The confinement and anion type effect on the physicochemical properties of ionic liquid/halloysite nanoclay ionogels

2.1 Materials

In this work, we used the following commercial reagents without further purification:

- Halloysite nanoclay, Hal (Sigma-Aldrich, USA, CAS Number 1332–58-7, molecular weight 294.19 g/cm³);

Ionic liquids:

• -

  1-butyl-3-methylimidazolium dicyanamide, BMImDCA (Sigma-Aldrich, Hong Kong, China. CAS Number: 448245–52-1, purity ≥ 97%, water impurity content 0.12 wt%);

• -

  1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide, BMImTFSI (Sigma-Aldrich, USA, CAS Number: 174899–83-3, for synthesis, water impurity content 0.03 wt%);

• -

  1-butyl-3-methylimidazolium trifluoromethanesulfonate, BMImOtf (Sigma-Aldrich, USA, CAS Number: 74899–66-2, for synthesis, water impurity content 0.07 wt%).

The water impurity content in the samples under study was determined by K. Fischer's method.

Stable IL-halloysite nanoclay ionogels were obtained by vibration (IKA VORTEX 4 basic vibration shaker, IKA-Werke GmbH & Co. KG, Germany) and ultrasonic (CT-431D2 ultrasonic cleaner, CTbrand Wahluen Electronic TOOL Co. Ltd., China) mixing of components in hermetically sealed capsules. The obtained mixtures had 2:1 IL: clay molar ratios which corresponded to 74.1 wt% BMImTFSI, 66.2 wt% BMImOtf, and 60.4 wt% BMImDCA in the ionogels.

2.2 Methods and apparatus

The visualization of the clay mineral and ionic liquid-clay ionogels was made by the scanning electron microscopy method using an Vega 3 SBH- TESCAN microscope (TESCAN, the Czech Republic) with an energy-dispersive X-ray spectroscopy (EDX) detector. The samples were loaded onto a carbon tape for testing. Then the samples were coated with a 1 nm thick carbon layer.

A Quanta Chrome Nova 1200 surface area analyzer (Quantachrome Corporation, Boynton Beach, FL, USA) was used to measure the volume of the adsorbed gas by the Brunauer-Emmett-Teller (BET) method in standard conditions of (STP: T = 273.15 K, 101.3 kPa), represented by adsorbed gas volume per 1 g of sample. Before testing, the halloysite samples had been kept in a vacuum at 105 °C for 7 h. The isotherm was an IUPAC IV type, with pore sizes in the mesoporous and macroporous range.

The halloysite particle size distribution and Z-potential measurements were carried out in distilled water at room temperature by a particle size and Z-potential analyzer Zetasizer Nano (Malvern Instruments Ltd., UK). Before the tests were made, the clay suspension in distilled water was sonicated and kept for 24 h.

The thermal stability of the clay and ionogel samples was determined with a TG 209 F1 thermal analyzer (NETZSCH, Germany). The samples weighing ∼ 10 mg were placed into a platinum capsule and heated in an argon flow (30 mL min⁻¹) at a rate of 10 °C min⁻¹ to a temperature of 800 °C. The accuracy of the mass and temperature measurement was ± 10⁻⁶ g and ± 0.1 °C, respectively. Each sample was scanned three times.

The phase behavior of the pure ionic liquids and ionogels (melting, crystallization, and glass transition) was studied with a DSC 204 F1 Phoenix calorimeter (NETZSCH, Germany). A sample of ∼ 10 mg was placed into a hermetically sealed platinum pan, cooled to − 120 °C with liquid nitrogen at a rate of 10 °C min⁻¹ and then heated to 200 °C at a rate of 10 °C min⁻¹. The measurements were made in an argon atmosphere. The accuracy of the mass and temperature measurement was ± 10⁻⁵ g and ± 0.1 °C, respectively.

The spectrophotometric measurements were made in a VERTEX 80v infrared Fourier-Transform spectrometer (Bruker, Germany). The FT-IR reflection spectra were recorded with a diamond crystal at room temperature in the region from 400 to 4000 cm⁻¹, with the spectral resolution equal to 2 cm⁻¹.

