One-Pot, Metal-Free Synthesis of Dimethyl Carbonate from CO₂ at Room Temperature

Abstract

Herein, we report on the metal-free, one-pot synthesis of industrially important dimethyl carbonate (DMC) from molecular CO₂ under ambient conditions. In this process, initially the CO₂ was chemisorbed through the formation of a switchable ionic liquid (SIL), [DBUH] [CH₃CO₃], by the interaction of CO₂ with an equivalent mixture of organic superbase 1,8-diazabicyclo-[5.4.0]-undec-7-ene (DBU) and methanol. The obtained SIL further reacted with methyl iodide (CH₃I) to form DMC. The synthesis was carried out in both dimethyl sulfoxide (DMSO) and methanol. Methanol is preferred, as it not only served as a reagent and solvent in CO₂ capture and DMC synthesis, but it also assisted in controlling the side reactions between chemical species such as CH₃I and [DBUH]⁺ cation and increased the yield of DMC. Hence, the use of methanol avoided the loss of captured CO₂ and favored the formation of DMC with high selectivity. Under the applied reaction conditions, 89% of the captured CO₂ was converted to DMC. DBU was obtained, achieving 86% recovery of its salts formed during the synthesis. Most importantly, in this report we describe a simple and renewable solvent-based process for a metal-free approach to DMC synthesis under industrially feasible reaction conditions.

1. Introduction

Dimethyl carbonate (DMC) is a commercially important organic carbonate, considering its valuable and widespread industrial applications. DMC has attracted much attention from researchers at both the academic and industrial levels, as it is an important and environmentally friendly precursor for phosgene-free polycarbonate synthesis, whereas polycarbonates are extensively used in construction materials, automobiles, and certain medical devices [1,2]. Additionally, due to its high dielectric constant, DMC is also used as an electrolyte in lithium ion batteries and could emerge as a sustainable alternative to fossil fuels [3]. As a result of having a high oxygen content (53%), high octane number, high biodegradability, and low toxicity, DMC is considered to be an efficient and alternative fuel additive compared to the environmentally hazardous methyl tert-butyl ether [4]. Hence, due to DMC’s various green and sustainable applications, the annual demand for this chemical has increased rapidly during the past decade. Therefore, several industrially applied processes produce enormous amounts of DMC every year.

Until now, several catalytic routes have been applied in order to synthesize DMC, whereby phosgenation, transesterification, oxidative carbonylation, and urea methanolysis have been extensively studied and industrially applied [4,5,6,7]. However, these catalytic routes are accompanied by the use of toxic raw materials or catalytic deactivation, or are less selective and reactive in nature. These are the main challenges that still remain in developing more industrially feasible processes. The conversion of CO₂ to alkyl carbonates such as linear and cyclic carbonates can be considered a sustainable approach, since CO₂ is widely known as a renewable, cheap, and safe C₁ source. Simultaneously, the use of CO₂ avoids the possible risk of explosion, since the process does not include the combination of CO and O₂, as in other synthetic approaches. Apart from this, considering the negative consequences of CO₂ on the environment, such as global warming, the conversion of CO₂ into fuels and fine chemicals is an attractive solution to many challenges related to energy and the environment [8]. As such, the direct conversion of CO₂ to DMC has been carried out using various types of catalytic reaction approaches, despite certain thermodynamic and kinetic limitations [9]. Metal oxides such as ZrO₂ and CeO₂ were found to be active catalysts in DMC synthesis. After detailed studies, it was concluded that having both acid and base sites and the special redox properties of ZrO₂ and CeO₂ assisted the selective synthesis of DMC and diethyl carbonate (DEC) in a direct CO₂ conversion route [10,11,12]. Metal-oxide-supported organotin catalysts with dehydrating agents and metal-oxide-supported hetero-polyacids have also been successfully applied in DMC synthesis [13,14,15]. However, considering the chemical inertness of CO₂ (even after applying a high-temperature and high-pressure reaction approach), catalyst deactivation, as well as hydrolysis of DMC by co-produced water (causing low yield), these catalytic systems are still less attractive for DMC synthesis. Besides the direct conversion of CO₂ to DMC, a two-step synthesis approach was also invented, whereby synthesis of DMC is carried out using reagents such as olefin–ethylene oxide, methanol, and CO₂ [4]. As such, several catalytic routes such as epoxide, chlorohydrin, and isobutene–epoxide were applied in DMC synthesis. Even though these synthesis approaches have been industrially applied in DMC synthesis, the synthesis of epoxides from olefins and subsequent conversion to cyclic carbonates is accompanied by the use of toxic or costly reagents, and also involves potential explosion hazards and a tedious work-up process for product separation.

