Green Synthesis of 2 ‐ Oxazolidinones by an Efficient and Recyclable CuBr/Ionic Liquid System via CO 2 , Propargylic Alcohols and 2 ‐ Aminoethanols

: With the aim of profitable conversion of carbon dioxide (CO 2 ) in an efficient, economical, and sustainable manner, we developed a CuBr/ionic liquid (1 ‐ butyl ‐ 3 ‐ methylimidazolium acetate) catalytic system that could efficiently catalyze the three ‐ component reactions of propargylic alcohols, 2 ‐ aminoethanols, and CO 2 to produce 2 ‐ oxazolidinones and α‐ hydroxy ketones. Remarkably, this catalytic system employed lower metal loading (0.0125–0.5 mol%) but exhibited the highest turnover number (2960) ever reported, demonstrating its excellent activity and sustainability. Moreover, our catalytic system could efficiently work under 1 atm of CO 2 pressure and recycle among the metal ‐ catalyzed systems.

Besides the efforts on direct condensation, researchers also developed alternative strategies that tried to circumvent the thermodynamic barrier of generating H2O. Among them, employing propargylic alcohols in the condensation of 2-aminoethanols and CO2 is a promising way that has been revealed as a thermodynamically feasible process. Moreover, α-hydroxyl ketones, a series of high-value compounds that are generally employed as key synthons for organic chemistry and biologically active fragments in pharmacological products, could be simultaneously synthesized together with 2oxazolidinones in this three-component process [54][55][56][57][58]. In this area, He et al. have achieved several milestones. Firstly, they employed 5 mol% of Ag2CO3 and 10 mol% of phosphine ligands (Xantphos) for this reaction, which could efficiently catalyze diverse substrates in CHCl3 at 60 °C under 1 MPa of CO2 [54]. Subsequently, a similar system containing 5 mol% of Ag2O and 30 mol% of 1,1,3,3-tetramethylguanidine was reported, which performed excellent activity under 1.0 MPa of CO2 at 80 °C in CH3CN [55].
In addition to the silver catalytic systems, they also established a cheaper and greener Cu(I) catalytic system, in which a competitive amount of CuI (5 mol%) was added together with 5 mol% of 1,10-phen and 10 mol% of t-BuOK [56]. This system could promote the three-component reaction under a relatively low CO2 pressure (0.5 MPa) at 80 °C. Recently, they synthesized a task-specific ionic liquid (IL), namely 1,5,7triazabicyclo[4.4.0]dec-5-ene trifluoroethanol ([TBD][TFE]), which could work under 1 atm of CO2 pressure at 80 °C [57]. Although great progress has been achieved for this strategy, several problems remained that blocked its further applications. For example, the only report of the metal-free catalyst [TBD][TFE] gave an acceptable catalytic performance, however, it was not commercially available and could be only obtained in laboratories by employing a rare organic base (TBD) through an anion exchange resin, which limited its large-scale application. In contrast, the metal-catalyzed systems employed simple and easily accessible materials as the catalysts, thus showing certain potential for practical applications. However, they still suffered from the disadvantages of high metal loading; elevated CO2 pressure; poor catalyst recyclability; and additions of ligands, bases, or other additives. Consequently, developments of simple, green, easily accessible, and recyclable catalytic systems that perform excellent activity under mild conditions are still highly desirable.
Generally, IL is considered an environmentally friendly and green solvent for its negligible vapor pressure as well as high thermal stability. Particularly, its physical and chemical properties can be easily adjusted by changing the cations and anions or introducing desired functional groups, which largely extend its availability in diverse fields such as gas adsorption, catalysis, extraction, sample preparation techniques, etc. Therefore, employment of IL together with the metal salts might be a potential methodology to develop the desired catalytic systems. Herein, we combined the green and versatile Cu salts with the commercially available imidazole-based ILs for the threecomponent reactions of propargylic alcohols, 2-aminoethanols, and CO2. After screening, an optimal CuBr/1-butyl-3-methylimidazolium acetate ([C4C1im][OAc]) catalytic system was obtained. This system proved to inherit the merits from both ILs and metal-catalyzed systems, which could efficiently promote the reaction under 1 atm of CO2 pressure with a lowermost metal loading in the absence of any ligands, bases, and additives. Moreover, this system behaved robustly in recyclability and sustainability. An unprecedented turnover number (TON) was achieved in this aspect. Table 1 describes the screening of catalytic systems for the three-component reaction, including copper salts and ionic liquids. 2-(benzylamino)ethanol (1a) and 2-methylbut-3yn-2-ol (2a) were used as the model substrates in the screening of the optimal catalytic systems for the three-component reaction, 3-(phenylmethyl)-2-oxazolidinone (3a) and 3hydroxy-3-methyl-2-butanone (4a) as products of 1a and 2a (

