Introduction of versatile and recyclable network poly (ionic liquid)s as an efficient solvent with desired properties for application in C-C cross-coupling reactions

doi:10.21203/rs.3.rs-1957171/v1

Download PDF

Research Article

Introduction of versatile and recyclable network poly (ionic liquid)s as an efficient solvent with desired properties for application in C-C cross-coupling reactions

https://doi.org/10.21203/rs.3.rs-1957171/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Synthesis of high viscosity, ecofriendly, and low melting point poly(ionic liquid)s (LMP-PIL) with network structure has been introduced by the reaction of TAIm[I] ionic liquid monomer with disodium salts of various open-chain carboxylic acids as well as sulfonic acids via a polycondensation on counter ions (PC-CI). Changes in methylene group numbers of carboxylate, as well as sulfonate, provide a broad spectrum of properties in terms of viscosity, density, and solubility for the resulting PILs. The physical properties of the PILs including, viscosity, melting point, boiling point, density, M¯_n , and DPI, were determined in this study. The ecofriendly PILs served as an efficient solvent and ligand for the Cu-catalyzed Heck and Sonogashira C-C coupling reactions. High to excellent yields were obtained for various coupling products under mild conditions at MP of the PILs. The prepared PILs could be recycled several times from the homogenous reaction mixture with preservation of the intrinsic properties. Also, a deep study was conducted on the mechanism of the coupling reactions.

Poly(ionic liquid)s

TAIm[I] ionic liquid

Heck reaction

Sonogashira coupling

Low melting point

Green solvent

Solvents play an undeniable and vital role in organic synthesis, so depending on the solubility of the reactant, catalyst, and reaction mechanism, any organic synthesis needs a solvent with certain intrinsic properties such as viscosity, density, melting point, boiling point, molecular weight, and polarity (dielectric constant), and the suitable solubility to achieve the highest reaction efficiency.^1,2 The choice of solvent for an organic reaction is usually based on the ability to create a suitable concentration of reactants (good solubility), inertness towards the reactants and products, the possibility of separating products from the reaction mixture (ease of work-up), high product efficiency, and lack of by-products formation.² Increasing attention to environmental issues also adds to the non-toxicity of the solvent and its recoverability, to this collection.^3,4 Large amounts of organic solvents used in organic reactions, enter them to the environment, which is not only toxic and expensive, but also are non-recyclable and non-biodegradable.⁵ These solvents not only disrupt the ecosystem, but also make them uneconomical to recover. Therefore, designing a cost-effective, environmentally friendly, recyclable solvent is essential for application in organic synthesis.

Pro-environmental researchers are always looking for the effective alternative green solvents instead of toxic and hydrocarbon ones, use of solvent-free conditions, and reducing the solvent amount in organic synthesis. Water, deep eutectic solvents (DES), ionic liquids (ILs), bio-based solvents,⁶ glycerol,⁴ PEG, scCO₂,⁷ and ethyl lactate⁸ are some of these alternative solvents developed by researchers.³

The high importance of Heck and Sonogashira C-C cross-coupling reactions in the synthesis of new compounds, preparation of vital drug molecules, as well as application in the preparation of valuable compounds in the fields of material science, natural products, pharmaceuticals, biologically active molecules, and heterocycles synthesis, has received much attention in organic synthesis.^9–12 The chemistry of these reactions is well known, and they are traditionally catalyzed by the Pd-based catalysts along with a toxic and expensive phosphine ligand. The high cost and toxicity of Pd have led to the consideration of Pd-free methods for the C-C bond formation. The formation of these bonds through the reactions of Heck and Sonogashira is of particular importance in the synthesis of biologically active molecules, natural products, conjugated polymers, dendrimers, and heterocycles, both in industry and in academia.^11,12 Therefore, researchers are always trying to come up with greener and less expensive methods for preparing these compounds; which (i) the use of alternative transition metals such as Cu, Co, Ag, and Rh, that are cheaper and more environmentally friendly,¹³ and (ii) the use of greener solvents are two major strategies to modify the protocols used in C-C bond formation from an environmental point of view. In addition, glycerol,⁴ bio-based solvents,⁶ ethyl lactate,⁸ and PEG⁷ were successfully utilized for the C-C cross-couplings as the green solvents. Cheng et al. also developed a gram scale approach for the synthesis of dimethyl sulfone, as a green and recyclable solvent with a low melting point for Cu-catalyzed C-C cross-coupling reactions.⁹ The application of lipids (fish oil, butter, vegetable oils, and waxes)¹⁴ and Cyrene (dihydrolevoglucosenone),¹⁵ as a dipolar aprotic solvent, was recently investigated and developed in C-C bond formation. Recently, Fantoni et al. used N-hydroxyethylpyrrolidone (HEP): H₂O mixture, as a recyclable biogenic solvent for the Heck-Cassar-Sonogashira cross-coupling reaction.¹⁶

ILs and Poly(ionic liquid)s (PILs) are another category of green solvents that, due to their multiple and reliable performance in organic synthesis, their application has expanded in many organic reactions and various industries.^17,18 PILs are a group of polyelectrolytes consisting of ionic repeating units on a polymer chain.¹⁹ They have unique properties in a polymer structure, which are different from their corresponding similar ionic liquids monomers. Depending on the cation and anion type, PILs have different properties, but they are common in ion conductivity, thermal/ electrochemical stability, and nonflammability.^17,18 In addition, their antimicrobial properties are well known.²⁰ The wide range of physicochemical properties of PILs resulting from structural diversity in the corresponding ionic liquid monomers, and also the possibility of its manipulation, has led to its use in various fields of science. Preparation of PILs with porous structure,²¹ magnetic PILs,²² and colloidal particles²³, are some of these manipulations.