Methyl iodide (CH₃I) and potassium metals containing base catalysts such as K₂CO₃, CH₃OK, or KOH were also successfully applied in the direct conversion of CO₂ to DMC, whereupon the applied reaction conditions were milder than in the case of the previously discussed catalyzed processes [16,17]. In this case, the previous studies explained that the methyl carbonate anion formed after interaction between the basic catalyst, methanol, and molecular CO₂, which further interacted equivalently with CH₃I upon formation of DMC. After mechanistic studies, it was confirmed that CH₃I played a role as a reagent in a similar way to methanol in DMC synthesis instead of as a promotor, while the base assisted in the activation of methanol molecules [15]. Even though this base-catalyzed direct CO₂ conversion approach emerged as an alternative to metal-oxide- and organotin-catalyzed processes, the catalyst deactivation and loss of the methoxide anion to the dimethyl ether (DME) as a byproduct caused the low reaction yield. Catalyst deactivation occurs during the course of the reaction, where hydrogen iodide formed in situ reacts with metal carbonates and forms corresponding metal iodides. This metal-carbonate-based reaction approach was also highly pressure-dependent, whereby the optimum yield of DMC was obtained at 40–60 bar of applied CO₂ pressure. In this regard, a high-pressure reaction was carried out to suppress DME formation. Recently, another CH₃I-promoted and CH₃OK-catalyzed approach has been applied for DMC synthesis, whereby various dehydrating agents were used to improve the yield of DMC through removal of water formed in situ. However, a mere 43% yield of DMC was obtained at 80 °C and under 40 bar of CO₂ pressure [18].

Switchable ionic liquids (SILs) are a special class of ionic liquids, whose synthesis is usually proceeded by capture of the acid gases [19,20,21]. The synthesis of SILs is not only a fascinating process in terms of acid gas capture and storage at ambient conditions, but it also introduces a new and efficient solvent system for organic synthesis and biomass processing, such as wood delignification [22,23]. In SIL synthesis, an equimolar mixture of organic superbases such as phosphazenes, amidines, or guanidines as well as proton sources such as alcohols, alkanolamines, or water interacts with acid gases such as CO₂ or SO₂ [24,25,26]. In this study, for the first time we report the transformation of chemisorbed CO₂ in the form of SIL to DMC by mimicking the previously discussed CH₃I-involved, base-catalyzed reaction approach. In contrast to previous CH₃I-involved processes, DMC synthesis in this report was carried out at ambient reaction conditions by implementing a single-solvent system such as methanol. As shown in the reaction Scheme 1, upon the actual synthesis, an equivalent mixture of organic superbase such as DBU and methanol activates CO₂ and the obtained [DBUH] [CH₃CO₃], i.e., SIL further interacts with an equivalent amount of CH₃I to form DMC. In this report, we described the synthetic protocol for DMC in terms of type of solvent, amount of CH₃I, reaction time, recovery of DBU, etc. The progress as well as the amount of chemical species formed in the reaction mixture were monitored and explained by means of nuclear magnetic resonance (NMR) spectroscopy analysis.

2. Materials and Methods

2.1. Chemicals and Methods

2.1.1. Chemicals

Methyl iodide (99.5%), 1,8-Diazabicyclo-[5.4.0]-undec-7-ene (DBU, 99.0%), D₂O (99.9 atom% D), and CDCl₃ (99.96 atom% D) were purchased from Sigma Aldrich, while methanol (≥99.0%) and dimethyl sulfoxide (≥99.0%) were purchased from VWR chemicals. All the purchased chemicals were used further without purification. A CO₂ gas bottle was obtained from AGA AB (Linde Group, Sweden) and used without further purification.

2.1.2. NMR Analysis

The composition of the reaction mixtures was confirmed by NMR analysis using a Bruker Avance 600 MHz spectrometer operated at ¹H and ¹³C frequencies of 600.17 and 150.91 MHz, respectively. NMR analyses were carried out in different solvents such as with crude reaction mixture and glass capillary filled with D₂O or in neat D₂O or CDCl₃. NMR data were further processed with TopSpin 3.2 software. After conducting the NMR analysis, the chemical species that were identified in the reaction composition for DMC synthesis process are shown in Figure 1.

2.2. Synthesis of DMC with and without DMSO Solvent

2.2.1. Synthesis of SIL [DBUH] [CH₃CO₃]

SIL, i.e., [DBUH] [CH₃CO₃], was synthesized by the interaction of CO₂ with DBU (0.75 mL, 4.9 mmol) and methanol (0.2 mL, 4.9 mmol) in DMSO solvent (4 mL). During the actual synthesis, CO₂ gas (100 mL/min) was bubbled under stirring, at room temperature, in DMSO solution containing equimolar amounts of DBU and methanol for 20 min. In the DMSO-free reaction mixture, CO₂ gas was bubbled in alcoholic solutions of DBU (25 or 50 vol.% of DBU in methanol, 0.5 mL or 1 mL of DBU in 1.5 or 1 mL of methanol, respectively), at room temperature for 20 min. In all cases, pale yellow colored clear solutions were obtained, and their compositions were further confirmed by NMR analyses.