Entry
[Cu] Salt Ionic Liquid After obtaining the best CuBr/[C4C1im][OAc] system, we continued to optimize its condition parameters ( Table 2). The reaction temperature was initially evaluated. In the beginning at 25 or 50 °C, the system was inactive without any products obtained (entries 1 and 2). However, the catalytic activity would increase along with the rising temperature from 50 to 100 °C (entries 2-4). A higher temperature of 120 °C was also tested; however, no obvious gain on the activity was observed (entry 5). Therefore, the suitable temperature was selected as 100 °C (entry 4). Furthermore, different amounts of [C4C1im][OAc] and CuBr were also tried. Surprisingly, increasing or decreasing the IL would lead to reduced yields (entries 6-7 vs. entry 3). Meanwhile, a lower CuBr loading of 0.25 mol% showed an unsatisfactory yield (entry 8). Due to 0.5 mol% of CuBr had given a satisfactory result under 1 bar of CO2, higher metal loadings or elevated CO2 pressure were not further investigated. Lastly, the ratio of 1a:2a was tuned to 1:1 while the yield was decreased (entry 9), indicating an excess amount of propargylic alcohols would be beneficial for this reaction. In conclusion, the most suitable reaction conditions were fixed as follows: 0.5 mol% of CuBr and 1.3 equiv. of [C4C1im][OAc] (based on 2-aminoethanols) under atmosphere CO2 pressure at 100 °C with the ratio of 1:1.5 (1a:2a) (entry 4). It is worth noting that 0.5 mol% is the lowest metal loading ever reported among the metal-catalyzed systems, even the generally more active Ag catalysts could not reach this level. Meanwhile, this is the first reported metal-catalyzed system that could efficiently work under 1 atm of CO2 pressure. Additionally, an experiment under the optimal conditions but without purging the system was performed; however, only moderate yields could be obtained (entry 10), indicating that lower CO2 partial pressure or lower CO2 purity was unfavorable for the reaction. Meanwhile, the purge operation was indeed necessary for obtaining high yields. After obtaining the suitable catalytic system as well as its optimal reaction conditions, we started to explore the substrate scope. The experimental data are listed in Table 3. Initially, different propargylic alcohols substituted by the alkyl, cycloalkyl, and aryl groups (2a-2e) were examined. Delightfully, all these substrates could be transformed into the desired products at satisfactory yields. Specifically, 2d or 2e with relatively bulky substituent groups required prolonged time for the conversion, implying that the steric effects of the substituents might influence the reactivity of the propargylic alcohols. On the other hand, a series of 2-aminoethanols were also introduced into the reaction (1a-1j). Obviously, the substituents in the phenyl rings would also affect the reactivity of those substrates containing aryl groups. Generally, aryl 2-aminoethanols with electrondonating groups such as -Me or -MeO would smoothly accomplish the reaction, while the electron-withdrawing group NO2 − in 1f largely limits its reactivity for this reaction (1a-1d vs. 1f). In addition, alkyl substituted 2-aminoethanols, 1g-1j were also applied to the reaction, and moderate to excellent yields could be obtained, indicating the broad substrate scope of this catalytic system. Furthermore, a gram-scale experiment was performed based on 1a and 2a. The result showed that the CuBr/[C4C1im][OAc] system still exhibited satisfactory activity for grams of substrates, implying its potential in practical applications. Besides catalytic activity, recyclability and sustainability were also important for comprehensively evaluating a catalyst. Herein, we explored the performance of the CuBr/[C4C1im][OAc] system in this aspect based on the model reaction of 1a and 2a under its optimal conditions. Owing to the advantage of the IL component that would retain the Cu salt during the extraction and separation, this catalytic system kept its excellent activity in the recycling assessment (as shown in Figure 1a), reflecting its stability and reusability (Table S2, supporting information). It is worth mentioning that this is the first metal-catalyzed system that could be reused for this three-component reaction. Subsequently, an experiment for evaluating the maximum turnover number (TON) was performed. To our delight, even when the metal loading reduced to an unprecedented level of 125 ppm, this catalytic system still exhibited considerable activity. Particularly, a TON of 2960 was obtained in this experiment (Figure 1b), indicating the excellent sustainability of this catalytic system. To our best knowledge, this is the highest TON ever reported for this three-component reaction ( Figure S1 and Table S1, supporting information).

