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).
Trang 1Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/chroma
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
Aqueousmicroemulsions(MEs),whereanoilcoexistswithwaterinthepresenceoftheanionicsurfactant sodiumdodecylsulphate(SDS),havebeenproposedasasolutiontodecreasetheamountoforganic sol-ventinthemobilephaseneededinreversed-phaseliquidchromatography(RPLC).However,theoilphase
ofatypicalMEisvolatile,toxicandflammable, andalthoughitisaddedinasmallamount,itwould
bedesirabletoavoiditfromanenvironmentalperspective.ThisisthereasonfortheproposalofPeng
etal.(J.Chromatogr.A1499(2017)132–139)toreplacetheoilinmicroemulsionliquidchromatography (MELC)bytheapolarionicliquid(IL)1-hexyl-3-methylimidazoliumhexafluorophosphate([C6C1IM][PF6]),
toanalyseneutralphenolicacidsatacidicpH.Basedonthisreport,anMELCprocedureishereproposed forβ-adrenoceptorantagonists,whicharebasiccompoundswhere thedominantspeciesiscationic.To verifytheformationofMEscontainingSDSandIL,andelucidatetheinteractionsbetweenthecationic basiccompoundswiththeSDSanion,and thecationandanionintheIL,anextensivestudywas car-riedoutwithseveralmethylimidazoliumILscontainingthecations[C2C1IM]+,[C4C1IM]+,or[C6C1IM]+, combinedwiththeanionsCl–,BF4–,orPF6–,using1-butanolasco-surfactant.Thebehaviourwas com-paredwiththatobservedinclassicalMELCwithoctane,micellar liquidchromatographywithSDSand 1-propanol,andRPLCwithmobilephasescontaininganILandacetonitrile
© 2022TheAuthor(s).PublishedbyElsevierB.V ThisisanopenaccessarticleundertheCCBY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/)
1 Introduction
Microemulsions (MEs) are thermodynamically stable and trans-
parent 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 sol-
vent performing as co-surfactant [ 1, 2] Oil-in-water (O/W) MEs are
made up of oil droplets dispersed in the aqueous medium contain-
ing 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 chro-
matographic mode, known as microemulsion liquid chromatogra-
phy (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 do-
decyl sulphate (SDS, anionic) and polyoxyethylene(23) lauryl ether
(Brij-35, non-ionic), the oils heptane, octane, cyclohexane, diiso-
∗ Corresponding author
E-mail address: celia.garcia@uv.es (M.C García-Alvarez-Coque)
propylether and ethyl acetate, and the alcohols 1-propanol, 1- butanol and 1-pentanol In MELC systems, the mobile phases re- quire lower concentration of organic solvent compared to con- ventional RPLC (below 1% and 10% v / v for the oil phase and co-
surfactant, respectively) Since any change in the nature and con- centration 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 ob- tain 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, con- sidering 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 alterna- tives 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]
https://doi.org/10.1016/j.chroma.2022.463142
0021-9673/© 2022 The Author(s) Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license
( http://creativecommons.org/licenses/by-nc-nd/4.0/ )
Trang 2The interest in ILs can be attributed to the wide range of inter-
actions with solutes (strong and weak ionic interactions, hydrogen
bonding, and van der Waals, dispersive, n- πand π-πinteractions)
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 ther-
mal stability, and low toxicity, which have led to the replacement
of conventional polluting solvents by ILs, which have earned the la-
bel 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 modu-
lated by the appropriate selection of the IL cation and anion
In the analytical field, ILs have been widely applied in sam-
ple preparation [ 14, 15] and chromatographic analysis [ 16, 17] They
have also been used immobilised in stationary phases in gas chro-
matography [ 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 characteris-
tics 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 tetrafluo-
roborate (BF 4–) and hexafluorophosphate (PF 6–), were proposed to
prepare ionic liquid-in-water (IL/w) MEs (also called aqueous IL-
based MEs), for the MELC analysis in acidic medium (pH = 2.5)
of hydrophilic phenolic compounds (danshensu, caffeic acid, proto-
