The structure and hydrogen-bond properties of N-alkyl-N-methyl-pyrrolidinium bis(trifluoromethylsulfonyl)imide and DMSO mixtures†

Yan-Zhen Zheng ^a, Hong Chen ^b, Yu Zhou ^b, Deng Geng ^c, Hong-Yan He *^d and Li-Ming Wu ^ae
aCollege of Animal Sciences (College of Bee Science), Fujian Agriculture and Forestry University, Fuzhou 350002, P. R. China
^bSchool of Chemistry and Chemical Engineering, Qingdao University, Qingdao 266071, P. R. China
^cKey Laboratory of Bioorganic Phosphorous Chemistry and Chemical Biology (Ministry of Education), Department of Chemistry, Tsinghua University, Beijing 100084, P. R. China
^dCAS Key Laboratory of Green Process and Engineering, State Key Laboratory of Multiphase Complex Systems, Beijing Key Laboratory of Ionic Liquids Clean Process, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, P. R. China. E-mail: hyhe@ipe.ac.cn
^eInstitute of Apicultural Research, Chinese Academy of Agricultural Sciences, Beijing 100093, P. R. China

First published on 19th November 2020

---

---

1. Introduction

Ionic liquids (ILs), which are salts at room temperature,¹ are regarded as a special class of novel solvents due to their unique properties, including nonflammability, negligible vapor pressure, and excellent ability to dissolve inorganic, organic and even polymeric materials.^2–8 Their attractive properties have opened a broad range of potential uses in the field of environmentally friendly solvents for chemical synthesis, electrolytes in battery and capacitor applications, extraction and processing of metals, radioactive materials and polymers, and gas separation related to fossil energy production.^2–8

Due to the large variety and feasible functionalization of anions and cations, there is great opportunity to design ILs with properties optimized for specific applications.^9–12 However, this task is significant and cannot overcome high viscosity also, which is the ubiquitous defect of neat ILs, resulting in the decrease of mass transfer rates, increase in pumping costs, and finally hindering the practical and large applications of ILs.^13–15 Fortunately, mixing ILs with molecular solvents is another simple way to achieve the modification of ILs with desired properties.^16,17 Furthermore, this operation can largely reduce the high viscosity of the system.^18–21 It has been found that mixing ILs with molecular solvents, such as water, DMSO, alcohols and acetonitrile can destroy the aggregation networks of neat ILs, which remarkably lowers the viscosity of the systems.^18–21 At a suitable concentration range, the desired properties of ILs are maintained and sometimes enhanced.^16,17 It was found that mixtures of ILs and aprotic cosolvents (dimethylformamide, dimethyl sulfoxide and 1,3-dimethyl-2-imidazolidinone) can enhance cellulose dissolution compared with pure ILs.^18,19,22 Mixtures of ILs and acetonitrile can better control the reaction rates.²³ It was found that mixing ILs and molecular solvents can enhance the efficiency of auto-partitioning extracted lipids and oil from lipid-bearing biomass.^24,25 An increased extraction efficiency in liquid–liquid extractions was also observed in mixtures of molecular cosolvents and ILs.^26,27 Thus, mixing ILs with molecular solvents can extend the practical applications of ILs and overcome the drawbacks of neat ILs.

The properties of any material depend on the structure of the system. Thus, it is very important to understand the structural features of IL-molecular solvent mixtures in depth. The unique properties of IL-molecular solvent mixtures are governed by the type and strength of interactions between their different constituents. Particularly, the hydrogen-bond interactions have been widely considered to have a strong influence on the physical and chemical properties of IL systems.^{19,22,28–34} Thus, numerous experimental and theoretical studies have been performed to understand the structure and hydrogen-bond properties of IL-molecular solvent systems. Especially, the combination of Fourier transform infrared (FTIR) spectroscopy and density functional theory (DFT) has been proven to be an effective tool in revealing the hydrogen-bond interactions of IL-molecular solvent mixtures.^32,35–44 Zhang et al. used FTIR, DFT and other experimental tools, and found that ion clusters (2cation–2anion complex) of IL can translate into ion pairs (cation–anion complex) in the N-butylpyridinium dicyanamide–dimethyl sulfoxide (DMSO) mixture.³⁵ Hydrogen bonds were found between DMSO and cation/anion, which were stronger between DMSO and cation. The hydrogen-bonded networks formed by the cations and anions are difficult to destroy when the DMSO concentration is low. Combining FTIR and DFT calculations, Köddermann et al. found that water molecules were strongly confined by hydrogen bonds in two ILs, namely 1-ethyl-3-methylimidazolium ethylsulfate ([EMIM][EtSO₄], hydrophilic IL) and 1-ethyl-3-methylimidazoliumbis(trifluoromethylsulfonyl)imide ([EMIM][Tf₂N], hydrophobic IL).³⁶ The v(O–H) of water is very sensitive to micro-environments changes. Thus, it can be used as an IR probe to characterize the polarity of ILs. Zhou et al. used FTIR and DFT calculations to investigate the structure and hydrogen-bond interactions in binary mixtures of 1-ethyl-3-methylimidazolium bis(fluorosulfonyl) imide ([EMIM][FSI]) and three polar solvents (CH₃OH, CH₃CN and DMSO).³⁷ CH₃OH and DMSO can break the [EMIM]⁺–[FSI]⁻ ion pair and form [EMIM]⁺–CH₃OH/DMSO complexes. In contrast, CH₃CN cannot break the ion pair in the studied concentration range.

