Self-assembly of a short-chain ionic liquid within deep eutectic solvents

Ionic liquids (ILs) and deep eutectic solvents (DESs) are receiving increased attention from both academic and industrial research due to their immense application potential. These designer solvents are environmentally friendly in nature with tunable physicochemical properties. In the present investigation, we have studied the aggregation behavior of a short-chain IL 1-butyl-3-methylimidazolium octylsulphate [Bmim][OS] within aqueous DESs using fluorescence, UV-vis, dynamic light scattering (DLS) and FT-IR spectroscopic techniques. We have prepared two DESs, ChCl–urea and ChCl–Gly, which are obtained by heating a mixture of an ammonium salt choline chloride with hydrogen bond donor urea or glycerol, respectively, in 1 : 2 molar ratios. The local microenvironment and size of the aggregates are obtained from steady state fluorescence (using pyrene and pyrene-1-carboxaldehyde as polarity probes) and DLS measurements, respectively. DLS results shows that IL [Bmim][OS] forms relatively larger micelles within the aqueous solution of DES ChCl–urea (avg. hydrodynamic radii = 209 nm) than compared to ChCl–Gly (avg. hydrodynamic radii = 135 nm). A significant decrease in the critical micelle concentration and increase in the aggregation number (Nagg) are observed within DES solutions as compared to that in water, thus indicating that the micellization process of the IL [Bmim][OS] is much favored in the DES solutions. Molecular interactions of [Bmim][OS] in DESs are revealed from FT-IR spectroscopic investigation. Furthermore, these systems were applied to study the IL-drug binding of the antidepressant drug promazine hydrochloride (PH).


Introduction
Ionic liquids (ILs) possess unusual physicochemical properties and bright application potential in various elds. 1 Surface active ILs as a novel class of surfactants are of signicant interest to researchers worldwide and have stimulated more signicance due to their self-assembling behavior. [2][3][4][5] It has been reported that these ILs can display surface activity when dissolved in water, denoted by a decrease in the surface tension. [6][7][8][9][10] It is noteworthy that the ILs have analogous properties to surface active agents and are capable of forming micellar nano-aggregates in aqueous solution. 11,12 ILs show impressive physicochemical properties, such as, high thermal stability, high electrical conductivity, low vapor pressure and low melting points, etc. 13,14 Furthermore, deep eutectic solvents (DESs) are emerging as new type of green solvents and analogs of ILs. 15 Indeed, a DES generally comprises of two or three components that self-associate through hydrogen bonding to form a eutectic mixture and possesses melting point below that of the isolated components, low cost, less toxicity, high conductivity, relatively low viscosity, non-ammability, environmentally friendliness and biodegradability. [16][17][18][19] The characteristics of DESs depend on its components, the ammonium salt and the hydrogen bond donors. [20][21][22] The simple structure of short-chain IL based surfactants has produced a signicant deviation in their micellar properties. 23 It is important to have a clear picture on the micellization and interfacial behaviour of short-chain IL based surfactants to concern them effectively in particular elds. [23][24][25] Several methods, such as, electrical conductivity, surface tension, dynamic light scattering (DLS), uorescence, UV-visible and NMR spectroscopic techniques has been successfully utilized to study their micellization and interfacial behavior. [24][25][26][27] Due to their structural exibility and outstanding properties, shortchain IL based surfactant systems have generated immense signicance, which is revealed in the increasing number of applications that have been reported in recent years. [25][26][27] Aggregation behavior of surface active compounds including ILs within DESs have now become growing research interest, the number of publications as dedicated to the use of DESs for this purpose is rapidly increasing. [27][28][29][30][31][32] Zhang et al. 30 have investigated the aggregation behavior of 1-alkyl-3-methylimidazolium chloride with DESs (choline chloride and glycerol in a 1 : 2 molar ratio) by different techniques including uorescence probe response, small angle X-ray scattering (SAXS) and FT-IR spectroscopy. They have presented a clear picture on the critical micellar concentration, micellar size and intermolecular interactions in IL/DES solutions using various spectroscopic techniques. Further, Pandey et al. 31 have studied the self-aggregation of an anionic surfactant sodium dodecyl sulphate (SDS) within DESs. They have used surface tension, DLS and SAXS, density and dynamic viscosity measurements, uorescence probe behavior of pyrene and 1,3-bis(1-pyrenyl)propane to characterize these molecular aggregates. Jackson et al. 22 have investigated the aggregation of alkyltrimethylammonium bromides in choline chloride: glycerol DES by means of surface tension, X-ray and neutron reectivity and small angle neutron scattering. Arnold et al. 32 have investigated the self-assembly of anionic surfactant sodium dodecyl sulfate (SDS) within DESs, choline chloride/urea using X-ray reectivity (XRR), small angle neutron scattering (SANS) and interestingly, the results propose that the micelle formation in DES solutions does not have the similar shape and size as those observed in water.
In the present investigation, we have studied for the rst time, the aggregation behaviour of a short-chain imidazoliumbased IL 1-butyl-3-methylimidazolium octylsulphate [Bmim] [OS] within aqueous solutions of deep eutectic solvents ChClurea and ChCl-Gly, respectively. We have investigated the role of DESs on the micellization process, i.e., critical micelle concentration (cmc), aggregation number, micellar size and polydispersity index (PDI). A detailed comparative study is performed on the aggregation behavior of IL [Bmim][OS] within the aforementioned two DESs solutions using various spectroscopic techniques. Further, these micellar solutions of [Bmim] [OS]-DESs are utilized to study the IL-drug binding of an antidepressant drug promazine hydrochloride (PH).