The application of PILs are well known in proton exchange membrane fuel cells,²⁰ polymer electrolytes,²⁴ composite materials,¹⁹ morphosynthesis of inorganic materials,¹⁷ 3D printing,²⁵ sensing applications,²⁶ and electrochromic devices.¹⁸ PILs are also good candidates for the composite materials science.¹⁹

Various methods have been developed for the preparation of PILs, among which (1) radical polymerization of ionic liquid monomers,²⁷ (2) creation of ionic sites on prefabricated polymers (modification/functionalization of polymers),²⁸ (3) condensation polymerization of bi-functional ionic liquids,²⁹ and (4) copolymerization of ionic liquid monomer with polyethylenes³⁰ are the main methods used (Scheme 1).

Recently, Ravula et al. synthesized PILs with dicationic pendant groups based on radical polymerization of norbornene as gas separation membranes.³¹ Also, in another recent report, the evaporation-assisted phase separation (EAPS) approach was served to the synthesize of PIL microbeads.³²

Despite extensive advances in the preparation of PILs, condensation polymerization on ionic monomers with condense of counter ions has not been reported so far. This synthesis approach allows the production of PILs with a low degree of polymerization, and unlike in the past, polymers with low melting points can be achieved (Scheme 2). PILs are usually prepared on the side branches of the pre-synthesized polyethylenes and, contrary to their name, are solid and insoluble, have high melting points, and therefore their use as a solvent has been restricted in organic reactions.

In this work, low melting point-PILs were prepared by polycondensation reaction of TAIm[I] ionic liquid monomer with disodium dicarboxylate and disodium disulfonate. To the best of our knowledge, this is the first report on the preparation of PILs via a polycondensation reaction, wherein NaI, is a component that is condensed by polymerization. The application of the resulting PILs was evaluated as a versatile solvent and ligand, that provides a suitable medium under mild conditions for the green Pd-free C-C cross-coupling reactions. The intrinsic properties of the ILs including, density, viscosity, average molecular weight, polydispersity index, melting point, and boiling point, were studied in this work.

Studies over the ILs properties

The structure and molecular weight of the PILs were studied by EDX and GPC analyses, respectively. Table 1 shows the elemental analysis of the PILs by EDX analysis. The prepared PILs using disodium dicarboxylate, and disodium disulfonate were briefly named PIL-DCA-n (n = 1-4) and PIL-DS-n (n = 1-4), respectively, where “n” represents the number of methylene groups in DCA or DS. The weight percentage ratio of C/N and C/O elements was completely in accordance with the ratios observed in one monomeric unit and confirmed their polymer structure. The presence of small amounts of Na and I in PILs indicate the terminals of the polymer chains.

GPC analysis of the PILs showed that they have a range of 13000-15000 g/mol, and that with increasing the chain length of dicarboxylate or disulfonate (n), the increased accordingly (Table 1). As shown in Table 1, the measured for PIL-DS-n was larger than PIL-DCA-n. PDI measurements for the PILs also showed that all had PDI> 2.

As shown in Table 1, PIL-DS-n (n = 1-4) and PIL-DCA-n (n = 1-4) have a PDI between 3.06-3.66 and 2.53-3.52, respectively. Specifically, PDI increased with increasing of the PILs. The results were completely consistent with the reported PDI of polymers that polymerize from several active centers bearing random branching.^33-35 The results showed that the PILs with similar chains (corresponding comparison of the PILs with each other), the PDI value for PIL-DS-n (n = 1-4) was higher than PIL-DCA-n (n = 1-4). According to the results, the degree of polymerization (DP) was 28 for PIL-DS-n (n = 1-4), 30 for PIL-DCA-1, and 31 for PIL-DCA-n (n=2-4). According to the results, the length of methylene groups in dicarboxylate or disulfonate gives various properties to the resulting PILs, so that a suitable PIL can be prepared for a specific purpose. All solvents have pH=7.2-7.6 with a pale-yellow appearance and were miscible in distilled water.

Table 2 shows the physical properties of the PILs, including viscosity, density, m.p., b.p., and solubility in water.

Due to the solidity of the PILs at room temperature, the viscosity was measured by the Brookfield model DV-III Ultra Programmable (Cone and Plate) Rheometer at the melting point of the PILs.³⁶ The highest viscosities related to PIL-DS-4 and PIL-DCA-4 was equal to 67 mPa·s (at mp = 110 °C) and 55 mPa·s (at mp = 98 °C), respectively. By reducing the chain length and increasing the melting point of the PILs, the viscosity decreased accordingly. PIL-DS-1 and PIL-DCA-1 have the lowest melting point. The results clearly showed that PIL-DS-n (n=1-4) had a higher viscosity than PIL-DCA-1 (In general, PIL-DS-n and PIL-DCA-n with similar chain lengths have a difference in viscosity of approximately 10 mPa·s).

A clear linear relationship and correlation were observed between viscosity and density parameters, so that with increasing viscosity in the PILs with longer chains (the number of dicarboxylate (or disulfonate) methylene groups in the PILs), the density also increases. As shown in Table 2, the density of the PILs were between 1.15-1.45 g/cm³for PIL-DCA-n (n = 1-4) and 1.22-1.65 g/cm³for PIL-DS-n (n = 1-4).

Measurements of melting points of the PILs showed that with increasing chain length for both PIL-DS-n and PIL-DCA-n, the melting point decreases, and the lowest melting point was related to PIL-DS-4 and PIL- DCA-4 equal to 95-100 and 100-110, respectively (Table 2). Higher melting points were obtained for PIL-DS-n (n = 1-4), which could be attributed to the sulfonate groups with a stronger interaction with TAIm cations. Also, the time interval of melting of PILs becomes smaller with increasing the chain length.

Optimization of reaction parameters

The reaction parameters over the coupling reaction of phenylacetylene with iodobenzene (model reaction) in poly(TAIm[DCA]) ILs and poly(TAIm[DS]) ILs as a solvent and ligand catalyzed by CuI and K₃PO₄ as a base reagent, respectively, were optimized. All the reactions were performed under an air atmosphere and without any ligand. At first, all the prepared solvents were studied in the model reaction. Solvent amount, reaction temperature, and base type were then optimized.