The specific conductivity (κ) of the considered ionic liquids and ionogels based on them was determined with a Solartron SI 1260A Impedance/Gain-Phase analyzer (Solartron Analytical, UK). The measurement was made with flat platinum electrodes in a designed hermetic cell. The alternating current frequency ranged from 1 to 100 kHz, the cell voltage equalled 10 mV. A 0.01 mol·dm⁻³ KCl solution was used to determine the cell constant value. The measurements were made in the range from − 20 to 80 °C. The cell temperature was maintained with an accuracy of ± 0.2 °C with the help of a LIOP FT-316–40 liquid cryothermostat (LOIP, Russia).

The X-ray diffraction studies of the crystal structure of the Hal powder and Hal samples modified by an ionic liquid were conducted in the angle range of 2Θ = 5–50° with a DRON-UM1 (Russia) diffractometer (CuK radiation = 0.154 nm) operating at a voltage of 40 kV and a current of 40 mA.

The electrochemical window was measured with a PI 50-Pro-3 pulse potentiostat with automatic data registration and processing in the PS Pack 2 programs (Elins, Russia) on a platinum electrode in a hermetically sealed three-electrode cell at a temperature of 25 °C. A platinum wire was used as the quasi-reference electrode, and a platinum sheet - as the counter electrode.

3.1 Structural characteristics of halloysite and its mixtures with ILs

Halloysite nanoclay (Al₂Si₂O₅(OH)₄·2H₂O), a natural layered alumosilicate, was used to reinforce the ionic liquids under study and obtain ionogels. The distinguishing structural feature of halloysite particles (hydrated kaolinite form) is the tubular shape of the alumosilicate sheets, with water of hydration and kaolinite OH-groups between them (Joussein et al., 2005; Kohyama et al., 1978; Yuan et al., 2012). The length of the nanoscrolls in the clay samples under study can exceed 1 μm with the outer diameter up to 0.1 μm (Fig. 1a). The results of the Hal particle size distribution are presented in Fig. 1(b). The experiment showed that the low-temperature adsorption–desorption isotherms of N₂ can be classified as type IV isotherms according to the IUPAC nomenclature (Fig. 1c) and are characteristic of mesoporous bodies. The total surface area (BET surface area) ∼ 57 m² g⁻¹ (∼16769 m² mol⁻¹). The pore size distribution is shown in Fig. 1d, which indicates that the diameter of most of the mesopores ∼ 10 nm and corresponds to the inner diameter of the nanoscrolls (Yu et al., 2016). The elemental composition of the Halloysite nanoclay is given in Table 1.

Close

Fig. 1

The SEM-image of Hal (a) and Hal particle size distribution (b); the isotherm of low-temperature nitrogen adsorption-desorption on Hal particles (c); the BJH pore size distribution for Hal (d).

Export to PPT

The inner surface of the kaolinite sheets in the nanoscrolls is positively charged, while the outer one is charged negatively (Yelleswarapu et al., 2010; White et al., 2012). The Z-potential of the investigated halloysite particles in distilled water is −13 mV.

Halloysite is a hydrophilic substance, and the hydrophilicity of an IL can affect the properties of the obtained ionogels and the clay structure. The role of the pore space, and halloysite hydrate moisture on the interaction with ionic liquids remains practically unknown. The existing concepts of the mechanisms of adsorption of organic molecules and ions on clays obtained at low loads on adsorbents (Droge and Goss, 2013a, 2013b) can reflect some stages of ionogel formation, where the adsorbat amount is very high, only with a certain degree of approximation

The hydrophilic character of ILs is known to depend mainly on the anion nature (Tomšík and Gospodinova, 2016; Huddleston et al., 2001). Based on our experiment results, the enumerated ionic liquids hygroscopicity corresponds to the series: DCA^– > Otf ^– > TFSI^–. The equilibrium moisture content in the ILs at 22 ± 0.5 °C and 60% relative air humidity in this case is 6.59, 1.59, and 0.18 wt%, respectively.

The ionic liquids used to prepare the ionogels have identical cations but different anions. Table 2 shows that the ions in the ILs under study have considerable differences in their sizes. When an IL is mixed with halloysite, the ions can easily penetrate and move inside the internal cylindrical cavity of the halloysite nanoscrolls and fill the macropore space.

Table 2 Ion volumes and characteristics of ILs with a BMIm⁺ cation.