2.2.2. Synthesis of DMC

In this synthesis, the reaction mixtures with different reaction compositions were used in order to improve the yield of DMC. In these regards, in the first approach DMC synthesis was accomplished with addition of CH₃I (based on equivalents of DBU) to DMSO solution containing SIL. In another reaction approach, the DMSO solution containing SIL was diluted initially with different amounts of methanol, and then CH₃I was added to initiate the formation of DMC. Finally, in the case of DMSO-free synthesis, the CH₃I was added to the alcoholic solution of SIL containing different amounts of methanol. As per requirement, in all reaction approaches, the reactions were quenched after the desired period, and the progress of the reaction was monitored by NMR analysis. All experiments for DMC synthesis were carried out at room temperature. In all experiments, CH₃I was added in amounts ranging from 1 to 5 equivalents based on the amount of DBU used in the reaction mixture.

2.2.3. Separation of DBU Salts from the Reaction Mixture

The DBU salts, namely [DBUH] [I] and [DBUCH₃] [I], formed during DMC synthesis were separated from the reaction mixture in order to facilitate the recovery of DBU. Upon actual separation, these DBU salts were precipitated after addition of ethyl acetate in the reaction mixture. Prior to ethyl acetate addition, NMR analysis was carried out to ensure complete conversion of the [CH₃CO₃]⁻ anion. The obtained mixture of white salts was further separated from the reaction mixture by filtration and washed 2–3 times with ethyl acetate. Next, this mixture was dried under high vacuum, at room temperature, and was further analyzed by means of NMR spectroscopy whereupon either D₂O or CDCl₃ was used as the internal reference.

2.2.4. Recovery of DBU from DBU Salts ([DBUH] [I] and [DBUCH₃] [I])

DBU was recovered from [DBUH] [I] and [DBUCH₃] [I] salt mixtures by using alkaline NaCl-saturated aqueous solution following a previously reported procedure [27,28]. Initially, NaCl saturated aqueous solution was prepared by mixing NaCl in water under stirring, and the required amount of solution was taken to make 4 wt% solution of NaOH in NaCl-saturated solution. The mixture of DBU salts comprising [DBUH] [I] and [DBUCH₃] [I] was added to the alkaline solution of NaCl, and the reaction mixture was further stirred for 30 min at room temperature. Ethyl acetate was used further to extract DBU from the aqueous phase, whereby DBU was further obtained through the removal of ethyl acetate from the organic phase. The purity of the as-recovered DBU was confirmed by means of NMR analysis, whereas the recovery percentage (weight %) of DBU was calculated from its recovered amount and the initial amount used for CO₂ capture in the reaction.

2.2.5. Amount of Chemical Species in the Reaction Mixture

The synthesis of DMC was accompanied with the formation of methanol in situ as well as consumption of [CH₃CO₃]⁻ anions. The formation of methanol during the reaction is discussed in the following sections. Quantification (in %) of DMC, methanol, and [CH₃CO₃]⁻ anions in the mixture was carried out using ¹³C NMR analysis. Equations (1)–(3) were used to calculate the amount of DMC, methanol, and [CH₃CO₃]⁻ anions in the reaction mixture, respectively. I_(DMC), I_(methanol), and I_(CH3CO3-₎ represent the integrated value of the carbon atom in the methyl groups of these chemical species in ¹³C NMR spectra. The equations were applied as such for the reactions carried out in DMSO solvent. However, in the case of DMSO-free reactions, i.e., where methanol was used as the reagent as well as solvent, the alternative value for methanol was the integrated value of the methyl group in the [DBUCH₃] [I] salt.

$$\%\mathrm{DMC}=\frac{\left[I\left(\mathrm{DMC}\right)/2\right]\times 100}{I\left(\mathrm{DMC}\right)/2+I\left({\mathrm{CH}}_3\mathrm{OH}\right)+I\left({\mathrm{CH}}_3{\mathrm{CO}}_{3^{-}}\right)}$$

(1)

$${\%\mathrm{CH}}_3\mathrm{OH}=\frac{I \left({\mathrm{CH}}_3\mathrm{OH}\right)\times 100}{I\left(\mathrm{DMC}\right)/2+I\left({\mathrm{CH}}_3\mathrm{OH}\right)+I\left({\mathrm{CH}}_3{\mathrm{CO}}_{3^{-}}\right)}$$

(2)

$${\%\mathrm{CH}}_3{\mathrm{CO}}_{3-}=\frac{\mathrm{I} \left({\mathrm{CH}}_3{\mathrm{CO}}_{3^{-}}\right)\times 100}{I\left(\mathrm{DMC}\right)/2+I\left({\mathrm{CH}}_3\mathrm{OH}\right)+I\left({\mathrm{CH}}_3{\mathrm{CO}}_{3^{-}}\right)}$$

