOF, although weobserved
an enhancement in current. This is because at higher concentra-
tions of the IL, the (BMIM) cation is likely to coat the capillary
wall. Modification of the capillary wall by the IL cation may lead
to a change in the EOF. Yanes et al. illustrated that when the
ionic liquid was used as an electrolyte, it attached to the capillary
wall, engendering anodic EOF.26In our case, the TRIS buffer was
Figure 5. Effect of increasing 1-butyl-3-methylimidazolium tetrafluo-
roborate (BMIMBF4) concentration on the eletroosmotic flow (EOF)
and the observed current for the separations in Figure 4a.
Figure 6. (a) Electropherograms showing the separation of seven phenols: (I) no surfactant; (II) 0.5% poly-SUS; (III), (IV), (V), and (VI) using
0.5% poly-SUS with 1, 2, 3, and 5 mM BMIMBF4, respectively. Conditions: 20 mM Na2B4O7/Na2HPO4, pH 9.2. Voltage, 20 kV; temp, 20 °C;
injection size, 30 mbar for 3 s; and detection λ, 254 nm. (b) Electropherograms showing the separation of seven phenols: (I) no surfactant; (II)
0.5% poly-SUS; (III), (IV), and (V) using 0.5% poly-SUS with 0.5, 2, and 5 mM EMIMPF6, respectively. Conditions same as in Figure 6a.
6094
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003

Page 7

responsible for the cathodic EOF. The addition of 1-2 mM of
BMIMBF4led to an increase in the current and a slight increase
in the EOF, which in turn led to shorter migration times and an
enhancement in the peak efficiency, especially the last three
analytes. At this low concentration of the ionic liquid, we suppose
that the ionic liquid cation moves with the buffer rather than
coating on the capillary wall. However, with the addition of 3mM
or higher of BMIMBF4, a reasonable amount of coating of the
cation on the capillary wall is possible, which causes a decrease
in the EOF value and leads to longer migration times as well as
poor peak efficiency of the more hydrophobic analytes, although
we observed an enhancement in the current. Ionic liquids are
complex solvent systems and the mechanism of separation when
ionic liquids are used as modifiers in MEKC is still not well-
understood. We are carrying out further studies, which involve
separating more hydrophobic analytes, such as PAHs, and using
fluorescence spectroscopy and other techniques to achieve a
better understanding of the interactions between ILs, polymeric
surfactants, and the analytes of interest.
Separation of Phenols. Seven chlorophenols were separated
using a plain buffer (i.e., 20 mM Na2B4O7/Na2HPO4, pH 9.2),
buffer + poly-SUS, and buffer + poly-SUS + ILs. The results
recorded while using the plain buffer are shown in electrophero-
gram I of Figure 6a,b. Six out of seven of the analytes were well-
resolved but peaks 4 and 5 (i.e., 2-chlorophenol and 2,4,6-
trichlorophenol) coeluted. The separation of the phenols was
possible even with plain buffer because the phenols are slightly
charged at pH 9.2 used in this study.
Poly-SUS was then added to the buffer to enhance the
separation of the phenols, and the corresponding data is shown
in electropherogram II in Figure 6a,b. The presence of the
surfactant increased the elution time of the analytes and gave a
better resolution ofthe six analytes, as compared tothat achieved
with the plain buffer. The ILs were used in order to improve the
resolution of peaks 4 and 5. Of the five ILs investigated,
onlyBMIMBF4 and EMIMPF6 enhanced the resolution of the
peaks, as shown in electropherogram V, Figure 6a, and electro-
pherogram III, Figure 6b, respectively. A similar trend was
observed for the migration times of the phenols as with the alkyl
aryl ketones when the IL BMIMBF4was used as the modifier.
When EMIMBF4, EMIMSO3F3, and EMIMCl were used as
modifiers, longer migration times of the analytes were recorded,
although there were no significant changes in the EOF. In
addition, the coeluting peaks (4 and 5) were not resolved (data
not shown).
Enantiomeric Separation of Binaphthyl Derivatives. The
binaphthyl derivatives examined in this study were BNP, BOH,
and BNA. These analytes have been separated in our laboratory
using different chiral surfactants.40,41The chirality of the bi-
naphthyl derivatives is attributed to the chiral plane (C2sym-
metry). This group of compounds has a varying degree of
hydrophobicity and charge states under the experimental condi-
tions used in this work. Using poly-L-SOLV surfactant in a 100
mM TRIS 10 mM sodium borate buffer (MEKC buffer), the S
and R enantiomers of each chiral analyte could only be resolved
individually, not in a mixture. The separation was also examined
when ILs were added to the MEKC buffer. Electropherogram I
in Figure 7 shows that BNA and BOH coeluted in the absence of
an IL in the MEKC buffer, whereas electropherograms II, III,
and IV are a display of the data collected when the opti-
mized concentration of three different ionic liquids EMIMPF6,
BMIMBF4, andEMIMBF4, respectively, wereaddedtotheMEKC
Figure 7. Electropherograms showing the separation of three binaphthyl derivatives using (I) no modifier, (II) 5 mM EMIMPF6, (III) 1 mM
BMIMBF4, and (IV) 2 mM EMIMBF4as modifiers. Conditions: 0.5% poly-L-SOLV, 100 mM TRIS + 10 mM sodium borate buffer, pH 10.0. Press.
Injection, 30 mbar for 3 s; temp, 15 °C; voltage, 30 kV; and detection λ, 254 nm.
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003
6095