Ionic liquid ion
Ion volume, nm3	0.150 (Tokuda et al., 2006)	0.232 (Slattery et al., 2007)	0.130 (Slattery et al., 2007)	0.056 (Slattery et al., 2007)
	0.198 (Marcus, 2015)	0.246 (Bica et al., 2013)	0.128 (Bica et al., 2013)	0.069 (Bica et al., 2013)
	0.199 (Slattery et al., 2007)	0.230 (Beichel et al., 2014)	0.126 (Beichel et al., 2014)	0.089 (Beichel et al., 2014)
	0.211 (Bica et al., 2013)
M*, g/mol		419.37	288.29	206.26
ρ*, g/cм3 (20 °C)		1.4409	1.3012	1.0621
Vm*, cм3/mol, (20 °C)		291.05	221.56	194.20

* Characteristics of the salts containing BMIm⁺ and a corresponding anion.

At the same time, large anions and cations of the ionic liquid cannot penetrate into the interlayer space of the considered HNTs. Fig. 2 (curve 1) shows the diffractogram of the powdered halloysite sample that was used in this work. The inter-turn spacing in the Hal scrolls under study equals 7.55 Å. The completely hydrated halloysite form is known to have a basal spacing of ∼ 10.1 Å (4 H₂O), while the partially dehydrated one contains 0.5–1.5 H₂O molecules, consists of chaotically arranged hydrated and unhydrated layers and has a basal spacing of 7.36–7.9 Å (Fig. 3a) (Grim, 1968). The results of the X-ray diffraction powder analysis of the halloysite under study describing the structural state of the partially dehydrated form of tubular halloysite - the so-called HNTs-7 Å) (Drits et al., 2018). The inter-turn spacing becomes smaller in the presence of BMImTFSI, BMImOtf, and BMImDCA and equals 7.47 Å, 7.42 Å and 7.41 Å, respectively. The hydrophilic anions, especially DCA^–, facilitate the removal of the residual water molecules from the inter-layer space (to the IL bulk), which makes them smaller (Fig. 3b).

Close

Fig. 2

X-ray diffraction (XRD) patterns of the Halloysite nanoclay (1) and the ionic liquids based on BMImTFSI (2), BMImOtf (3), and BMImDCA (4).

Export to PPT

Close

Fig. 3

Changes in the structure of (a) halloysite (4 H₂O) due to partial dehydration upon drying and (b) halloysite (0.5–1.5 H₂O) due to dehydration upon contact with hydrophilic IL.

Export to PPT

The SEM-images of the ionogels, Fig. 4, show that the ionic liquids are not adsorbed by the clay completely. Besides, it indicates a noticeable destruction of the HNTs in the hydrophilic ILs, especially in BMImDCA (Fig. 4c), which may be caused by their more complete dehydration than in the initial halloysite.

Close

Fig. 4

Scanning electron microscope (SEM) images of the ionogels containing BMImTFSI (a), BMImOtf (b), BMImDCA (c), and ionogel in container (d).

Export to PPT

3.2 FT-IR spectra

To obtain more information on the behavior and interaction of ionic liquids entrapped by halloysite, we examined the infrared spectra (4000–400 cm⁻¹) of the obtained ionogels (Hal–BMIm[X], X - DCA⁻, Otf ⁻ and TFSI⁻) in comparision with the spectra on the initial compounds.

Fig. 5a-c shows FT-IR spectra in the range 3800–2800 cm⁻¹, where the main bands of halloysite and ionic liquids are highlighted and assigned according to (Kloprogge, 2016; Yuan, 2016; Cheng et al., 2010; Moschovi et al., 2012; Vitucci et al., 2014; Dhumal et al., 2014). The peak positions are also indicated. The frequencies marked with an asterisk were determined by the fitting method (Ramenskaya et al., 2020).