Activation of the Hydroxyl Group
According to the previous literature, activation of hydroxyl groups in propargylic alcohols is the initial step of the three-component reaction, which could be monitored by the shape and chemical shift of the hydroxyl signal in 1 H NMR [55,59,60]. Generally, this weak acidic proton of the hydroxyl group required relatively strong bases to activate it [61,62], and the OAc − in normal acetate salts could not afford this activation [63]. However, from the following experiment, we verified that OAc  (Figure 2b). However, in the 2a/1a system, the sharp peak was still maintained, indicating that 2-aminoethanol was invalid for this activation (Figure 2c). Interestingly, once CO2 was introduced into the 2a/1a system, the hydroxyl peak was changed into a relatively obtuse shape, implying 2aminoethanol together with CO2 also showed slight activated ability for the hydroxyl proton (Figure 2d and Figure S4 of the supporting information). In consequence, [C4C1im][OAc] plays a vital role in the activation of the hydroxyl group, which initiates the following proposed mechanism.

Proposed Catalytic Mechanism
According to the previous publications [25,54,55,57,[63][64][65][66][67], a probable catalytic mechanism of the CuBr/[C4C1im][OAc] system was proposed for the three-component reaction (Scheme 1a), which might contain two steps: (1) propargylic alcohols are combined with CO2 to generate the key cyclic carbonates, D; (2) D react with aminoethanols to give 2-oxazolidinones and α-hydroxyl ketones ( Figure S2 and S3, supporting information). In step 1, the OAc − anion initially activates the hydroxyl group of the propargylic alcohol and CO2 [68,69], which is favorable for the following attack of the hydroxyl oxygen to the carbon center of the CO2, generating intermediate B. Then, the metal catalyst activates the triple bond so that the negative oxygen in intermediate B can attack the carbon of this triple bond intramolecularly and form intermediate C. Finally, the catalyst is released from the five-membered ring through the returning of the proton, giving the important intermediate cyclic carbonate D. Then step 2 occurs, in which the nitrogen of the aminoethanol attacks the carbon in D and breaks the C-O bond, resulting in the breakage of the five-membered ring and the generation of E. E is converted to F due to its unstable enol structure. Finally, the hydroxyl oxygen attacks the adjacent carbonyl carbon with the aid of the catalysts. A five-membered ring of 2-oxazolidinone is generated by releasing an α-hydroxy ketone molecule.
Interestingly, besides the general mechanism of the Cu salt, another Cu species might also exist in our catalytic system. According to our previous reports [25,67], the basic OAc − in [C2C1im][OAc] might interact with the imidazole cation, leading to the chemical equilibrium with the free N-heterocyclic carbenes (NHCs) and the corresponding HOAc. Once Ag salts are involved, the NHCs might be coordinated in situ and form the NHC-Ag complexes. Therefore, we speculated that similar NHC-Cu complexes might also exist in this Cu-catalyzed system (Scheme 1b).  Subsequently, based on the experimental results and our previous study [67], we speculated a probable mechanism involving the bis-NHC-Cu complex (Scheme 2). The main parts were consistent with the mechanism in Scheme 1a. Particularly, when the bis-NHC-Cu complex enters the catalytic cycle, one NHC might drop and participate in the interaction between OAc − and the hydroxyl proton. Meanwhile, the remaining [Cu] species perform the same role as the normal Cu salt. Scheme 2. Proposed catalytic mechanism involving the bis-NHC-Cu complex.