(3)

3. Results and Discussion

Being chemically inert, CO₂ was initially activated via the synthesis of SIL, which is considered as a well-practiced process for the capture of CO₂. In a consecutive step, the obtained SIL further reacted with CH₃I to form DMC. In order to achieve the maximum yield upon DMC synthesis, reaction parameters such as the type of solvent system, amount of CH₃I, and reaction time were examined. In case of variations in the solvent system, the two-step DMC synthesis approach mentioned in the reaction Scheme 1 was carried out. Initially, this was tried in the DMSO solvent and further explored in a similar synthesis protocol in methanol. In the case of SIL, i.e., [DBUH] [CH₃CO₃] synthesis, after bubbling the CO₂ in the DMSO solution of DBU and methanol, the composition of the obtained pale-yellow solution was confirmed by ¹H and ¹³C NMR analysis. As shown in Figure 2, the carbon atoms at position 6 and 9, respectively, in DBU were shielded by 5–6 ppm, whereas the peak for carbon at position 7 was de-shielded by 5 ppm after CO₂ capture. In addition, the peak at chemical shift 47.44 ppm for the methyl group disappeared, corresponding to methanol, and new peaks for the methyl group and the carbonyl carbon in the [CH₃CO₃]⁻ anion appeared at 50.06 and 155.97 ppm, respectively. This confirmed the formation of SIL, i.e., [DBUH] [CH₃CO₃], after interaction of DBU and methanol with CO₂. Regarding the ¹H NMR spectrum, it also confirmed the formation of SIL since changes in the chemical shifts for the protons, especially at positions 6 and 9, were observed in the case DBU, and a new peak for the protons in the methyl group of [CH₃CO₃]⁻ anion was also observed (Figure S1).

As shown in the reaction Scheme 1, after the synthesis of SIL, CH₃I was added to the DMSO solution of SIL to synthesize DMC. In this case, in order to gain insight into the progress of the reaction, CH₃I was added as 0.25, 0.5, 0.75, and 1 equivalents based on the amount of DBU used previously during the SIL synthesis. After 15 min of reaction time, for each and every reaction, the reaction mixture was further analyzed by ¹³C NMR analysis, and the obtained spectra are shown in Figure S2 and Figure 3, respectively.

As shown in Figure 3, after addition of different equivalents of CH₃I to the reaction mixture, the peaks belonging to the methyl group and carbonyl carbon in the [CH₃CO₃]⁻ anion, at the chemical shifts 30.45 and 155.98 ppm, respectively, steadily decreased. On the other hand, new peaks appeared at the chemical shifts 40.09, 47.44, 53.74, 154.51, 164.15, and 164.87 ppm, respectively, whereby the intensities of these peaks increased with increasing amounts of CH₃I. As observed during the synthesis of SIL, the peak with the chemical shift 47.44 ppm is related to the methyl group of methanol. Further, after comparing the NMR spectra of both reaction mixtures (with addition of 1 eq. CH₃I) and commercially available DMC, signals at 53.74 and 154.51 ppm, respectively, belonging to the carbon atoms in CH₃- and carbonyl group of the DMC were observed (Figure S2). Hence, as mentioned in the proposed reaction Scheme 1, it is possible to synthesize DMC after the synthesis of SIL followed by its equivalent interaction with CH₃I. ¹H NMR spectra of the reaction mixture also confirmed the formation of DMC in the reaction mixture. Similar to the case of ¹³C NMR analysis, upon an increase in the amount of CH₃I, the amount of DMC steadily increased (Figures S3 and S4). Simultaneously, the peak intensity for the free protons of the methyl group in methanol, at 3.45 ppm, along with an unknown signal at 3.55 ppm, also increased in the reaction mixture upon increasing the amount of CH₃I. In order to identify the unknown peaks in both ¹H and ¹³C NMR, as well as to gain insights into the formation of methanol in the reaction mixture, two-dimensional NMR (2D NMR) analysis of the reaction mixture was carried out further.

In Figure 4a, the correlation peak observed in the HSQC analysis at 3.46/47.44 ppm belongs to the methyl group of methanol; hence, as shown in ¹H and ¹³C NMR previously, 2D NMR also confirmed that a free methanol molecule formed during DMC synthesis. Usually, interaction of SIL with protic acids is the pathway for the room temperature decomposition of [CH₃CO₃]⁻ anions upon regeneration of methanol through the release of CO₂. This was further confirmed by the equivalent interaction of para-toluene sulfonic acid and SIL in a DMSO solution. After the addition of acid in the DMSO solution of SIL, CO₂ bubbles were suddenly released in the reaction mixture, and the formation of methanol was confirmed by ¹H NMR analysis (Figure S5). Hence, during DMC synthesis, it seems that the acidic moieties formed in- situ in the reaction mixture, thus allowing further decomposition of the [CH₃CO₃]⁻ anion in the SIL composition.

