Journal of Chromatography A 1674 (2022) 463142

Contents lists available at ScienceDirect

Journal of Chromatography A
journal homepage: www.elsevier.com/locate/chroma

Interactions of basic compounds with ionic liquids used as oils in
microemulsion liquid chromatography
N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel, M.C. García-Alvarez-Coque∗
Departament de Química Analítica, Universitat de València, c/Dr. Moliner 50, Burjassot, Spain

a r t i c l e

i n f o

Article history:
Received 14 February 2022
Revised 10 May 2022
Accepted 10 May 2022
Available online 13 May 2022
Keywords:
Oil-in-water microemulsion liquid
chromatography
Sodium dodecyl sulphate
Ionic liquids
β -adrenoceptor antagonists

a b s t r a c t
Aqueous microemulsions (MEs), where an oil coexists with water in the presence of the anionic surfactant
sodium dodecyl sulphate (SDS), have been proposed as a solution to decrease the amount of organic solvent in the mobile phase needed in reversed-phase liquid chromatography (RPLC). However, the oil phase

of a typical ME is volatile, toxic and flammable, and although it is added in a small amount, it would
be desirable to avoid it from an environmental perspective. This is the reason for the proposal of Peng
et al. (J. Chromatogr. A 1499 (2017) 132–139) to replace the oil in microemulsion liquid chromatography
(MELC) by the apolar ionic liquid (IL) 1-hexyl-3-methylimidazolium hexafluorophosphate ([C6 C1 IM][PF6 ]),
to analyse neutral phenolic acids at acidic pH. Based on this report, an MELC procedure is here proposed
for β -adrenoceptor antagonists, which are basic compounds where the dominant species is cationic. To
verify the formation of MEs containing SDS and IL, and elucidate the interactions between the cationic
basic compounds with the SDS anion, and the cation and anion in the IL, an extensive study was carried out with several methylimidazolium ILs containing the cations [C2 C1 IM]+ , [C4 C1 IM]+ , or [C6 C1 IM]+ ,
combined with the anions Cl– , BF4 – , or PF6 – , using 1-butanol as co-surfactant. The behaviour was compared with that observed in classical MELC with octane, micellar liquid chromatography with SDS and
1-propanol, and RPLC with mobile phases containing an IL and acetonitrile.
© 2022 The Author(s). Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license
( />
1. Introduction
Microemulsions (MEs) are thermodynamically stable and transparent colloidal solutions that contain an organic phase (oil) and
an aqueous phase (water), the latter stabilised by a surfactant
above its critical micelle concentration (CMC) and an organic solvent performing as co-surfactant [1,2]. Oil-in-water (O/W) MEs are
made up of oil droplets dispersed in the aqueous medium containing the surfactant, so they have high water content, low viscosity
and high solubilising power that makes them suitable as mobile
phases in reversed-phase liquid chromatography (RPLC). This chromatographic mode, known as microemulsion liquid chromatography (MELC), provides unique selectivity for both hydrophilic and
hydrophobic substances and has gained relevance in recent years
[3,4].
Common reagents in O/W MELC are the surfactants sodium dodecyl sulphate (SDS, anionic) and polyoxyethylene(23) lauryl ether
(Brij-35, non-ionic), the oils heptane, octane, cyclohexane, diiso-

∗

Corresponding author.
E-mail address: (M.C. García-Alvarez-Coque).


propylether and ethyl acetate, and the alcohols 1-propanol, 1butanol and 1-pentanol. In MELC systems, the mobile phases require lower concentration of organic solvent compared to conventional RPLC (below 1% and 10% v/v for the oil phase and cosurfactant, respectively). Since any change in the nature and concentration of the reagents (surfactant, oil and co-surfactant) can
significantly affect the chromatographic behaviour of the solutes
[4,5], a detailed systematic investigation is usually required to obtain successful separations.
In general, the replacement of harmful and volatile solvents,
traditionally used in many processes, has generated great interest
in recent years. Ideally, the best solvent would be no solvent, considering health hazards, waste generation and treatment, as well as
economic reasons [6]. Since the absence of solvent is not always
possible, several more environmentally friendly alternatives have
been proposed to decrease the impact and overall risk of chemical
exposure to conventional organic solvents. Amongst these alternatives are ionic liquids (ILs) [7], which are salts with low melting
points (usually below 100 °C), formed by a bulky organic cation
associated with a smaller inorganic/organic anion to get electrical
neutrality [8–10].

/>0021-9673/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
( />

N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

The interest in ILs can be attributed to the wide range of interactions with solutes (strong and weak ionic interactions, hydrogen
bonding, and van der Waals, dispersive, n-π and π -π interactions).
All these interactions give rise to interesting solvation properties,
compared to conventional organic solvents [11]. Other attractive
features of ILs are their low volatility and flammability, high thermal stability, and low toxicity, which have led to the replacement
of conventional polluting solvents by ILs, which have earned the label of benign or green solvents. Some recent reports have shown,
however, that some ILs (as those based on fluorinated anions), are
not as safe and non-toxic as claimed [12,13], but their toxicity (as
is the case for other physico-chemical properties) can be modulated by the appropriate selection of the IL cation and anion.