catechualdehyde, rosmarinic acid and salvianolic acid B) in Dan-
shen 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 alkyl-
methylimidazolium ILs with alkyl chains of diverse lengths and as-
sociated with the anions Cl –, BF 4– and PF 6– (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
2 Experimental
2.1 Reagents
Seven β-adrenoceptor antagonists (atenolol, acebutolol, carte-
olol, 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-3-
methylimidazolium hexafluorophosphate ([C 2C 1IM][PF 6]), 1–butyl–
3-methylimidazolium hexafluorophosphate ([C 4C 1IM][PF 6]), 1- hexyl-3-methylimidazolium hexafluorophosphate ([C 6C 1IM][PF 6]), 1–butyl–3-methylimidazolium tetrafluoroborate ([C 4C 1IM][BF 4]), 1-hexyl-3-methylimidazolium tetrafluoroborate ([C 6C 1IM][BF 4]), and 1-hexyl-3-methylimidazolium chloride ([C 6C 1IM][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 (Loughbor- ough, 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 (p K a ≥ 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 Mi- cron 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 in- jector (Series 1260 Infinity II), a thermostatted column compart- ment (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 de- tected at 254 nm The retention data were obtained at 25 °C, using isocratic conditions with a flow rate of 1 ml/min Duplicate injec- tions 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 fol- lows: 150 mm × 4.6 mm i.d., 5 μm particle size, 120 ˚A mean pore diameter, 175 m ²/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 m 2/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 analyt- ical 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 evap- oration risk of organic solvents in the mobile phases with addi- tives makes recycling possible, as long as a sufficiently low num- ber 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
Trang 3stationary phase Over the weekend, the column was maintained
with 2-propanol
3 Results and discussion
3.1 Formation of transparent mixtures of [C 6 C 1 IM][PF 6 ], SDS and
1-butanol
In a recent study, Peng et al assayed alkyl-methylimidazolium
ILs formed with [C 4C 1IM] +, [C 6C 1IM] + and [C 8C 1IM] +, associated
to BF 4–, PF 6– and bis[(trifluoromethyl) sulfonyl] imide (NTf 2–) to
be used in MELC [26] These authors observed that due to the low
solubility of [C 6C 1IM][PF 6] 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, [C 6C 1IM][PF 6] and 1-butanol The effect of the ad-
dition of [C 6C 1IM][PF 6] was checked in the 0.01–0.10 M range, in
solutions containing a fixed amount of SDS (0.10 M) and varying 1-
butanol (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 sev-
eral 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 [C 6C 1IM][PF 6],
phase separation was not so clearly observed, being only evidenced
by the observation of a whitish drop of IL falling through the so-
lution However, in most tested mixtures, a clear solution was ob-
tained
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 [C 6C 1IM][PF 6]
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 [C 6C 1IM][PF 6] required a
higher concentration of SDS to obtain stable solutions
Maximal concentrations of 0.10 M and 0.25 M were tested for
[C 6C 1IM][PF 6] and SDS, respectively It should be noted that the
concentration range for [C 6C 1IM][PF 6] 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 sol-
ubilisation 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 [C 6 C 1 IM][PF 6 ] in the presence
of 0.10 M SDS, and (b) SDS and [C 6 C 1 IM][PF 6 ] 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, [C 6 C 1 IM][PF 6 ] 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 [C 6C 1IM] +), associated to chaotropic anions (such as PF 6–), 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 surfac-
3
Trang 4Fig 2 Change in retention at increasing concentration of SDS, in the presence of
8.15% 1-butanol and: (a) 0.01 M [C 6 C 1 IM][PF 6 ], and (b) 1.14% octane Solute identity:
( ) acebutolol, ( ◦) atenolol, ( ♦ ) carteolol, ( ) metoprolol, ( ●) oxprenolol, ( ) propra-
nolol, and ( ) timolol (acknowledgement is given to The Royal Society of Chemistry
for the reproduction of Fig 2 a from Ref 27 )