Close

Fig. 5

FT-IR spectra of the pure Hal (blue line), pure ionic liquid BMIm[X] (red line) and Hal/BMIm[X] (black line) ionogels in the ranges of OH and CH stretching modes (a-c) and CN stretching modes (d), X: DCA⁻ (a), Otf ⁻ (b) and TFSI⁻ (c); (e) histogram of shifts of stretching modes, Δνⁱ = νⁱ (pure Hal or ILs) – νⁱ (ionogel); * - the peak positions was determined by fitting. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Export to PPT

The initial halloysite shows two distinct peaks at 3697 and 3622 cm⁻¹ attributed to the stretching vibration from the inner OH-groups (νSi–OH). The wide band centered on 3480 cm⁻¹ is due to the OH stretching modes (νOH) of the inter-layer and surface-absorbed H₂O. The spectra of the pure ionic liquids show several bands between 3000 and 3200 cm⁻¹ assigned to the aromatic C–H stretching modes νC2–H and νC4,5–H on the imidazolium ring. The band at a lower frequency corresponds to the C2–H bond due to the more positive charge on the C2 atom than on the C4 and C5 atoms of the imidazolium ring. The peaks between 2800 and 3000 cm⁻¹ are attributed to the C–H stretching modes of the aliphatic chain (νCH_2.3). The aromatic stretching modes of BMImDCA appear in lower frequency range compared to BMImOtf and BMImTFSI due to stronger electrostatic and hydrogen interactions of the cation and anion. The “acid” H atom at C2 is more favorable for the formation of hydrogen bonds, leading to the red shift of the νC2–H mode. As follows from Fig. 5a-c, the hydrogen interaction force, proportional to this shift, decreases in the following order: DCA⁻ > Otf⁻ > TFSI⁻.

The infrared spectra of ionogels include bands of both ionic liquids and halloysite which have a number of differences from those in the initial components. The observed shifts of the frequency modes are shown in Fig. 5e. As can be seen, the stretching mode of absorbed H₂O exhibits multidirectional shifts of different magnitudes depending on the type of anion. In addition, the relative intensity of the OH (h_OH/h_SiOH) band in ionogels, compared to pure Hal, decreases by a factor of 2 for TFSI⁻ and increases by 2 and 6 times for Otf⁻ and DCA⁻, respectively. Since both the frequency and spectral intensity of the νOH mode depend on the IL – H₂O interaction, the water molecules in the DCA ionogel are more bound to the ionic liquid than the Otf and TFSI ionogels (Cammarata et al., 2001; Ramenskaya et al., 2018). All ionic liquids exhibit insignificant blue shifts (1–3 cm⁻¹) for CH stretching modes due to interaction with clay. However, in the Hal–BMImDCA ionogel, a red shift about 2 cm⁻¹ was found for the νC2–H mode, which can be explained by the interaction of the BMIm⁺ cation with the DCA⁻ anion due to the confinement effect (Ramenskaya and Grishina, 2016; Singh et al., 2010).

Fig. 6a-d shows FT-IR spectra of the initial compounds and ionogels in the range 1800–400 cm⁻¹. Typical groups of halloysite bands are highlighted; the bands corresponding to the anion or cation are indicated by arrows. The peak positions are also indicated, the frequencies marked with an asterisk were determined by the fitting method.

Close

Fig. 6

FT-IR spectra (1800–400 cm⁻¹) of the pure Hal (blue line), pure ionic liquid BMIm[X] (red line) and Hal/BMIm[X](black line) ionogels, X: DCA⁻ (a), Otf ⁻ (b) and TFSI⁻ (c); (e) histogram of shifts of stretching (bending) modes, Δνⁱ(δ)ⁱ = νⁱ(δ)ⁱ (pure Hal or ILs) – νⁱ(δ)ⁱ (ionogel).

Export to PPT

The original halloysite shows three peaks at 1090, 1034, and 912 cm⁻¹ related to the Si–O and Si–O–Si stretching modes and the Al–OH bending mode, respectively; several peaks below 800 cm⁻¹ are also typical of OH translation vibrations of the pure halloysite OH units (Kloprogge, 2016; Yuan, 2016; Cheng et al., 2010). The low-frequency range of the BMImOtf and BMImTFSI ionic liquids mainly reflects the anion spectral features, such as ν(SO₂) at 1347, 1329, and 1123 cm⁻¹ from TFSI^–, ν(SO₃) at 1251 and 1027 cm⁻¹ from Otf ⁻, ν(CF₃) from TFSI⁻ at 1127 and 1276 cm⁻¹, and from Otf ⁻ at 1223 and 1150 cm⁻¹ (Moschovi et al., 2012; Vitucci et al., 2014; Dhumal et al., 2014). The DCA⁻ anion absorbs between 2300 and 2000 cm⁻¹ (Fig. 5d), therefore, in the low-frequency range, stretching and bending modes of CH bonds from various groups of the BMIm⁺ cation can be observed (Grishina et al., 2019). In this range, the infrared spectra of ionogels are very similar to the spectra of pure ILs and also have halloysite bands. The observed shifts of the frequency mode are shown in Fig. 6e.