Materials
1-Butyl-3-methylimidazolium octylsulphate IL, potassium bromide, choline chloride, as ammonium salt and urea, glycerol as hydrogen bond donors were purchased from Sigma Aldrich Pvt. Ltd. with high purity and utilized for the synthesis of DESs without further purication. All the aqueous solutions were prepared using millipore water. Chemical structure of IL 1butyl-3-methylimidazolium octylsulphate, urea, glycerol, pyrene, pyrene-1-carboxyaldehyde, promazine hydrochloride and cetylpyridinium chloride are represented in Scheme 1.

Methods
Fluorescence spectra were performed on ''Cary Eclipse spectrophotometer'' (Agilent Technologies). UV-vis absorption spectra were measured on Cary-60 UV-Vis spectrophotometer (Varian). FT-IR spectra were recorded on Nicolet iS10 spectrometer (Thermo sher) by using KBr pellets. Dynamic light scattering were performed by Malvern Zeta Sizer Nano (Nano Zs 90 UK).

Preparation and characterization of DESs
In this study, ammonium salt choline chloride (ChCl) and two different hydrogen bond donors (HBDs) namely; urea and glycerol were selected to synthesize the DESs, with different compositions. The deep eutectic solvents were synthesized by mixing the choline chloride salt with different HBDs in 1 : 2 mole fraction of salt at 425 K until a homogenous and colorless liquid appeared. In this study, two types of DESs were synthesized in 1 : 2 ratios of quaternary ammonium salts (choline chloride) with hydrogen bond donors (glycerol and urea, respectively). The eutectic mixtures were prepared by stirring of two components at 425 K until a homogeneous transparent liquid was formed. The structures of the synthetic DESs were conrmed by FT-IR spectral results as shown in the Fig. 1 and data are shown as Table 1.
An earlier work of D'Agostino et al. 33 used the pulsed eld gradient (PFG) NMR spectroscopic technique to investigate the self-diffusion of molecular and ionic species in aqueous solution of choline chloride (ChCl) based DESs. From the NMR spectrum of aqueous ChCl-Gly at 13 wt% water content, it is shown that the NMR peak positions are (in ppm): and for ChCl-urea at 1 wt% water content, the NMR peak positions are (in ppm): It is noteworthy that in aqueous ChCl-urea, the amine species of Ch + and HBD show a stronger interaction with water as water is added to the system. Whereas, in the case of ChCl-Gly, water has little effect on both hydroxyl proton diffusion of Ch + and HBD. Furthermore, Mantle et al. 34 have studied the self diffusion coefficients of the liquid mixtures in a noninvasive way at equilibrium which is measured by pulsed eld gradient (PFG)-NMR technique and they have observed that the inter-and intra-dipolar interactions are responsible to origin an enhance effect on the line shapes of the NMR spectrum.