Page 8

buffer. The migration times of the analytes varied depending on
the IL used. EMIMPF6provided enhanced separation of all three
analytes. However, no enantiomeric separation was observed for
BOH, as shown in Figure 7 electropherogram II. A shift in the
baseline was observed for this separation. The cause of this shift
was not completely discernible but it could be responsible for the
poor enantiomeric separation of BNA and BOH. Both BMIMBF4
and EMIMPF6 facilitated the separation of all three chiral
compounds with baseline resolution of the R (+) and S (-)
enantiomers for each compound. The enantiomeric recognition
was, however, based on the interactions of the analytes with the
chiral poly-L-SOLV surfactant. The resolutions of each analyte
under each different buffer solution are recorded in Table 1.
CONCLUSIONS
Successful separations of two achiral mixtures and one chiral
mixture have been achieved using ILs as pseudostationary phase
modifiers. The results obtained were reproducible, and the ILs
were found to be stable in the background electrolyte solutions.
The separation of the analyte mixtures was dependent on the
interactionoftheanalytes with thepolymeric surfactants, whereas
the ionic liquids influenced the analytes' elution time and peak
efficiency. Longer migration times of the more hydrophobic
analytes were recorded when high concentrations ofthe ILs were
used as modifiers.
ACK NOWLEDGMENT
I.M.W. acknowledges the Philip W. West Endowment, the
National Institutes ofHealth, andtheNational ScienceFoundation
for their support. The authors thank Serigne Thiamand Kimberly
Hamilton for helpful discussions regarding these studies.
NOTE ADDED AFTER ASAP
Article posted on the Web 9/20/03. Some reference citations
were corrected and the article was reposted on 10/7/03.
Received for review April 14, 2003. Accepted August 13,
2003.
AC034386I
(40) Haddadian, F.; Billot, E.; Shamsi, S.; Warner, I. M. J. Chromatogr., A 1999,
858, 219.
(41) Yarabe H. H.; Billot, E.; Warner, I. M. J. Chromatogr., A 2000, 875, 179.
T able 1. Resolution V alues of the Binaphthyl
Derivatives S eparated in Figure 6 under the Optimized
Conc entrations of EMIMPF6, BMIMBF4, and EMIMBF4
electropherogram
no. in Figure 6
amt, type of
modifier
BNP
resolution
BNA
resolution
BOH
resolution
I
II
III
IV
0 mM
5 mM EMIMPF6
1 mM BMIMBF4
3 mM EMIMBF4
3.28
5.03
4.51
2.97
0.67
2.03
2.95
1.28
1.76
6096
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003