In the analytical field, ILs have been widely applied in sample preparation [14,15] and chromatographic analysis [16,17]. They
have also been used immobilised in stationary phases in gas chromatography [18,19] and liquid chromatography [20–22], and as
mobile phase additives in the hydro-organic mobile phases used in
RPLC [22,23]. In RPLC analysis, ILs lose their physical characteristics as solvents, being just salts that dissociate in aqueous medium
[24]. A relevant advantage of the addition of ILs to the mobile
phase is that ion exchange interactions of cationic solutes with
residual anionic silanols, which are present in conventional silica
stationary phases, are minimised. This improves peak performance,
which has been explained by the adsorption of the IL cation and
anion onto the stationary phase, creating an asymmetric bilayer,
positively or negatively charged that masks the silanols. The effect
is stronger for ILs with a cation of larger size [25].
Recently, alkyl-methylimidazolium ILs, associated to tetrafluoroborate (BF4 – ) and hexafluorophosphate (PF6 – ), were proposed to
prepare ionic liquid-in-water (IL/w) MEs (also called aqueous ILbased MEs), for the MELC analysis in acidic medium (pH = 2.5)
of hydrophilic phenolic compounds (danshensu, caffeic acid, protocatechualdehyde, rosmarinic acid and salvianolic acid B) in Danshen samples (a traditional Chinese medicinal herb) [26]. The
procedure produced excellent selectivity and adequate resolution.
In this work, the effect of the addition to IL/w MEs of alkylmethylimidazolium ILs with alkyl chains of diverse lengths and associated with the anions Cl– , BF4 – and PF6 – (the most common in
RPLC [23]), on the retention and peak profile behaviours of cationic
basic compounds (β -adrenoceptor antagonists), is investigated at
acidic medium. The results are compared with those found with
MELC mobile phases containing SDS, 1-butanol and octane as oil,
and with RPLC mobile phases containing SDS and 1-propanol, or
ILs and acetonitrile.

3-methylimidazolium hexafluorophosphate ([C4 C1 IM][PF6 ]), 1hexyl-3-methylimidazolium hexafluorophosphate ([C6 C1 IM][PF6 ]),
1–butyl–3-methylimidazolium tetrafluoroborate ([C4 C1 IM][BF4 ]),
1-hexyl-3-methylimidazolium tetrafluoroborate ([C6 C1 IM][BF4 ]),
and 1-hexyl-3-methylimidazolium chloride ([C6 C1 IM][Cl]), all from
Sigma. Molar concentrations were used for the surfactant and
ILs. Volumetric fractions (expressed as percentage) were used for

acetonitrile, 1-butanol and octane.
The mobile phases considered in our study contained: (i) SDS,
1-butanol and IL, (ii) SDS, 1-butanol and octane (data taking from
Ref. [27]), (iii) SDS and 1-propanol (data taking from Ref. [28]), or
(iv) IL and acetonitrile (data taking from Ref. [28]). The pH of the
MELC mobile phases with either IL or octane was fixed at 1.35 with
0.05% trifluoroacetic acid from Thermo Fisher Scientific (Loughborough, UK). The pH of the other mobile phases was buffered at 3.0
with 0.01 M citric acid monohydrate and sodium hydroxide from
Panreac (Barcelona). The pH metre was calibrated with aqueous
buffers, while the pH of the mobile phases was always fixed in the
presence of the organic solvent. β -Adrenoceptor antagonists have
a strong basic character (pKa ≥ 9), which means that at the acidic
pH of the mobile phases the cationic species are dominant.
The solutions of the β -adrenoceptor antagonists and mobile
phases were filtered through 0.45 μm Nylon membranes from Micron Separations (Westboro, MA, USA). Nanopure water obtained
with an Adrona system (Riga, Latvia) was used throughout.
2.2. Apparatus and columns
An Agilent (Waldbronn, Germany) chromatograph was used,
equipped with a quaternary pump (Series 1200), an automatic injector (Series 1260 Infinity II), a thermostatted column compartment (Series 1290 Infinity II), and a diode array detector (Series
1100). The β -adrenoceptor antagonists were monitored at 225 nm,
except for timolol, which was detected at 300 nm. Uracil was detected at 254 nm. The retention data were obtained at 25 °C, using
isocratic conditions with a flow rate of 1 ml/min. Duplicate injections of 20 μl were made.
The chromatographic system was controlled with an OpenLAB
CDS LC Chemstation (Agilent B.04.03). The mathematical treatment
was carried out with Excel (Microsoft Office 2010, Redmond, WA,
USA). Chromatographic peaks were processed with the MICHROM
software to obtain the peak parameters (retention times and peak
half-widths) [29].
An XTerra-MS C18 column from Waters (Milford, MA, USA),
which replaces one out of every three silanols with a methyl group,

was used with the MELC mobile phases of SDS, 1-butanol and
IL or octane, and the mixtures of IL and acetonitrile, and SDS
and 1-propanol. The characteristics of the column were as follows: 150 mm × 4.6 mm i.d., 5 μm particle size, 120 A˚ mean
pore diameter, 175 m²/g surface area, and 12 wt% total carbon. A
Kromasil C18 column from Análisis Vínicos (Ciudad Real, Spain)
with 150 mm × 4.6 mm i.d., 5 μm particle size, 110 A˚ average
pore diameter, 320 m2 /g surface area, and 19% carbon load, was
also used when working with micellar mobile phases and mobile
phases containing ILs and acetonitrile [28]. In all cases, the analytical columns were preceded by similar 30 mm guard columns for
mobile phase protection.
Mobile phases were recycled between runs and also during
analysis to reduce reagent consumption and wastes. The low evaporation risk of organic solvents in the mobile phases with additives makes recycling possible, as long as a sufficiently low number of injections is made. The mobile phase was renewed each
week when the composition was not changed. A small flow rate
of 0.1 ml min–1 was used between analyses. The chromatographic
system was periodically rinsed with pure water and methanol or
2-propanol (around 30 ml) to remove surfactant and IL from the

2. Experimental
2.1. Reagents
Seven β -adrenoceptor antagonists (atenolol, acebutolol, carteolol, metoprolol, timolol, oxprenolol, and propranolol), all from
Sigma (St. Louis, MA, USA) were used as probe compounds. The
drugs were dissolved in 1 mL of methanol from VWR International
(France), with the aid of an Elmasonic S15 H ultrasonic bath from
Elma (Singen, Germany), and diluted with water. The concentration
of the stock solutions of the probe compounds, which were stable
during at least two months at 4 °C, was approximately 100 μg/mL.
These solutions were diluted with water to a final concentration
of 20 μg/mL, prior to injection into the chromatograph. Uracil from
Acros Organics (Geel, Belgium) was used as hold up time marker.
The reagents used to prepare the mobile phases were sodium

dodecyl sulphate from Merck (99% purity, Darmstadt, Germany),
acetonitrile and 1-butanol from Scharlab (Barcelona, Spain),
octane from Alfa Aesar (Kandel, Germany), and the ILs 1-ethyl-3methylimidazolium hexafluorophosphate ([C2 C1 IM][PF6 ]), 1–butyl–
2


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

stationary phase. Over the weekend, the column was maintained
with 2-propanol.