The unknown peak at chemical shift 3.56 ppm appeared as a singlet and showed an integration value identical to that of the methyl group in methanol. Thus, it seems that the unknown peak at this position belongs to the methyl group. Further, it was also observed that this methyl group showed correlation peaks at 3.56/40.13 and 3.56/164.87 ppm in HSQC and HMBC analyses, respectively (Figure 4). As shown in both ¹H and ¹³C NMR analyses, two types of DBU salts formed in the reaction mixture after addition of 1 eq. of CH₃I in the reaction mixture (Supplementary Materials S2 and S4). The peaks for the carbon atoms at position 7 (i.e., C-7) in these two salts appeared at 164.15 and 164.87 ppm, respectively, in the ¹³C NMR spectra. One of the two DBU salts is protonated (i.e., [DBUH] [I]) since DMC formed in the reaction mixture as shown in the reaction Scheme 1b. Further, since the newly appearing methyl group at the chemical shift 3.56 ppm has a long range coupled with the carbon atom C-7 at chemical shift 164.87 ppm in HMBC spectra, the peak appearing at 164.17 ppm belongs to the C-7 atom in the [DBUH] [I]. In other words, apart from the formation of [DBUH] [I], the methylated salt of DBU (i.e., [DBUCH₃][I]) also formed in the reaction mixture during DMC synthesis, whereby the CH₃⁻ group (3.56 ppm) was attached to the sp²-hybridized N atom of DBU. The peaks for the C-7 atoms obtained at chemical shifts 164.15 and 164.87 ppm in the protonated and methylated DBU, respectively, are shown as C-7^# and C-7^## in ¹³C NMR spectra (Figure 3). Hence, during the synthesis of DMC, a part of CH₃I reacted with the [DBUH]⁺ cation of the SIL through an interaction between the CH₃ group and sp²-hybridized N atom to form [DBUCH₃][I] (reaction Scheme 2). In this case, an equivalent number of protons was released in the reaction mixture, which subsequently decarboxylated [CH₃CO₃]⁻ anions of the SIL to facilitate the formation of methanol and CO₂. Hence, under the used reaction compositions and applied reaction conditions, as shown in the reaction Scheme 2 and Scheme 3, the [CH₃CO₃]⁻ anion is distributed to DMC and methanol.

Further, in order to confirm that the captured CO₂ during SIL synthesis became a part of DMC, labeled CO₂ (i.e., ¹³C-enriched CO₂ (tracer technique)) was used during the two-step DMC synthesis process. The existence of ¹³C-enriched CO₂ in SIL as well as DMC was further confirmed by ¹³C NMR and HMBC analyses, respectively. As shown in Figure 5b, SIL formed after the reaction of DBU and methanol with ¹³C-enriched CO₂, and the high-intensity peak for the carbonyl carbon in [CH₃CO₃]⁻ anion was observed at the chemical shift of 154.56 ppm. After an addition of CH₃I in the DMSO solution of SIL, the high-intensity peak for the carbonyl carbon atom in DMC was observed, and this was further confirmed with the help of ¹³C NMR of the commercially available DMC (Figure 5c,d). Being ¹³C-enriched, the intensities of the peaks for the carbonyl carbon atoms in both [CH₃CO₃]⁻ and DMC were found significantly higher than the corresponding intensities of the C-7’, C-7^#, and C-7^## in DBU cations formed in the reaction mixture. On the other hand, as shown in Figure 3, the intensity of the peak for the carbonyl carbon was comparable with the intensities of these sp²-hybridized carbons in the DBU cation when no ¹³C-enriched CO₂ was used. Further, HMBC analysis also revealed that a correlation peak was observed for the protons of the methyl group and ¹³C-enriched carbonyl carbon atom in DMC (Figure S6a). In addition, due to having a ¹³C-enriched carbon atom, the doublet peak was observed for the protons for the methyl group in DMC, whereas when no ¹³C-enriched CO₂ was used, a singlet peak was observed for these protons (Figure S6b) [29]. Therefore, these above-mentioned NMR analysis studies revealed that the ¹³C-enriched CO₂ became a part of DMC formed after the reaction of ¹³CO₂ enriched SIL with CH₃I. Hence, it can be concluded that DMC can be synthesized from CO₂ following a two-step synthetic approach, mentioned in the reaction Scheme 1.

Besides confirming the origin of carbonyl carbon in DMC, in the ¹³C NMR spectra of the reaction mixture, the peak for molecular CO₂ at 123.01 ppm was also observed (Figure 5c) [30]. Hence, as shown in the reaction Scheme 3, this control experiment further proved that CO₂ was released in the reaction mixture through the decomposition of [CH₃CO₃]⁻ anions upon DMC synthesis.