tant and IL ions in the mobile phase and adsorbed on the station-
ary phase) complicate the interpretation of the retention mecha-
nism
The retention factors for the β-adrenoceptor antagonists ob-
tained with mobile phases containing 0.01 M [C 6C 1IM][PF 6], 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
Table 1
Half-width plots parameters for several chromatographic systems: slopes of the left ( m A ) and right ( m B ) half-widths, sum of slopes and slope ratio
mA mB mA + m B mB / m A
SDS/1-butanol/IL [C 2 C 1 IM][PF 6 ] 0.028 0.025 0.053 0.9 [C 4 C 1 IM][PF 6 ] 0.027 0.028 0.055 1.0 [C 6 C 1 IM][PF 6 ] 0.031 0.033 0.064 1.0 [C 6 C 1 IM][BF 4 ] 0.026 0.039 0.065 1.5
IL/acetonitrile without SDS b
[C 2 C 1 IM][PF 6 ] 0.026 0.038 0.064 1.5 [C 4 C 1 IM][PF 6 ] 0.026 0.040 0.066 1.5 [C 2 C 1 IM][BF 4 ] 0.018 0.022 0.040 1.2 [C 4 C 1 IM][BF 4 ] 0.020 0.022 0.042 1.1 [C 6 C 1 IM][BF 4 ] 0.022 0.017 0.039 0.8
a Acetonitrile-water, from Ref [32]
b From Refs [ 25 , 32 ]
ME formed by SDS, 1.14% octane and 8.15% 1-butanol ( Fig.2b), re- ported in Ref [27] The trend produced by increasing the SDS con- centration is similar, but with lower retention when octane is used instead of [C 6C 1IM][PF 6] 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 sys- tems of SDS and IL on the retention of the β-adrenoceptor an- tagonists, 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 us- ing the same IL cation (1-hexyl-methylimidazolium) ([C 6C 1IM][Cl], [C 6C 1IM][BF 4], and [C 6C 1IM][PF 6]) was studied, and on the other, the effect of different alkyl lengths in the IL cation using the same
IL anion (hexafluorophosphate) ([C 6C 1IM][PF 6], [C 4C 1IM][PF 6] and [C 2C 1IM][PF 6]) 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% ace- tonitrile and 0.05 M NaCl, NaBF 4 or NaPF 6, was measured [30]:
Cl – showed low affinity to the C18 stationary phase ( ∼2.5 μmol), whereas the affinity of BF 4– and PF 6– 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 increas- ing concentration of the ILs, being the effect stronger as the alkyl chain in the IL increased: [C 2C 1IM] + < [C 4C 1IM] + < [C 6C 1IM] + This decreasing trend was also observed in mobile phases contain- ing the ILs without SDS, when combined with the anions BF 4–and
Cl –, which are weakly adsorbed ( Fig 3b) This can be explained
by considering that the stronger adsorption of the more hydropho- bic 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 op- posite effect) Furthermore, a stronger adsorbed IL anion would at- tract the cationic solutes (also increasing the retention)
Trang 5Fig 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 [C 6 C 1 IM][PF 6 ], [C 6 C 1 IM][BF 4 ], and [C 6 C 1 IM][Cl] Assayed ILs: ( ) C 2 C 1 IM][PF 6 ], ( ) [C 4 C 1 IM][PF 6 ], ( ) [C 4 C 1 IM][BF 4 ], ( ●) [C 6 C 1 IM][PF 6 ], ( ) [C 6 C 1 IM][BF 4 ], and ( ◦) [C 6 C 1 IM][Cl]
The decreased retention of basic solutes at a higher concentra-
tion of IL, in the range of 0 to 0.03 M, suggested that the interac-
tion 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 hin-
der the adsorption of the IL anion (even for PF 6–) In Fig.3a, note
that in the presence of SDS, the retention times for [C 6C 1IM][PF 6],
[C 6C 1IM][BF 4] and [C 6C 1IM][Cl] are identical In the presence of
SDS, the behaviour for [C 4C 1IM][PF 6] and [C 4C 1IM][BF 4] will prob-
ably 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
on the stationary phase As commented above, once the station- ary phase is saturated with SDS, the amount of surfactant in the mobile phase (forming micelles) is increased This makes the elu- tion strength stronger due to the attraction of the cationic so- lutes 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 reten- tion is globally smaller due to the presence of the organic solvents
or IL
A comparison of the trends in retention at increasing concen- tration 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 [C 2C 1IM][PF 6], [C 4C 1IM][PF 6], [C 4C 1IM][BF 4], [C 6C 1IM][BF 4], and [C 6C 1IM][Cl], since the solubility
of [C 6C 1IM][PF 6] in the absence of SDS was too low
5
Trang 6Fig. 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) [C 6 C 1 IM][PF 6 ], (b) [C 4 C 1 IM][PF 6 ], (c) [C 2 C 1 IM][PF 6 ], (d) [C 4 C 1 IM][BF 4 ], and (e) [C 4 C 1 IM][Cl]
