Effective Absorption of Dichloromethane Using Carboxyl-Functionalized Ionic Liquids

Abstract

Dichloromethane (DCM) is recognized as a very harmful air pollutant because of its strong volatility and difficulty to degrade. Ionic liquids (ILs) are considered as potential solvents for absorbing DCM, while it is still a challenge to develop ILs with high absorption performances. In this study, four carboxyl-functionalized ILs—trioctylmethylammonium acetate [N₁₈₈₈][Ac], trioctylmethylammonium formate [N₁₈₈₈][FA], trioctylmethylammonium glycinate [N₁₈₈₈][Gly], and trihexyl(tetradecyl)phosphonium glycinate [P₆₆₆₁₄][Gly]—were synthesized for DCM capture. The absorption capacity follows the order of [P₆₆₆₁₄][Gly] > [N₁₈₈₈][Gly] > [N₁₈₈₈][FA] > [N₁₈₈₈][Ac], and [P₆₆₆₁₄][Gly] showed the best absorption capacity, 130 mg DCM/g IL at 313.15 K and a DCM concentration of 6.1%, which was two times higher than the reported ILs [Beim][EtSO₄] and [Emim][Ac]. Moreover, the vapor–liquid equilibrium (VLE) of the DCM + IL binary system was experimentally measured. The NRTL (non-random two-liquid) model was developed to predict the VLE data, and a relative root mean square deviation (rRMSD) of 0.8467 was obtained. The absorption mechanism was explored via FT-IR spectra, ¹H-NMR, and quantum chemistry calculations. It showed a nonpolar affinity between the cation and the DCM, while the interaction between the anion and the DCM was a hydrogen bond. Based on the results of the study of the interaction energy, it was found that the hydrogen bond between the anion and the DCM had the greatest influence on the absorption process.

1. Introduction

Dichloromethane (DCM) is an excellent organic solvent [1,2] which is widely used in film, metal manufacturing, and pharmaceutical production fields. The annual world production of DCM exceeds 5 × 10⁵ tons [3], and due to its highly volatile nature, approximately 77% of the emitted DCM is released into the atmosphere [4]. The release of DCM into the atmosphere affects the environment by producing highly toxic phosgene and carbon monoxide and destroying the ozone layer [5]. DCM also poses a serious threat to human health [6] particularly as risk of cancer [7]. Therefore, it is urgent to develop DCM recovery technology.

To date, various recovery methods for DCM have been developed, such as the adsorption [8,9,10,11] and absorption methods [12,13,14,15]. Nevertheless, it cannot be ignored that the adsorption method has the problems of low separation selectivity and limited adsorption capacity of the adsorbent [16]. The absorption method is considered a promising DCM capture technology due to its good separation selectivity, the applicability of the gas source, and its simple process. At present, silicone oil and several organic solvents, such as bis(2-ethylhexy) phthalate (DEHP), bis(2-ethylhexy) adipate (DEHA) [17], and polydimethylsiloxane (PDMS) [18], are used to capture DCM. However, these absorbents showed low capacity for soaking up DCM compared with other absorbents, which makes high solvent consumption achieve deep removal. Moreover, organic solvents have volatile properties, resulting in resource waste. It is of great importance to search for and develop more efficient, greener, and less volatile absorbents to overcome the above problems.

