Design and optimization of the reaction systemOur examination started with the synthesis of the organocatalyst precursor—NHC salt (A) containing sterically crowded groups localized at the terminal nitrogen atoms in imidazole rings, according to the methodology described in our previous paper19 (Fig. 2):Figure 2Structure of organocatalyst precursor A.It has been proved that conversion of α,β-unsaturated aldehydes into saturated esters could occur in the presence of simple NHC precursors and a strong base14. For this reason we decided to apply similar conditions to the reaction of cinnamaldehyde (1a) and benzyl alcohol (2a) in the presence of the bulky NHC salt (A). The addition of equimolar ratio of reagents to a toluene solution of 10 mol% of A and 10 mol% of KHMDS at 100 °C, resulted in a complete conversion of the substrates after 2 h, as revealed by the GC–MS technique. The 1H NMR analysis of the reaction mixture showed a selective formation of the expected product P1 in a quantitative yield (Fig. 3).Figure 3Esterification of cinnamaldehyde (1a) with benzyl alcohol (2a).This achievement prompted us to continue our studies. A series of tests were performed to select optimal temperature, the concentration of NHC precursor and the appropriate solvent. All optimization tests were performed on the above model reaction. The results are summarized in Table 1.Table 1 Optimization of the type of solvent, temperature and concentration of NHC precursor (A).Solvent screening showed that the process can be carried out both in toluene, ethylene glycol diacetate (EGDA), methyl isobutyl ketone (MIBK) and acetone. This is especially important in ecological aspects as these solvents, except toluene, are classified as green solvents20. Because of its density, boiling point and price, acetone was selected as the most suitable medium for the tested process. As presented in Table 1, the reaction that proceeded at 100 °C gave the same results as the ones run at 80 °C and 60 °C (Table 1, Entries 1–3, 5–7, 9–11, 18–20, 22–24). Further decrease in temperature led to a reduction in the product yield, even though the reaction time was increased to 48 h (Table 1, Entry 4). Next, the optimum concentrations of the NHC precursor and KHMDS were established. The best results were achieved by carrying out the process in the presence of 5 mol% A and 5 mol% base (Table 1, Entry 11). Lowering the concentration of NHC salt to 1 mol % resulted in a decrease in substrate conversion, despite the extension of the reaction time (Table 1, Entry 13). Carrying out the reaction without addition of KHMDS led to a conversion in trace amounts, which confirmed that a base is indispensable for the effective reaction course (Table 1, entry 16). Finally, we found that the atmosphere in which the process was conducted had a significant impact on the efficiency of the esterification reaction. This process must be carried out under dry argon, using standard Schlenk-line and vacuum techniques. Otherwise, the main product of the reaction practically does not form (Table 1, Entry 17). We did not observe significant differences between the process carried out in the presence of carbene generated in situ and the reaction catalyzed by freshly isolated free carbene (Table 1, Entry 11 vs Entry 14).In the next step, the catalytic activity of the known, less sterically crowded precursors of NHC carbenes (IMes, IPr) was also verified. Surprisingly, these salts were found inactive, as the conversion of substrates reached only ca. 5% and no products of esterification were observed (Table 2).Table 2 Optimization of the type of NHC precursor and base.Scope of the reactionHaving the optimized conditions in hand, the range of substrates was extended to determine the versatility of the method. We tested the reactivity of commercially available α,β-unsaturated aldehydes (1a-j) toward selected benzyl (2a) and alkyl (2b-f, 2i-l) alcohols as well as phenols (2 g, 2 h) (Fig. 4).Figure 4Substrate scope and overview of the obtained products. Isolated yields are given in parentheses.In the first series of experiments, we checked the reactivity of a broad range of commercially available α,β-unsaturated aldehydes (1a-j) toward benzyl alcohol (2a), obtaining a wide variety of esters (P1-P10) in very good isolated yields with low organocatalyst concentrations and under mild reaction conditions. Fortunately, alkyl and aryl enals could be readily adopted in this protocol. No meaningful difference in the efficiency of the process for aryl aldehydes with both electron-withdrawing and electron-donating substituents was noted. Next, the catalytic properties of NHC salt A were evaluated in the esterification of a series of alcohols (2a-j) with cinnamaldehyde (1a). As shown in Fig. 4, the proposed method can be successfully applied to all alkyl- and aryl-substituted primary alcohols. Disappointingly, secondary and tertiary alcohols were found unsuitable for our catalytic system. When we used these compounds, their conversions were below 10%. The alcohols having unsaturated bonds in their structures were also applied in the reactions with 1a and exhibited high activity leading to desired products (P11 and P12). This opens up the possibility of synthesizing esters that may contain groups susceptible to further modification. Analogously, we carried out experiments using phenols instead of alcohols. As a result of the reactions, we observed the formation of esters, as confirmed by the analysis of their isolates from the post-reaction mixture (P16, P17). In this instance, as in other studies reported by our research group17, the course of the reaction depended on the type of the organocatalyst used. A significant steric hindrance promotes intramolecular proton transfer, leading to the formation of an ester rather than an SMA product (See section: Mechanistic studies). We isolated all products (P1-P17) in order to