Cyclic alkyl quaternary ammonium-based ILs, N-alkyl-N-methyl-pyrrolidinium ([CnMPyr]⁺), were first reported in 1999.⁴⁵ Compared to imidazolium-based ILs, these ILs possess certain irreplaceable features, namely, high aliphaticity, high localization of the positive charge and greater flexibility.^46,47 Their straightforward synthesis, recyclable nature, low viscosity, low cost and low toxicity properties^48–50 have facilitated their use in various extraction and biological processes, organic synthesis, heat stabilizers, catalysis and fuel cell devices.^50–52 DMSO is a solvent with versatile, powerful, biodegradable and low toxicity properties.⁵³ It can be miscible with various inorganic and organic substances.⁵³ Recently, mixtures of DMSO and pyrrolidinium-based ILs were found to be highly effective solvents to dissolve cellulose.⁵⁴ These mixtures were also demonstrated to be excellent electrolytes in LiO₂ batteries for the oxygen evolution reaction and oxygen reversible reduction reaction.⁵⁵

The knowledge on the structure and hydrogen-bond interaction properties of IL-molecular solvent mixtures is essential for their chemical application. There are many reports regarding the microstructure and interaction properties of IL–DMSO mixtures. However, to the best of our knowledge, most of these studies are limited to imidazolium-based ILs and DMSO.^{37,43,56–59} The molecular interactions between pyrrolidinium-based ILs and DMSO have not been widely explored. Thus, this work was carried out to study the structure and interaction properties of pyrrolidinium-based ILs with DMSO. In this regard, four N-alkyl-N-methyl-pyrrolidinium bis(trifluoromethylsulfonyl)imide ([CnMPyr][Tf₂N], n = 3, 4, 6 and 8) ILs were selected. The structures of the studied ILs are presented in Fig. 1. FTIR and DFT calculations were employed to gain insight into the structure and interaction features in pyrrolidinium-based ILs and DMSO mixtures. Excess infrared absorption spectroscopy^{35,37,40–42,56,57,60,61} and two-dimensional correlation spectroscopy (2D-COS)^38,62,63 can readily enhance the spectra resolution and give more information of the structure properties. Thus, they were employed to extract the structure properties from the original IR spectra of the mixtures.

2. Experimental

2.1. Materials

The four [CnMPyr][Tf₂N] ILs (>99%) were obtained from Shanghai Cheng Jie Chemical Co. LTD (Shanghai, China). Deuterated DMSO (DMSO-d₆, >99.8 of deuterium) was used instead of DMSO to avoid the overlap of v(C–H) between the ILs and DMSO, which was purchased from Cambridge Isotopes Laboratories. Before sample preparation, a vacuum oven was applied to remove water from the ILs. DMSO-d₆ was used without any treatment.

2.2. Sample preparation

A series of binary IL–DMSO-d₆ mixtures was prepared by weighing. The detail procedure is as follows: the volume of IL was calculated first based on the density and molar mass of the IL and DMSO-d₆. Then, the corresponding volumetric IL was transferred to a volumetric flask with a pipette and the weight of the IL could be obtained by weighing. Finally, DMSO-d₆ was used for the volumetric determination and its mass was obtained by weighed. DMSO-d₆ was miscible with the studied ILs across the entire concentration range at room temperature. The mole fractions of DMSO-d₆ (x(DMSO-d₆)) in the [C3MPyr][Tf₂N]–DMSO-d₆ mixtures were 0.1001, 0.1983, 0.2983, 0.3993, 0.5003, 0.5994, 0.7000, 0.7998 and 0.9000. That in the [C4MPyr][Tf₂N]–DMSO-d₆ mixtures were 0.0990, 0.1992, 0.2989, 0.4000, 0.5001, 0.5997, 0.7000, 0.7999 and 0.9000. That in the [C6MPyr][Tf₂N]–DMSO-d₆ mixtures were 0.0992, 0.1995, 0.2988, 0.3990, 0.5001, 0.6000, 0.7000, 0.7999 and 0.8999. That in the [C8MPyr][Tf₂N]–DMSO-d₆ mixtures were 0.1001, 0.1990, 0.2993, 0.3992, 0.5000, 0.5999, 0.7003, 0.7999 and 0.9000.

2.3. FTIR spectroscopy measurements

FTIR spectra were recorded on a Nicolet 5700 FTIR spectrometer (Thermo Nicolet Analytical Instruments, Madison, USA) in the range of 4000 cm⁻¹ to 650 cm⁻¹ with a resolution of 2 cm⁻¹, an average of 32 parallel scans and a zero-filling factor. The Nicolet 5700 FTIR spectrometer was used with an MCT detector and a horizontal attenuated total reflectance (ATR) cell, which attached a trapezoidal ZnSe crystal (2.43 refractive index, 45° angle of incidence and 12 reflections). In the FTIR measurement, each sample was directly dripped above the ATR cell. The penetration depth (de) of this work was corrected using the equation devised by Hansen.⁶⁴ A WYA-2S Abbe refractometer (Shanghai YiCe Apparatus & Equipment co., LTD, Shanghai, China) was used to collect the refractive indexes of each sample including pure liquids and the binary mixtures.

2.4. Excess spectrum analysis

Excess spectra can enhance the resolution and were measured for each original IR spectrum. For a binary liquid system, the data processing equation for an excess spectrum is as follows:^60,61

	(1)

where A, C_i, d, ε_i* and x_i are the IR absorbance, the molarity of component i, the penetration depth, the molar absorption coefficient of the pure component i and the mole fraction of component i, respectively.

2.5. Two-dimensional correlation spectroscopy analysis

A standard 2D correlation analysis based on the algorithm developed by Noda^62,63 was carried out with a program written in our laboratory on the Matlab software. The modified component-normalization method was used to remove the linear contribution to the absorbance by concentration.^65,66 The average of all the excess spectra for each binary system on the studied concentration range was used as the reference spectrum.

2.5. Quantum chemical calculations

The geometrical and vibrational properties of the DMSO-d₆ monomer, DMSO-d₆ dimer, [CnMPyr]⁺ cation, [Tf₂N]⁻ anion, [CnMPyr]⁺–DMSO-d₆ complexes, [Tf₂N]⁻–DMSO-d₆ complex, [CnMPyr]⁺–[Tf₂N]⁻ complexes, 2[CnMPyr]⁺–2[Tf₂N]⁻ complexes, [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complexes and 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes were calculated using the m06-2x/6-311g(d,p), ωB97XD/6-311g(d,p), B3LYP-D3(BJ)/6-311g(d,p) and m06-2x/6-311++g(d,p) methods. The initial structures of each interaction complex were based on the most stable geometries of the corresponding monomer/complex. For example, the initial structures of [C4MPyr]⁺–[Tf₂N]⁻–DMSO-d₆ were used the most stable structures of the [C4MPyr]⁺–[Tf₂N]⁻ complex and DMSO-d₆ monomer. DMSO-d₆ monomer was placed around all the possible interaction sites of the optimized [C4MPyr]⁺–[Tf₂N]⁻ structure to obtain the most stable geometry of the [C4MPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complex. All the calculated IR frequencies were confirmed to be positive, verifying that the optimized structure was a true minimum for each geometry. The single point energy at the M06-2X/6-311+g(2d,p) level was calculated to confirm the most stable geometry for each complex. Because the optimization and vibration analysis results are less sensitive on the basis sets than energy calculation, we used higher basis sets to calculate the single point energy. All the DFT calculations were performed using the Gaussian 16 program.⁶⁷