3. Results and discussion
3.1. Formation of transparent mixtures of [C6 C1 IM][PF6 ], SDS and
1-butanol
In a recent study, Peng et al. assayed alkyl-methylimidazolium
ILs formed with [C4 C1 IM]+ , [C6 C1 IM]+ and [C8 C1 IM]+ , associated
to BF4 – , PF6 – and bis[(trifluoromethyl) sulfonyl] imide (NTf2 – ) to
be used in MELC [26]. These authors observed that due to the low
solubility of [C6 C1 IM][PF6 ] in water it can replace the oil to form a
ME composed of SDS, IL and 1-butanol. This IL was found suitable
to analyse a group of neutral phenolic compounds, based on the
short analysis time and good separation selectivity. In this work,
this ME has been taken as starting point for the analysis of the
cationic β -adrenoceptor antagonists.
To verify the conditions for the formation of a transparent
medium to be used as mobile phase, or the presence of two well
differentiated phases, we prepared several mixtures with different
amounts of SDS, [C6 C1 IM][PF6 ] and 1-butanol. The effect of the addition of [C6 C1 IM][PF6 ] was checked in the 0.01–0.10 M range, in

solutions containing a fixed amount of SDS (0.10 M) and varying 1butanol (2–12% v/v), or fixed 1-butanol (8.15% v/v) and varying SDS
(0.02–0.25 M). Once the reagents were mixed, the mixtures were
allowed to stand for at least 12 h, and then centrifuged. When
transparent mixtures were obtained, they were left to rest for several weeks to check its long-term stability. The formation of clear
and stable solutions was visually verified at room temperature, at
least during two weeks.
In a previous study [5], the formation of an emulsion containing
SDS, with increasing octane or decreasing 1-butanol, gave rise to
an upper phase that increased in thickness and turned whitish, an
effect that was more intense at the smallest assayed concentrations
of the surfactant. When the oil was replaced with [C6 C1 IM][PF6 ],
phase separation was not so clearly observed, being only evidenced
by the observation of a whitish drop of IL falling through the solution. However, in most tested mixtures, a clear solution was obtained.
The composition of the transparent mixtures, and of mixtures
giving rise to the possible formation of an emulsion, is represented
in Fig. 1a and 1b. Using 0.10 M SDS (Fig. 1a), stable mixtures were
always formed, with a maximal concentration of [C6 C1 IM][PF6 ]
close to 0.08 M at both lower (1.81% ) and upper (12.7%) extreme
concentrations of 1-butanol. This means that the surfactant was
able to solubilise the IL without the need of a large amount of
co-surfactant. When the concentration of 1-butanol was fixed at
8.15% v/v (Fig. 1b), increasing amounts of [C6 C1 IM][PF6 ] required a
higher concentration of SDS to obtain stable solutions.
Maximal concentrations of 0.10 M and 0.25 M were tested for
[C6 C1 IM][PF6 ] and SDS, respectively. It should be noted that the
concentration range for [C6 C1 IM][PF6 ] in the mobile phase used in
RPLC is usually narrow and with an upper value below 0.04 M
to avoid high viscosity. The ability of SDS to solubilise this IL
can be explained by the formation of a stable ME, where the IL
would act as oil (IL/w ME). However, the formation of a neutral

ion pair or any other structure between the anionic SDS micelles
and the alkyl-methylimidazolium cation must also be considered.
This could also explain the secondary role of 1-butanol in the solubilisation of the IL.
The results in Fig. 1a and 1b should be compared with those
shown in previous work with an SDS/octane/1-butanol system,
where the role of the co-surfactant (1-butanol) was relevant for the
solubilisation of octane [5]. At 0.10 M and 0.18 M SDS concentra-

Fig. 1. Concentration range for: (a) 1-butanol and [C6 C1 IM][PF6 ] in the presence
of 0.10 M SDS, and (b) SDS and [C6 C1 IM][PF6 ] in the presence of 8.15% 1-butanol.
The circles correspond to the compositions that gave rise to the formation of clear
solutions, whereas the crosses correspond to the compositions that produced phase
separation.

tions, a high concentration of 1-butanol solubilised higher amounts
of octane.
3.2. Retention behaviour of basic compounds with mobile phases
containing SDS, [C6 C1 IM][PF6 ] and 1-butanol
In a chromatographic system with mobile phases containing
SDS and IL, the stationary phase should probably be coated by
layers of surfactant monomers, IL cation and, to a lesser extent,
IL anion. Alkyl-methylimidazolium cations with sufficiently long
alkyl chains (such as [C6 C1 IM]+ ), associated to chaotropic anions
(such as PF6 – ), have been reported to be significantly adsorbed
on the stationary phase [30]. The adsorbed reagents, which are
ionic, change the nature of the stationary phase from an apolar
(hydrophobic) to a polar charged (hydrophilic) surface. The charge
sites in the stationary phase produced by this adsorption serve as
ion exchangers for cationic solutes. The multiple possible effects
(interactions of the anionic surfactant and IL cation and anion with

the stationary phase, and of the cationic solutes with the surfac3


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142
Table 1
Half-width plots parameters for several chromatographic systems: slopes of the left
(mA ) and right (mB ) half-widths, sum of slopes and slope ratio.