ILs have various favorable characteristics, such as low vapor pressure [19,20,21] and designable structure, and have attracted much attention in various fields. For the past few years, many studies have been carried out on the absorption of DCM by ILs, and some progress has been made. Castillo et al. [22] determined the Henry’s law constants (K_H) of DCM in 23 kinds of hydrophobic ILs at 298 K and infinite dilution by the static headspace method, and found that the K_H of DCM in ILs was 14–50 times lower than that in water. Wu et al. [23] synthesized six ILs with imidazolium-based cation and one IL with pyridyl-based cations as absorbents to capture DCM. Among them, [Bmim][SCN] had an excellent absorption capacity of 1.46 g/g at 303.15 K and DCM concentration of 60% (volume fraction). According to Shi et al. [24], the most remarkable DCM absorption was obtained by using [Beim][EtSO₄] with a capacity 0.426 g/g under an absorption temperature of 313.15 K and a DCM concentration of 37.7%. Gui et al. [14] investigated the removal ratio of DCM by using [Emim][Ac], and the results demonstrated that the removal ratio could reach 91.82% under 20 °C and ambient pressure with a gas flow rate and IL flow rate of 1 L/min and 20 mL/min, respectively. It was also pointed out that the strong interaction between anion [Ac]⁻ and DCM plays the main role in the separation ability of ILs. In the study of DCM absorption with ILs, Wu et al. [25] established a framework for multiscale research based on the COSMO-RS model. Among the screened ILs, [EMPIP][Ac] was found to be the most effective IL for absorption of DCM, and the K_H was 3.29 kPa. They also proposed that noncyclic cations, such as quaternary amine and quaternary phosphonium bases, were more conducive to DCM absorption. Moreover, the K_H values of other kinds of anions were relatively higher than those of carboxylate-based anions.

It can be seen from the above research that ILs have good performance in the capture of DCM. However, the absorption capacity of DCM is still not very high and needs to be further improved. At present, most reported ILs are conventional ILs with short-chain cyclic cations, such as imidazolium-based cations. In fact, noncyclic cations, such as quaternary ammonium and quaternary phosphonium and carboxylate-based anions have low K_H values, according to the simulation results from Wu et al. [25]. Therefore, four carboxyl-functionalized ILs were synthesized in this study: [N₁₈₈₈][Ac] (trioctylmethylammonium acetate), [N₁₈₈₈][FA] (trioctylmethylammonium formate), [N₁₈₈₈][Gly] (trioctylmethylammonium glycinate), and [P₆₆₆₁₄][Gly] (trihexyl(tetradecyl)phosphonium glycinate), respectively. Several factors, such as IL structure and absorption temperature, were studied to evaluate their impact on absorption performance. Then the vapor–liquid equilibrium (VLE) experiment of DCM + IL systems was carried out, and a comparison of the experimental data and the data predicted by the NRTL model was conducted to verify whether this model was suitable for the system in this study. Finally, spectroscopy analysis and quantum chemical calculations were performed to reveal the absorption mechanism.

2. Materials and Methods

2.1. Materials

Nitrogen (N₂, >99.99%) was purchased from Henan Yuanzheng Special Gas Co., Ltd., (Henan, China). Trioctylmethylammonium chloride ([N₁₈₈₈][Cl], 97%) was purchased from Shanghai Maclin Biochemical Technology Co., Ltd., (Shanghai, China). Trihexyl(tetradecyl)phosphonium chloride ([P₆₆₆₁₄][Cl], 97%) was purchased from Jiangsu Aikang Biomedical Research and Development Co., Ltd., (Jiangsu, China). Anhydrous ethanol (≥99.7%) was obtained from Wuxi Yatai United Chemical Co., Ltd., (Wuxi, China). The following chemicals were also used in this study: DCM (≥99.5%), sodium acetate (99%), sodium formate (≥98.5%), glycine (99%), and Ambersep 900(OH) ion exchange resin, which were all purchased from Shanghai Titan Technology Co., Ltd., (Shanghai, China). All the chemicals used in this study are listed in Supporting Information (Table S3).

2.2. Synthesis of ILs

[N₁₈₈₈][Gly] and [P₆₆₆₁₄][Gly] were prepared via the anion exchange method, as reported in the literature [26,27]. Firstly, a solution of [N₁₈₈₈]Cl in ethanol flowed through a chromatography column containing anion exchange resin to obtain the [N₁₈₈₈]OH-ethanol solution. The flow rate was controlled to ensure the complete exchange of chloride. Then 1.1 mol equivalents glycine was added to the solution flowing down the chromatography column. The mixed solution was stirred with using a magnetic heating agitator for 24 h at a temperature of 313.15 K. Then the ethanol was moved at 333.15 K under vacuum and acetonitrile was used to remove unreacted glycine. Finally, the obtained light yellow viscous liquid was dried for 48 h in vacuum at 343.15 K. [P₆₆₆₁₄][Gly] was synthesized using the same method.