To reveal the chemical origin of the hydrogen bond between the ILs and DMSO, the atoms in molecular (AIM) approach was performed on the most stable geometries of the studied complexes at the M06-2X/6-311g(d,p) level.⁶⁸ Three topological descriptors at the bond critical point (BCP) related with the bond properties, namely, the electron density (ρ_BCP), the Laplacian of the electron density (∇²ρ_BCP) and the energy density (H_BCP), were analysed. The search of the BCP at the hydrogen bond in the IL–DMSO-d₆ interaction complexes and the topological descriptors collected were used the Multiwfn 3.5 suite.⁶⁹

3. Results and discussion

To study the molecular interactions between DMSO and the four pyrrolidinium-based ILs with the same anion, the absorption region related with the anion and DMSO-d₆ was used as the probe. Due to the overlap of the absorptions of the IL and DMSO below 2000 cm⁻¹, the vibrational frequencies related with the anion and SO of DMSO could not be used. Thus, only the v(C–D) of DMSO-d₆ was selected for the investigation. In this work, the v_s(C–D) of the methyl group on DMSO-d₆ was used due to its less complicated assignment compared with the v_as(C–D).

3.1. Original IR spectra of v_s(C–D) region

The ATR-FTIR spectra in the 2150 cm⁻¹–2100 cm⁻¹ region of the [C3MPyr][Tf₂N]–DMSO-d₆, [C4MPyr][Tf₂N]–DMSO-d₆, [C6MPyr][Tf₂N]–DMSO-d₆ and [C8MPyr][Tf₂N]–DMSO-d₆ systems in the studied concentration range are shown in Fig. 2A1–D1, respectively. The detail fraction of DMSO-d₆ for each original IR spectrum is labelled in the excess spectrum in Fig. 2A2–D2 using the same colour as the corresponding original IR spectrum. As can be seen in Fig. 2A1–D1, all the [CnMPyr][Tf₂N] do not have any absorption peaks in the wavenumber region of 2150 cm⁻¹–2100 cm⁻¹, and the band at around 2124 cm⁻¹ is completely ascribed to the symmetrical stretching vibration (v_s(C–D)) of the methyl groups of DMSO-d₆.⁷⁰ Thus, the spectral changes of the v_s(C–D) region can reflect the structure and hydrogen-bond property changes of DMSO-d₆ and can be used to study the structure and interaction properties of the [CnMPyr][Tf₂N]–DMSO-d₆ systems. As can be seen in Fig. 2A1–D1, v_s(C–D) gradually shifted to a higher wavenumber with an increase in the content of [CnMPyr][Tf₂N]. The hydrogen bond related with the methyl group is an “improper blue-shift hydrogen bond’’.³⁵ The blue-shift of v(C–H) of the methyl group indicates the enhancement of the related hydrogen bond. Thus, the hydrogen bond involving the methyl group of DMSO-d₆ is strengthened in the [CnMPyr][Tf₂N]–DMSO-d₆ mixtures compared with that in pure DMSO-d₆. The main cause may be that the methyl groups of DMSO interact with the oxygen atom of DMSO. However, in the IL–DMSO mixtures, they can also interact with the [Tf₂N]⁻ anion. The interaction between [Tf₂N]⁻ and the methyl groups is stronger than that between the methyl groups and the oxygen atom of DMSO. When x(DMSO-d₆) was reduced to 0.1, v_s(C–D) shifted to 2130.5 cm⁻¹, 2130.0 cm⁻¹, 2129.5 cm⁻¹ and 2129.3 cm⁻¹ for [C3MPyr][Tf₂N]–DMSO-d₆, [C4MPyr][Tf₂N]–DMSO-d₆, [C6MPyr][Tf₂N]–DMSO-d₆, and [C8MPyr][Tf₂N]–DMSO-d₆, respectively, which indicates that the interactions between [CnMPyr][Tf₂N] and DMSO-d₆ are weakened with an increase in the alkyl chain.

3.2. Excess spectra analysis of v_s(C–D) region

The excess spectra in the 2150 cm⁻¹–2100 cm⁻¹ region of the [CnMPyr][Tf₂N]–DMSO-d₆ systems are shown in Fig. 2A2–D2. As shown in Fig. 2A2–D2, the negative band (around 2122 cm⁻¹) at a lower wavenumber and positive band (around 2131 cm⁻¹) at a higher wavenumber are observed in each excess spectrum of the four IL–DMSO-d₆ systems. The appearance of the positive and negative excess peaks reflect the departure of the absorptions of the binary mixtures from that of the corresponding ideal mixtures, which illustrates that the mixing process of [CnMPyr][Tf₂N] and DMSO-d₆ is non-ideal and interaction complexes form between [CnMPyr][Tf₂N] and DMSO-d₆.^{35,37,40–42,56,57,60,61}

In an excess spectrum, the appearance of different position peaks is related with different associations in the mixture. Specially, the positive and negative signs of the excess peaks indicate the increase/appearance and decrease/disappearance of the related species in the mixture compared with the pure liquid, respectively.^{35,37,40–42,56,57,60,61} In this work, the negative excess peaks are related with the decrease in the species existing in pure DMSO-d₆. The positive excess peaks are related with the appearance of an interaction complex between [CnMPyr][Tf₂N] and DMSO-d₆. For mixtures containing stable complexes, the excess spectra will present a fixed position in the concentration range. However, in our work, with an increase in [CnMPyr][Tf₂N], both the negative and positive peaks gradually shifted to a higher wavenumber for the four [CnMPyr][Tf₂N]–DMSO-d₆ systems. The peak positions (cm⁻¹) of the positive and negative bands in Fig. 2A2–D2 of the different concentrations are collected in Table S1 in the ESI.† Thus, the positive and negative peaks in Fig. 2A2–D2 may not only be composed of one band and are overlapped with the absorptions of the different complexes.