The DCA⁻ anion shows noticeable blue shifts: (i) up to 4 cm⁻¹ for the stretching mode νN≡C (Fig. 5d), apparently caused by the interaction of the anion with both halloysite and water; (ii) up to 5 cm⁻¹ for the bending mode δ_as^ip ring (Fig. 6b), caused by the enhanced stacking of imidazolium rings, due to the confinement effect (Dupont, 2004). No changes are observed for the bands from the Otf⁻ and TFSI⁻ anions, such as ν_as(SO₃) (Fig. 6c), ν_as(SO₂) and ν_as(S-N-S) (Fig. 6d), which are usually sensitive to the confinement effect (Fujii et al., 2008; Ferreyra et al., 2012; Blach et al., 2013). Since the confinement effect depends on the pore size, the Otf and TFSI ionic liquids appear to be predominantly located inside the halloysite macropores, where the confinement effect is not observed (Singh et al., 2010). At the same time, the blue shifts about 2 cm⁻¹ in the asymmetric stretching modes ν_as(CF₃) indicates the interaction of these anions with halloysite. All halloysite modes exhibit significant red shifts up to 13 cm⁻¹, which are cased not only by the interaction with ionic liquid, but also with water molecules. It is clear that hydrophilic and hydrophobic anion of the ionic liquid has different effect on the processes of hydration and dehydration of the active groups of halloysite.

Thus, the type of anion affects the interaction of water molecules of halloysite with both the ionic liquid and the functional groups of halloysite. This ability decreases in accordance with an increase in hydrophobisity and anion size in the following order: DCA^– > Otf ^– ≫ TFSI⁻. When ionic liquids are captured by halloysite, the hydrophilic DCA^– anion can easy interacts with water molecules. In the presence of the hydrophobic Otf ^– and TFSI⁻ anions, water molecules interact predominantly with groups such as Si–O, Si–O–Si and Al–OH. Both the cation and the anion interact with halloysite due to electrostatic forces, which leads to a weakening of ionic interactions (Grishina et al., 2019; Zhang et al., 2017). The confinement effect is observed for the BMImDCA ionic liquid due to the small size of the DCA^– anion, which facilities the penetration of ionic liquid molecule into the small pores of halloysite. The BMImOtf and BMIMTFSI ionic liquids, due to their large size, are mainly located inside the macropores of halloysite.

3.3 TG and DSC studies

Fig. 7 shows the DTG curves for the samples of pure halloysite and ionic liquids, as well as the Hal-BMImX ionogels. The peaks observed in region (I) are associated with the loss of adsorbed water; those observed in region (II) are caused by halloysite dehydroxylation and destruction of the ionic liquids (Yuan, 2016; Ngo et al., 2000). The destruction of the bulk and confined BMImDCA ionic liquid by halloysite is a three-step weight-loss process (Grishina et al., 2019). Unlike the one-step decomposition of the bulk BMImOtf and BMImTFSI, the weight-loss in the confined ILs is a two-step process. The values of the peak decomposition temperatures (T_d) and mass loss (Δm) under heating are shown in Table 3.

Close

Fig. 7

TG (a) and DTG (b) curves of the initial Hal (1), initial BMImX (2, 4, 6) and Hal-BMImX (3, 5, 7), X: DCA⁻ (2, 3), Otf ⁻ (4, 5), and TFSI⁻ (6, 7) nanocomposites.

Export to PPT

In region (II), all the ionic liquids confined by halloysite show a decrease in the peak temperatures compared to the bulk salt and the original clay. The depression of the decomposition temperature (ΔT_d = T_d(bulk) − T_d(pore)) depends on the anion, for example, ΔT_d of the first peak changes in the following order: BMImTFSI (94 °C) > BMImOtf (31 °C) > BMImDCA (14 °C). In region I, the DTG trace of the Hal-BMImDCA and Hal-BMImOtf ionogels show multidirectional shifts in the dehydration temperature in comparison with the original components, and an increase in the dehydration temperature of the Hal-BMImDCA ionogel is associated with the interaction of the DCA⁻ anion with water. The mass loss observed during the dehydration of the initial components and the resulting composites suggests that the ionic liquids entrapped by halloysite replace some of the free water in the host-clay, and this ability decreases in the following order: BMImDCA < BMImOtf ≪ BMImTFSI. Consequently, the thermal stability of the BMIm⁺-based ionic liquid entrapped by halloysite decreases due to the confinement and the effect of the interaction with both the clay wall and the water molecules absorbed by the halloysite.