[N₁₈₈₈][Ac] and [N₁₈₈₈][FA] can also be synthesized by using the above method. However, since the anion exchange method was complicated and difficult to use on a large scale, the two ILs were prepared using the ion exchange method. This synthesis route was relatively simple, and the generated byproduct NaCl can be easily removed during preparation [28]. The detailed synthesis method was as follows: firstly, sodium acetate (0.077 mol) was added to the ethanol solution of [N₁₈₈₈][Cl] (0.07 mol). The reaction was stirred for 24 h at 323.15 K in water. Afterward, a sand core funnel was used to remove the byproduct NaCl. The ethanol was removed at 333.15 K. Finally, IL was dried for 48 h at 343.15 K. A pale yellow viscous liquid [N₁₈₈₈][Ac] was obtained. In addition, [N₁₈₈₈][FA] was synthesized similarly. The structure of these four ILs was shown in Figure 1, and the general preparation route can be found in Figure S1.

2.3. Characterization of ILs

The ¹H-NMR spectra of the ILs were measured by using a Bruker 600 spectrometer with deuterated dimethyl sulfoxide (DMSO˗d₆) as the solvent. FT-IR spectra in the range of 500~4000 cm⁻¹ were obtained by using a Fourier transform infrared spectrometer (Thermo Fisher Scientific, MA, USA). The thermal decomposition of the ILs was determined by using TGA Q5000 V3.15 with a heat rate of 10 K/min under nitrogen atmosphere.

The density of the ILs was measured using an Anton Paar DMA 5000 M densitometer with a standard uncertainty of 0.000007 g∙cm⁻³, and the Anton Paar Lovis 2000 ME micro viscometer was used to measure the viscosity of ILs with an uncertainty of 0.5%. The experiments were conducted at atmospheric pressure and temperature from 313.15 K to 353.15 K with intervals of 10 K. The mass fraction of the water in the ILs was determined using a Karl Fischer moisture meter (Model C20s, Mettler Toledo Switzerland, Greifensee, Switzerland), and the water contents of these four ILs were lower than 2000 ppm. The detailed information can be found in Figures S2–S11 and Tables S1 and S2.

2.4. Apparatus and Procedure of Gas Absorption

The absorption experiment was performed using the bubble method. During the experiment, as carrier gas, nitrogen flowed into a three-necked flask containing DCM. The temperature of the condensing tube was adjusted by a cryostat cooler, and stable DCM vapor with different concentrations was obtained by adjusting the temperature of the condensing tube. The DCM gas concentration was analyzed using a gas chromatograph (GC-7960 plus, Tengzhou Allen Analytical Instrument Co., Ltd., Zaozhuang, China) equipped with a packed column (type: DNP, size 3 m × 3 mm). At the beginning of each experiment, the IL sample (about 5.0 g) was weighed using an electronic analytical balance (Type PL403, Mettler Toledo Switzerland) with a resolution of 0.0001 g [29] and placed in an absorption bottle. The absorption bottle was put in a large beaker of water, and the water level was always kept higher than the IL level in the absorption bottle. The temperature was controlled by a magnetic heating stirrer. The exhaust gas passing through the IL entered an ethanol absorption bottle and absorbed the remaining DCM. When the concentration of the DCM gas in the inlet and outlet of the absorption bottle was nearly identical, the experiment was considered to have reached absorption equilibrium. The saturated absorption capacity (S_AC, unit mg/g) of the IL at different DCM concentrations and absorption temperatures was measured experimentally with an electronic balance with 0.0001 g precision. In addition, S_AC can be calculated from Equation (1):

$$S_{AC}=\frac{m_3-m_2}{m_2-m_1}\times 1000$$

(1)

where m₁ represents the mass of the empty absorption bottle, m₂ represents the total mass of the absorption bottle and IL before absorption, and m₃ represents the total mass of the absorption bottle and of the IL after absorption. Each experiment was carried out three times, and the average value of the three experiments was taken as the final experimental result. The experimental procedure is shown in Figure S12.