3.3. Two-dimensional correlation spectroscopy analysis of the excess spectra in the v_s(C–D) region

The above section illustrated that the excess bands in Fig. 2A2–D2 are overlapped with different peaks. Thus, to extract the complicated information from the excess bands, 2D-COS analysis was carried out in the v_s(C–D) region for the excess spectra of the four IL–DMSO-d₆ systems in the whole concentration range. The 2D-COS analysis results are presented in Fig. 3A1–D2.

In our work, the 2D-COS analysis was based on the perturbation of the IL content in the mixtures. With the help of this powerful tool, the overlapped peaks of the complex spectra could be simplified. There are more correlation bands in the asynchronous spectrum (Fig. 3A2–D2) than in the synchronous spectrum (Fig. 3A1–D1, respectively), because the asynchronous cross peaks result from the relative dissimilarity of the intensity variation behaviour. In the synchronous spectra shown in Fig. 3A1–D1, the strong auto peak indicates the susceptibility of the C–D stretching vibration to change in spectral intensity with the perturbation of the IL content. One negative cross peak at around (2133 cm⁻¹, 2124 cm⁻¹) was observed in each synchronous spectrum, which indicates that the changes in the two species related with 2133 cm⁻¹ and 2124 cm⁻¹ are opposite.^38,62,63 In the synchronous spectra, we also observed that the intensities of the auto peak at around (2133 cm⁻¹, 2133 cm⁻¹) and the cross peak gradually weakened from the [C3MPyr][Tf₂N]–DMSO-d₆ system to [C8MPyr][Tf₂N]–DMSO-d₆. The weakened auto peak indicates that this auto peak is more stable with the lengthened alkyl chain by the perturbation of the IL content. That of the cross peak may be induced by the more stable intensity of the peak at 2133 cm⁻¹ or the increased intensity of that at 2124 cm⁻¹. Combining these reasons, the more stable 2133 cm⁻¹ peak with the lengthened alkyl chain length by the perturbation of the IL content is the main cause of the weakening of the auto peak at around 2133 cm⁻¹ and the cross peak.

The asynchronous spectrum shown in Fig. 3A2–D2 is more complicated compared with the synchronous spectrum, which indicates that DMSO-d₆ is not homogenously mixed with the ILs and different small aggregates of DMSO-d₆ molecules were found. Two cross peaks at around (2121 cm⁻¹, 2135 cm⁻¹) and (2124 cm⁻¹, 2128 cm⁻¹) were observed in the asynchronous spectra, which indicate that DMSO-d₆ molecules are at least in four different molecular environments in the mixtures.

3.4. The assignments of the cross peaks in the asynchronous spectra

It is clear that the wavenumbers of 2121 cm⁻¹ and 2124 cm⁻¹ identified from the asynchronous spectra correspond to the negative peak (around 2122 cm⁻¹) of the excess spectra in Fig. 2A2–D2 and are related with the decrease in the species appearance in pure DMSO-d₆. According to the literature,^53,56,57 DMSO-d₆ mainly exists as a monomer and dimer in pure DMSO liquid. Thus, the negative peak in the excess spectra is derived from the decrease in the DMSO dimer and monomer.

The peaks at 2135 cm⁻¹ and 2128 cm⁻¹ correspond to the positive peak of the excess spectra (around 2131 cm⁻¹) in Fig. 2A2–D2, which are not present in pure DMSO-d₆. They appear upon the addition of the ILs and must be related with the interaction complexes of the ILs and DMSO. In the IL–cosolvent binary mixtures, the interaction complexes of the IL and cosolvent may be ion cluster–cosolvent, ion pair–cosolvent, cation–cosolvent, and anion–cosolvent.^{35,37,39–42,56,57} Thus, to determine the origin of the positive excess peaks and reveal the interaction complexes in the [CnMPyr][Tf₂N] and DMSO mixtures, we applied theoretical calculations to assign the excess peaks. Similar to the previous studies,^{35,37,39–42,56,57} the two cations-two anions–DMSO-d₆ (2[C3MPyr]⁺–2[Tf₂N]⁻–DMSO-d₆, 2[C4MPyr]⁺–2[Tf₂N]⁻–DMSO-d₆, 2[C6MPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ and 2[C8MPyr]⁺–2[Tf₂N]⁻–DMSO-d₆) and cation–anion–DMSO-d₆ ([C3MPyr]⁺–[Tf₂N]⁻–DMSO-d₆, [C4MPyr]⁺–[Tf₂N]⁻–DMSO-d₆, [C6MPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and [C8MPyr]⁺–[Tf₂N]⁻–DMSO-d₆) were used to represent the ion cluster–cosolvent interaction complexes and ion pair–cosolvent interaction complexes, respectively. The cation–DMSO-d₆ ([C3MPyr]⁺–DMSO-d₆, [C4MPyr]⁺–DMSO-d₆, [C6MPyr]⁺–DMSO-d₆ and [C8MPyr]⁺–DMSO-d₆) and anion–DMSO-d₆ ([Tf₂N]⁻–DMSO-d₆) represented the cation–cosolvent complexes and anion–cosolvent complex, respectively. The calculated frequencies of v_s(C–D) of DMSO-d₆ in the most stable structures of the abovementioned complexes are presented in Table 1.