Characterization
All the nuclear magnetic spectra were obtained by a Bruker Avance Ⅲ HD spectrometer. 1 H NMR was recorded at 500 MHz in CDCl3 (7.26 ppm) or DMSO-d6 (2.51 ppm), and 13 C NMR was recorded at 126 MHz in CDCl3 (77.16 ppm) or DMSO-d6 (39.52 ppm). High-resolution mass spectra were conducted by a Bruker Daltonics micro TOF-QII mass spectrometry instrument given in per charge (m/z).

Materials
CO2 at a purity of 99.999% was purchased from the Xiang Yun Gas Company. Unless specifically mentioned, all the raw materials, including propargylic alcohols, copper salts, and ionic liquids, were obtained from Sigma-Aldrich, Aladdin, TCI, Macklin, Alfa Aesar, etc.

Three-Component Reactions of Propargylic Alcohols, 2-Aminoethanols, and CO2
Propargylic alcohols (7.5 mmol), 2-aminoethanols (5 mmol), CuBr (0.025 mmol), and [C4C1im][OAc] (6.5 mmol) were added into a reaction tube equipped with a magnet bar. The gas inside the tube was replaced by CO2 (99.999%) three times to confirm that this system was completely under the atmosphere of 1 atm of CO2. Then the tube was heated in an oil pot at 100 °C for 12 h. When the reaction was completed, the mixture was extracted by diethyl ether (5 × 10 mL). Finally, the upper layers were collected and evaporated by the rotary evaporator. The obtained raw products were further separated and purified by column chromatography. For the recyclability investigation, the lower layer (recovered CuBr and [C4C1im][OAc]) was directly reused for the next round after drying under vacuum at 100 °C for 3 h.

Conclusions
In summary, we have developed a CuBr/[C4C1im][OAc] catalytic system that can efficiently produce 2-oxazolidinones and α-hydroxy ketones through the threecomponent reactions of propargylic alcohols, 2-aminoethanols, and CO2 in a convenient and green manner. Particularly, this system exhibited excellent catalytic activity under 1 bar of CO2 with only 0.0125-0.5 mol% of CuBr. Furthermore, the robust recyclability and sustainability of this system were also demonstrated with an unprecedented TON of 2960, the highest ever reached. In further mechanistic investigations, we detected an NHC-Cu complex during the experimental process, which was eventually identified as a bis-NHC-Cu configuration by the HRMS.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4344/11/2/233/s1, Figure S1. The literatures reported for the three-component reactions; Figure S2. 1H NMR of the control experiment mixture (red) and the pure cyclic carbonate (blue); Figure S3. 1H NMR of pure 4a (green), pure 3a (red) and the control reaction mixture (blue); Figure S4. Investigations on the activation of hydroxyl protons in the presence of 1 atm of CO2; Table S1. TON reported in the previous literatures; Table S2. Exploration of metal leaching in the recycling experiments.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Acknowledgments:
The authors would like to express their deep appreciation to the State Key Lab of Advanced Technology for Materials Synthesis and Processing for their financial support.

Conflicts of Interest:
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Sample Availability: Samples of the compounds are available from the authors.