Studies over poly(TAIm [DCA])ILs and poly(TAIm [DS])ILs have shown that poly(TAIm [DS]) performs better than poly(TAIm [DCA]) with the similar chains (Figure 1a). As shown in Figure 1a, the highest efficiency belongs to PIL-DS-3 with 94% for 6 hours, while PIL-DCA-3 (with similar chain length) gave 90% efficiency. The chain length of the dicarboxylate or disulfonate also affected the performance of the PILs. Increasing the chain length from three to five, increased the efficiency for both poly(TAIm[DCA]) ILs and poly(TAIm[DS]) ILs, but PILs with six methylene groups gave less efficiency than PILs with the five groups. It seems that increasing the chain length also causes more solubility of organic reagents in the PILs that act as solvents. The higher the methylene groups in dicarboxylate and disulfonate, the higher the percentage of lipophilic groups in the resulting PIL, increasing the efficiency in the coupling products.

The results of the PILs characterization showed that the chain length (methylene groups in DS or DCA) has a direct effect on their physical properties and reaches the optimal value in a specific chain length. Increasing the chain length increases the density and viscosity, and on the other hand, reduces the melting point in the PILs. The solvent amount parameter also had a significant effect on the coupling efficiency. Studies have shown that 2.5 mg of PIL-DS-3 produces the highest efficiency for 6 h (120 °C), and at higher dilutions, the efficiency decreases (Figure 1b). Also, in amounts less than 2.5 mg, the efficiency decreased, and in the amounts of 1 and 1.5 mg of PIL-DS-3, 86% efficiency was observed. One of the advantages of PILs was their applicability at high temperatures due to their high boiling points. As shown in Figure 1c, at 120 °C, the highest efficiency was achieved for 6 h. Temperatures up to 140 °C had no effect on the efficiency, but at 160 °C, the efficiency dropped to 88%. Also, the recovery studies on PIL-DS-3 showed that (as will be shown in the next section), the high temperatures up to 120 °C, did not effect on solvent properties and reflected its high stability.

Another advantage of the PILs was their ionic network structure, making it possible to dissolve salts such as K₃PO₄ in such reactions. Figure 1d shows the effect of different bases on the model Sonogashira reaction efficiency. HMTA, K₃PO₄, Cs₂CO₃, and K₂CO₃ produced moderate to good efficiency for this reaction, and the highest of which belonged to K₃PO₄ with 94% yield. Bases such as KOH and DABCO made low efficiency of 20% and 46%, respectively.

Catalytic activity

Upon obtaining the best conditions for the coupling reactions, various derivatives of aryl halides were used to prepare Heck and Sonogashira products. Tables 3 and 4 show the results of this study for Sonogashira and Heck reactions, respectively. All the reactions were performed in PIL-DS-3 without any ligand agent. As shown in Table 3, good to excellent efficiency has been reported for all derivatives. Accordance to the previous reports on the Pd-catalyzed,¹⁰ as well as Cu-catalyzed cross-coupling reactions,^11,12 iodobenzene produced higher efficiencies in less time than bromobenzene and chlorobenzene. The highlight point was the achievement of moderate to good efficiencies for aryl chlorides. As shown in Tables 3 and 4, moderate to good efficiency was obtained for Sonogashira and Heck cross-coupling reactions for 18-20 h and 14-26 h, respectively. The highest efficiency belonged to 4-nitro aryl halides with efficiencies between 90-96 (6d) and 82-90 (8e) for Sonogashira and Heck reactions, respectively. In addition, aryl halides bearing electron-withdrawing substituents, accelerate the coupling and increase efficiency in both reactions. The observations reinforce the mechanism involving the steps of oxidative-addition and reductive-eliminations, in agreement with the literature.^10-12,39,40

Another advantage of the proposed method was the non-formation of the Glaser-Hay homo-coupling product despite the high applied temperature (120 °C). According to the GC-MS results, no homo-coupling products were detected.

Similar results were obtained for the Heck cross-coupling reaction under optimized conditions (Table 4). As shown in Table 4, the Heck cross-coupling reaction by styrene (8l, 8m, 8n, 8o) gave more efficiency than n-butyl acrylate. Another advantage of the method was the high selectivity of the reaction compared to the halide type. As shown in both Tables 3 (entry 5) and 4 (entry 6), Heck or Sonogashira coupling for 4-(chloro- or bromo-) aryl halide substrates occurs quite selectively in the iodine position, and no coupling by-product was observed.

Control experiments

Non-use of any ligands (especially phosphine ligands) indicates the possibility of proper coordination of Cu ions by poly(TAIm [DS]) and, or poly (TAIm [DCA]). To elucidate the ligand effect of the PILs, the Sonogashira model reaction was studied in different solvents including, DMSO, PIL-DS-3, H₂O, DMF, glycerol, EtOH, MeOH, CH₃CN, THF, and NMP (N-methyl-2-pyrrolidone) in the presence as well as in the absence of PIL-DS-3. In addition, this property was compared with the performance of a phosphine ligand (PPh₃). The results clearly showed that PIL-DS-3 was an effective reagent/ligand in the coupling reaction. As shown in Figure 2, in the absence of PPh₃or PIL-DS-3, no efficiency was observed for most studied solvents. Only for DMF and glycerol, 10 and 15% efficiencies were obtained, respectively, in agreement with the previous reports.⁹ This small amount can be attributed to the slight coordinating effect of these two solvents on the copper ions. Most importantly, the efficiency of PIL-DS-3 was almost similar to the powerful PPh₃ ligand, which confirms the function of PIL-DS-3 as an effective ligand for Cu coordination in the coupling reactions.