---

In Table 1, it can clearly be seen that the calculated frequency of v_s(C–D) for the four [CnMPyr][Tf₂N]–DMSO-d₆ systems ranges in the following sequences: DMSO-d₆ dimer < [Tf₂N]⁻–DMSO-d₆ < DMSO-d₆ monomer < [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ < [CnMPyr]⁺–DMSO-d₆ < 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ (M062X/6-311g(d,p)) or DMSO-d₆ dimer < [Tf₂N]⁻–DMSO-d₆ < DMSO-d₆ monomer < [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ < 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ < [CnMPyr]⁺–DMSO-d₆ (ωB97XD/6-311g(d,p), B3LYP-D3(BJ)/6-311g(d,p) and m062x/6-311++g(d,p)). The v_s(C-D) of the DMSO-d₆ monomer is larger than that of the DMSO-d₆ dimer, which indicates that the negative excess peaks at around 2121 cm⁻¹ and 2124 cm⁻¹ correspond to the absorptions of the DMSO dimer and monomer, respectively. In the four IL–DMSO-d₆ systems, the calculated frequencies of v_s(C–D) of the 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆, [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and [CnMPyr]⁺–DMSO-d₆ complexes are larger than that of the DMSO dimer and DMSO monomer. Thus, the positive excess peaks may be related with the ion cluster-DMSO-d₆, ion pair-DMSO-d₆ and cation–DMSO-d₆ complexes. If the positive excess peak is assigned to the absorption of v_s(C–D) of the cation–DMSO-d₆ complex, since the cation and anion coexist in the mixtures, there should also be a positive excess peak related with the absorption of the [Tf₂N]⁻–DMSO-d₆ complex. According to Table 1, the v_s(C–D) of [Tf₂N]⁻–DMSO-d₆ is larger than that of the DMSO-d₆ dimer, but lower than that of the DMSO-d₆ monomer. The excess peaks related with the [Tf₂N]⁻–DMSO-d₆ complex may appear to have a negative sign. This result is contradictory with the fact that the excess peak related with IL–DMSO-d₆ is positive. Thus, it seems impossible that the positive excess peak is related with v_s(C–D) of the cation–DMSO-d₆ complex. It is more possible that it originated from the absorptions of v_s(C–D) of ion cluster-DMSO-d₆ complexes and ion pair-DMSO-d₆ complexes. The calculated v_s(C–D) of 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ is larger than that of [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆. Thus, 2135 cm⁻¹ and 2128 cm⁻¹ correspond to the absorptions of the ion cluster-DMSO-d₆ complex and ion pair-DMSO-d₆ complex, respectively.

Thus, with the help of quantum chemical calculations, the observed frequencies in the 2D-COS analysis results of the excess spectra centred at around 2121 cm⁻¹, 2124 cm⁻¹, 2128 cm⁻¹ and 2135 cm⁻¹ could be assigned. They are related with the absorptions of the DMSO-d₆ dimer, DMSO-d₆ monomer, ion pair-DMSO-d₆ and ion cluster-DMSO-d₆, respectively. Both the negative and positive excess peaks in Fig. 2A2–D2 gradually shift to a higher wavenumber. Thus, the species present in pure DMSO gradually decrease from DMSO dimer to DMSO monomer with an increase in ILs. This is because the DMSO dimer would break into DMSO monomer with the addition of ILs. Also, the amount of DMSO dimer would reduce before that of DMSO monomer. The blue-shift of the positive excess peak indicates that the interaction complexes of IL–DMSO related with ion cluster-DMSO gradually increase, while that of the ion pair-DMSO decrease with an increase in ILs. This is mainly ascribed to the strong electrostatic interaction between the cation and anion. With an increase in ILs, the cation and anion would gather into larger interaction complexes, which induce an increase in the ion cluster-DMSO complex and decrease in the ion pair-DMSO complex. The cross peaks at around (2133 cm⁻¹, 2124 cm⁻¹) in the synchronous spectrum shown in Fig. 3A1–D1 are negative, which indicates that the changes in the associations related with 2133 cm⁻¹ and 2124 cm⁻¹ are opposite.^38,62,63 This is easy to understand. The increase in the ion cluster/ion pair-DMSO complexes in the mixtures is accompanied with a decrease in the DMSO dimer and monomer. Thus, the content changes of the ion cluster/ion pair-DMSO complexes are opposite with that of the DMSO dimer and monomer.

3.5. Hydrogen-bond properties between [CnMPyr][Tf₂N] and DMSO

In Section 3.4, we found that DMSO-d₆ interacts with the four [CnMPyr][Tf₂N] ILs in the form of ion pair-DMSO-d₆ complexes and ion cluster-DMSO-d₆ complexes. Thus, to study the hydrogen bond between DMSO and the four ILs, the most stable geometries of the 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes and [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complexes optimized by m06-2x/6-311g(d,p) were analysed, as shown in Fig. 4 and 5, respectively.

The distance is a reliable indicator characterizing the formation of a hydrogen bond, which is smaller than the sum of the van der Waals atomic radii of the hydrogen atom and the electronegative atom indicates the presence of a hydrogen bond.⁷¹ The sum of the van der Waals atomic radii between hydrogen and oxygen, that between hydrogen and nitrogen and that between hydrogen and fluorine are 2.5 Å, 2.6 Å and 2.45 Å, respectively.⁷¹ Careful examination was paid to the distances between the hydrogen atoms and the possible hydrogen-bond acceptors in the most stable geometries of the 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes and [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complexes shown in Fig. 4 and 5. The identified hydrogen bonds in the eight structures are marked by dashed lines and the distances of the hydrogen bonds between [CnMPyr][Tf₂N] and DMSO-d₆ are labelled. The distances of the remaining hydrogen bonds between the cation and the anion in the complexes are presented in Tables S2 and S3 in the ESI.† As can be seen in Fig. 4 and 5, the hydrogen bonds are present in the four [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complexes and the four 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes. The oxygen atom and the hydrogen atoms in the methyl groups of DMSO simultaneously interact with the ILs. The most stable geometries of the four [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complexes are similar with each other. As can be seen in these geometries in Fig. 4, the SO group of DMSO interacts with H32, H33, H35 and H37 in [C3MPyr]⁺–[Tf₂N]⁻–DMSO-d₆, [C4MPyr]⁺–[Tf₂N]⁻–DMSO-d₆, [C6MPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and [C8MPyr]⁺–[Tf₂N]⁻–DMSO-d₆, respectively. These hydrogen atoms are attached to the methylene group close to the nitrogen atom of the cation. The methyl groups of DMSO interact with the oxygen atom and the nitrogen atom of the anion.

