Chinese Journal of Chemical Physics  2017, Vol. 30 Issue (5): 521-528

#### The article information

Yan-tao Wang, Kun-lan Li, Li-gang Wei, Ying-chong Ma

Probing Molecular Interactions in 1-Butyl-3-methylimidazolium Chloride-Water and 2, 6-Dimethoxyphenol Mixtures Using Attenuated Total Reflection Infrared Spectroscopy
1-丁基-3-甲基咪唑氯盐-水和2，6-二甲氧基苯酚混合物中分子相互作用的衰减全反射红外光谱研究
Chinese Journal of Chemical Physics, 2017, 30(5): 521-528

http://dx.doi.org/10.1063/1674-0068/30/cjcp1611203

### Article history

Accepted on: July 12, 2017
Probing Molecular Interactions in 1-Butyl-3-methylimidazolium Chloride-Water and 2, 6-Dimethoxyphenol Mixtures Using Attenuated Total Reflection Infrared Spectroscopy
Yan-tao Wanga,b, Kun-lan Lia, Li-gang Weia, Ying-chong Maa
Dated: Received on November 2, 2016; Accepted on July 12, 2017
a. School of Light Industry and Chemical Engineering, Dalian Polytechnic University, Dalian 116034, China;
b. Laboratory of EA4297 TIMR, University of Technology of Compiègne, Compiègne 60200, France
*Author to whom correspondence should be addressed. Kun-lan Li, E-mail:likl@dlpu.edu.cn; Li-gang Wei, E-mail:wei_ligang@hotmail.com, Tel: +86-411-86323726
Abstract: Molecular interactions of the ternary mixtures of 1-butyl-3-methylimidazolium chloride ([C4C1im]Cl)-water-2, 6-dimethoxyphenol (2, 6-DMP, a phenolic monomer lignin model compound) were investigated in comparison with the [C4C1im]Cl-water binary systems through attenuated total reflection infrared spectroscopy. Results indicated that the microstructures of water and [C4C1im]Cl changed with varying mole fraction of [C4C1im]Cl (xIL) from 0.01 to 1.0. This change was mainly attributed to the interactions of [C4C1im]Cl-water and the self-aggregation of [C4C1im]Cl through hydrogen bonding. The band shifts of C-H on imidazolium ring and the functional groups in 2, 6-DMP indicated that the occurrence of intermolecular interactions by different mechanisms (i.e., hydrogen bonding or π-π stacking) resulted in 2, 6-DMP dissolution. In the case of xIL=0.12, the slightly hydrogen-bonded water was fully destroyed and [C4C1im]Cl existed in the form of hydrated ion pairs. Interestingly, the maximum 2, 6-DMP solubility (238.5 g/100 g) was achieved in this case. The interactions and microstructures of [C4C1im]Cl-water mixtures influenced the dissolution behavior of 2, 6-DMP.
Key words: Lignin model compound    Ionic liquid-water    Molecular interactions
Ⅰ. INTRODUCTION

Ionic liquid (IL)-water mixtures have attracted significant attention as a novel solvent system in the pretreatment of lignocellulosic biomass because of their low cost, ease of IL recycling, and low viscosity [1-5]. The presence of water strongly affects the physical and chemical properties of ILs, such as viscosity, electrical property, polarity, solvation, and solubility [6-10]. IL-water mixtures with an appropriate water content (60 wt%-90 wt%) are the efficient solvents for the dissolution and removal of lignin [1-5]. To understand the underlying reasons of lignin dissolution in the IL-water mixtures at the molecular level, the molecular interactions in IL-water and IL-water-lignin mixtures were investigated to elucidate the mechanisms of lignin dissolution. This study may shed light on why lignin prefers to dissolve in IL-water mixtures rather than in pure IL.

IL-water interactions have been investigated by spectroscopy, molecular dynamics, and thermodynamics [11-26]. Attenuated total reflectance infrared (ATR-IR) spectroscopy is a non-invasive technique and highly sensitive to different molecular circumstances. It has been used to analyze molecular interactions between water and IL [11-14]. Cammarata et al. [11] found through IR spectroscopy that water molecules absorbed by imidazolium-based ILs from air mainly interacted with anions and existed in symmetric anion-HOH-anion hydrogen-bonded complexes. Kim et al. [27] investigated the aqueous mixtures of 1-butyl-3-methylimidazolium-based ILs with different anions ([C$_4$C$_1$im]X, X=Cl$^-$, Br$^-$, I$^-$, and BF$_4$$^-) by using the ATR-IR and proton nuclear magnetic resonance (NMR) analyses. Upon the introduction of water to screen the electrostatic forces and separate the ions, both IR and NMR spectra revealed that the hydrogen-bonding strength followed the order C2-H\cdotsCl>C2-H\cdotsBr>C2-H\cdotsI. The relative positions of halide anions with respect to the imidazolium ring were different from that of the BF_4$$^-$ anion because of the natures of the anion-cation interactions.