The performance of PIL-DS-3 as a separate solvent was also studied. As shown in Figure 2 (yellow column), PIL-DS-3 produced 94% efficiency (the highest efficiency observed among the solvents tested) without any external ligand agent. The addition of PPh₃ to the reaction mixture also did not effect on the reaction efficiency. These results also reaffirmed the function of PIL-DS-3 as a solvent as well as an effective ligand in the coupling reactions.

To study the performance of the PILs with network structure prepared in this work, two non-polymeric structures were prepared using 1-hexane sulfonic acid and heptatonic acid (Table 5). For this purpose, the TAIm[I] IL was reacted separately with 1-hexane sulfonic acid and heptatonic acid under the same conditions, and then the resulting ionic liquids were studied as a ligand and solvent for the preparation of 6a from iodobenzene and phenyl ethylene, as five separate control experiments (Table 5). In addition, for a more accurate comparison, the performance of TAIm[I] IL (2), 1-hexane sulfonic acid, and heptatonic acid were also studied. For better comparison, the efficiency of all reactions was studied for 6 h (similar to the time considered for PIL-DS-3, Table 5), in the presence of 2.5 mg of the reagents (Table 5). TAIm[I] IL (2) gave only 53% efficiency for 6a, which, in accordance with the previous reports,^40,41 confirms the coordinating and solvent effect of TAIm[I] IL (2) (Table 5, entry 1). The lack of efficiency for 1-hexane sulfonic acid and heptatonic acid was another proof of this claim (Table 5, entry 3). 9a and 9b produced 45% and 60% efficiency for 6a, respectively. The results well showed the superiority of the performance of the network structure of PILs over the non-polymeric ones (Table 5, entries 4,5). In agreement with the results of examining the effect of different PILs on the coupling reaction efficiency (Figure 1a), 9b provides higher efficiency than 9a; Sulfonate groups seem to have more compatible properties to the coordination of the transition metals as well as the role of solvent.

The results showed that the network structure of PILs provides a suitable environment for organic reactions. Scheme 3 shows the possible interactions between the PILs prepared in this work with the network structure in the Sonogashira reaction. The ionic structure of PIL causes the solubility of salts such as CuI and K₃PO₄. Also, the methylene groups in disulfonate and dicarboxylate form the lipophilic moieties in PILs, responsible for the solubility of organic molecules such as phenylacetylene and aryl halide. In addition, as shown in Scheme 3, the PILs also acts as a ligand, in agreement with the previous reports,^40,41 and catalyze the reaction by coordinating Cu species.

Recyclability of the ILs: Mechanistic view

To evaluate the recoverability of the PILs, PIL-DS-3 was individually studied in the reaction of the Sonogashira model. The advantage of the PILs was their ability to miscible in water in any proportion, which makes it possible to recover it from the aqueous medium with a simple solvent-solvent extraction (Scheme 4). Scheme 4 shows how the PIL-DS-3 was recovered from the mixture. Solvent-solvent extraction of the reaction mixture by water: ethyl acetate causes the extraction of copper salt along with PIL-DS-3 into the aqueous phase (Scheme 4, Cu/PIL-DS-3 mixture); and the coupling product also enters the organic phase. The presence of copper salt in the aqueous phase was confirmed by ICP-MS analysis. According to the results of ICP-MS (average of 3 repetitions), the extracted aqueous phase mixture contains 32.6 wt% Cu, which, compared to the initial amount of copper salt used at the beginning of the reaction, no loss due to its transfer to other phase was observed. In addition, the presence of copper in this phase was also confirmed. Extraction of this mixture with n-BuOH causes the separation of dissolved copper salt and PIL-DS-3. EDX analysis of the extracted organic phase after removing n-BuOH under reduced pressure, confirmed the presence of PIL-DS-3 in this phase (EDX analysis (mean of 5 points) Wt%: C 32.93, N 17.36, O 23.98, I 8.55, Na 1.93, S 15.25).

As shown in Figure S1, the presence of elements such as S, Na, I, and O in the corresponding EDX spectrum, confirms the successful extraction of PIL-DS-3. High-resolution XPS Cu2p analysis of the aqueous phase containing copper showed that most copper centers have an oxidation state of +1, and a small amount was in a +2 oxidation state (Figure 3). In each cycle, Cu^(I) species appear to oxidize to Cu^(II), upon contact with molecular oxygen in the air. The results were completely consistent with the previous reports on Cu^(I)-based catalysts and the presence of a mixture of Cu^(I) and Cu^(II).^42-45 Also, previously, Hu et al., reported a mixture of Cu^(I) and Cu^(II) in fiber-polyquaterniums@Cu^(I) as a polymer-supported Cu complex catalyst.⁴² However, despite the formation of Cu^(II) species, no significant reduction in efficiency was observed during successive cycles. The results also showed that in the absence of PIL-DS-3 or PPh₃, Cu^(II) species were inactive in the reaction (Figure 3). To study this issue in more depth, the Sonogashira model reaction was performed in the presence of Cu(OAc)₂.H₂O salt. Also, for a more accurate comparison, the reaction was performed both in the presence and absence of PIL-DS-3. Figure 3a shows the Cu2p-XPS in situ spectrum of the reaction in the presence of PIL-DS-3, wherein a mixture of Cu^{(I),(II),(III)} species were detected. According to these results, the concentration of Cu^(II) species was much lower than Cu^(I) and Cu^(III), whereas, in the absence of PIL-DS-3 (in DMSO and in the absence of PPh₃ ligand agent (Fig. 3b)), no efficiency was observed for 6a. As shown in Figure 3b, Cu^(II) was the only species present in the reaction and did not undergo any oxidation or reduction. In contrast, in the presence of PIL-DS-3, Cu^{(I),(II),(III)} species were detected after 3.5 h of the reaction. According to these results, the concentration of Cu^(II) species was much lower than Cu^(I) and Cu^(III) species.