Without additivea
SDS/1-butanol/IL
[C2 C1 IM][PF6 ]
[C4 C1 IM][PF6 ]
[C6 C1 IM][PF6 ]
[C6 C1 IM][BF4 ]
[C6 C1 IM][Cl]
SDS/1-butanol/octane
IL/acetonitrile without SDSb
[C2 C1 IM][PF6 ]
[C4 C1 IM][PF6 ]
[C2 C1 IM][BF4 ]
[C4 C1 IM][BF4 ]
[C6 C1 IM][BF4 ]
[C2 C1 IM][Cl]
[C4 C1 IM][Cl]
[C6 C1 IM][Cl]
a
b


mA

mB

mA + mB

mB /mA

0.021

0.047

0.068

2.3

0.028
0.027
0.031
0.026
0.023
0.043

0.025
0.028
0.033
0.039
0.027
0.044


0.053
0.055
0.064
0.065
0.049
0.087

0.9
1.0
1.0
1.5
1.2
1.0

0.026
0.026
0.018
0.020
0.022
0.017
0.019
0.020

0.038
0.040
0.022
0.022
0.017
0.023
0.019

0.016

0.064
0.066
0.040
0.042
0.039
0.041
0.039
0.036

1.5
1.5
1.2
1.1
0.8
1.3
1.0
0.8

Acetonitrile-water, from Ref. [32].
From Refs. [25,32].

ME formed by SDS, 1.14% octane and 8.15% 1-butanol (Fig. 2b), reported in Ref. [27]. The trend produced by increasing the SDS concentration is similar, but with lower retention when octane is used
instead of [C6 C1 IM][PF6 ]. Observe that with either IL or octane, the
decrease in the retention factors at the highest SDS concentrations
is no more relevant for all probe compounds.

3.3. Effect of the IL cation and anion on retention
In order to gain more insight on the effect of hybrid systems of SDS and IL on the retention of the β -adrenoceptor antagonists, several mobile phases containing 0.05 M SDS, 8.15%

1-butanol, and alkylimidazolium ILs with different cations and
anions (and therefore, different solubility in water [30]) were
assayed. On the one hand, the effect of different anions using the same IL cation (1-hexyl-methylimidazolium) ([C6 C1 IM][Cl],
[C6 C1 IM][BF4 ], and [C6 C1 IM][PF6 ]) was studied, and on the other,
the effect of different alkyl lengths in the IL cation using the same
IL anion (hexafluorophosphate) ([C6 C1 IM][PF6 ], [C4 C1 IM][PF6 ] and
[C2 C1 IM][PF6 ]). The selected concentrations were 0.01 and 0.03 M
for all ILs. In previous work, the amount of the anions adsorbed
on a Kromasil C18 column with mobile phases containing 30% acetonitrile and 0.05 M NaCl, NaBF4 or NaPF6 , was measured [30]:
Cl– showed low affinity to the C18 stationary phase (∼2.5 μmol),
whereas the affinity of BF4 – and PF6 – was moderate (∼15 μmol)
and strong (∼32 μmol), respectively.
Fig. 3a depicts the effect of the addition of different ILs, in the
presence of 0.05 M SDS and 8.15% 1-butanol, on the behaviour
of metoprolol, which shows intermediate retention amongst the
studied β -adrenoceptor antagonists (similar trends were observed
for the other compounds). The retention decreased with increasing concentration of the ILs, being the effect stronger as the alkyl
chain in the IL increased: [C2 C1 IM]+ < [C4 C1 IM]+ < [C6 C1 IM]+ .
This decreasing trend was also observed in mobile phases containing the ILs without SDS, when combined with the anions BF4 – and
Cl– , which are weakly adsorbed (Fig. 3b). This can be explained
by considering that the stronger adsorption of the more hydrophobic IL cation with a longer alkyl chain repels the cationic solutes
significantly [28]. Note that the IL cation dissolved in the mobile
phase will also repel the cationic solute, but this would be shifted
towards the stationary phase, increasing the retention (i.e., the opposite effect). Furthermore, a stronger adsorbed IL anion would attract the cationic solutes (also increasing the retention).

Fig. 2. Change in retention at increasing concentration of SDS, in the presence of
8.15% 1-butanol and: (a) 0.01 M [C6 C1 IM][PF6 ], and (b) 1.14% octane. Solute identity:
( ) acebutolol, (◦) atenolol, (♦) carteolol, ( ) metoprolol, (●) oxprenolol, ( ) propranolol, and ( ) timolol (acknowledgement is given to The Royal Society of Chemistry
for the reproduction of Fig. 2a from Ref. 27).


tant and IL ions in the mobile phase and adsorbed on the stationary phase) complicate the interpretation of the retention mechanism.
The retention factors for the β -adrenoceptor antagonists obtained with mobile phases containing 0.01 M [C6 C1 IM][PF6 ], 8.15%
1-butanol, and SDS in the range 0.05–0.25 M, are depicted in
Fig. 2a. As observed, the addition of an increasing concentration of
surfactant produced the expected decrease in retention, since there
is a maximal amount of surfactant adsorbed on the C18 column
that attracts the cationic solutes, while the concentration of SDS
micelles in the mobile phase (which also interact with the solutes)
increases [31]. Therefore, the cationic solutes undergo a progressive
distribution into an increased volume of microemulsion droplets
(micelles containing IL in its core or surface), which increases the
elution strength.
The observed behaviour must be compared with the changes
in retention observed for the β -adrenoceptor antagonists with the
4