2.5. VLE Experiments

The vapor pressures of the DCM + IL binary system at different DCM concentrations and temperatures were measured by a VLE experimental device (Type DPCY-6C, Jiangsu Nanjing Nanda Wanhe Technology Co., Ltd., Nanjing, China), as shown in Figure S13.

Before the experiment, the IL were poured into a 500 mL beaker and then dried under vacuum for 12 h under the set conditions. Later, the pretreated IL was sealed for use. When performing this experiment, the airtightness of the experimental device should first be examined, and then the condensing device should be turned on. In our study, approximately 18 mL of the mixed solution of DCM + IL binary system was added to the U-shaped tube. Then we turned on the vacuum pump to vacuum the system, discharging the air between the liquid in the glass ball and the liquid in the U-shaped tube. After pumping air for a while, the liquid in the U-balance tube bulged out upward in a bubble state. When the bubbling had lasted for several minutes, we closed the atmospheric valve and connected the vacuum pump valve, and the air was considered clean. Then we opened the valve of the atmosphere slowly and let a small amount of air into the system, until the liquid level of both arms of the U-shaped tube was equal and maintained for about 20 min, which can be considered to reach VLE at this time. Finally, the value of the pressure indicator, atmospheric pressure, and temperature of the thermostatic water bath was recorded. The temperature reading of the experimental apparatus was accurate to 0.01 K, and the pressure was accurate to 0.01 kPa. Two repetitions of each experiment were performed, and the average value of the two experiments was taken as the final experimental result.

2.6. Computational Methods

Gaussian 09, revision D.01 [30] was used for quantum chemistry calculations on the basis of density functional theory (DFT). First, Gaussview was used to generate the initial structure of the DCM as well as the anion and cation of ILs, then the geometric optimization of DCM, ILs, and their complex was carried out under the B3LYP/6-311+G(d, p) [31,32] basis set. To improve the accuracy of dispersion effects [33], Grimme’s DFT-D3 dispersion correction was applied. The potential energy of the structures was at a minimum after optimizing, and the frequency check did not identify an imaginary frequency. Finally, the interaction energy of IL with DCM was calculated using Equation (2) [34]:

$$\Delta E\left(\mathrm{kJ}·{\mathrm{mol}}^{-1}\right)={\mathrm{E}}_{\mathrm{IL}-\mathrm{DCM}}-{\mathrm{E}}_{\mathrm{IL}}-{\mathrm{E}}_{\mathrm{DCM}}+{\mathrm{E}}_{\mathrm{BSSE}}$$

(2)

where E_IL−DCM signifies the energy of complex of IL and DCM, E_IL and E_DCM denote the energy of IL and DCM, respectively, and E_BSSE stands for the energy correct of basis set superposition error (BSSE) by the counterpoise method [35].

3. Results and Discussion

3.1. DCM Capture with Carboxyl-Functionalized ILs

3.1.1. Effect of IL Structure on Absorption Capacity

The saturated absorption capacity in four carboxyl-functionalized ILs, including [N₁₈₈₈][Ac], [N₁₈₈₈][FA], [N₁₈₈₈][Gly], and [P₆₆₆₁₄][Gly], was investigated under a gas concentration of 6.1% and an atmospheric pressure of 313.15 K, as shown in Figure 2a. The experimental results followed the order of [P₆₆₆₁₄][Gly] (130.62 mg/g) > [N₁₈₈₈][Gly] (101.42 mg/g) > [N₁₈₈₈][FA] (89.69 mg/g) > [N₁₈₈₈][Ac] (86.49 mg/g). In order to evaluate the absorption performance of DCM by the ILs designed in this work, we compared two ILs [Emim][Ac] [14] and [Beim][EtSO₄] [24] that have been reported to have better absorption of DCM. The absorption capacity of [Emim][Ac] and [Beim][EtSO₄] was tested under the same conditions in this study, as shown in Figure 2b. Among the four ILs, it is noteworthy that the absorption capacity of [P₆₆₆₁₄][Gly] was more than two times that of [Beim][EtSO₄] and [Emim][Ac].