Compared with IL-water mixtures, ternary systems based on IL-water mixtures have been given less attention. Cláudio et al. [28] found that the ILs as hydrotropes enhance the solubility of hydrophobic substances in aqueous media. They speculated that the formation of IL-biomolecule aggregates was a crucial factor that influenced solubilization. Miki et al. [25] measured the excess partial enthalpy of 1-propanol in 1-butyl-3-methylimidazolium chloride ([C$_4$C$_1$im]Cl)-water mixtures. The effects of [C$_4$C$_1$im]Cl on the molecular organization of water in the mixtures were evaluated by thermodynamic analysis. They found that water molecules are attracted to the delocalized positive charge of the imidazolium ring in water-rich regions. The bulk of water is influenced such that the global probability of hydrogen bonding is reduced.

Investigation of molecular interactions during lignin dissolution in IL-water mixtures has been limited by the complexity of lignin. Most studies about lignin were performed from model compounds. 2, 6-Dimethoxyphenol (2, 6-DMP), a phenolic monomer model compound, was used in this work. The molecular interactions in the ternary mixtures of [C$_4$C$_1$im]Cl-water-2, 6-DMP were investigated by using ATR-IR spectroscopy in comparison with that of the [C$_4$C$_1$im]Cl-water binary mixtures. This study provides a general understanding of lignin dissolution in IL-water mixtures.

Ⅱ. EXPERIMENTS A. Chemicals and preparation of solutions

The IL, [C$_4$C$_1$im]Cl (purity$>$99% and water content $<$0.4%), was purchased from Lanzhou Institute of Chemical and Physics, and used without any further purification. The water used was double distilled and further treated using a CSR-1-20 high-purity water system (specific electronic conductivity, $\sim$0.055 ${\rm{ \mathsf{ μ} }}$S/cm). 2, 6-DMP (purity$>$99%) and D$_2$O were purchased from Sigma Aldrich Co. The chemical structures of [C$_4$C$_1$im]Cl and 2, 6-DMP with the notation of atoms are shown in FIG. 1. The aqueous solution of [C$_4$C$_1$im]Cl was gravimetrically prepared with the mole fraction of [C$_4$C$_1$im]Cl ($x_{\textrm{IL}}$) ranging from 0.01 to 0.80. Furthermore, 2, 6-DMP was added according to the specific mole fraction. The uncertainty in the mole fraction of the mixture was estimated to be less than $\pm$0.001.

 FIG. 1 Chemical structures of (a) [C$_4$C$_1$im]Cl and (b) 2, 6-DMP.
B. Spectroscopic measurements

ATR-IR (Perkin Elmer 94416) measurements were performed at 303.15 K. The IR absorbance spectra of the mixtures were obtained from 650 cm$^{-1}$ to 4000 cm$^{-1}$ with 4 cm$^{-1}$ resolution and averaged from 16 readings. A small droplet of each sample was placed on top of the ZnSe crystal. Single-reflection ATR (incident angle 45$^{\circ}$) ensured that the IR bands of the vibrational modes of hydroxyl from water or the 2, 6-DMP were not off-scale because of the extremely short effective path length. The advantage of ATR-IR spectroscopy had been widely utilized to analyze aqueous solutions. Each ATR-IR spectrum was corrected for evanescent wave penetration depth, and proper normalization was needed for the comparison. The second derivatives of the IR spectra were directly obtained using an online processing system.

C. Solubility of 2, 6-DMP in [C$_4$C$_1$im]Cl-water mixtures

The 2, 6-DMP was gradually added to 2 g of the [C$_4$C$_1$im]Cl-water mixtures in a small screw-capped flask, each of which was equipped with a magnetic stir bar. The suspensions were stirred in a water bath at 303.15 K. Dissolution of the 2, 6-DMP was observed by the naked eye (homogeneous solutions were clear and no undissolved particles were found). Each test was performed thrice to measure the error limit.

Ⅲ. RESULTS AND DISCUSSION A. [C$_4$C$_1$im]Cl-water binary mixtures

The ATR-IR absorption spectra of aqueous [C$_4$C$_1$im]Cl with different $x_{\textrm{IL}}$ are plotted in FIG. 2 and FIG. S1 in the supplementary materials. The attribution and analysis of the IR spectra of pure water [29] and [C$_4$C$_1$im]Cl [30] had been previously studied. In this work, we mainly focused on the stretching vibration regions of OH and CH$_x$ on imidazolium ring which represent the structural changes in water and [C$_4$C$_1$im]Cl.

 FIG. 2 (a) ATR-IR spectra (3800-2700 cm$^{-1}$) of [C$_4$C$_1$im]Cl-water mixtures with different $x_{\textrm{IL}}$ and (b) the second derivatives of the IR spectra in the range of 3800-3200 cm$^{-1}$.