The amounts (%) of DMC, methanol, and [CH₃CO₃]⁻ anions in the reaction mixture after each addition of CH₃I were calculated using Equations (1)–(3), respectively. As described in the experimental section, the equations correspond to the integrated peak area of the carbon atom in the methyl group of these chemical species obtained from the ¹³C NMR spectra. The number of [CH₃CO₃]⁻ anions was also taken into consideration because both DMC and methanol formed upon consumption of the [CH₃CO₃]⁻ anion (reaction Scheme 2). Representative integrations of the required signals in ¹³C NMR spectra are shown in Figure S7. The obtained values for the amounts of DMC, methanol, and [CH₃CO₃]⁻ anions after calculation are shown in Figure 6.

After sequential addition of CH₃I in DMSO solution of SIL, the number of [CH₃CO₃]⁻ anions steadily decreased and were completely consumed after addition of 1 eq. of CH₃I. At the same time, the amounts of both DMC and methanol increased and reached values corresponding to 62 and 38% in the reaction mixture, respectively. In this case, it seems that as a result of the direct interaction between CH₃CO₃⁻ anions and CH₃I, a higher amount of DMC can be obtained. On the other hand, methanol synthesis possibly proceeds through a three-step process, namely interaction of CH₃I with DBUH⁺, intermolecular proton transfer, and decarboxylation of [CH₃CO₃]⁻ anions to methanol. These intermediate steps might have slowed down the formation of methanol compared to DMC. Further, after addition of 1 eq. of CH₃I and complete conversion of [CH₃CO₃]⁻ anions, the reaction mixture was added to the ethyl acetate, and the thus-obtained white solid was analyzed by ¹H NMR analysis after separation from the reaction mixture. As shown in Figure 7, the NMR spectra confirm that the obtained white solid is a physical mixture of iodide salts of protonated and methylated DBU, i.e., [DBUH] [I] and [DBUCH₃] [I]. The occurrence of [DBUCH₃] [I] in the mixture of salts was also confirmed by HMBC analysis (Figure S8).

Hence, as shown in the reaction Scheme 1, DMC formed after equivalent interaction between SIL and CH₃I. In this case, even though CO₂ is a chemically inert species, the highly basic (Brønsted base) characteristics of DBU further facilitated the activation of CO₂ under ambient reaction conditions. Further, CH₃I has negligible steric hindrance, and since I⁻ is a good leaving group, its interaction with the [DBUH]⁺ cation and [CH₃CO₃]⁻ anion of the SIL was further promoted. Hence, as a result of having the reagents with these special chemical characteristics, the overall synthesis proceeded with lower energy consumption in terms of temperature and pressure. However, as described previously, the reaction is not atom efficient, as simultaneous formation of methanol and CO₂ also proceeds after decomposition of [CH₃CO₃]⁻ anions (reaction Scheme 2 and Scheme 3). This side-reaction not only involves the loss of CH₃I reagent, but it also releases equally important CO₂, which is needed upon DMC synthesis. Since this side-reaction occurred due to the interaction between the [DBUH]⁺ cation and CH₃I, it is necessary to avoid their interaction in order to increase the yield of DMC in this synthetic approach.

Considering these aspects, prior to the addition of CH₃I, the DMSO solution of SIL was initially diluted with methanol, whereby the total reaction mixture volume of methanol was maintained as 6, 12, 25, or 50 vol.%. In these diluted reaction mixtures, three equivalents of CH₃I (based on DBU in the composition) were added, and after 1 h the reaction mixture was analyzed by NMR analysis. As shown in Figure 8, peaks related to the methylated DBU cation steadily decreased as the amount of methanol in the reaction mixture increased from 6 to 50 vol.%. The amounts of DMC, methanol, and [CH₃CO₃]⁻ anions in the reaction mixture were calculated using Equations (1)–(3), respectively. As the reaction mixture comprised an excess quantity of methanol, the integrated value of the peak belonging to CH₃- group in [DBUCH₃]⁺ cation was used for methanol. This was decided based on the fact that equal integrated values were observed for the methyl group in both methanol and [DBUCH₃]⁺ cations when the reaction was previously carried with only DMSO as a solvent (Figure S7). The calculated amounts of DMC, methanol, and [CH₃CO₃]⁻ anions observed in the reaction mixture are shown in Figure S9.