For the strong hydrophobic cations [N₁₈₈₈]⁺ and [P₆₆₆₁₄]⁺, the absorption capacity of [P₆₆₆₁₄][Gly] not only significantly improved the absorption capacity as compared to [N₁₈₈₈][Gly] but also reduced the time to reach absorption saturation. Nonpolar affinity [25] dominates the interaction between the cations and DCM, which may be the result of the higher nonpolar affinity between [P₆₆₆₁₄]⁺-DCM than [N₁₈₈₈]⁺-DCM. In addition, it is possible that increasing the alkyl chain length of the cations increased the steric hindrance between the ions, making it easier for the DCM molecules to approach the anions. Simultaneously, comparing [N₁₈₈₈][Ac] with [Emim][Ac], due to the nonpolar affinity of DCM with noncyclic cations, it exhibited higher absorption capacity for DCM than aromatic cations. [N₁₈₈₈][FA] had better absorption performance than [N₁₈₈₈][Ac], indicating that the length of the anionic alkyl chain was not conducive to DCM absorption, which verified the simulation calculation results of Wu et al. [25]. Under the same experimental conditions, it was clear from the results that among the three anions studied, [N₁₈₈₈][Gly] had the greatest DCM absorption capacity. The reason may be that [Gly]⁻ contains both amino and carboxyl groups, providing more active sites than [FA]⁻ and [Ac]⁻ [36]. These experimental results were further verified by quantum chemistry calculations.

3.1.2. Effect of Temperature and Gas Concentration of DCM on the Absorption Capacity

The absorption experiments of [P₆₆₆₁₄][Gly] at different temperatures were conducted under the condition of 6.1% DCM concentration. As shown in Figure 3a, the absorption capacity of [P₆₆₆₁₄][Gly] decreased obviously as the temperature increased, which is also the common trend of most gas absorption; that is, it had higher solubility at low temperatures. In detail, the absorption capacity decreased from 130.62 to 60.29 mg/g, and the time for DCM to reach absorption saturation was shortened from 6 to 2 h as the temperature increased from 313.15 to 353.15 K. The results demonstrated that the absorption process was significantly impacted by temperature. Theoretically, the viscosity of [P₆₆₆₁₄][Gly] decreased when the temperature increased (Figure S10), the mass transfer rate accelerated [37], and the time of DCM to reach absorption saturation was shortened.

Figure 3b displays the change in the absorption capacity of [P₆₆₆₁₄][Gly] as a function of DCM gas concentration, and the gas concentrations were 4.2%, 6.1%, and 8.1%. At the absorption temperature of 313.15 K, when the inlet concentration increased from 4.2% to 8.1%, the DCM absorption capacity of [P₆₆₆₁₄][Gly] increased from 100.71 to 144 mg/g. Actually, in the absorption process with a specific absorption temperature, the partial pressure of DCM gas on the side of the gas film will increase when the inlet vapor concentration increases, and increasing the concentration gradient between gas and liquid phases boosts the mass transfer force [12]. This is conducive to the absorption of DCM in the ILs. According to the two-film theory [38], the higher gas concentration provides a higher force during the gas–liquid absorption process, and this effect facilitates the absorption of DCM in [P₆₆₆₁₄][Gly].

3.1.3. Absorption-Desorption Cycles of [P₆₆₆₁₄][Gly]

In the practical application of absorbents, the regeneration performance is directly related to the operation cost and equipment investment.

According to previous reports in the literature, a higher temperature is beneficial to gas desorption from ILs [39]. In this study, since there was only DCM, the absorbed ILs were regenerated by passing nitrogen. The saturated absorbent was blown off with 100 mL/min N₂ [40] for 10 h at 373.15 K under laboratory conditions. Different methods were used to obtain the mass change in the absorbent before and after desorption. When the mass of the absorbent before and after desorption was almost the same, the absorbent was considered to be desorbed completely. It can be seen from Figure 4a, the absorption capacity had no apparent loss after five cycles of absorption/desorption on [P₆₆₆₁₄][Gly], confirming that it had outstanding thermostability and could be reused. The FT-IR spectra of the regenerated [P₆₆₆₁₄][Gly] and fresh [P₆₆₆₁₄][Gly] are shown in Figure 4b. The FT-IR spectra of the regenerated [P₆₆₆₁₄][Gly] was consistent with the fresh IL, which proved that its structure did not change after recycling.