For the spectra of O-H in water molecules shown in FIG. 2(a), a broad peak of bulk water (3000-3800 cm$^{-1}$) represented the nonhomogeneous environment of water molecules [31]. However, limit information was directly obtained from the original IR spectra of water due to the effects of hydrogen-bonding and Fermi resonance [11]. The second derivative of the IR spectrogram can be employed to search for overlapping peak positions [32-36], the results are shown in FIG. 2(b). The peaks at around 3250 cm$^{-1}$ were attributed to "ice-like" water with tetrahedral structures through strong hydrogen bonding, which existed in the all the investigated $x_{\textrm{IL}}$ range. The peaks at around 3420 cm$^{-1}$ (corresponding to "liquid-like" water, $x_{\textrm{IL}}$$<0.25) and at around 3640 cm^{-1} (corresponding to slightly hydrogen-bonded water, x_{\textrm{IL}}$$<$0.12) were only observed in the water-rich region. This result implied that the water with "liquid-like" and slightly hydrogen-bonded structures was gradually destroyed with addition of [C$_4$C$_1$im]Cl. When the $x_{\textrm{IL}}$ was increased to 0.60, some other structures of water are observed in FIG. 2(b). The two bands (ca. 3650 and 3570 cm$^{-1}$) could be separately assigned to the antisymmetric and symmetric stretching vibrations of the water monomers existing in ILs to form symmetric 1:2-type hydrogen-bonded complexes of anion-HOH-anion [11], whereas the component bands at 3450 cm$^{-1}$ may be attributed to the O-H stretching vibration mode from three-and two-coordinate water molecules. "Liquid-like" water was further decomposed into oligomers and even into monomers. Furthermore, the existence of monomeric water molecules was verified in the original IR spectra in the case of $x_{\textrm{IL}}$=0.80.

The main characteristic absorption bands of the [C$_4$C$_1$im]$^+$ cation as a function of $x_{\textrm{IL}}$ are illustrated in FIG. 3. D$_2$O was used to produce the mixtures because the shoulder of the O-H peak overlapped with the C-H$_x$ peaks. The vibration frequencies of C4, 5-H and C2-H stretching ($v_{\textrm{C4, 5-H}}$ and $v_{\textrm{C2-H}}$), antisymmetric stretching, symmetric stretching and Fermi resonance of the alkyl chain CH$_3$ on imidazolium ring ($v_{\textrm{as}(\textrm{CH}_3)}$, $v_{\textrm{s}(\textrm{CH}_3)}$, and $v_{\textrm{FR}(\textrm{CH}_3)}$) are plotted as a function of $x_{\textrm{IL}}$ (FIG. S2 in supplementary materials). Notably, these bands had a similar tendency to redshift with increasing $x_{\textrm{IL}}$. The band frequencies were the highest in the water-rich region (0$<$$x_{\textrm{IL}}$$<$0.06). This result can be attributed to the completely "free" positively charged cations that had a strong electron-withdrawing effect and enlivened the C-H bonds. As the $x_{\textrm{IL}}$ was increased from 0.06 to 0.50, $v_{\textrm{C4, 5-H}}$, $v_{\textrm{C2-H}}$, $v_{\textrm{as}(\textrm{CH}_3)}$, $v_{\textrm{s}(\textrm{CH}_3)}$, and $v_{\textrm{FR}(\textrm{CH}_3)}$ were respectively redshifted by 7, 43, 9, 7, and 8 cm$^{-1}$. The shift was due to the formation of ion pairs or small ion clusters. The decreased distance between the cations and the anions weakened the effect of the positive charge of the cations. At the same time, enhancement of the hydrogen bonds between Cl$^-$ and C2-H also resulted in the redshift. The frequencies of the C-H groups slightly decreased when the $x_{\textrm{IL}}$ was increased to over 0.50. The addition of [C$_4$C$_1$im]Cl resulted in the complete destruction of slightly hydrogen-bonded water in the case of $x_{\textrm{IL}}$$>0.12, whereas other states of water (i.e., monomer and small cluster) still remained in the case of x_{\textrm{IL}}$$>$0.50 (FIG. 2(b)). The aggregates of [C$_4$C$_1$im]Cl formed and the water molecules that functioned as bridges were speculated. The imidazolium cation is also characterized by nine ring vibrations that can easily couple with one of the $\delta$$_{\textrm{CH}} or \gamma$$_{\textrm{CH}}$ modes having the same symmetry. The in-plane R1 and R2 modes were situated at 1568 cm$^{-1}$ [37, 38]. The IR spectra of R1/R2 imidazolium ring vibrations (FIG. 3(b)) followed similar trends as that of water and CH$_x$ with the increase of $x_{\textrm{IL}}$.