Trang 7Without surfactant ( Fig.3b), the retention was significantly af-
fected 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 an-
ions is stronger and also the saturation of the stationary phase to-
wards these ions As discussed above, the adsorption on the sta-
tionary phase of the cation in the IL increases with increasing
length of its alkyl chain, whereas the adsorption of PF 6– is signif-
icantly stronger compared to BF 4– and Cl – This explains the simi-
lar trend in retention with mobile phases containing [C 6C 1IM][BF 4]
and [C 6C 1IM][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 BF 4– and an IL with shorter
alkyl length ([C 4C 1IM][BF 4]), in the absence of SDS, results in a
nearly constant retention at increasing amount of the IL This be-
haviour is produced by the smaller adsorption of [C 4C 1IM] +, com-
pared to [C 6C 1IM] + (both with a decreasing effect on retention),
the latter being more competitive with respect to the adsorption
of BF 4–(which would increase the retention)
For [C 4C 1IM][PF 6] added to mobile phases without SDS, the
combined effect of cation and anion resulted in increased reten-
tion 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 ([C 4C 1IM] +) and
anion (PF 6–), 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
[C 2C 1IM][PF 6] is interesting, since the lower adsorption of an IL
cation with shorter alkyl length ([C 2C 1IM] +) is combined with an
anion that shows strong adsorption (PF 6–) In this case, the reten-
tion 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 [C 4C 1IM][PF 6] and [C 2C 1IM][PF 6], the re-
tention 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 depend-
ing on the values of the left and right peak half-widths The ob-
servation 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 optimi-
sation 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 ex-
perience 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 4for 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 1collects the char-
acteristics of the plots: the slopes of the left ( m A) and right ( m B)
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 com-
pared with the results obtained with mobile phases of acetonitrile-
water, 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 hydro- organic RPLC with acetonitrile-water This can be explained by the masking effect of the free anionic silanols in the silica-based sta- tionary phases by the ionic additives (SDS and IL)
In the presence of IL, the peaks were significantly more sym- metrical compared to acetonitrile-water mixtures, especially for [C 2C 1IM][PF 6], [C 4C 1IM][PF 6] and [C 6C 1IM][PF 6], in the presence
of SDS, and for [C 4C 1IM][BF 4] and [C 4C 1IM][Cl] without SDS ( B/A = 0.9–1.1, see Table1) The parameters of the half-width plots
in Fig.4should 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.5represents the plots for particular mo- bile 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
Trang 8Fig 6 Experimental chromatograms obtained for mixtures of β-adrenoceptor antagonists with: (a) 0.10 M SDS, 0.01 M [C 6 C 1 IM][PF 6 ], 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 [C 6 C 1 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 4 f from Ref [27] )
3.5 Retention of basic compounds with SDS / IL mobile phases
without organic solvent
We must remember that the purpose of adding 1-butanol is
to help with the stabilisation of MEs, but when an IL is used in-
stead of an apolar organic solvent as octane, the presence of 1-
butanol 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 [C 6C 1IM][PF 6]/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
of 0.114 M SDS/1.14% octane/8.15% 1-butanol, optimised in previ- ous work ( Fig.6b) It was thus evident that the co-surfactant (1- butanol) did not help to achieve the needed chromatographic res- olution for the basic compounds when [C 6C 1IM][PF 6] was added instead of octane Meanwhile, the studies in Section 3.3indicated that the separation was dominated by attraction of the cationic so- lutes to the adsorbed SDS monomers and repulsion from the ad- sorbed 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., attrac- tion 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