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

Fig. 3. Retention behaviour of metoprolol in different RPLC systems containing: (a) 0.05 M SDS and 8.15% 1-butanol, with increasing concentration of diverse ILs, (b) several
ILs in the presence of fixed 15% acetonitrile, and (c) SDS in the presence of fixed 15% 1-propanol. In (b), the retention times are identical for [C6 C1 IM][PF6 ], [C6 C1 IM][BF4 ],
and [C6 C1 IM][Cl]. Assayed ILs: ( ) C2 C1 IM][PF6 ], ( ) [C4 C1 IM][PF6 ], ( ) [C4 C1 IM][BF4 ], (●) [C6 C1 IM][PF6 ], ( ) [C6 C1 IM][BF4 ], and (◦) [C6 C1 IM][Cl].

on the stationary phase. As commented above, once the stationary phase is saturated with SDS, the amount of surfactant in the
mobile phase (forming micelles) is increased. This makes the elution strength stronger due to the attraction of the cationic solutes to the anionic micelles. A similar behaviour is observed
with the mobile phases that contain an increased concentration
of SDS in the presence of fixed amounts of IL and 1-butanol
(Fig. 2a), or octane and 1-butanol (Fig. 2b), although the retention is globally smaller due to the presence of the organic solvents

or IL.
A comparison of the trends in retention at increasing concentration of IL, in the presence of SDS (Fig. 3a) (MELC with IL), and
without SDS (Fig. 3b) (RPLC with IL), can also help to interpret the
possible interactions. We should indicate that the behaviour could
only be studied in the presence of [C2 C1 IM][PF6 ], [C4 C1 IM][PF6 ],
[C4 C1 IM][BF4 ], [C6 C1 IM][BF4 ], and [C6 C1 IM][Cl], since the solubility
of [C6 C1 IM][PF6 ] in the absence of SDS was too low.

The decreased retention of basic solutes at a higher concentration of IL, in the range of 0 to 0.03 M, suggested that the interaction of the cationic basic compounds with the imidazolium cations
(electrostatic repulsion with the adsorbed IL cation) should prevail
over the association with the adsorbed IL anions on the stationary
phase (which would cause the attraction of the basic compounds),
whose concentration also changes when the IL is added to the
mobile phase. Therefore, the strongly adsorbed SDS should hinder the adsorption of the IL anion (even for PF6 – ). In Fig. 3a, note
that in the presence of SDS, the retention times for [C6 C1 IM][PF6 ],
[C6 C1 IM][BF4 ] and [C6 C1 IM][Cl] are identical. In the presence of
SDS, the behaviour for [C4 C1 IM][PF6 ] and [C4 C1 IM][BF4 ] will probably be also similar.
Fig. 3c shows the retention of metoprolol with a mobile phase
with SDS in the 0.01–0.15 M range, containing also 15% 1-propanol.
The high retention at low concentration of the surfactant reveals
the attraction of the cationic solutes towards the adsorbed SDS
5


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

Fig. 4. Half-width plots (left, A (◦) and right, B (●)), built with the data obtained for the set of β -adrenoceptor antagonists with mobile phases containing 0.05 M SDS, 8.15%
1-butanol, and the following ILs at 0.01 and 0.03 M concentrations: (a) [C6 C1 IM][PF6 ], (b) [C4 C1 IM][PF6 ], (c) [C2 C1 IM][PF6 ], (d) [C4 C1 IM][BF4 ], and (e) [C4 C1 IM][Cl].


6


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

Without surfactant (Fig. 3b), the retention was significantly affected by the presence of specific IL cations and anions, which
should be explained by their particular adsorption capability on
the C18 stationary phase. The adsorption of some cations and anions is stronger and also the saturation of the stationary phase towards these ions. As discussed above, the adsorption on the stationary phase of the cation in the IL increases with increasing
length of its alkyl chain, whereas the adsorption of PF6 – is significantly stronger compared to BF4 – and Cl– . This explains the similar trend in retention with mobile phases containing [C6 C1 IM][BF4 ]
and [C6 C1 IM][Cl], where the retention decreases with increasing IL,
which is the behaviour observed in the presence of SDS (Fig. 3a).
Meanwhile, the combined effect of BF4 – and an IL with shorter
alkyl length ([C4 C1 IM][BF4 ]), in the absence of SDS, results in a
nearly constant retention at increasing amount of the IL. This behaviour is produced by the smaller adsorption of [C4 C1 IM]+ , compared to [C6 C1 IM]+ (both with a decreasing effect on retention),
the latter being more competitive with respect to the adsorption
of BF4 – (which would increase the retention).
For [C4 C1 IM][PF6 ] added to mobile phases without SDS, the
combined effect of cation and anion resulted in increased retention at low IL concentration and decreased retention at higher
concentration (Fig. 3b). The interpretation of this behaviour is not
easy, due to the significant amount of both cation ([C4 C1 IM]+ ) and
anion (PF6 – ), adsorbed on the stationary phase and dissolved in
the mobile phase, giving rise to repulsion and attraction of the
cationic solutes, respectively. In this regard, the trend observed for
[C2 C1 IM][PF6 ] is interesting, since the lower adsorption of an IL
cation with shorter alkyl length ([C2 C1 IM]+ ) is combined with an
anion that shows strong adsorption (PF6 – ). In this case, the retention increased at least until reaching the maximal concentration
tested, which indicates that the adsorption of the anion (which