 FIG. 3 (a) ATR-IR spectra of C4, 5-H and C2-H stretching, antisymmetric stretching, and Fermi resonance of the alkyl chain CH$_3$ on imidazolium ring, and (b) IR spectra of R1/R2 imidazolium ring vibrations as functions of $x_{\textrm{IL}}$.

According to the above ATR-IR analyses, the molecular interactions and microstructure of [C$_4$C$_1$im]Cl-water mixtures varied depending largely on the $x_{\textrm{IL}}$. This variation can affect the dissolution of lignin (or its model compounds) in the [C$_4$C$_1$im]Cl-water mixtures.

B. [C$_4$C$_1$im]Cl-water-2, 6-DMP ternary mixtures

Solubility of 2, 6-DMP in [C$_4$C$_1$im]Cl-water mixtures was firstly investigated at 303.15 K. The results are shown in FIG. 4. The 2, 6-DMP was hardly dissolved in pure water. The solubility of 2, 6-DMP increased monotonically with the increase of $x_{\textrm{IL}}$ from 0.01 to 0.12. The solubility reached a maximum (238.5 g/100g) at $x_{\textrm{IL}}$=0.12, and then slowly decreased as the $x_{\textrm{IL}}$ was increased from 0.12 to 1.0. Interestingly, the solubility of 2, 6-DMP followed a similar tendency to that of organosolv lignin in [C$_4$C$_1$im]Cl-water mixtures. The maximum solubility of organosolv lignin was also achieved at around $x_{\textrm{IL}}$=0.12, as indicated in our previous work [5]. The differences of the microstructure and interactions of the [C$_4$C$_1$im]Cl-water mixtures led to the varying solubility of lignin (or 2, 6-DMP). The ATR-IR analyses of [C$_4$C$_1$im]Cl-water-2, 6-DMP will help us understand the dissolution mechanisms of lignin (or 2, 6-DMP) at the molecular level.

 FIG. 4 Solubility of 2, 6-DMP in [C$_4$C$_1$im]Cl-water mixtures as a function of $x_{\textrm{IL}}$ at 303.15 K.

The ATR-IR analyses of the ternary systems were conducted to probe the molecular interactions between [C$_4$C$_1$im]Cl-water mixture and 2, 6-DMP. The ATR-IR spectra are shown in FIG. 5 and FIG. 6. The second derivative technique was applied to find the exact peak positions of the functional groups; an example is shown in FIG. S3 in supplementary materials.

 FIG. 5 ATR-IR spectra of [C$_4$C$_1$im]Cl-water-2, 6-DMP mixtures at different $x_{\textrm{IL}}$ (molar ratio of 2, 6-DMP to [C$_4$C$_1$im]Cl ($n$$_{\textrm{MC}}/n$$_{\textrm{IL}}$)=0.2).
 FIG. 6 ATR-IR spectra of (a) the [C$_4$C$_1$im]Cl-water-2, 6-DMP mixtures ($x_{\textrm{IL}}$=0.80) with different 2, 6-DMP content and (b) pure 2, 6-DMP.

In comparison with the [C$_4$C$_1$im]Cl-water binary system (FIG. 2), the ATR-IR spectra of the ternary mixtures (FIG. 5) and the second derivative spectrogram in the 3200-3400 cm$^{-1}$ range (FIG. S4 in supplementary material) revealed that the water state was scarcely changed after adding the 2, 6-DMP. Notably, the exact peak position of phenolic hydroxyl stretching vibration could not be easily found, even by using the second derivation method. However, with the addition of 2, 6-DMP, the original spectrum within the range of 3200-3300 cm$^{-1}$ was changed, which can be assigned to the phenolic hydroxyl stretching vibration mode (FIG. 6(a)). In comparison with the ATR-IR spectrum of pure 2, 6-DMP (FIG. 6(b)), the peak position of phenolic hydroxyl was apparently red shifted by ca.250 cm$^{-1}$, implying that strong hydrogen bonds formed with phenolic hydroxyl groups in the solution. This result was in agreement with the COSMO-RS simulation analysis by Casas et al. [39]. The main force involving the lignin dissolution process in IL was the hydrogen bonding, whereas the contribution of the electrostatic forces and van der Waals interactions were secondary and insignificant, respectively. Besides, Lateef et al. [40] supposed that ILs could interact with the terminal OH groups of lignin, subsequently disrupting the internal network within lignin molecules that results in the dissolution of lignin.

The band shifts ($\Delta v$) of the C-H and imidazole ring of the [C$_4$C$_1$im]$^+$ cation with a mole ratio of 2, 6-DMP to [C$_4$C$_1$im]Cl ($n$$_{\textrm{MC}}/n$$_{\textrm{IL}}$)=0.2 are depicted in FIG. 7(a) and (b). $\Delta v$=$v_\textrm{T}$-$v_\textrm{B}$, where $v_\textrm{T}$ is the frequency for the ternary system and $v_\textrm{B}$ is the frequency for the binary system. The frequency changes of the main functional groups of 2, 6-DMP are also shown in FIG. 7(c) and (d). The band shifts of these typical groups presented various trends in the different $x_{\textrm{IL}}$ ranges.