Fig. 8 shows DSC curves of the first heating/cooling cycles of the samples of bulk and confined BMImX ionic liquids in halloysite. The temperatures of the observed phase transitions, such as melting (T_m), glass transition (T_g), and crystallization (T_c) or cold crystallization (T_cc) for the first and second cycles, are presented in Table 3.

Close

Fig. 8

DSC curves of bulk BMImX (1, 2, 5, 6, 9, 10) and BMImX confined in halloysite (3, 4, 7, 8, 11, 12), X⁻: DCA⁻ (1–4), Otf ⁻ (5–8), and TFSI⁻ (9–12); the solid line shows the first heating cycle, the dashed line - the cooling cycle.

Export to PPT

The DSC curves of the confined and bulk ionic liquids seem rather similar, but have some differences. Unlike BMImTFSI, confined BMImDCA shows a slight decrease in T_g caused by further amorphisation in the presence of free halloysite water. The evident difference between the confined and bulk ionic liquids is the change in the freezing and melting points. For example, a noticeable increase in T_c by 14 °C and in T_cc by 4 and 18 °C was observed in the BMImOtf and BMImTFSI ionic liquids, respectively, in both scanning cycles. At the same time, T_m shows a minor decrease (∼2 °C) for BMImTFSI and a multidirectional change (ΔT_m = T_m(bulk) − T_m(pore) = 7 and − 4 °C in the first and second heating cycles, respectively) for the more hydrophilic BMImOtf. It should be noted that the melting point became lower in most of the silica-confined ILs, which was explained by the simultaneous effects of the strong interaction of the ILs with the silica wall and confinement (Néouze et al., 2006; Kanakubo et al., 2006; Liu et al., 2010). A melting point increase was observed in the ILs included in carbon materials, such as CNT (Im et al., 2012; Chen et al., 2007). This was largely due to the structural transformation of the chaotically distributed ions in the bulk phase into an ordered arrangement inside the CNT channel, which led to the anomalous phase transition from the liquid to crystallites with a high melting point (Zhang et al., 2017).

Thus, the imidazolium ILs confined in a halloysite matrix undergo thermal decomposition more easily than the bulk ILs and the process is controlled by the type of paired anions. The anomalous phase and thermal behavior is due to the change in the local ionic arrangement and interaction with clay and free water.

3.4 Ionic conductivity

Ionic liquids modified by inorganic fillers have the properties of solid electrolytes retaining high electric conductivity and a wide electrochemical stability window (Deb and Bhattacharya, 2017; Shimano et al., 2007; Horowitz et al., 2016; Li et al., 2016). ILs fill the nanopores, penetrate the interlayer spaces, and get adsorbed on the surface of porous materials but the emerging confinement effects can change the IL properties preserving the liquid-like dynamics and mobility of the ions (Lee et al., 2009).

The structure and size of the anion are among the factors significantly affecting the electric conductivity (κ) of ionic liquids. For example, work (Leys et al., 2010) shows that the electric conductivity of 1-butyl-3-methylimidazolium salts with Cl⁻, Br⁻, I⁻, [NCS]⁻, [N(CN)₂]⁻, and [BF₄]⁻ −anions increases with the anion size increasing. This regularity, however, does not work in the anion group considered by us, and ILs with the smallest anion volume have the highest ionic conductivity. Fig. 9 shows the temperature dependences of specific electric conductivity of the studied ionic liquids and ionogels based on them in the Arrhenius plots. A temperature decrease leads to a noticeable deviation of this dependence from linearity, which is typical of glass-forming substances (Wasserscheid and Welton, 2002). It is evident that the temperature behaviour of the conductivity of the ILs and ionogels is similar, and it can be characterized by the Vogel-Fulcher-Tammann (VFT) equation related to the ordinary en masse diffusion (vehicle mechanism) (Wasserscheid and Welton, 2002; Garaga et al., 2017):

Close

Fig. 9

Dependence of log κ (1/T) (Arrhenius plot) of the ILs (1) and Hal-IL mixtures (2): a - with BMImTFSI; b - with BMImOtf; c - with BMImDCA.