Besides the geometric evaluation criterion, the formation of the hydrogen bond should also meets the topological standards in the AIM theory as follows: (1) the presence of a BCP at the hydrogen bond, (2) ρ_BCP in the range of 0.002 a.u. to 0.040 a.u. and (3) the ∇²ρ_BCP from 0.02 a.u. to 0.15 a.u.⁷² We carefully analysed the hydrogen bonds identified by geometric criterion in Fig. 4 and 5. All the identified hydrogen bonds have BCPs. The three topological properties (ρ_BCP, _∇²ρ_BCP and H_BCP) at the BCPs of the hydrogen bonds between [CnMPyr][Tf₂N] and DMSO-d₆ at the most stable geometry of the [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes obtained using the AIM theory by the M06-2X/6-311g(d,p) method are presented in Tables 2 and 3, respectively. Those between cation and anion in [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes are shown in Tables S2 and S3 in the ESI,† respectively. As can be seen in these tables, the ρ_BCP and ∇²ρ_BCP are all positive. More importantly, the values of ρ_BCP in Tables 2 and 3 are in the range of 0.016–0.012 a.u. and 0.020–0.009 a.u., respectively, and that in Tables S1 and S2 (ESI†) are in the range of 0.016 au to 0.009 au. The values of ∇²ρ_BCP are in the range of 0.073–0.040 a.u., 0.078–0.030 a.u., 0.060–0.033 a.u. and 0.060–0.028 a.u., as shown in Tables 2, 3 and Tables S2, S3 (ESI†) respectively. All the ρ_BCP and ∇²ρ_BCP values are within the range for conventional hydrogen bonds, which further verifies the identified hydrogen bonds in the most stable geometries of the [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes shown in Fig. 4 and 5.

---

---

Besides assisting with the identification of hydrogen bonds, the topological properties at the BCPs of the hydrogen bonds can also be used to study the hydrogen-bond properties. In the AIM theory, the larger the ∇²ρ_BCP, the stronger the hydrogen bond.⁶⁸ In Fig. 4 and 5, the oxygen atom of DMSO prefers to interact with the ring hydrogen atoms of the methylene group attached to the nitrogen atom. Besides, in Table S2 (ESI†), the values of ∇²ρ_BCP at the BCPs of the hydrogen bonds between H29, H30, H39, and H41 and the oxygen atoms are the largest. In Table S3 (ESI†), that between H29, H30, H35, and H93 and the oxygen are larger than the other hydrogen bonds. As can be seen in Fig. 4 and 5, these hydrogen atoms are attached to the ring methylene groups close to the nitrogen atom. Thus, the ring hydrogen atoms of the methylene group directly attached to the nitrogen atom are the preferred interaction sites of the [CnMPyr]⁺ cations.

In the AIM theory, the sign of ∇²ρ_BCP is closely related with the type of hydrogen bond.⁷³ That dominated by the contraction of electron density towards each nucleus of a hydrogen bond is a closed-shell interaction, which comes with a positive ∇²ρ_BCP. On the other hand, a negative ∇²ρ_BCP indicates the shared-type interaction of the hydrogen bond, which is related with the concentration of the charge towards the hydrogen bond interaction line. The values of ∇²ρ_BCP in Tables 2, 3 and Tables S2 and S3 (ESI†) are all positive values. Thus, the hydrogen bonds in the [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes are all closed-shell interactions.

According to the literature, the sign of H_BCP can indicate the dominant electrostatic or covalent interactions in a hydrogen bond.⁷⁴ A positive H_BCP indicates that the electrostatic interaction is dominant in the hydrogen bond. In contrast, the covalent interaction is dominant in hydrogen bond with a negative H_BCP. All the H_BCP values in our work were positive, which illustrates that the hydrogen bonds in the [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ and 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes are electrostatically dominant.

According to the classification by Rozas et al.,⁷⁵ the hydrogen bond can be divided into three strengths: weak hydrogen bond (∇²ρ_BCP > 0 and H_BCP > 0), medium hydrogen bond (∇²ρ_BCP > 0 and H_BCP < 0) and strong hydrogen bond (∇²ρ_BCP < 0 and H_BCP < 0). Both ∇²ρ_BCP and H_BCP in Tables 2, 3 and Tables S2, S3 (ESI†) are positive, which implies that the hydrogen bonds in the [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complexes and 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes are weak strength.

4. Conclusions

In this work, we studied the structure and hydrogen-bond properties in four [CnMPyr][Tf₂N]–DMSO-d₆ (n = 3, 4, 6 and 8) binary mixtures using FTIR and DFT calculations. The v_s(C–D) region of the methyl groups in DMSO-d₆ was the focus and used to study the structure and hydrogen-bond properties of the four systems. The v_s(C–D) gradually shifted to a higher wavenumber with an increase in [CnMPyr][Tf₂N], which indicates that the hydrogen bond involving the methyl group of DMSO-d₆ is strengthened in the [CnMPyr][Tf₂N]–DMSO-d₆ mixtures compared with pure DMSO-d₆. In the excess spectra of v_s(C–D), a positive peak at a higher wavenumber and negative peak at a lower wavenumber were observed for the four systems, which indicate that the mixing process is non-ideal and interaction complexes form between [CnMPyr][Tf₂N] and DMSO-d₆. The excess peaks gradually shifted to a higher wavenumber, which illustrates that they are overlapped with the absorptions of different complexes. With the help of 2D-COS analysis, the overlapped peaks in the excess spectra were identified. The negative excess peak is composed of two peaks with wavenumbers of 2121 cm⁻¹ and 2124 cm⁻¹, and the positive excess peak is overlapped with two peaks at around 2135 cm⁻¹ and 2128 cm⁻¹. Using the DFT calculations, these peaks were assigned. From lower wavenumber to higher wavenumber, they are related with the absorptions of the DMSO dimer, DMSO monomer, ion pair-DMSO and ion cluster-DMSO, respectively. In the mixing processes, the species present in pure DMSO gradually decrease from DMSO dimer to DMSO monomer with an increase in ILs. Besides, the interaction complexes of IL–DMSO related with ion cluster-DMSO gradually increased, while the ion pair-DMSO decreased due to the strong electrostatic interaction between the cation and anion.

To study the hydrogen bond between DMSO and the four ILs, the most stable geometries of the 2[CnMPyr]⁺–2[Tf₂N]⁻–DMSO-d₆ complexes and the [CnMPyr]⁺–[Tf₂N]⁻–DMSO-d₆ complexes were carefully analysed. The ring hydrogen atoms of the methylene group directly attached to the nitrogen atom are the preferred interaction sites of the [CnMPyr]⁺ cations. All the hydrogen bonds in the identified complexes are closed-shell, electrostatically dominant and weak strength.

These studies on the structure and hydrogen bonding interactions between [CnMPyr][Tf₂N] and DMSO-d₆ provide in-depth information for understanding the physical properties of mixtures of [CnMPyr][Tf₂N] and DMSO-d₆. This is important for chemical applications and may also shed light for the study of other ionic liquids with cosolvents in the future, particularly when the identification of solution species is a concern.