 FIG. 7 Band shifts of the typical groups on imidazolium cation and 2, 6-DMP as a function of $x_{\textrm{IL}}$ in the case of $n$$_{\textrm{MC}}/n$$_{\textrm{IL}}$=0.2 (a) C-H of imidazolium cation, (b) R1/R2 imidazolium ring vibrations; (c) benzene skeleton vibration of 2, 6-DMP, (d) stretching of methoxy group ($v_{\textrm{s}(\textrm{CH}_3)}$) and phenolic hydroxyl bending ($\delta_{\textrm{s}(\textrm{O-H})}$) of 2, 6-DMP.

In the case of 0.01$<$$x_{\textrm{IL}}$$<$0.06, the redshifts of the stretching bands of butyl side chain were evidently larger than those of C2-H and C4, 5-H (FIG. 7(a)). After the addition of 2, 6-DMP, the $\Delta v$$_{\textrm{C2-H}} and \Delta v$$_{\textrm{C4, 5-H}}$ basically remained unchanged, but the $\Delta$ $v_{\textrm{s}(\textrm{CH}_3)}$ was decreased by more than 2 cm$^{-1}$. The above results indicated that the hydrophobic butyl chain was superior to C2-H or C4, 5-H for interacting with 2, 6-DMP. This phenomenon may be due to the fact that C2-H and C4, 5-H were connected to water by hydrogen bonds, whereas the butyl chain was relatively "free" in the mixtures.

The changes in the R1/R2 vibrations of the imidazole ring were also studied (FIG. 7(b)). At a low $x_{\textrm{IL}}$ (0.01$<$$x_{\textrm{IL}}$$<$0.06), the R1/R2 ring vibrations were redshifted by more than 4 cm$^{-1}$ ($x_{\textrm{IL}}$=0.01) after adding the 2, 6-DMP into the solution. At the same time, the benzene skeleton vibration of 2, 6-DMP appeared at high frequencies (FIG. 7(c)). It was attributed to the interactions between the imidazolium cation and the benzene ring via $\pi$-$\pi$ stacking. In addition, the color changes of the solutions in water-rich regions (0.01$<$$x_{\textrm{IL}}$$<$0.06) after adding 2, 6-DMP attracted our attention (FIG. S5 in supplementary materials). The imidazolium cation can polarize the benzene ring of the 2, 6-DMP, which led to the delocalization of conjugated electrons, thus causing the color change of the solutions.

In view of the frequency changes in the methoxy group of 2, 6-DMP (FIG. 7(d)), we inferred that a dispersion force (interactions with non-polar fragments [39]) existed between the butyl chain of cations and the methoxy group of 2, 6-DMP. When lignin model compounds (2-methoxyphenol and 3, 5-dimethoxybenzyl alcohol) were dissolved in IL, the apparent $^{13}$C-NMR chemical shifts of methoxy group were also noticed by Pu et al. [41].