Trang 9According to this, a mixture containing only SDS and [C 6C 1IM][Cl]
in aqueous solution was prepared, which gave rise to a clear solu-
tion 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 [C 6C 1IM][Cl] (without or-
ganic solvent) Thus, the retention times for atenolol, carteolol, ace-
butolol, 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 [C 6C 1IM][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 con-
taining 0.10 M SDS and 0.02 M [C 6C 1IM][Cl], without organic sol-
vent The separation suggests that the aqueous mixtures of SDS
and [C 6C 1IM][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
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 co-
surfactant) 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
([C 6C 1IM][PF 6]), and the anionic surfactant SDS (quite unusual for
the preparation of IL/w MEs) was combined with 1-butanol as co-
surfactant The authors developed an analytical procedure for neu-
tral 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 posi-
tively charged, was investigated The research was focused on the
effect on retention times and peak profiles produced by imida-
zolium ILs with alkyl chains of increasing length (with n =2, 4 and
6), associated to Cl –, BF 4–, or PF 6– The research group had previ-
ously developed a detailed work on the interactions of cationic so-
lutes with RPLC C18 columns using mobile phases containing aque-
ous 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 acetoni-
trile 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 reten-
tion to remain constant or increase with a maximum at a particu-
lar 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 [C 4C 1IM][BF 4] and [C 4C 1IM][Cl], in-
dicating an efficient masking of the silanol effect
The formation of transparent and stable MEs prepared with sur- factant (SDS), co-surfactant (1-butanol), and apolar solvent (IL or octane), to be used in RPLC, was found less dependant on the con- centration 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 [C 6C 1IM][PF 6] 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 finan- cial interests or personal relationships that could have appeared to influence the work reported in this paper
Acknowledgements
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 INV- PREDOC16F1–384313
References
[1] J Sjöblom, R Lindbergh, S.E Friberg, Microemulsions: phase equilibria, charac- terization, structures, applications and chemical reactions, Adv Colloid Inter- face Sci 95 (1996) 125–287
[2] C Stubenrauch, Microemulsions: Background, New concepts, Applications, Per- spectives editor, Wiley-Blackwell, Hoboken, NJ, USA, 2009
[3] E Peris-García, N Pankajkumar-Patel, S Carda-Broch, M.J Ruiz-Angel, M.C Gar- cía- 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 Gar- cía-Alvarez-Coque, Modulation of retention and selectivity in oil-in-water mi- croemulsion 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 sus- tainability, 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 Doma ´nska, 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 environ- mental 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 en- hanced 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 tech- niques, 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 selectiv- ity of ionic liquid stationary phases for enhanced separation of nonpolar an- alytes in kerosene using multi-dimensional gas chromatography, Anal Chem
86 (2014) 3717–3721
Trang 10[20] H Xian, H Peng, X Wang, D Long, R Ni, J Chen, S Li, Z Zhang, J Peng, Prepa-
ration 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 chro-
matography, 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 chro-
matography, 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 addi-
tives in reversed-phase liquid chromatography for the analysis of basic com-
pounds, 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 separa- tion 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, Mar- cel 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-Al- varez-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