Conflicts of interest

The authors declare no actual or potential conflicts of interests.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21703035), Hong Kong Scholar Program (XJ2018022), the National Science Fund for Excellent Young Scholars (21922813) and the Fujian Agriculture and Forestry University Foundation for excellent youth teachers (xjq201715).

References

1. K. R. Seddon, J. Chem. Technol. Biotechnol., 1997, 68, 351–356 CrossRef CAS .
2. A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–2502 CrossRef CAS .
3. R. Hagiwara and Y. Ito, J. Fluorine Chem., 2000, 7 Search PubMed .
4. R. Gomes de Azevedo, J. M. S. S. Esperança, V. Najdanovic-Visak, Z. P. Visak, H. J. R. Guedes, M. Nunes da Ponte and L. P. N. Rebelo, J. Chem. Eng. Data, 2005, 50, 997–1008 CrossRef .
5. D. Villemin1 and M. A. Didi, Orient. J. Chem., 2014, 29, 1267–1284 CrossRef .
6. M. L. Dietz and J. A. Dzielawa, Chem. Commun., 2001, 2124–2125 RSC .
7. J. Sayago, F. Soavi, Y. Sivalingam, F. Cicoira and C. Santato, J. Mater. Chem. C, 2014, 2, 5690–5694 RSC .
8. J. F. Wishart, Energy Environ. Sci., 2009, 2, 956–961 RSC .
9. R. D. Rogers and K. R. Seddon, Science, 2003, 302, 792–793 CrossRef .
10. P. J. Carvalho, V. H. Álvarez, B. Schröder, A. M. Gil, I. M. Marrucho, M. Aznar, L. M. N. B. F. Santos and J. A. P. Coutinho, J. Phys. Chem. B, 2009, 113, 6803–6812 CrossRef CAS .
11. T. Welton, Chem. Rev., 1999, 99, 2071–2084 CrossRef CAS .
12. N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37, 123–150 RSC .
13. H. Weingärtner, Angew. Chem., Int. Ed., 2008, 47, 654–670 CrossRef .
14. P. Bonhote, A. P. Dias, N. Papageorgiou, K. Kalyanasundaram and M. Grätzel, Inorg. Chem., 1996, 35, 1168–1178 CrossRef CAS .
15. K. Fumino, T. Peppel, M. Geppert-Rybczyńska, D. H. Zaitsau, J. K. Lehmann, S. P. Verevkin, M. Köckerling and R. Ludwig, Phys. Chem. Chem. Phys., 2011, 13, 14064–14075 RSC .
16. A. M. Smith, A. A. Lee and S. Perkin, Phys. Rev. Lett., 2017, 118, 096002 CrossRef .
17. V. Govinda, P. M. Reddy, P. Attri, P. Venkatesu and P. Venkateswarlu, J. Chem. Thermodyn., 2013, 58, 269–278 CrossRef CAS .
18. R. Rinaldi, Chem. Commun., 2011, 47, 511–513 RSC .
19. J.-M. Andanson, E. Bordes, J. Devémy, F. Leroux, A. A. H. Pádua and M. F. Costa Gomes, Green Chem., 2014, 16, 2528–2538 RSC .
20. A. Heintz, A review, J. Chem. Thermodyn., 2005, 37, 525–535 CrossRef CAS .
21. E. Gomez, B. Gonzalez, A. Dominguez, E. Tojo and J. Tojo, J. Chem. Eng. Data, 2006, 51, 696–701 CrossRef CAS .
22. A. Xu, Y. Zhang, Y. Zhao and J. Wang, Carbohydr. Polym., 2013, 92, 540–544 CrossRef CAS .
23. S. T. Keaveney, T. L. Greaves, D. F. Kennedy and J. B. Harper, J. Phys. Chem. B, 2016, 120, 12687–12699 CrossRef CAS .
24. G. Young, F. Nippgen, S. Titterbrandt and M. J. Cooney, Sep. Purif. Technol., 2010, 72, 118–121 CrossRef CAS .
25. G. Severa, G. Kumar and M. J. Cooney, Int. J. Chem. Eng., 2013, 2013, 617274 Search PubMed .
26. C. Manohar, T. Banerjee and K. Mohanty, J. Mol. Liq., 2013, 180, 145–153 CrossRef CAS .
27. Q. Yang, H. Xing, B. Su, K. Yu, Z. Bao, Y. Yang and Q. Ren, Chem. Eng. J., 2012, 181, 334–342 CrossRef .
28. K. Fumino, A. Wulf and R. Ludwig, Angew. Chem., Int. Ed., 2008, 47, 8731–8734 CrossRef CAS .
29. K. Fumino, A. Wulf and R. Ludwig, Phys. Chem. Chem. Phys., 2009, 11, 8790–8794 RSC .
30. K. Dong and S. Zhang, Chem. – Eur. J., 2012, 18, 2748–2761 CrossRef CAS .
31. A. Mele, C. D. Tran and S. H. D. P. Lacerda, Angew. Chem., Int. Ed., 2003, 42, 4364–4366 CrossRef CAS .
32. L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton, Phys. Chem. Chem. Phys., 2001, 3, 5192–5200 RSC .
33. R. Hayes, S. Imberti, G. G. Warr and R. Atkin, Angew. Chem., Int. Ed., 2013, 52, 4623–4627 CrossRef CAS .
34. Y. Zhao, X. Liu, J. Wang and S. Zhang, J. Phys. Chem. B, 2013, 117, 9042–9049 CrossRef CAS .
35. Y. Zhang, H. He, S. Zhang and M. Fan, ACS Omega, 2018, 3, 1823–1833 CrossRef CAS .
36. T. Köddermann, C. Wertz, A. Heintz and R. Ludwig, Angew. Chem., Int. Ed., 2006, 45, 3697–3702 CrossRef .
37. Y. Zhou, S. Gong, X. Xu, Z. Yu, J. Kiefer and Z. Wang, J. Mol. Liq., 2020, 299, 112159 CrossRef CAS .
38. L. Zhang, Z. Xu, Y. Wang and H. Li, J. Phys. Chem. B, 2008, 112, 6411–6419 CrossRef CAS .