A break point was observed in C-H stretching and R1/R2 ring vibrations at $x_{\textrm{IL}}$=0.12. Break points are indicative of qualitative changes in solution structures due to the formation of ion pairs. Interestingly, the maximum 2, 6-DMP solubility was also achieved at this $x_{\textrm{IL}}$. In the case of ion pairs being formed (0.06$<$$x_{\textrm{IL}}$$<$0.50), the 2, 6-DMP interacted with both [C$_4$C$_1$im]$^+$ and Cl$^-$. These interactions further increased the distances between the cations and the anions. The increased distances may result in the blueshift of $v_{\textrm{C2-H}}$, $v_{\textrm{C4, 5-H}}$ (FIG. 7(a)). The band redshifts of butyl side chain can be interpreted as the decrease in dispersion force. It corresponded to the decrease in "free" hydrated ions. The decrease in the frequency shifts of the R1/R2 imidazole ring may represent the weakened $\pi$-$\pi$ stacking effect (FIG. 7(b)). Meanwhile, the apparent redshift of phenolic hydroxyl bending ($\delta$$_{\textrm{s}(\textrm{O-H})}) of 2, 6-DMP was approximately 5 cm^{-1} (FIG. 7(d)), indicating that the hydrogen-bonding between water (or Cl^- anions) and the phenolic hydroxyl groups was increased. In the case of x_{\textrm{IL}}$$>$0.50, the aggregate state of the IL was formed. The [C$_4$C$_1$im]$^+$ cations could shield the Cl$^-$ anions and the water molecules from interacting with the 2, 6-DMP. These events directly led to the blueshift of phenolic hydroxyl groups because of the lack of hydrogen bonds (FIG. 7(d)). Ji et al. [42] investigated the mechanism of lignin dissolution in 1-allyl-3-methylimidazolium chloride (AmimCl) using 1-(4-methoxyphenyl)-2-methoxyethanol (LigOH) as lignin model compound. Their theoretical study indicated that hydrogen bonds decreased with the increase of $x_{\textrm{IL}}$ (relative to H$_2$O), which was in agreement with our previous results about solubility parameters calculation [5]. However, they proposed that the addition of 1-3 mol H$_2$O (based on $n$$_{\textrm{AmimCl}}) significantly decreased the solubility of lignin in IL because water competed with the IL to form hydrogen bonds with the lignin. This finding was inconsistent with our results about solubility of lignin [5] or 2, 6-DMP (FIG. 4) in [C_4C_1im]Cl-water mixtures. It may be attributed to the difference of IL type, which needs further research. C. Possible mechanisms of 2, 6-DMP dissolution in [C_4C_1im]Cl-water mixtures According to the above ATR-IR analyses, the possible states of water and [C_4C_1im]Cl in the binary mixtures and the schematic diagram are speculated; these are shown in FIG. S6 in supplementary materials. The [C_4C_1im]Cl-water mixtures can be divided into three ranges: First, adding small amount of [C_4C_1im]Cl to bulk water destroyed the slightly hydrogen-bonded water clusters to a certain extent. The hydrated cations and anions cannot bind with each other (0<$$x_{\textrm{IL}}$$<0.06); Second, adding more [C_4C_1im]Cl to the mixture further destroyed slightly hydrogen-bonded water until the size and structure of water cluster stabilized (x_{\textrm{IL}}=0.12). Hydrated cations and anions existed in the form of hydrated ion pairs. Furthermore, the distance between the cations and the anions gradually decreased with increasing x_{\textrm{IL}} from 0.06 to 0.50. Finally, the [C_4C_1im]Cl clusters formed and the water molecules functioned as bridges in the case of x_{\textrm{IL}}$$>$0.50; the state of water dramatically changed (FIG. 2), wherein the relatively weaker hydrogen-bonded liquid-like water clusters decomposed into monomers or oligomers of water molecules.

The variation of 2, 6-DMP solubility with increasing $x_{\textrm{IL}}$ (FIG. 4) seemed to depend on the microstructure changing trend of the [C$_4$C$_1$im]Cl-water mixtures. The possible molecular interactions in the [C$_4$C$_1$im]Cl-water-2, 6-DMP ternary system are illustrated in FIG. 8. In water-rich regions (especially at $x_{\textrm{IL}}$$<0.06), the relatively "free" ions could lead to the strong interactions between [C_4C_1im]Cl and 2, 6-DMP. However, the ion concentration was so limited that water clusters hindered the 2, 6-DMP dissolution. Even when the hydrated ion pairs formed, the cations and anions were more "free" than that involved in the aggregates. In addition, the slightly hydrogen-bonded water clusters were completely destroyed in the case of x_{\textrm{IL}}=0.12. It may be one of the main reasons that the maximum solubility of 2, 6-DMP was achieved in this case. After the formation of ion pairs or aggregates (especially at x_{\textrm{IL}}$$>$0.12), the compact structure of [C$_4$C$_1$im]Cl formed through interactions between cations and anions and the "active" sites that interacted with 2, 6-DMP decreased, which may result in the decrease of the solubility of 2, 6-DMP.

 FIG. 8 Schematic diagram of the molecular interactions between [C$_4$C$_1$im]Cl-water and 2, 6-DMP at different $x_{\textrm{IL}}$.
Ⅳ. CONCLUSION

The molecular interactions of [C$_4$C$_1$im]Cl-water-2, 6-DMP ternary mixtures and the [C$_4$C$_1$im]Cl-water binary mixtures were investigated via ATR-IR spectroscopy. The results indicated that 2, 6-DMP solubility depended on the variation of molecular interactions and microstructure of [C$_4$C$_1$im]Cl-water mixtures. The possible mechanisms of 2, 6-DMP were also proposed. To a certain extent, lignin dissolution in the IL-water mixtures was believed to follow a similar mechanism. However, the dissolution of lignin is rather more complex than that of 2, 6-DMP because of its amorphous polymer structure. This study contributes to understanding the lignin dissolution in IL-water mixtures. Further investigation with the aid of NMR technique or molecular dynamics simulation is necessary to elucidate the detailed mechanisms of 2, 6-DMP (or lignin) dissolution in IL-water mixtures.