Figure S9 depicts that dilution imparts a positive effect on the DMC formation, where the amount of DMC increased from 62% (reaction proceeded with no dilution with methanol) to 81 and 86% with 12 and 25 vol.% of methanol used in the reaction composition, respectively. However, further dilution with 50 vol.% methanol did not improve the DMC yield, and only 16% yield was obtained and 76% of [CH₃CO₃]⁻ anions remained unreacted. In this case, having 50 vol.% of methanol in the reaction mixture could not allow efficient interaction between the [CH₃CO₃]⁻ anion and CH₃I. Most importantly, upon an increase in the amount of methanol, the intensity of the peaks for the carbon atoms in DBU and the methyl group in [DBUCH₃] [I] salt steadily decreased. In other words, instead of using neat DMSO solution of SIL, its diluted solution using methanol allowed more selective interaction between CH₃I and [CH₃CO₃]⁻ anions, whereas it seems that the side-reaction for the interaction between [DBUH]⁺ and CH₃I was significantly hampered.

The influence of the amount of CH₃I on the formation of DMC in the reaction mixture was further examined in a diluted reaction mixture, whereupon excess methanol was used with SIL and DMSO. In the actual study, the different equivalents of CH₃I (on the basis of DBU in the composition) such as 1, 3, or 5 were added to the reaction mixture in which 88 vol.% of DMSO solution of SIL and 12 vol.% of methanol were present in the reaction composition. As shown in Figure S10, with increasing amounts of CH₃I from 1 to 3 equivalents, the amount of DMC also increased from 62 to 83% in the reaction mixture. This could be observed since it resulted in a shift in the equilibrium towards DMC with more equivalents of CH₃I compared to SIL amount. However, further increase in the CH₃I to 5 equivalents did not improve the amount of DMC in the reaction mixture.

As it was observed that dilution with methanol resulted in an increase in the amount of DMC, this synthesis approach was further upgraded by replacing DMSO with methanol. During the actual experimental process, 25 or 50 vol.% solutions of DBU in methanol were prepared and used for CO₂ capture through SIL formation. Further, DMC was synthesized after addition of CH₃I (3 equivalents based on amount of DBU) in the resultant alcoholic solution of SILs, and the obtained ¹³C NMR spectra are shown in S11. Here, along with the influence of the amount of DBU in its alcoholic solution (25 and 50 Vol.%), the influence of the reaction time was also examined in terms of the formation of DMC. As shown in Figure 9a,b in 1 h of reaction time, 13 and 58% of DMC formed when 25 and 50 vol.% of DBU were used, respectively. As observed previously, the use of excessive amounts of methanol resulted in lower amounts of DMC in the reaction mixture. On the other hand, in a comparatively concentrated system, i.e., with 50 vol.% of DBU, a higher amount of DMC was obtained as a result of effective interaction between [CH₃CO₃]⁻ anions and CH₃I. In addition, a similar trend was also observed in the case of methanol formation, whereby higher amounts of methanol were found (7%) with 50 vol.% DBU compared to 25 vol.% DBU-containing reaction mixture. Further, for 5 h reaction period, with 25 vol.% DBU-containing reaction mixture, merely 54% of the [CH₃CO₃]⁻ anion reacted and 49% of DMC was formed. On the other hand, in the reaction mixture in which 50 vol.% of DBU was used, 96% of [CH₃CO₃]⁻ anion was consumed, whereas 10 and 86% of methanol and DMC were formed, respectively. DBU is a well-known organocatalyst, and various organic transformations have been performed where it acts either as a strong Brønsted base or good nucleophile. In non-protic solvents the [DBUH]⁺ cation is usually involved in a proton exchange with DBU, and there DBU can perform both as a nucleophile as well as a base in the reaction. Since DMSO is a non-protic solvent (not acting as a proton source), during this proton exchange it further allows the interaction between DBU formed in situ with CH₃I. On the other hand, in protic solvents, good nucleophilic characteristics of DBU can be diminished, and it transforms to a weak base as the nucleophilic nitrogen becomes engaged in hydrogen bonding with protic solvents [31]. Since methanol is a protic solvent, there is a lower probability of DBU forming from its protonated form, and nucleophilic interaction of DBU (and hence [DBUH]⁺ cation) with CH₃I is kept under control. Hence, instead of DMSO, the use of methanol as a solvent efficiently controlled the side-reactions and further improved the yield of DMC in the reaction mixture.

In order to complete conversion of [CH₃CO₃]⁻ anions in the reaction mixture with 50 vol.% of DBU, the reaction time was further increased from 5 h to 7 and 10 h. The corresponding ¹³C NMR spectra for this study are shown in Figure S12. As shown in Figure S13, the [CH₃CO₃]⁻ anion was completely consumed in the reaction mixture, whereas the amount of DMC increased from 86 to 89% in 10 h of reaction time. Simultaneously, the amount of methanol did not significantly increase, and 11% was found in the reaction mixture. Since no formation of other methylated products (such as dimethyl ether) except methanol was observed, DMC formation reached 89% selectivity in 10 h. Hence, in this reaction approach, as mentioned previously, methanol not only became a part of the CO₂ capture process, but it also assisted in the synthesis of DMC with high selectivity. In this regard, excess methanol in the reaction composition significantly influenced the interactions between the chemical species and favored the formation of more DMC.