attracts the cationic solutes to the stationary phase) is dominant.
Note that, in contrast, for [C4 C1 IM][PF6 ] and [C2 C1 IM][PF6 ], the retention always decreases in the presence of SDS, with the addition
of IL.
3.4. Effect of the IL cation and anion on the peak profiles
Peak profiles in liquid chromatography are characterised by
their height, position, width and skewness, the two latter depending on the values of the left and right peak half-widths. The observation of the trend of peak half-widths is useful to evaluate
the interaction kinetics of the solutes with the stationary phase.
Also, equations that allow predicting the profiles of the peaks in
the chromatograms can be obtained, which are useful for optimisation purposes. Fortunately, simple correlations can be established
between peak half-widths and retention times, which in isocratic
elution can be approximated to straight-lines. When all solutes experience the same kinetics, such plots can be obtained with the
half-widths/retention time data obtained with a mobile phase of
fixed or variable composition [32,33]. When the solutes experience
different resistance to mass transfer to/from the column, the plots
will exhibit significant scattering.
Half-width plots for the set of β -adrenoceptor antagonists are
plotted in Fig. 4 for mobile phases containing SDS/1-butanol and
five ILs with diverse cations and anions. The plots were drawn
with the information obtained for the set of solutes eluted with
mobile phases of variable composition. Table 1 collects the characteristics of the plots: the slopes of the left (mA ) and right (mB )
half-widths, the sum of slopes (which describes the relationship
of the width with the retention times) and the ratio of slopes
(which is related to the asymmetry). The values should be compared with the results obtained with mobile phases of acetonitrilewater, SDS/1-propanol, and IL/acetonitrile [28]. The presence of the
additives (SDS and/or IL), in all cases, gave rise to a significant im-

Fig. 5. Half-width plots (left, A (◦) and right, B (●)), built with the data obtained for
the set of β -adrenoceptor antagonists with mobile phases containing: (a) 0.114 M
SDS, 0.28% octane, and 8.15% 1-butanol, and (b) 0.156 M SDS, 0.28% octane, and
8.15% 1-butanol.


provement in the peak profiles with respect to the classical hydroorganic RPLC with acetonitrile-water. This can be explained by the
masking effect of the free anionic silanols in the silica-based stationary phases by the ionic additives (SDS and IL).
In the presence of IL, the peaks were significantly more symmetrical compared to acetonitrile-water mixtures, especially for
[C2 C1 IM][PF6 ], [C4 C1 IM][PF6 ] and [C6 C1 IM][PF6 ], in the presence
of SDS, and for [C4 C1 IM][BF4 ] and [C4 C1 IM][Cl] without SDS
(B/A = 0.9–1.1, see Table 1). The parameters of the half-width plots
in Fig. 4 should also be compared with those obtained for an MELC
mobile phase with octane, where the mean value of B/A = 1.0. For
comparison purposes, Fig. 5 represents the plots for particular mobile phases in MELC with octane: 0.28% octane/8.15% 1-butanol in
the presence of 0.114 M or 0.156 M SDS, for which the B/A values
were 1.1 and 0.9, respectively.
7


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

Fig. 6. Experimental chromatograms obtained for mixtures of β -adrenoceptor antagonists with: (a) 0.10 M SDS, 0.01 M [C6 C1 IM][PF6 ], and 8.15% 1-butanol, (b) 0.114 M SDS,
1.14% octane, and 8.15% 1-butanol, and (c) 0.10 M SDS and 0.02 M [C6 C1 IM][Cl]. Solute identity: (1) Atenolol, (2) carteolol, (3) acebutolol, (4) metoprolol, (5) oxprenolol, and
(6) propranolol. Column: XTerra-MS C18 (150 mm × 4.6 mm i.d., 5 μm) (acknowledgement is given to The Royal Society of Chemistry for the reproduction of Fig. 4f from
Ref. [27]).

3.5. Retention of basic compounds with SDS / IL mobile phases
without organic solvent

of 0.114 M SDS/1.14% octane/8.15% 1-butanol, optimised in previous work (Fig. 6b). It was thus evident that the co-surfactant (1butanol) did not help to achieve the needed chromatographic resolution for the basic compounds when [C6 C1 IM][PF6 ] was added
instead of octane. Meanwhile, the studies in Section 3.3 indicated
that the separation was dominated by attraction of the cationic solutes to the adsorbed SDS monomers and repulsion from the adsorbed IL cation in the stationary phase. Therefore, the possibility
of removing the alcohol from the mobile phase was considered.

We thought that the combined effect of both reagents (i.e., attraction to the anionic SDS monomer and repulsion to the IL cation)
would be able to modulate the separation of the analytes, and
yield an adequate separation without the need of the co-surfactant.

We must remember that the purpose of adding 1-butanol is
to help with the stabilisation of MEs, but when an IL is used instead of an apolar organic solvent as octane, the presence of 1butanol does not seem to be so relevant in the formation of clear
mixtures useful for RPLC (see Section 3.1). On the other hand,
the retention of β -adrenoceptor antagonists using mobile phases
containing 0.10 M SDS/0.01 M [C6 C1 IM][PF6 ]/8.15% 1-butanol was
too short (below 10 min), and a significant overlap of the peaks
of the set of compounds was observed (Fig. 6a). The separation
was indeed poorer compared to that achieved with the mixture
8


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

According to this, a mixture containing only SDS and [C6 C1 IM][Cl]
in aqueous solution was prepared, which gave rise to a clear solution able to be used as mobile phase with an RPLC column.
It should be noted that the retention times for the β adrenoceptor antagonists are rather long with aqueous mobile
phases containing only micellar SDS or [C6 C1 IM][Cl] (without organic solvent). Thus, the retention times for atenolol, carteolol, acebutolol, metoprolol, oxprenolol and propranol eluted with 0.10 M
SDS (without IL) from the XTerra column were 9.9, 14.3, 16.8, 34.8,
57.1 and 83.5 min, respectively, while with 0.02 M [C6 C1 IM][Cl]
(without SDS), atenolol and carteolol eluted at 3.5 min and
11.6 min, respectively, and metoprolol, oxprenolol and propranolol
needed above 60 min.
Fig. 6c shows the chromatogram obtained for the mixture of six
β -adrenoceptor antagonists, using an isocratic mobile phase containing 0.10 M SDS and 0.02 M [C6 C1 IM][Cl], without organic solvent. The separation suggests that the aqueous mixtures of SDS

and [C6 C1 IM][Cl] can be successful in the separation of mixtures
of the β -adrenoceptor antagonists, with a favourable effect on the
resolution and an analysis time below 30 min. However, the most
remarkable aspect is that the separation was achieved in aqueous
medium, using an IL with chloride as anion, without the need an
organic solvent in the mobile phase.