The results were completely in accordance with the proposed mechanism (that will be described in the next section) and confirmed the presence of Cu^(I) and Cu^(III)species (Scheme 4). In another separate reaction, the model reaction was performed to prepare 6a in the absence of PIL-DS-3 and the presence of pph₃ ligand. Figure 3c shows the high resolution XPS Cu2p in situ analysis of the reaction mixture after 3.5 h. Compared to the XPS spectrum in the presence of PIL-DS-3 (Figure 3a), the concentrations of Cu^{(I),(II),(III)} species were different in the presence of PPh₃, and in this case, the dominant species were Cu^(II) (Figure 3c). As shown in Figure 2, the efficiency under this condition was less than PIL-DS-3. These differences can be attributed to the quantitative differences between PPh₃ and PIL-DS-3 used in the two reactions and their functional differences. However, the results confirmed the reductive capability and ligand function of PIL-DS-3 in Cu-catalyzed C-C coupling reaction under homogenous conditions.

The results of various analyzes showed that the structure and properties of the Cu/PIL-DS-3 mixture did not change, and it was subsequently recyclable. For this purpose, the recycling of Cu/PIL-DS-3 mixture, along with Cu salt and PIL-DS-3, were studied separately (see Scheme 4). Figure 4 shows the results for eight consecutive recoveries for all three experiments. The efficiency of all reactions was measured over the same time of 6 h. A slight decrease was observed for the Cu/PIL-DS-3 mixture and also the recycled PIL-DS-3 by the end of the 8^th cycle.

According to the results, PIL-DS-3 can be recovered up to at least 8 times, even in a mixture as Cu/PIL-DS-3, and the efficiency drop in each cycle cannot be observed/detected and measured, so after 8 consecutive recyclings of Cu/PIL-DS-3, the reaction yield decreased from 94% to 92% (only 2% efficiency drop), and from 94% to 90% for the recycled PIL-DS-3. The highest efficiency loss was observed for the recycled Cu, which decreased to 86% yield at the end of the 8^th cycle of 6a.

These results were quite consistent with the oxidation of some Cu^(I) species to Cu^(II) ions during successive recovery cycles. Also, in agreement with the previous reports, Cu^(I) and Cu^(III) were active species in copper-catalyzed Sonogashira reactions.^46-48 For this reason, during successive cycles with the formation of Cu^(II) species (reduction in mol% of active species), the efficiency decreases. The results of in situ XPS (after 2^th, 4^th, and 8^th cycles for Cu/PIL-DS-3 mixture and the recovered Cu salt) were also completely consistent with this theory. As shown in Figure 5a, in Cu/PIL-DS-3 mixture, copper species often were in a +1 oxidation state (~934 eV), which in cycles 4^th to 6^th, its intensity was reduced, and the intensity of peaks related to Cu^(II) species (~936 eV) was increased. However, the recovered salt showed completely different behavior in successive cycles (Figure 5b), so that, at the end of the 8^th cycle, the copper species often have an oxidation state of +2. These results were consistent with the efficiency drop observed in successive cycles for the recovered copper salts. These results also showed that PIL-DS-3 in the Cu/PIL-DS-3 mixture acts as an inhibitor of oxidation or a reducing agent relative to copper species, which prevents the formation of Cu⁽^II) species.

The results also showed that the Cu/PIL-DS-3 mixture, largely maintains the oxidation state of Cu^(I) species and prevents their oxidation. Thus, the Cu/PIL-DS-3 mixture can be used for consecutive cycles with minimal efficiency loss. These results also show high reproducibility of the reaction and cause PILs to be considered as the cost-effective solvents.

Also, due to the high stability of the PILs, it is possible to use them many times and with high reliability. For this purpose, the physical properties of the recovered PIL-DS-3 were measured after 2^th, 4^th, 6^th, and 8^th cycles and compared with the freshly prepared one. The results do not show any significant changes in the PIL properties during different cycles as well as with the freshly prepared one. Only a very slight decrease in PIL-DS-3 density was observed during successive recovery cycles. As shown in Table S1, very little difference was observed for PIL-DS-3 in terms of average M_w, PDI, and viscosity. Due to the fact that the PIL was exposed to 120 °C for 6 h in each cycle, the lack of change in the properties of the PIL indicates its thermal and mechanical stability, in such a way that it becomes a reliable and reproducible solvent in organic reactions. In order to prove the absence of copper in the recovered PIL-DS-3 (or contamination of the PIL with CuI salt), the residue in 2^th, 4^th, 6^th, and 8^th cycles was analyzed by ICP-MS to detect Cu traces. The results in these cycles did not detect any trace of Cu. These results demonstrate that PIL-DS-3 was recovered purely in each cycle and validates the proposed recovery procedure (Scheme 4). This analysis also showed that Cu coordination to PIL-DS-3 did not result in the formation of a stable heterogeneous complex, and in agreement with the mercury test (that will be discussed in the next section), the reaction took place in a completely homogeneous environment. This results in the effective recovery of the PIL from the reaction medium and its consecutive recycling while retaining all the properties.

Reaction mechanism

Due to the strong σ-donor and weak π-acceptor nature of the bis-N-heterocyclic carbene (NHC), they were successfully employed as an efficient ligand for Cu-catalyzed organic reactions.^49-51 Also, their application in C-C cross-coupling reaction is well understanding.^40,41,52 In addition, in the case of poly(TAIm [DS]) ILs, this coordination can also be performed through sulfone groups;⁹ As higher efficiency was observed in optimization experiments for poly(TAIm [DS]) IL than for poly (TAIm [DS]) IL (Figure 1).