develop a universal method for their separation from the reaction mixture.In the optimized reaction systems, we have also checked the possibility of multiesters formation. Relevant tests were performed using cinnamaldehyde (1a) and catechol (3a) at the molar ratio 2:1 (Fig. 5).Figure 5Bis-esterification of cinnamaldehyde (1a) with catechol (3a).To our great satisfaction, the reaction led to the expected disubstituted derivative and product P18 was isolated in a pure form in 86% yield. It is worth mentioning that we performed an additional reaction starting from equimolar amounts of reagents 1a and 3a in an attempt to achieve the selective esterification of only one hydroxyl group of 3a. Unfortunately, this trial gave mixtures of mono- and disubstituted product at the 6:4 ratio. Decrease in the reaction temperature from 80 to 25 °C or addition of diol in small portions did not improve this selectivity.To highlight the utility of our procedure for the coupling between polyols and enals, we conducted experiments with the use of two different types of SQs depicted below (Fig. 6). We turned our attention to functionalization of compounds of this kind because, according to our knowledge, there are no literature reports on the functionalization of SQs containing hydroxyl groups with enals. Moreover, the materials based on SQs have the unique properties determining the directions of their versatile applications21, Particularly attractive is the possibility of using them in diverse areas of medicine, e.g. as drug delivery platforms to specific target sites22,23,24,25,26,27,28,29,30 or as the functional group carriers in photodynamic therapy and bioimaging31,32.Figure 6Structures of SQs used as substrate.Finally, we obtained two novel hybrid materials of structures depicted in Fig. 7. We also made attempts to functionalize the other hydroxyl group(s) in the products P19 and P20, at the secondary carbon atoms. In order to do this, we run the reaction of SQ-2OH using twofold excess of 1a and the reaction of DDSQ-4OH using fourfold excess of 1a. No expected products were obtained, which is consistent with the method limitation given in Fig. 4Figure 7Structures of the obtained ester moieties containing SQs.Application of selected product in hydrosilylation processNext, we have focused on potential applicability of the selected material. For further studies, we chose derivative P11 containing a functional group susceptible to subsequent modifications. This choice permitted designing compounds that are excellent synthons for the synthesis of complex molecules of well-defined structures and interesting properties. Thus, P11 bearing terminal unsaturated C≡C bond was verified in hydrosilylation processes (Fig. 8).Figure 8Usability of P11 in hydrosilylation process. Isolated yields are given in parentheses.The proposed transformation was carried out under optimal conditions for terminal alkenes, established in our previous work33. As a result, we obtained a variety of organic derivatives with good to excellent isolated yields, ranging from 89 to 97%. It should be emphasized that in each reaction we observed a quantitative conversion of reagents. In the reaction systems proposed, neither side products nor other isomers of products P21-P23 were observed to form.The successful functionalization of P11 prompted us to probe the feasibility of a one-pot procedure leading to products P21-P25. To accomplish this goal, we performed esterifications of 1a with 2b and subsequently use the obtained derivative P11 in transformations depicted in Fig. 9.Figure 9One-step synthesis of P21-P25. The isolated yields are given in parentheses.The preparative scale of synthesisIn the next step, we conducted a larger number of tests to reliably determine the scalability of the described method. The results, which are summarized in Table 3, indicated that the proposed methodology has a significant application potential. The conducted tests have clearly demonstrated that the described method allows the production of esters on a large scale, while maintaining high efficiency. This synthetic pathway permitted reduction of the amount of catalyst by half, while the amount of base by three times when compared to the amounts needed in the previously described methods, and does not require any excess of substrates. Additionally, what is important from the environmental perspective, only small amounts of the green solvent-acetone-are needed, rather than harmful toluene.Table 3 Scaled-up synthesis of P1.In accordance with the standard practice of our laboratory, all catalytic tests were repeated three times and the obtained results indicated a high reproducibility of the method. To better verify the reproducibility of yields and reaction conditions, the model reaction was repeated five times. The standard deviation for the obtained yield values was calculated to be around 2. The results of the statistical analysis demonstrate good reproducibility and reliability of the described method (See ESI).Mechanistic studiesFinally, based on our research of thioester synthesis from α,β-unsaturated aldehydes and alcohols catalyzed by bulky NHC16 and the deuterium-labelling experiment (See ESI), the following mechanism of this process is proposed (Fig. 10).Figure 10Proposed mechanism of esterification reaction.The reaction mechanism commences from the free carbene (NHC), which undergoes the reaction with α,β-unsaturated aldehyde. This reaction can lead to Breslow intermediate (A-I), which remains in equilibrium with its homoenolate form (A-II). 13,34,35 In the next step a direct proton transfer between the hydroxyl group and γ carbon atom can occur. 