3.2. VLE of ILs and DCM

3.2.1. Reliability of Experimental Equipment

First of all, a check of the experimental device was carried out. Then the VLE experiment of DCM (1) + [Emim][Ac] (2) (x₁ = 0.9) was carried out, and the experimental values were compared with the data reported in the literature [14]. The comparison of vapor pressure values is shown in Figure S14, and it can be intuitively seen from the Figure that there is no significant deviation between the experimental values and the literature data, indicating that the device can be used to accurately measure the VLE data of the DCM + IL mixed solutions.

3.2.2. Vapor Pressure Data of DCM + ILs Binary System

The vapor pressure of four binary systems DCM + [N₁₈₈₈][Ac], DCM + [N₁₈₈₈][FA], DCM + [N₁₈₈₈][Gly] and DCM + [P₆₆₆₁₄][Gly] at different DCM molar fractions and temperatures were tested as illustrated in Figure 5. An increase in the value of the vapor pressure accompanies the rise in temperature. The larger the molar fraction of IL, the smaller the vapor pressure values at the same temperature. At the same DCM molar fraction and temperature, the vapor pressures of the binary systems are as follows: DCM + [P₆₆₆₁₄][Gly] < DCM + [N₁₈₈₈][Gly] < DCM + [N₁₈₈₈][FA] < DCM + [N₁₈₈₈][Ac], further indicating that [P₆₆₆₁₄][Gly] has the tremendous potential for DCM absorption. The VLE data of the DCM + IL binary system were predicted on the basis of the NRTL model (Supporting Information Section S3).

Table 1. The binary NRTL parameters fitted.

Binary System	α12	g12–g22	g21–g11	ARD	rRMSD
		J·mol−1	J·mol−1
DCM + [N1888][Ac]	0.0686	−23,396.2	13,497.93	0.0081	0.6620
DCM + [N1888][FA]	0.9900	210.1503	−264.229	0.0105	0.7840
DCM + [N1888][Gly]	0.9900	274.86	−342.058	0.0121	0.8150
DCM + [P66614][Gly]	0.9900	358.2333	−422.142	0.0152	1.0900

summarizes the fitted parameters; the average relative deviation (ARD) [41,42] and rRMSD [43] are defined as Equations (3) and (4), respectively.

$$\mathrm{ARD}\left(\mathrm{P}\right)=\left(\frac{1}{\mathrm{n}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}\left|p^{\mathit{NRTL}}-p^{\mathit{exp}}\right|{/p}^{\mathit{exp}}\right)$$

(3)

$$\mathrm{rRMSD}\left(\mathrm{P}\right)=\sqrt{\frac{1}{\mathrm{n}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\left(\frac{p^{\mathit{exp}}-p^{\mathit{NRTL}}}{p^{\mathit{exp}}}\right)}^2}$$

(4)

The experimental vapor pressure data and the data predicted by the NRTL model were summarized in (Tables S4–S7) with an ARD of less than 2% and an overall rRMSD of 0.8467. These results indicate that the NRTL model was suitable for the DCM + IL binary system in this study, and provides an excellent thermodynamic method for designing and simulating the DCM absorption process.

3.3. Mechanism of DCM Absorption

3.3.1. FT-IR and ¹H-NMR Analysis

The interactions between the DCM and the [P₆₆₆₁₄][Gly] were studied by the FT-IR spectra and ¹H-NMR. The FT-IR spectra of the DCM, fresh [P₆₆₆₁₄][Gly] and [P₆₆₆₁₄][Gly] after absorbing DCM, were recorded and are displayed in Figure 6a. It can be seen that a new absorption peak at 756 cm⁻¹ was observed in the FT-IR spectra after the DCM absorption compared with the fresh [P₆₆₆₁₄][Gly], which was the C-Cl stretching vibration peak in DCM. The infrared spectrum also showed that the stretching vibration peak of C=O in the [Gly]⁻ shifted from 1574 cm⁻¹ to 1595 cm⁻¹, indicating that there was a C-H∙∙∙O hydrogen bonding interaction [44,45] between the [Gly]⁻ and the DCM. At the same time, there may be a C-H∙∙∙Cl hydrogen bonding interaction [24], as the bending vibration peak of -CH₂ in the [P₆₆₆₁₄]⁺ at 1305 cm⁻¹ moved to 1309 cm⁻¹.