Supplementary materials: ATR-IR spectra of [C$_4$C$_1$im]Cl-water binary mixtures, band shifts of C4, 5-H, C2-H and alkyl chain CH$_3$ on imidazolium ring, typical bands of the mixtures analyzed by second derivative technique, the second derivative spectrogram of the ternary system, color changes of the ternary mixtures with varying the $x_{\textrm{IL}}$, schematic diagram of the states of water and [C$_4$C$_1$im]Cl in the binary mixtures were given.

Ⅴ. ACKNOWLEDGMENTS

This work was supported by the National Natural Science Foundation of China (No.21106011 and No.21276034) and the Program of Science and Technology of Liaoning Province (No.201602058), and China Scholarship Council.

Reference
 [1] D. B. Fu, and G. Mazza, Bioresour. Technol. 102 , 7008 (2011). DOI:10.1016/j.biortech.2011.04.049 [2] X. D. Hou, N. Li, and M. H. Zong, Bioresour. Technol. 136 , 469 (2013). DOI:10.1016/j.biortech.2013.02.118 [3] A. Brandt, M. J. Ray, T. Q. To, D. J. Leak, R. J. Murphy, and T. W. Welton, Green Chem. 13 , 2489 (2011). DOI:10.1039/c1gc15374a [4] L. G. Wei, K. L. Li, Y. C. Ma, and X. Hou, Ind. Crops Prod. 37 , 227 (2012). DOI:10.1016/j.indcrop.2011.12.012 [5] Y. T. Wang, L. G. Wei, K. L. Li, Y. C. Ma, N. N. Ma, S. Ding, L. L. Wang, D. Y. Zhao, B. Yan, W. Y. Wan, Q. Zhang, X. Wang, J. M. Wang, and H. Li, Bioresour. Technol. 170 , 499 (2014). DOI:10.1016/j.biortech.2014.08.020 [6] J. J. Wang, Y. Tian, Y. Zhao, and K. L. Zhuo, Green Chem. 5 , 618 (2003). DOI:10.1039/b303735e [7] J. A. Widegren, E. M. Saurer, K. N. Marsh, and J. W. Magee, J. Chem. Thermodyn. 37 , 569 (2005). DOI:10.1016/j.jct.2005.04.009 [8] R. A. Brown, P. Mollet, E. McKoon, C. A. Eckert, C. L. Liotta, and P. G. Jessop, J. Am. Chem. Soc. 123 , 1254 (2001). DOI:10.1021/ja005718t [9] V. Najdanovic-Visak, J. M. S. S. Esperanc, L. P. N. Rebelo, M. N. da Ponte, H. J. R. Guedas, K. R. Seddon, H. C. de Souza, and J. Szydlowski, J. Phys. Chem. B 107 , 12797 (2003). DOI:10.1021/jp034576x [10] S. Ding, L. G. Wei, K. L. Li, and Y. C. Ma, Chin. J. Chem. Phys. 29 , 497 (2016). DOI:10.1063/1674-0068/29/cjcp1601003 [11] L. Cammarata, S. G. Kazarian, P. A. Salter, and T. Welton, Phys. Chem. Chem. Phys. 3 , 5192 (2001). DOI:10.1039/b106900d [12] T. Köddermann, C. Wertz, A. Heintz, and R. Ludwig, Angew. Chem. Int. Ed. 45 , 3697 (2006). DOI:10.1002/(ISSN)1521-3773 [13] H. C. Chang, J. C. Jiang, W. C. Tsai, G. C. Chen, and S. H. Lin, J. Phys. Chem. B 110 , 3302 (2006). DOI:10.1021/jp0560009 [14] A. Dominguez, - Vidal, N. Kaun, M. J. Ayora-Cañada, and B. Lendl, J. Phys. Chem. B 111 , 4446 (2007). DOI:10.1021/jp068777n [15] S. Saha, and H. Hamaguchi, J. Phys. Chem. B 110 , 2777 (2006). DOI:10.1021/jp053817t [16] G. M. Sando, K. Dahl, and J. C. Owrutsky, J. Phys. Chem. B 111 , 4901 (2007). DOI:10.1021/jp067143d [17] Y. A. Gao, N. Li, L. Q. Zheng, X. T. Bai, L. Yu, X. Y. Zhao, J. Zhang, M. W. Zhao, and Z. Li, J. Phys. Chem. B 111 , 2506 (2007). DOI:10.1021/jp068299g [18] A. Mele, and C. D. Tran, S. H. De Paoli Lacerda, Angew. Chem. Int. Ed. 42 , 4364 (2003). DOI:10.1002/anie.v42:36 [19] T. Singh, and