The control experiments also confirmed the ligand role of the IL towards the suitable coordination of Cu for the C-C coupling in agreement with the literature. Therefore, Cu ions could be adsorbed on the ILs networks via the coordination to NHC ligands in accordance with the previous reports.⁵³ This coordination was simply depicted at Scheme 5, as a plausible reaction mechanism for the Sonogashira coupling reaction (intermediate I) that provides various active sites for the reaction. Then, Cu-acetylide species (intermediate II) is formed on the IL network by the addition of acetylene and in the presence of a base (here K₃PO₄).⁹ One proof of the formation of this intermediate was the lack of significant efficiency in the absence of the basic agent (Table 5, control experiments). The literature confirmed the formation of this intermediate by H-abstraction of phenylacetylene by a base reagent.^9,11,39 In following, by the oxidation-addition reaction, aryl halide (Ar-X, Scheme 5) is added to the Cu-acetylide intermediate (Intermediate III). A strong proof for this step as an oxidation-addition reaction, comes from the XPS in situ analysis of the homogenous mixture, wherein a mixture of Cu^(I) and Cu^(III),⁹ was evident as the reaction is proceeding (Fig. 6). This behavior was also observed by various groups,^54-56 confirmed the presented mechanism. As shown in Figure 6, as the reaction progresses, the intensity of the peaks corresponding to Cu^(III)in ~938 eV and ~956 eV increases, and as it approaches the end of the reaction, its intensity decreases. As shown in Fig. 6, according to the XPS spectra of the reaction mixture, it was shown that the reaction takes place in the presence of a mixture of Cu^(I) and Cu^(III) ions. But the predominant species at the end of the reaction were Cu^(I), which reflects the reductive properties of PILs (Fig. S1, Fig. 5).

This capability was also demonstrated in the presence of Cu(OAc)₂ salt, where all copper elements were initially in the +2 oxidation state (Figure 3). In the end, by a reductive elimination process, the desired product was obtained. Also, by removing X by the base (here K₃PO₄), the catalyst returns to the catalytic cycle with its original structure. According to the previous results in this work (Figure5), copper species in the recovered PIL-DS-3/Cu mixture often were almost in a +1 oxidation state and, therefore, returned to the cycle with similar properties, and were also recoverable. The reactions were performed entirely in the air atmosphere; in this way, to investigate the effect of the presence or absence of molecular O₂ in the coupling reactions, the reaction was performed under Aratmosphere. The results did not show any change in the reaction efficiency under Ar atmosphere and reflect the lack of dependence of the coupling reaction on molecular O₂ and the absence of the formation of oxidized copper species (CuO and Cu₂O) under air atmosphere.

Mercury (0) test was used to evaluate and confirm the homogeneity of the reaction and the lack of formation of heterogeneous species in the reaction (due to the interaction of PILs with Cu ions). In this test, 240 mmol of Hg(0) was added to the Sonogashira model reaction mixture once at the beginning, and once after 3 h of reaction time (Conversion (GC)= 62%). Mercury poisons heterogeneous catalytic centers including, Cu, Pd, and Ni (immobilized on a solid substrate) and blocks them from reacting,^40,57 while it does not effect on homogeneous catalytic systems. Each reaction was performed under optimal conditions and repeated twice, and the progress of the reaction was monitored by GC. According to the results, Hg(0) had no effect on the efficiency of each reaction, and the product was formed with kinetics similar to normal (in the absence of Hg(0)) (94-97% conversion).

In addition, the three-phase test for the homogeneous performance of PIL-DS-3 was also studied (Scheme 6).

PIL-DS-3 was reacted in the presence of phenylacetylene and Wang resin-bounded 4-iodebenzoate under optimal conditions. If the PIL acts heterogeneously in the reaction mixture, no reaction takes place, due to the presence of three different phases in the reaction.^39,40,57After hydrolysis of Wang resin, the efficiency was almost the same as the main reaction for product 6a (92% isolated yield). The formation of 6a confirms the reaction on the Wang resin and subsequently confirms the homogeneous performance of PIL-DS-3 in the coupling reactions.

A novel PC-CI approach was used to prepare of a variety of PILs with versatile solvent properties. Poly(TAIm[DCA]) and poly(TAIm[DS]) ILs could be efficiently served as a green ligand and solvent for the Heck and Sonogashira coupling reactions. The PILs were prepared with DPI = 28–31 and Mw ~ 13000- ~14000 g.mol^-1. The properties of the solvents included viscosity, density, solubility, and efficiency in the coupling reactions, depending on the methylene groups of dicarboxylate and disulfonate, so that a wide range of properties could be obtained by using a variety of the disodium dicarboxylate or disodium disulfonate. The results confirmed the homogenous performance of the PILs in the coupling reactions, which was responsible for the high to excellent efficiencies observed for the C-C Heck and Sonogashira coupling reactions under mild conditions. The method was versatile towards a wide variety of substrates, even aryl chlorides, bearing electron-donating or electron-withdrawing groups. Lipophilic methylene groups in dicarboxylate or disulfonate along with ionic moieties in the PILs with a network structure were responsible for the high solubility of organic compounds as well as mineral salts. Mechanism studies revealed that Cu^(I) and Cu^(III) species were involved in the reaction with direct interference of PILs as an effective ligand. No change in intrinsic properties and or activity of the PILs was observed for 8 consecutive recycling, which was accomplished by simple solvent-solvent extraction. High boiling point, low melting points, recycle ability with preservation of properties, ecofriendly, high solubility towards organic molecules, formation of a homogenous mixture composed of organic molecules and inorganic salts, ability to coordinate to transition metals, and ligand property were some of the advantages of PILs developed in this work, which suggest a versatile and ecofriendly solvent for the application in organic synthesis.