36,37,38 In this case there is no need to use alcohol excess or any additive, facilitating proton transfer, which was inevitable in the earlier developed systems. Finally, intermediate A-III reacts with alcohol to form the reaction product after the imidazole elimination and catalyst regeneration.Importance of the developed methodOur studies have brought about an efficient and eco-friendly synthetic pathway for obtaining esters which are important and valuable compounds in modern organic chemistry. In our opinion, the presented results are very attractive from the cognitive, economic and ecological points of view and the procedure proposed is a metal-free alternative to the hitherto applied catalytic methods described in literature. As shown below the system described and optimized proved to be very attractive when compared to the known organocatalytic protocols for obtaining esters. To make the comparison credible, we based it on the results obtained using similar reagents, i.e. α,β-unsaturated aldehyde and alkyl alcohol (Table 4).Table 4 Comparison of the proposed method with the best known organocatalytic protocols.The tabulated comparison shows that in the case of a catalytic system proposed by us (A), it is possible to use smaller amounts of the organocatalyst and the base when compared to those needed in the systems based on non-bulky salts (D-F). Our system also ensures the highest product formation efficiency in the shortest time (24 h vs 40 h) and at the lowest reaction temperature (60 °C vs. 100 °C). These aspects demonstrate the superiority of the proposed method over the known procedures, both financial and environmental. Additionally, the systems described in literature have been tested only for methanol, ethanol and butanol, whereas the system based on precursor A has been examined for a wide range of alcohols and aldehydes. Therefore, the proposed system can be applied to a greater number of substrates than the other systems described in literature so far.A significant advantage of the described method is also its pro-ecological nature. A comparison of its environmental impact with those of the other known methods was conducted using The Green Degree Method (GDM)39. Within this study, the catalyst, the base and the solvent have been examined taking into account such aspects as global warming potential (GWP), ozone-depleting potential (ODP), photochemical ozone creation potential (PCOP), acidification potential (AP), eutrophication potential (EP), ecotoxicity potential to water (EPW), ecotoxicity potential to air (EPA), human carcinogenic toxicity potential to water (HCPW), human non-carcinogenic toxicity potential to water (HNCPW). In the described method, acetone is used as the solvent, for which the calculated Green Degree (GD) is -2.3072 (-[GWP], -[ODP] 0.094 [PCOP], -[AP], -[EP], 0.00962 [EPW], 0.0132 [EPA], -[HCPW], 0.006538 [HNCPW]). For comparison, toluene, used as a solvent in the other synthesis methods of esters, described in literature, has a GD of -5.4919 (-[GWP], -[ODP] 0.637 [PCOP], -[AP], -[EP], 1.6269 [EPW], 0.0025 [EPA], -[HCPW], -[HNCPW]). This value is more than twice as high as for acetone. Safety data sheets and literature sources concerning the employed supersteric precursor salt (A) and the base (KHMDS) do not report them as hazardous or harmful to the environment. They do not undergo bioaccumulation and are not considered toxic. Therefore, their environmental impact is negligible and can be omitted when determining the overall environmental impact. Considering all aspects, it can be concluded that the proposed method does not have a significant negative impact on the environment and is the most environmentally friendly among all the presented synthesis routes of esters from alcohols and α,β-unsaturated aldehydes.Because of full elimination of the need of using metals, oxidants and any other additives, besides the application of a stoichiometric ratio of reagents, low catalyst loading, eco-friendly solvent and full chemoselectivity, the proposed method can be used for the synthesis of products of the cosmetic, food and pharmaceutical industries. In these areas, there is a need to move away from traditional catalytic methods due to the particularly stringent requirements regarding the purity and quality of products. 5 The proposed strategy is, for example, an ideal solution to improve the production of esters from polysaccharides, which find applications as non-toxic materials for food packaging or prebiotics. 40 Moreover, we have demonstrated that the proposed approach can be easily extended to various types of polyhedral oligomeric silsesquioxanes (SQ), which offers a great chance of using the designed materials as non-toxic drug nanocarriers. Recent development of efficient drug delivery platforms to specific target sites in a controlled manner have attracted wide interest because it is critical for implementation of current advances in diverse areas of medicine. From this point of view, SQs have many benefits. 1 Because of their nanoscale size and high charge density, SQ units can be easily transferrable through vascular pores, which considerably improves the uptake by tissues. Unlike traditional organic compounds, SQ derivatives are colorless, nonvolatile, odorless, non-toxic and can be easily functionalized through introduction of reactive organic groups to the vertex positions via various chemical methods. After suitable modifications, the SQ units gain water-solubility which allows the SQ-based drug delivery systems to be taken orally. Additionally, they can be successfully used as the functional group carriers in photodynamic therapy and bioimaging. 2 Their great advantage is also high thermal stability, simple synthesis and well-defined 3D structure that eliminates the that arise from the polydispersity of linear polymers. Therefore, the SQs derivatives described in the manuscript seem particularly attractive for applications in the medical and pharmaceutical fields, because they contain the reactive functional groups susceptible to further modification (including deposition of therapeutic substances on them).