The formation of transparent and stable MEs prepared with surfactant (SDS), co-surfactant (1-butanol), and apolar solvent (IL or
octane), to be used in RPLC, was found less dependant on the concentration of co-surfactant, when octane was replaced with an IL.
Furthermore, SDS allowed more concentrated solutions of the ILs,
which suggested the formation of stable structures. In view of this
behaviour and the fact that the addition of 1-butanol to the ME
formed with SDS and [C6 C1 IM][PF6 ] yielded too short retention
times, and poor resolution in the separation of the group of β adrenoceptor antagonists, the elimination of the co-surfactant from
the mobile phase was proposed, the detailed study of which will
be the subject of future work.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.
Acknowledgements
Work
supported
by
Grant
PID2019–106708GB-I00
funded by MCIN (Ministery of Science and Innovation of
Spain)/AEI/10.13039/50110 0 011033. Ester Peris-García thanks
the University of Valencia for the post-doctoral grant UV INVPREDOC16F1–384313.

4. Conclusions

In the literature, ILs seem ideal for replacing organic solvents
used as oil phase in MEs, due to their attractive physico-chemical
properties and lower toxicity. However, reported MEs formed by
IL, water and surfactant (and in some cases, an alcohol as cosurfactant) are usually prepared with non-ionic surfactants, such
as Brij-35 and Triton X-100, instead of the anionic SDS [34]. The
work by Peng et al., published in 2017 [26], pioneered the use
of MEs in RPLC, where the oil was replaced with an insoluble IL
([C6 C1 IM][PF6 ]), and the anionic surfactant SDS (quite unusual for
the preparation of IL/w MEs) was combined with 1-butanol as cosurfactant. The authors developed an analytical procedure for neutral phenolic acids.
In this work, the feasibility of using the IL/w ME recommended
by Peng et al. as mobile phase, for the RPLC analysis of a group
of basic compounds (β -adrenoceptor antagonists), which are positively charged, was investigated. The research was focused on the
effect on retention times and peak profiles produced by imidazolium ILs with alkyl chains of increasing length (with n = 2, 4 and
6), associated to Cl– , BF4 – , or PF6 – . The research group had previously developed a detailed work on the interactions of cationic solutes with RPLC C18 columns using mobile phases containing aqueous solutions of imidazolium ILs, in the presence of acetonitrile.
Here a comparison is made of the effect of the cation and anion in
diverse ILs, in the presence of SDS and 1-butanol, with respect to
our previous work with mobile phases containing IL and acetonitrile instead of 1-butanol in the absence of SDS. The study gives
some insight on the retention mechanisms.
The anionic surfactant SDS was found to compete with the IL
anions for column adsorption, the behaviour being similar to that
found without SDS, when an IL cation showing strong adsorption
is associated with a weakly adsorbed anion. In these situations, the
retention decreased by addition of an increasing concentration of
IL. Meanwhile, in the absence of SDS, the addition of an IL with a
weakly adsorbed cation or a strongly adsorbed anion makes retention to remain constant or increase with a maximum at a particular concentration of IL. On the other hand, with all tested ILs, the
peak profiles of the basic compounds were improved, but the effect
was stronger in the presence of SDS. The peaks were completely
symmetrical (B/A = 1.0–1.1) for [C4 C1 IM][BF4 ] and [C4 C1 IM][Cl], indicating an efficient masking of the silanol effect.

References

[1] J. Sjöblom, R. Lindbergh, S.E. Friberg, Microemulsions: phase equilibria, characterization, structures, applications and chemical reactions, Adv. Colloid Interface Sci. 95 (1996) 125–287.
[2] C. Stubenrauch, Microemulsions: Background, New concepts, Applications, Perspectives editor, Wiley-Blackwell, Hoboken, NJ, USA, 2009.
[3] E. Peris-García, N. Pankajkumar-Patel, S. Carda-Broch, M.J. Ruiz-Angel, M.C. García- Alvarez-Coque, Oil-in-water microemulsion liquid chromatography, Sep.
Purif. Rev. 49 (2020) 89–111.
[4] N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel, S. Carda-Broch, M.C. García-Alvarez-Coque, Modulation of retention and selectivity in oil-in-water microemulsion liquid chromatography, J. Chromatogr. A 1592 (2019) 91–100.
[5] N. Pankajkumar-Patel, E. Peris-García, O. Schiopu, M.J. Ruiz-Angel, J.J. Baeza-Baeza, M.C. García-Alvarez-Coque, Performance and modelling of retention in
microemulsion liquid chromatography, J. Chromatogr. A 1634 (2020) 461651.
[6] R.A. Sheldon, The E factor 25 years on: the rise of green chemistry and sustainability, Green Chem. 19 (2017) 18–43.
[7] M. Freemantle, An Introduction to Ionic Liquids, RSC Publishing, Cambridge UK,
2009.
[8] T. Welton, Room-temperature ionic liquids: solvents for synthesis and catalysis,
Chem. Rev. 99 (1999) 2071–2084.
[9] C.F. Poole, Chromatographic and spectroscopic methods for the determination
of solvent properties of room temperature ionic liquids, J. Chromatogr. A 1037
(2004) 49–82.
´
[10] U. Domanska,
General review of ionic liquids and their properties, in: Ionic
Liquids in Chemical Analysis. M. Koel, Ed., CRC Press, New York, 2009,
pp. 1–71.
[11] H. Olivier-Bourbigou, L. Magna, D. Morvan, Ionic liquids and catalysis: recent
progress from knowledge to applications, Appl. Catal. Gen. 373 (2010) 1–56.
[12] D. Zhao, Y. Liao, Z. Zhang, Toxicity of ionic liquids, Clean 35 (2007) 42–48.
[13] G. Cevasco, C. Chiappe, Are ionic liquids a proper solution to current environmental challenges? Green Chem. 16 (2014) 2375–2385.
[14] M. Germán-Hernández, V. Pino, J.L. Anderson, A.M. Afonso, A novel in situ
preconcentration method with ionic liquid-based surfactants resulting in enhanced sensitivity for the extraction of polycyclic aromatic hydrocarbons from
toasted cereals, J. Chromatogr. A 1227 (2012) 29–37.
[15] M.J. Trujillo-Rodríguez, P. Rocío-Bautista, V. Pino, A.M. Afonso, Ionic liquids in
dispersive liquid-liquid microextraction, Trends Anal. Chem. 51 (2013) 87–106.
[16] A. Berthod, M.J. Ruiz-Angel, S. Carda-Broch, Ionic liquids in separation techniques, J. Chromatogr. A 1184 (2008) 6–18.