Materials and instrumentation

All reagents, solvents, and chemicals were supplied from Merck commercial corporation with analytical grade and used as received without further purification. Triton™ X-100 (t-Oct-C₆H₄-(OCH₂CH₂)_xOH, x= 9-10; Mol-Gew. micellar avg. mol wt 80,000, and average mol wt 625) with laboratory-grade was purchased from Sigma. Polymer-bound 4-iodobenzene (Wang resin) was provided from NovaBiochem with 100 – 300 mesh containing 0.64 – 1.1 mmol/g loading and cross-linked with 1% divinylbenzene. All solvents were dried before use. The progress of reactions was monitored by thin-layer chromatography (TLC) polygram SILG/UV 254 plates or by GC-MS instrument on an Agilent GC 6890 (MSD 5973, U.S.A.) equipped with N₂ as a carrier gas. Anisole was used as an internal standard, for quantitative analyses. ¹HNMR spectra were recorded on a Bruker Avance DPX 250 MHz spectrometer using DMSO-d₆ as solvent and TMS as an internal reference. FTIR analyses were performed on a JASCO FT/IR 4600 spectrophotometer using KBr dick. Energy-dispersive X-ray spectroscopy (EDX) analysis was used to study the elemental analysis of the samples on a field emission scanning electron microscope (FESEM) JEOL 7600F, equipped with a spectrometer of energy dispersion of X-ray from Oxford instruments. The viscosity of the PILs was measured according to a previously reported protocol³⁶ using a Brookfield model DV-III Ultra Programmable (Cone and Plate) Rheometer. The PILs density was determined by compressing the PILs into a cubic mold made of aluminum with a length, width, and height of 1 cm. Then, by weighing the resulting compact cube and using the r=m/V equation, the density of PILs was determined. Density measurements were repeated three times for each PILs, and the final result was reported as average.

Preparation of 1,1',1''-(1,3,5-triazine-2,4,6-triyl)tris(3-methyl-1H-imidazol-3-ium) iodide (TAIm[I]) (2)

TAIm[I] ionic liquid was synthesized and characterized according to a previously described procedure in two steps (1) synthesis of cyanuric iodide followed by (2) the preparation of TAIm[I] IL by the reaction of cyanuric iodide with 1-methyl imidazole.^40,41,58 TAIm[I] IL (as a dark yellow oil, M_w=705.0 g/mol, density=1.84 g/mL, viscosity=11.82 cP, (Table S2)) was characterized by FTIR and EDX analyses (Fig. S2).

Typical procedure for the preparation of poly(TAIm[dicarboxylic acid]) ionic liquid (PIL-DCA-n(n=1-4))

Typically, TAIm[I] IL monomer was polymerized by 3- to 6-open chains of dicarboxylic acids disodium salts in one step. Typically, for the preparation of PIL-DCA-3, a mixture of TAIm[I] IL monomer (5.0 mmol), Glutaric acid disodium salt (5.0 mmol), 5.0 mL n-BuOH, and 1.0 ml of Triton X-100 (as a surfactant) was added to a 25-mL round-bottomed flask. The mixture was degassed by N₂ bubbling for 7 min, then sealed to prevent any gas leakage. The reactor was placed into an oil bath, and the reaction temperature was adjusted to 120 °C under an N₂ atmosphere. As the reaction progressed, the formation of white NaI sediments appeared. The reaction was subjected to the conditions as mentioned earlier for 8 h. Then, the mixture was decanted and allowed to cool to room temperature. The resulting sediments (PIL- DCA-3) were decanted and washed several times with cooled diethyl ether and water, then dried and stored at room temperature.

PIL-DCA-1, PIL-DCA-2, and PIL-DCA-4 were also prepared by disodium salts of malonic acid, succinic acid, and Adipic acid, respectively, using the same procedure described for PIL-DCA-3. Fig. S3 shows the ¹HNMR of PIL-DCA-1.

Preparation of poly([TAIm[disulfonate]) ionic liquid (PIL-DS-n (n=1-4))

As the same procedure described for the preparation of PIL-DCA-n (n=1-4), PIL-DS-1, PIL-DS-2, and PIL-DS-4 were also prepared by disodium salts of methanedisulfonic acid, ethane-1,2-disulfonic acid, propane-1,3-disulfonic acid, butan-1,4-disulfonic acid, respectively. Fig. S4 shows the ¹HNMR of PIL-DS-1.

Cu-catalyzed C-C Heck and Sonogashira cross-coupling reactions in poly(TAIm[DS]) IL

The coupling reactions were performed in PIL-DS-3 as a premium solvent and ligand. For the Sonogashira cross-coupling, a homogenous mixture of aryl halide (1.0 mmol), phenylacetylene (1.0 mmol), and K₃PO₄ (1.2 mmol) in 2.5 mg of PIL-DS-3 was stirred at 120 °C for an appropriate time. The reaction progress was monitored by TLC at various time intervals. After reaction completion, 5.0 mL of distilled water and 5.0 mL of EtOAc were added to the mixture after cooling to ambient temperature. This solvent-solvent extraction transfers the coupling products and PIL-DS-3 to the organic and aqueous phases, respectively. Thus, the coupling product was obtained after removing the solvent under reduced pressure. The products were purified by column chromatography. On the other hand, to recover PIL-DS-3, 7.0 ml of n-BuOH was added to the aqueous phase, and the PIL-DS-3 was extracted to n-BuOH; Then, PIL-DS-3 was recovered by removing n-BuOH.

The Heck cross-coupling reactions were performed by the identical procedure described for Sonogashira coupling in PIL-DS-3 using 1.0 mmol of styrene (or n-butyl acrylate).

The main text of the article should appear here with headings as appropriate.

Ethical Approval

Not applicable for this work.

Competing interests

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.

Authors' contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Indrajit Patra, Duaa Abdul Rida Musa, Yasser Fakri Mustafa, Zukhra Ziyatovna Yakhshieva, and Jihad M Hadi. The first draft of the manuscript was written by Milad Kazemnejadi and Reena Solanki and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This research received no specific grant from any funding agency.

Availability of data and materials

The authors confirm that the data supporting the findings of this study are available within the article.