A. Kumar, J. Phys. Chem. B 111 , 7843 (2007). DOI:10.1021/jp0726889 [20] A. L. Rollet, P. Porion, M. Vaultier, I. Billard, M. Deschamps, C. Bessada, and L. Jouvensal, J. Phys. Chem. B 111 , 11888 (2007). DOI:10.1021/jp075378z [21] S. Rivera-Rubero, and S. Baldelli, J. Am. Chem. Soc. 126 , 11788 (2004). DOI:10.1021/ja0464894 [22] S. Rivera, - Rubero, and S. Baldelli, J. Phys. Chem. B 110 , 15499 (2006). DOI:10.1021/jp062694p [23] S. Baldelli, J. Phys. Chem. B 107 , 6148 (2003). DOI:10.1021/jp027753n [24] H. Katayanagi, K. Nishikawa, H. Shimozaki, K. Miki, P. Westh, and Y. Koga, J. Phys. Chem. B 108 , 19451 (2004). DOI:10.1021/jp0477607 [25] K. Miki, P. Westh, K. Nishikawa, and Y. Koga, J. Phys. Chem. B 109 , 9014 (2005). DOI:10.1021/jp046309c [26] W. J. Li, Z. F. Zhang, B. X. Han, S. Q. Hu, Y. Xie, and G. Y. Yang, J. Phys. Chem. B 111 , 6452 (2007). DOI:10.1021/jp071051m [27] S. Cha, M. Q. Ao, W. Sung, B. Moon, B. Ahlstr, P. Johansson, Y. Ouchi, and D. Kim, Phys. Chem. Chem. Phys. 16 , 9591 (2014). DOI:10.1039/C4CP00589A [28] A. F. M. Cláudio, M. C. Neves, K. Shimizu, J. N. C. Lopes, M. G. Freire, and J. A. P. Coutinho, Green Chem. 17 , 3948 (2015). DOI:10.1039/C5GC00712G [29] Y. J. Marechal, J. Chem. Phys. 95 , 5565 (1991). DOI:10.1063/1.461630 [30] Z. P. Liu, S. P. Huang, and W. C. Wang, J. Phys. Chem. B 108 , 12978 (2004). DOI:10.1021/jp048369o [31] Y. Jeon, J. Sung, D. Kim, C. Seo, H. Cheong, Y. Ouchi, R. Ozawa, and H. Hamaguchi, J. Phys. Chem. B 112 , 923 (2008). DOI:10.1021/jp0746650 [32] J. P. Wu, M. J. Wang, and J. P. W. Stark, J. Quant. Spectrosc. RA 102 , 228 (2006). DOI:10.1016/j.jqsrt.2006.02.009 [33] P. W. Labuschagne, S. G. Kazarian, and R. E. Sadiku, Spectrochim. Acta A 78 , 1500 (2011). DOI:10.1016/j.saa.2011.01.040 [34] Y. Zhu, C. H. Xu, J. Huang, G. Y. Li, Q. Zhou, X. H. Liu, S. Q. Sun, and J. H. Wang, J. Mol. Struct. 1069 , 96 (2014). DOI:10.1016/j.molstruc.2013.11.054 [35] P. I. Haris, Biochim. Biophys. Acta 1828 , 2265 (2013). DOI:10.1016/j.bbamem.2013.04.008 [36] X. Yao, Y. Peng, Q. Zhou, P. G. Xiao, and S. Q. Sun, J. Mol. Struct. 974 , 161 (2010). DOI:10.1016/j.molstruc.2010.02.064 [37] C. Roth, S. Chatzipapadopoulos, D. Kerlé, F. Friedriszik, M. Lütgens, S. Lochbrunner, O. Kühn, and R. Ludwig, New J. Phys. 14 , 105026 (2012). DOI:10.1088/1367-2630/14/10/105026 [38] J. C. Lassègues, J. Grondin, D. Cavagnat, and P. Johansson, J. Phys. Chem. A 113 , 6419 (2009). DOI:10.1021/jp903160r [39] A. Casas, J. Palomar, M. V. Alonso, M. Oliet, S. Omar, and F. Rodriguez, Ind. Crops Prod. 37 , 155 (2012). DOI:10.1016/j.indcrop.2011.11.032 [40] H. Lateef, S. Grimes, P. Kewcharoenwong, and B. Feinberg, J. Chem. Technol. Biot. 84 , 1818 (2009). DOI:10.1002/jctb.v84:12 [41] Y. Q. Pu, N. Jiang, and A. J. Ragauskas, J. Wood Chem. Technol. 27 , 23 (2007). DOI:10.1080/02773810701282330 [42] W. Y. Ji, Z. D. Ding, J. H. Liu, Q. X. Song, X. L. Xia, H. Y. Gao, H. J. Wang, and W. X. Gu, Energy Fuel. 26 , 6393 (2012). DOI:10.1021/ef301231a
1-丁基-3-甲基咪唑氯盐-水和2，6-二甲氧基苯酚混合物中分子相互作用的衰减全反射红外光谱研究

a. 大连工业大学轻工与化学工程学院, 大连 116034;
b. 贡比涅技术大学可再生资源综合利用实验室, 贡比涅 60200