Export to PPT

κ = κ₀ exp [-k_κ/(T-T₀)],

where κ₀ is the limiting conductivity, k_κ is a constant related to the Arrhenius activation energy, T₀ is the ideal glass transition temperature, which is usually several tens of degrees lower than Tg measured by DSC (Table 3). The results of the calculation of the VFT equation parameters for the considered systems are given in Table 4. Interestingly, the patterns of changes in the VFT equation parameters of the hydrophobic BMImTFSI and hydrophilic ILs into gel are opposite to each other.

Since the clay particles are uniformly distributed in the ionic liquid, the relation of the electric conductivity of the obtained mixture with that of the pure electrolyte looks as follows (Dukhin, 1975): F = κ_IL/κ_IG, where F is the structural resistance of the mineral filler, κ_IL and κ_IG are the specific electric conductivities of the bulk IL and the ionogel, respectively. Fig. 10 shows temperature dependences of the F value that are radically different for the ILs with high and low hydrophilicity. In the first case, the structural resistance produced by the halloysite filler is several times higher than unity in the region of negative temperatures and gradually decreases when the temperature becomes higher. In salts with higher hydrophilicity (BMImOtf and BMImDCA), the F(T) dependences have the opposite character. Moreover, the ionic conductivity of the hybrid material with BMImDCA is higher in the region of negative temperatures than that of the bulk BMImDCA. Despite this difference, as the temperature rises, the F values become closer to each other and at 80 °C F ≈ 1.4 in all the systems studied.

Close

Fig. 10

Temperature dependence of the structural resistance coefficient (F = κ_IL/κ_IG) of the of the halloysite filler in the ionogels based on BMImTFSI (1), BMImOtf (2), and BMImDCA (3).

Export to PPT

3.5 Electrochemical window

Application of ionic liquids in electrochemical processes is based on their high electrochemical stability comparable with that of many known organic solvents (Zhang and Bond, 2005) but ILs are more environmentally friendly because of their low vapour pressure. The electrochemical stability window (electrochemical window) - a potential region without significant background current - can reach 4–6 V in case of a suitable cation–anion combination, with the cathode stability limits usually determined by the cation, and the anode stability limit - by the anion. But the counter-ion can affect these characteristics too (Zhang et al., 2006; Zhang et al., 2009; Kroon et al., 2006). TFSI⁻ and Otf⁻ anions are oxidized at relatively high anodic potentials, which assumes that the ionic liquids based on these anions have high stability (Galiński et al., 2006; Ignat’ev et al., 2012). The electrochemical window for BMImDCA is 3–3.7 V, according to works (Yuan et al., 2018; Zec et al., 2016). The researchers also note that the size of the electrochemical window depends on the electrode material used. Fig. 11 shows the voltammograms of the studied ILs and ionogels based on them measured on a platinum electrode. The range of potentials in all the systems exceeds 4 V at the current density ± 1 mA cm⁻². However, there are certain differences between the obtained electrochemical window characteristics in the pure ILs and ILs in the ionogels. The differences consist in the narrowing of the electrochemical window (in BMImOtf, Fig. 11b) and appearance of a well-pronounced redox process (in BMImDCA, Fig. 11c) inside the IL electrochemical window. Water impurities in ionic liquids are known to considerably reduce the electrochemical window (O’Mahony et al., 2008), with the water electrochemical activity expressed in the form of two interrelated irreversible anodic and cathodic current peaks (Grishina et al., 2008). It was suggested that hygroscopic ILs could dehydrate halloysite and absorb the released water molecules. An additional experiment consisting in introducing water into BMImDCA showed that the increase in the content of this impurity in the IL from 0.11 wt% (the initial level) to 11.5 wt% made the anodic current peak more than 100 times higher and it also increased the cathode current peak associated with him. At the same time, the electrochemical characteristic of the hydrophobic BMImTFSI in the free state and as an ionogel component did not change a lot (Fig. 11a).

Close

Fig. 11

Cyclic voltammograms of the Pt electrode in ionic liquids (solid line) and ionogels based on them (dashed line). The potential scan rate is 5 mV s⁻¹, the temperature is 25 °C.

Export to PPT