To further understand the absorption mechanism, the ILs before and after the absorption of the DCM were characterized by ¹H-NMR spectra. From Figure 6b, it can be seen that the peak corresponding to the DCM appeared at 5.76 ppm, and there was no evident chemical shift in the ¹H-NMR spectra of [P₆₆₆₁₄][Gly] before and after the DCM absorption, which means that no new substance was formed after the [P₆₆₆₁₄][Gly] absorption of DCM.

3.3.2. Model and Calculation Section

Effect of Anion and Cation Structures on DCM Absorption

The affinity strength between the molecules was analyzed utilizing the σ-profile obtained from the COSMO model in this work. Since the [Gly]⁻, [N₁₈₈₈]⁺, and [P₆₆₆₁₄]⁺ are not included in the built-in database of COSMOthermX Version 19.0.4 [14], the optimized anions were imported into the COSMObase through Gaussian 09, revision D.01. Figure 7 shows the σ-profiles of [Ac]⁻, [FA]⁻, [Gly]⁻, [N₁₈₈₈]⁺, [P₆₆₆₁₄]⁺, and DCM. The σ-profile consists of three distinct regions: the hydrogen bond donor (HBD) region (σ < −0.0084 e/Ų), the nonpolar region (−0.0084 e/Ų < σ < 0.0084 e/Ų), and the hydrogen bond acceptor (HBA) region (σ > 0.0084 e/Ų) [46].

Most peaks of the σ-profile are in the nonpolar region for DCM, and it has a high attraction for other nonpolar components. A faint peak can be seen in the HBD region in addition to the nonpolar zone, suggesting that the hydrogen bonds may be formed between the DCM and the other HBA. It can be seen from Figure 7 that the nonpolar area is the main distribution of the σ-profile of [N₁₈₈₈]⁺ and [P₆₆₆₁₄]⁺, suggesting that the interaction is mainly a nonpolar affinity between the cations and the DCM. It also can be seen that the peak of [P₆₆₆₁₄]⁺ was higher than that of [N₁₈₈₈]⁺, indicating that the nonpolar affinity between [P₆₆₆₁₄]⁺-DCM is higher than that of [N₁₈₈₈]⁺-DCM. For the σ-profile of anions [Ac]⁻, [FA]⁻, and [Gly]⁻, they are mainly in the HBA region and the nonpolar region, while the strong peak of the σ-profile appears around the 0.02 e/Ų HBA region, indicating that the interactions between the anions and the DCM are mainly hydrogen bonds.

The calculated results of the interaction energy of the anion and cation of the ILs with the DCM are shown in Figure 8, and the optimized structures are presented in Figure S16. The interaction energies of [N₁₈₈₈]⁺-DCM and [P₆₆₆₁₄]⁺-DCM were −37.26 kJ∙mol⁻¹ and −37.73 kJ∙mol⁻¹, respectively, while the interaction energies of [Ac]⁻-DCM, [FA]⁻-DCM, and [Gly]⁻-DCM were −78.67 kJ∙mol⁻¹, 76.64 kJ∙mol⁻¹, and −74.81 kJ∙mol⁻¹, respectively. The distances of O∙∙∙H were 2.10 Å, 2.25 Å, 2.19 Å, 2.19 Å, 2.26 Å, and 2.12 Å. These bond distances were greater than the covalent bond length of O-H (0.96 Å) and shorter than the Van der Waals distance between atom O and atom H (2.72 Å), which was within acceptable standards for a hydrogen bond [47]. The distances of Cl∙∙∙H were around 2.9 Å for cation-DCM systems, which also falls into the range 2.7 Å–3.0 Å of C-H∙∙∙Cl hydrogen bond [48], suggesting a weak interaction between the cation and the DCM. In general, the interaction energy of the anion–DCM was more negative than that of the cation–DCM, which indicates that the anion played a crucial role in the absorption of the DCM.