Fluorescence
Steady-state uorescence experiments are carried out using an Agilent Technology spectrouorometer. An excitation wavelength of 334 nm is used and emission spectra are scanned between 350 nm to 450 nm. The excitation slit and emission slit width were kept at 2.5 nm. The concentration of probe pyrene (1.2 Â 10 À4 M) and pyrene-1-carboxyaldehyde (1.

Dynamic light scattering
The size of amphiphilic micelle was observed by means of dynamic light scattering (DLS) method. DLS measurements were performed with Malvern Zeta Sizer Nano and intensity of the scattered light was maintained at 90 and temperature at 298 K.

Fourier transform infrared spectroscopy
FT-IR spectra of IL 1-butyl-3-methylimidazolium octylsulphate and mixture of choline chloride, urea, glycerol, DESs were obtained using a Nicolet iS10 (Thermo Fisher Scientic Instrument, Nadison, USA) spectrophotometer. All IR spectra were achieved by standard 32 examines at 4 cm À1 declaration more the spectral range of 4000-400 cm À1 . Deep eutectic solvent (DES) mixture with ionic liquid (IL) 1-butyl-3methylimidazolium octyl sulphate was delivered over 0.1 g pre-weighed nely ground IR grade KBr for DRS-FTIR scan. The KBr was dried around 100 C, for 5-10 minutes, prior to spectral Scheme 1 Structures of IL 1-butyl-3-methylimidazolium octylsulphate, choline chloride, glycerol, urea, promazine hydrochloride, cetylpyridinium chloride, pyrene and pyrene-1-carboxyaldehyde. scan to remove water aberration. The FTIR was purged for 30 minutes with >99.99% analytical grade nitrogen gas using external purge kit (iS10 iZ10 model, Thermo Fisher Scientic), to minimize atmospheric interferences. The dried KBr was then lled over the sample cup and analyte was carefully delivered over it. Diffuse reectance accessory with IL/DES/KBr beam splitter and deuterated, L-alanine doped triglycinesulphate (DLaTGS) detector was employed in the present work. The soware OMNIC 9.1, automatically performs the spectral scaling and the resultant absorptions. The instrument was calibrated as all spectra were obtained by averaging 32 scans at 4 cm À1 resolution over the spectral range of 4000-400 cm À1 using the auto gain function and slit set at 100 without ATR/DRS modication for wavelength dependence.

UV-visible spectrophotometer
UV-vis absorption spectra were measured using a Varian Cary Eclipsed-60 spectrophotometer. The absorption spectra of the aqueous solution of sort-chain IL + pyrene + DESs mixtures were collected against the reference solutions at 300 K temperature. Pyrene was used as the probe with a concentration of 0.002 (mol dm À3 ) in all experiments.