39. T. Köddermann, C. Wertz, A. Heintz and R. Ludwig, ChemPhysChem, 2006, 7, 1944–1949 CrossRef .
40. Y.-Z. Zheng, Y. Zhou, G. Deng, R. Guo and D.-F. Chen, Spectrochim. Acta, Part A, 2020, 226, 117624 CrossRef CAS .
41. Y.-Z. Zheng, Y. Zhou, G. Deng, R. Guo and D.-F. Chen, Spectrochim. Acta, Part A, 2020, 226, 117641 CrossRef .
42. Q.-G. Zhang, N.-N. Wang, S.-L. Wang and Z.-W. Yu, J. Phys. Chem. B, 2011, 115, 11127–11136 CrossRef CAS .
43. B. A. Marekha, K. Sonoda, T. Uchida, T. Tokuda, A. Idrissi and T. Takamuku, J. Mol. Liq., 2017, 232, 431–439 CrossRef CAS .
44. J. Kiefer, M. M. Molina and K. Noack, ChemPhysChem, 2012, 13, 1213–1220 CrossRef CAS .
45. D. R. MacFarlane, P. Meakin, J. Sun, N. Amini and M. Forsyth, J. Phys. Chem. B, 1999, 103, 4164–4170 CrossRef CAS .
46. B. M. Quinn, Z. Ding, R. Moulton and A. J. Bard, Langmuir, 2002, 18, 1734–1742 CrossRef CAS .
47. S.-I. Hsiu, C.-C. Tai and I.-W. Sun, Electrochim. Acta, 2006, 51, 2607–2613 CrossRef CAS .
48. D. R. MacFarlane, J. Golding, S. Forsyth, M. Forsyth and G. B. Deacon, Chem. Commun., 2001, 1430–1431 RSC .
49. C. Samorì, T. Campisi, M. Fagnoni, P. Galletti, A. Pasteris, L. Pezzolesi, S. Protti, D. Ravelli and E. Tagliavini, ACS Sustainable Chem. Eng., 2015, 3, 1860–1865 CrossRef .
50. I.-W. Sun, H. P. Wang, H. Teng, S.-G. Su, Y.-C. Lin, C.-W. Kuo, P.-R. Chen and T.-Y. Wu, Int. J. Eelctrochem. Sci., 2012, 9748–9764 CAS .
51. O. U. Ahmed, F. S. Mjalli, T. Al-Wahaibi, Y. Al-Wahaibi and I. M. AlNashef, Ind. Eng. Chem. Res., 2015, 54, 6540–6550 CrossRef CAS .
52. P. F. Requejo, E. Gómez, N. Calvar and Á. Domínguez, Ind. Eng. Chem. Res., 2015, 54, 1342–1349 CrossRef CAS .
53. Z.-W. Yu and P. J. Quinn, Biosci. Rep., 1994, 14, 259–281 CrossRef CAS .
54. D. Kasprzak, E. Krystkowiak, I. Stępniak and M. Galiński, Eur. Polym. J., 2019, 113, 89–97 CrossRef CAS .
55. A. Khan and C. Zhao, Electrochem. Commun., 2014, 49, 1–4 CrossRef CAS .
56. Y.-Z. Zheng, Z.-R. Shen, Y. Zhou, R. Guo and D.-F. Chen, J. Chem. Thermodyn., 2019, 131, 441–448 CrossRef CAS .
57. Y.-Z. Zheng, H.-Y. He, Y. Zhou and Z.-W. Yu, J. Mol. Struct., 2014, 1069, 140–146 CrossRef CAS .
58. T. Takamuku, T. Tokuda, T. Uchida, K. Sonoda, B. A. Marekha, A. Idrissi, O. Takahashi, Y. Horikawa, J. Matsumura, T. Tokushima, H. Sakurai, M. Kawano, K. Sadakane and H. Iwase, Phys. Chem. Chem. Phys., 2018, 20, 12858–12869 RSC .
59. T. Takamuku, H. Hoke, A. Idrissi, B. A. Marekha, M. Moreau, Y. Honda, T. Umecky and T. Shimomura, Phys. Chem. Chem. Phys., 2014, 16, 23627–23638 RSC .
60. Q. Li, N. Wang, Q. Zhou, S. Sun and Z. Yu, Appl. Spectrosc., 2008, 62, 166–170 CrossRef CAS .
61. Q. Li, G. Wu and Z. Yu, J. Am. Chem. Soc., 2006, 128, 1438–1439 CrossRef CAS .
62. I. Noda, Appl. Spectrosc., 1993, 47, 1329–1336 CrossRef CAS .
63. I. Noda and Y. Ozaki, Two-Dimensional Correlation Spectroscopy: Applications in Vibrational and Optical Spectroscopy, John Wiley & Sons, 2005 Search PubMed .
64. W. N. Hansen, Spectrochim. Acta, 1965, 21, 815–833 CrossRef CAS .
65. N. Wang, Q. Jia, Q. Li and Z. Yu, J. Mol. Struct., 2008, 883–884, 55–60 CrossRef CAS .
66. Z.-W. Yu, L. Chen, S.-Q. Sun and I. Noda, J. Phys. Chem. A, 2002, 106, 6683–6687 CrossRef CAS .
67. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, Gaussian 16 Rev. C.01, Wallingford, CT, 2016 Search PubMed .
68. R. F. W. Bader, Atoms in Molecules: A Quantum Theory, Oxford University Press, Oxford, New York, 1994 Search PubMed .
69. T. Lu and F. Chen, J. Comput. Chem., 2012, 33, 580–592 CrossRef CAS .
70. W. D. Horrocks and F. A. Cotton, Spectrochim. Acta, 1961, 17, 134–147 CrossRef CAS .
71. L. Pauling, The Nature of the Chemical Bond and the Structure of Molecules and Crystals: An Introduction to Modern Structural Chemistry, Cornell University Press, 1960 Search PubMed .
72. U. Koch and P. L. A. Popelier, J. Phys. Chem., 1995, 99, 9747–9754 CrossRef CAS .
73. H. Roohi, A.-R. Nowroozi and E. Anjomshoa, Comput. Theor. Chem., 2011, 965, 211–220 CrossRef CAS .
74. L. F. Pacios, J. Phys. Chem. A, 2004, 108, 1177–1188 CrossRef CAS .
75. I. Rozas, I. Alkorta and J. Elguero, J. Am. Chem. Soc., 2000, 122, 11154–11161 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d0cp03640d

This journal is © the Owner Societies 2020