The reaction mixture in which 50 vol.% of DBU was used was further added to ethyl acetate after 10 h reaction to separate the salts of DBU forming during the reaction. The composition of these salts was confirmed by ¹³C NMR analysis. As shown in Figure 10a,b, both DBU salts, i.e., [DBUH] [I] and [DBUCH₃][I], were selectively separated from the reaction mixture. Based on the intensity of the peaks belonging to both salts, a major part of the salt mixture was [DBUH] [I], whereas its methylated analogue was present in much lower quantities. In order to recover the DBU from the mixture of salts, 4 wt.% NaOH in saturated solution of NaCl was used. The purity and recovery of the DBU was controlled by ¹H NMR analysis and gravimetric measurements, respectively. As shown in Figure S14, a pure form of DBU was recovered following alkaline solvent treatment. Besides that, 86% recovery of DBU was achieved from the recovered DBU salts.

Hence, as a part of capture and storage, CO₂ is stored chemically through the formation of DMC, where initially the SIL was prepared to activate chemically inactive CO₂; subsequently, it reacted with CH₃I to form DMC. In this reaction approach, a reaction yield >85% was obtained under mild reaction conditions. Even though the use of alcoholic solution further led to slow reaction kinetics, it simultaneously improved the yield of DMC through control over loss of chemisorbed CO₂ through side-reactions. Previous reports based on metal carbonates and oxides were accompanied with the use of high reaction temperature and pressure to reach the optimized DMC yield. In addition, in these cases, the catalyst and product decomposition (due to formation of water during the reaction) and formation of by-products (e.g., dimethyl ether in metal carbonate based process) significantly influenced DMC yield. Simultaneously, the regeneration of catalysts or promoters was not carried out in these metal carbonate and oxide-based processes. On the other hand, in this report, the industrially feasible, single-solvent-based and metal-free synthetic approach was established to obtain DMC with high yields and selectivity, at room temperature and under atmospheric pressure. This synthetic approach also demonstrated the recovery of organic superbase, which was considered as an important moiety in the synthesis. The current organic superbase-induced DMC synthesis approach is also accompanied with the use of equivalent amounts of CH₃I as the methylating reagent. This substrate is unfortunately considered as a toxic chemical moiety to the ecosystem as well as human beings. However, CH₃I is considered as a more active reagent and a more powerful methylating agent compared to others such as DMC, methanol, and dimethyl sulphate [32]. Additionally, the reaction with CH₃I usually proceeds under mild reaction conditions and is frequently active without any catalyst, whereas in the case of other methylating agents, the reaction proceeds under harsh reaction conditions and also requires catalysts and sometimes results in low yields. If CH₃I is handled safely during its use by considering all the necessary safety precautions, then it can be explored in bulk implementation for DMC synthesis, at industrial scale. Further, this reaction approach involving an organic superbase was carried out with easily available reagents as well as a solvent system, and this fundamental study for DMC synthesis can be extended to synthesize a wide range of industrially important alkyl carbonates. In this case, several alcoholic moieties as well as alkyl halides ranging from lower to high molecular weight can be applied.

4. Conclusions

The industrially important dimethyl carbonate (DMC) was synthesized from CO₂ following switchable ionic liquid (SIL) involved in a one-pot and metal-free reaction approach under ambient conditions. CO₂ was activated through its equivalent interaction with organic superbase DBU and methanol, and the resultant SIL reacted with CH₃I and formed DMC. Initially, this synthetic approach was proven in DMSO solvent, where 62% of chemisorbed CO₂ was converted to DMC. The yield of DMC was further increased to 89% with the DMSO-free reaction approach where synthesis was carried out with methanol as a solvent. Based on NMR analysis, it was observed that with DMSO as a solvent, along with DMC synthesis, as a side reaction, the CH₃I also simultaneously reacted with protonated DBU in SIL and formed methylated DBU cations. Hence, released protons further decarboxylated [CH₃CO₃⁻] anions in unreacted SIL to form methanol and released CO₂ in the reaction mixture. However, the use of methanol as a solvent further improved the yield of DMC where the loss of CO₂ was avoided by control over the side reactions. Therefore, in this regard, methanol not only performed as a reagent in CO₂ capture and DMC synthesis processes, but it also simultaneously assisted in the efficient conversion of chemisorbed CO₂ to DMC. DBU salts such as [DBUH] [I] and [DBUCH₃] [I] were also converted to pure DBU, where 86% recovery was achieved. Hence, metal-free, single-solvent-based, and one-pot synthesis of DMC was performed under ambient conditions. As a main achievement, in this fundamental study anthropogenic CO₂ was not only captured but also converted to a valuable chemical.