Determinations of critical micelle concentration (CMC)
(I) UV-visible spectroscopy. The UV-vis absorption spectroscopy is a simple and accurate technique to determine the cmc of various amphiphilic molecules. Pyrene is used as UV probe in our studies. When the concentration of pyrene is changed, the absorption peak heights changed largely as shown in Fig. 2 2B) and in the presence of two types of DESs (Fig. 2B). The pre-micellization red-shis can be attributed to the formation of the IL aggregates just below the cmc. The interaction between these pre-micelles and pyrene result in the red-shis. Usually, the cmc is determined based on the UV absorption peak, where the centre of the sigmoid is regarded as the cmc. The cmc is dened as the concentration of amphiphilic molecules above which micelles formation takes place and all additional amphiphilic molecules added to the system go to the micelles. Aer reaching the cmc, there results in a drastic change in the physicochemical properties of the surfactant solution. The cmc is an important characteristic of a amphiphilic molecule. In general, a typical plot of absorbance versus  Table 1. Table  1, clearly shows the cmc of pure IL in water is larger than compared to their values within aqueous mixture of 5 wt% DESs. The results clearly show that the cmc of IL [Bmim][OS] within ChCl-urea DES solution is less than in ChCl-Gly solution.
The formation of the [Bmim][OS] cumulative at the concentration below the cmc, we examined the pre-micellization red shis. The UV spectra reveal that the red shis for the strongest UV peak (P2) occur only at the IL concentration 0.1 M. This strong peak is approved to pyrenes that are located at the palisade layer of the micelles and as a result, the red shi of this peak is certied to the close interactions between the IL hydrophilic groups and the p electron clouds of pyrene. As a result, the wavelength l max red shi indicates the formation of micelles. Thus, the ChCl-Gly micelles apply much stronger interactions with the pyrene p electrons than the ChCl-urea micelle.
(II) Fluorescence spectroscopy. Fluorescence probes are usually used to achieve various micellar characteristics, such as cmc, aggregation number (N agg ), size and shape, among others. The behavior of a uorescence probe in a micellar solution depends on the properties of the micellar solution (e.g., nature of the bulk solvent, properties of the micelles, nature of amphiphilic molecules) as well as on the molecular structure of the uorophore. We have used two uorescence probes, pyrene and PyCHO to obtain information on DESs added aqueous [OS] is higher in water than compared to that in aqueous solution of 5 wt% DESs (ChCl-urea > ChCl-Gly) and the values are in good agreement with those achieved by uorescence (pyrene as probe) methods. The cmc of [Bmim][OS] in the aqueous solution and in DESs obtained is given in Table 2.
(B) Behavior of PyCHO. PyCHO is used as a uorescence probe to study the micellization behavior of ILs. PyCHO probe shows distinctive structural (Scheme 1) difference that has found usefulness in studies of the solution and interfacial polarity. The uorescence spectra were recorded keeping xed concentration of PyCHO (4.1 Â 10 À7 mol L À1 ). PyCHO probe has two types of closely-lying excited singlet states (n-p* and p-p*), both of which show emission in solution. In nonpolar solvents, the emission from PyCHO is highly structured and weak arising from the n-p* state. Change of the dipolarity from non-polar to polar medium the p-p* state is brought below the n-p* state through, solvent relaxation to become the emitting state. This is obvious by a broad reasonably instance emission that red shi with increasing solvent dielectric.
PyCHO uorescence spectra are collected in the presence of varying amount of IL [Bmim][OS] in aqueous 5 wt% DESs solutions. Fig. 4 (S1) shows a hypsochromic shi in l max from pre to the post-micellar region for each addition of signifying, as expected, the increased hydrophobicity of the cybotactic region of the average probe upon micelle formation. Sigmoidal ts to the data are presented using board lines. Leaning among the curves clearly imply early onset of micelle formation in the presence of 5 wt% DESs. It is suggested by our data that PyCHO uorescence intensity may be more sensitive to the changes in the probe cybotactic region than l max . Again as earlier, cmc could be evaluated from the sigmoidal nature of the changes and they are found to be statistically similar to those obtained from pyrene I I /I III .

Determination of micellar aggregation number
For the determination of aggregation number (N agg ) of micelles, we have employed steady-state uorescence quenching method.    (1) and these values are listed in Table 3. It can be seen that the N agg of [Bmim][OS] decreased on going to ChCl-Gly and ChCl-urea DESs.
The strength of the hydrophobic environment of short-chain based IL can be estimated by the Stern-Volmer quenching constant (K sv ) was calculated using the following eqn (2); The Stern-Volmer quenching constant (K sv ) can be probable from the reached slope values of the plots ln F 0 /F Q versus [CPC]. The calculated K sv values are illustrated in Table 3. K sv are explained the hydrophobicity of micellar solutions and its utilized to decrease the uorophore. Table 3, are clearly show the K sv value of pure IL are lesser compared to mixed of IL-DESs (ChCl-urea > ChCl-Gly). The result shows that the broad behavior of the micellar aggregation of [Bmim] [OS] in ChClurea is comparable to that in water, i.e. the micellization of [Bmim] [OS] in ChCl-urea is mostly determined by the solvophobic effect, similar to the micellization of IL molecules in water caused by the hydrophobic effect. 32