[17] A. Berthod, M.J. Ruiz-Angel, S. Carda-Broch, Recent advances on ionic liquid
uses in separation techniques, J. Chromatogr. A 1559 (2018) 2–16.
[18] F.
Destaillats,
M.
Guitard,
C.
Cruz-Hernández,
Identification
of
6-monoinsaturated fatty acids in human hair and nail samples by GC-MS
using ionic liquid coated capillary column, J. Chromatogr. A 1218 (2011) 9384.
[19] L.W. Hantao, A. Najafi, C. Zhang, F. Augusto, J.L. Anderson, Tuning the selectivity of ionic liquid stationary phases for enhanced separation of nonpolar analytes in kerosene using multi-dimensional gas chromatography, Anal. Chem.
86 (2014) 3717–3721.
9


N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel et al.

Journal of Chromatography A 1674 (2022) 463142

[20] H. Xian, H. Peng, X. Wang, D. Long, R. Ni, J. Chen, S. Li, Z. Zhang, J. Peng, Preparation and evaluation of a mixed-mode stationary phase with imidazolium and
carboxyl group for high performance liquid chromatography, Microchem. J. 150
(2019) 104131.
[21] R. Ni, J. Peng, M. Wen, H. Peng, Z. Zhang, S. Li, X. Wang, J. Chen, D. Long,
H. Xian, Preparation, characterization and evaluation of an imidazolium ionic
liquid copolymer stationary phase for mixed-mode chromatography, Anal.
Methods 12 (2020) 324–332.
[22] M.J. Ruiz-Angel, S. Carda-Broch, M.C. García-Alvarez-Coque, Ionic liquids as
mobile phase additives and immobilised on stationary phases in liquid chromatography, Ionic liquids: New Insights and Recent Developments, Elsevier,

Waltham, MA, 2022 S. Carda-Broch, M.J. Ruiz-Angel, Eds..
[23] M.C. García-Alvarez-Coque, M.J. Ruiz-Angel, A. Berthod, S. Carda-Broch, On the
use of ionic liquids as mobile phase additives in high-performance liquid chromatography, Anal. Chim. Acta 883 (2015) 1–21.
[24] P. Sun, D.W. Armstrong, Ionic liquids in Analytical Chemistry, Anal. Chim. Acta
661 (2010) 1–16.
[25] M.T. Ubeda-Torres, C. Ortiz-Bolsico, M.C. García-Alvarez-Coque, M.J. Ruiz-Angel,
Gaining insight in the behaviour of imidazolium-based ionic liquids as additives in reversed-phase liquid chromatography for the analysis of basic compounds, J. Chromatogr. A 1380 (2015) 96–103.
[26] L.Q. Peng, J. Cao, L.J. Du, Q.D. Zhang, Y.T. Shi, J.J. Xu, Analysis of phenolic acids
by ionic liquid-in-water microemulsion liquid chromatography coupled with
ultraviolet and electrochemical detector, J. Chromatogr. A 1499 (2017) 132–139.

[27] N. Pankajkumar-Patel, E. Peris-García, M.J. Ruiz-Angel, M.C. García-Alvarez–
Coque, Comparison of surfactant-mediated liquid chromatographic modes with
sodium dodecyl sulphate for the analysis of basic drugs, Anal. Methods 12
(2020) 2443–2452.
[28] J.J. Fernández-Navarro, M.C. García-Alvarez-Coque, M.J. Ruiz-Angel, The role of
the dual nature of ionic liquids in the reversed-phase liquid chromatographic
separation of basic drugs, J. Chromatogr. A 1218 (2011) 398–407.
[29] J.R. Torres Lapasió, MICHROM Software, Marcel Dekker, New York, 20 0 0.
[30] A. Berthod, M.J. Ruiz-Angel, S. Huguet, Non-molecular solvents in separation methods: dual nature of room temperature ionic liquids, Anal. Chem. 77
(2005) 4071–4080.
[31] A. Berthod, M.C. García-Alvarez-Coque, Micellar Liquid Chromatography, Marcel Dekker, New York, 20 0 0.
[32] J.J. Baeza-Baeza, M.J. Ruiz-Angel, M.C. García-Alvarez-Coque, S. Carda-Broch,
Half-width plots, a simple tool to predict peak shape, reveal column kinetics
and characterise chromatographic columns in liquid chromatography: state of
the art and new results, J. Chromatogr. A 1314 (2013) 142–153.
[33] J.J. Fernández-Navarro, J.R. Torres-Lapasió, M.J. Ruiz-Angel, M.C. García-Alvarez-Coque, Silanol suppressing potency of alkyl-imidazolium ionic liquids on
C18 stationary phases, J. Chromatogr. A 1232 (2012) 166–175.
[34] M. Hejazifar, O. Lanaridi, K. Bica-Shröder, Ionic liquid based microemulsions: a
review, J. Mol. Liq. 303 (2020) 112264.


10