Interaction Energy Analysis

The optimized structures of the [P₆₆₆₁₄][Gly]-DCM, [N₁₈₈₈][Gly]-DCM, [N₁₈₈₈][FA]-DCM, and [N₁₈₈₈][Ac]-DCM are illustrated in Figure 9.

Table 2. Bond distance, bond angle, and interaction energy of ILs–DCM.

IL-DCM	Atomic Number	Bond Distance (Å)	Bond Angle (deg)	ΔE/(kJ/mol)
[N1888][Ac]-DCM	88C-92H···84O	2.14	149.39	−55.37
	88C-92H···83O	2.68	107.17
	56C-59H∙∙∙90Cl	2.87	128.22
[N1888][FA]-DCM	85C-88H···83O	2.18	147.53	−58.58
	32C-36H···86Cl	2.82	162.71
	56C-59H···86Cl	2.87	129.49
[N1888][Gly]-DCM	90C-93H···84O	2.10	152.83	−62.24
	90C-93H···85O	2.69	109.85
	32C-36H···91Cl	2.81	161.45
	56C-59H···91Cl	2.87	128.35
[P66614][Gly]-DCM	111C-115H···106O	1.91	171.15	−78.28
	3C-36H···113Cl	2.89	162.93

lists the interaction energies of the IL-DCM as well as other information. The bond angle ∠X-H∙∙∙Y was larger than 90° [49], and the hydrogen bonds were within the accepted norm. There was a C-H∙∙∙O hydrogen bond interaction in [P₆₆₆₁₄][Gly], and the bond distance and bond angle was 1.91 Å and 171.15°, respectively. In addition to the hydrogen bond with [Gly]⁻ in the anion, a weaker hydrogen bond also formed between the DCM and [P₆₆₆₁₄]⁺, which was C-H∙∙∙Cl (2.89 Å). As shown in Table 2, the interaction energy of the IL-DCM follows the order of [P₆₆₆₁₄][Gly]-DCM(−78.28 kJ/mol) > [N₁₈₈₈][Gly]-DCM(−64.24 kJ/mol) > [N₁₈₈₈][FA]-DCM(−58.58 kJ/mol) > [N₁₈₈₈][Ac]-DCM(−55.37 kJ/mol). The magnitude of the interaction energy indicates the strength of the interaction between the DCM and the absorbents. The intensity of the interaction between the absorbents and the DCM was positively correlated with the absorption capacity. In general, [P₆₆₆₁₄][Gly]-DCM has the largest interaction energy among the four complexes, which explains the results of the absorption experiments.

4. Conclusions

In this study, four carboxyl-functionalized ILs were synthesized, and their absorption capacity were measured by a set of absorption experiments. The results showed that the [P₆₆₆₁₄][Gly] exhibited a higher absorption capacity of 130.62 mg DCM/g IL at 313.15 K and 6.1% DCM concentration than the reported ILs [Emim][Ac] and [Beim][EtSO₄]. Subsequently, the vapor pressure data of the DCM + IL binary systems were measured by a VLE experimental device and predicated by the NRTL model. According to the fitted results, the model worked adequately for the VLE of these binary systems. The order of the vapor pressure of the DCM + IL binary system under the same conditions was DCM + [P₆₆₆₁₄][Gly] < DCM + [N₁₈₈₈][Gly] < DCM + [N₁₈₈₈][FA] < DCM + [N₁₈₈₈][Ac]. In addition, the [P₆₆₆₁₄][Gly] exhibited good recyclability, with the absorption capacity showing no significant decrease after five cycles of absorption–desorption. The absorption mechanism was identified through FT-IR, ¹H-NMR and quantum chemistry calculations, indicating that the hydrogen bond played a significant role in the absorption of DCM by [P₆₆₆₁₄][Gly]. In conclusion, [P₆₆₆₁₄][Gly] was found to be a promising, efficient, and green absorbent for DCM.