Particle size from dynamic light scattering
Dynamic light scattering (DLS) technique was employed to investigate the micro structural changes taken place within the IL-added DESs-based systems. Successive measurements were made within a cell of 3 mL for normalization analysis. The average hydrodynamic radius distribution of the DESs solution (0.1 g mL À1 ), [Bmim][OS] solution (0.2 mg mL À1 ) and DES-rich phase aer extraction (diluted 5 times) are shown in Fig. 6. DLS was useful technique to substantiate the evidence of aggregate formation in aqueous imidazolium based IL [Bmim] [OS] in presence and absence of 5 wt% aqueous DESs and also to study the variation of size of aggregates in the system. Fig. 6 shows the scattering intensity for the given diameter (D), PDI measured at room temperature of aqueous [Bmim][OS] and in the presence of DESs.
The bimodal distribution and PDI is observed and shown in Table 3

Fourier transform infrared (FTIR) spectroscopy
Infrared spectroscopy is a standard analytical tool for assessing liquid structures. The intra-molecular vibrational modes of the ions composing the materials are oen quite sensitive to their local potential energy environment. FT-IR spectroscopy is a characteristic technique to analyze the strength of Hydrogen bond interactions and identify the structure of DESs between hydrogen bond donor (HBD) and hydrogen bond acceptor (HBA). In the present study, FT-IR spectral response is used to achieve the hydrogen bond interactions taking place within ILadded DES solutions based on the spectral shis. The FT-IR spectra of the [Bmim][OS]/ChCl-Gly/ChCl-urea micellar solutions are shown in Fig. 7. The FT-IR spectra of 1-butyl-3-methylimidazolium octylsulphate (symmetric and asymmetric stretching CH 2 vibration of alkyl chains) at 2856. 19  [OS] molecules distributed in the aggregates closely mutually and supports the dissociation of head group counter ions in the surface of aggregates, resulting in a closer arrangement of micelles.

Ionic liquid-drug binding of promazine hydrochloride
UV-visible absorption spectroscopic technique is a constructive tool to probe IL-drug binding. 35 In the absorption spectra of drug, absorbance at 300 nm increases upon addition of micellar solution ([Bmim][OS] (100 mM)-ChCl-Gly/ChCl-urea) (Fig. 8). The peak positioned around 300 nm is red shied. Based on the peak shi and increase in absorbance, it can be concluded that all mixture can form complex with the PH drug since ChCl-Gly and ChCl-urea have almost no absorption band through the wavelength range (300-600 nm) (Fig. 8). The absorption band for the promazine hydrochloride (10 mM) drug was observed at l max ¼ 300 nm.
The binding constants for drug-IL complexes were estimated from Benesi-Hildebrand (B-H) equation. The change in absorbance is depended on the concentration of drug, according to the following eqn (3),  (Fig. 8), which further indicates the formation of 1 : 1 complex between drug (PH) and IL representation on Scheme 2. The values of the binding constant obtained from the intercept-to-slope ratio of the Bensei-Hildebrand plot ( Fig. 8 and S2 †) for drug/IL complexes show that the ChCl-urea (6 mol dm À1 ) shows more binding affinity towards PH than ChCl-Gly (5 mol dm À1 ) as they readily interact and form hydrogen bond with water. Owing to substituted hydroxyl group, it reduces the unfavorable interactions.

Conclusions
We  [OS] in aqueous DESs and the intermolecular hydrogen-bond interaction plays a positive role to promote the micellization process. PH drug shows more binding affinity and most capable action is shown by ChCl-urea over ChCl-Gly. The present work clearly shows the tendency to form self assembled nanostructures by short-chain imidazolium IL within aqueous DESs and this would serve for potential application of IL-and DES-based systems in drug delivery, aggregation, colloidal systems and novel ways for researchers to explore new ndings.

Conflicts of interest
There are no conicts to declare.