R.A. Sheldon, Green Chem. 7, 267 (2005)
F.P. Byrne, S. Jin, G. Paggiola, T.H. Petchey, J.H. Clark, T.J. Farmer, A.J. Hunt, C. Robert McElroy, J. Sherwood, Sustainable Chem. Processes 4, 1 (2016)
W.M. Nelson, Green solvents for chemistry: perspectives and practice. Oxford University Press. 2003, 6-30; 133-196.
A.E. Díaz-Álvarez, J. Francos, P. Croche, V. Cadierno, Curr. Green Chem. 2014, 1, 51-65.
B. Nanda, M. Sailaja, P. Mohapatra, R.K. Pradhan, B.B. Nanda, Mater. Today: Proc. 47, 1234 (2021)
M.O. Sydnes, Curr. Green Chem. 6, 96-104 (2019)
S. Paul, K. Pradhan, A. R Das, Curr. Green Chem. 3, 111 (2016)
S. Chandrasekhar, C. Narsihmulu, S.S. Sultana, N.R. Reddy, Org. Lett. 4, 4399 (2002)
S. Cheng, W. Wei, X. Zhang, H. Yu, M. Huang, M. Kazemnejadi, Green Chem. 22, 2069 (2020)
A.R. Sardarian, M. Kazemnejadi, M. Esmaeilpour, Dalton Trans. 48, 3132 (2019)
M.A. Nasseri, Z. Rezazadeh, M. Kazemnejadi, A. Allahresani, Dalton Trans. 49, 10645 (2020)
M.A. Nasseri, M. Kazemnejadi, A. Allahresani, M. HussainZadeh, New J. Chem. 45, 7741 (2021)
F. Mohjer, P. Mofatehnia, Y. Rangraz, M.M. Heravi, J. Organomet. Chem. 936, 121712 (2021)
A. Gevorgyan, K.H. Hopmann, A. Bayer, Green Chem. 23, 7219 (2021)
D. Kong, A.V. Dolzhenko, Sustainable Chem. Pharm. 25, 100591 (2022)
W. Cabri, T. Fantoni, S. Bernardoni, A. Mattellone, G. Martelli, L. Ferrazzano, P. Cantelmi, D. Corbisiero, A. Tolomelli, F. Vacondio, F. Ferlenghi, ChemSusChem 14, 2591 (2021)
M.R. Gao, J. Yuan, M. Antonietti, Chem. - Eur. J. 23, 5391 (2017)
B. Yang, G. Yang, Y.M. Zhang, S.X.A. Zhang, J. Mater. Chem. C 9, 4730 (2021)
S.Y. Zhang, Q. Zhuang, M. Zhang, H. Wang, Z. Gao, J.K. Sun, J. Yuan, Chem. Soc. Rev. 49, 1726 (2020)
H.A. Elwan, R. Thimmappa, M. Mamlouk, K. Scott, J. Power Sources 510, 230371 (2021)
R.V. Barrulas, M. Zanatta, T. Casimiro, M.C. Corvo, Chem. Eng. J. 411, 128528 (2021)
S. Sadjadi, J. Mol. Liq., 2021, 323, 114994.
Y. Men, D. Kuzmicz, J. Yuan, Curr. Opin. Colloid Interface Sci. 19, 76 (2014)
A.S. Shaplov, R. Marcilla, D. Mecerreyes, Electrochimica Acta, 175, 18 (2015)
H. Nulwala, A. Mirjafari, X. Zhou, Eur. Polym. J. 108, 390 (2018)
R. Guterman, M. Ambrogi, J. Yuan, Macromol. Rapid Commun. 37, 1106 (2016)
W. Zhang, J. Yuan, Macromol. Rapid Commun. 37, 1124 (2016)
J. Yuan, D. Mecerreyes, M. Antonietti, Prog. Polym. Sci. 38, 1009 (2013)
T.O. Magalhaes, A.S. Aquino, F. Dalla Vecchia, F.L. Bernard, M. Seferin, S.C. Menezes, R.O.S.A.N.E. Ligabue, S.A.N.D.R.A. Einloft, Rsc Adv. 4, 18164 (2014)
Y. Yang, J. Zheng, S. Man, X. Sun, Z. An, Polym. Chem. 9, 824 (2018)
S. Ravula, K.E. O’Harra, K.A. Watson, J.E. Bara, Membranes 12, 264 (2022)
Q. Lei, F. He, X. Zhao, J. Yin, Macromol. Chem. Phys. 223, 2100379 (2022)
J.W. Peeters, A.R. Palmans, E.W. Meijer, C.E. Koning, A. Heise, Macromol. Rapid Commun. 26, 684 (2005)
Q. Luo, Z. Duan, X. Li, L. Gu, L. Ren, H. Zhu, X. Tian, R. Chen, H. Zhang, Q. Gong, Z. Gu, Adv. Funct. Mater. 32, 2110408 (2022)
Q. Jiang, Y. Zhang, Q. Jiang, Y. Du, L. Jiang, W. Huang, X. Xue, H. Yang, B. Jiang, J. Polym. Sci. 59, 404 (2021)
J.M. Crosthwaite, M.J. Muldoon, J.K. Dixon, J.L. Anderson, J.F. Brennecke, J. Chem. Thermodyn. 37, 559 (2005)
H. Rodriguez, J.F. Brennecke, J. Chem. Eng. Data 51, 2145 (2006)
C.C. Ho, J.W. Evans, P.K. Wright, J. Micromech. Microeng. 20, 104009 (2010)
M. Kazemnejadi, Z. Rezazadeh, M.A. Nasseri, A. Allahresani, M. Esmaeilpour, Green Chem. 21, 1718 (2019)
Q. Min, P. Miao, D. Chu, J. Liu, M. Qi, M. Kazemnejadi, Catal. Lett. 151, 3030 (2021)
Y. Zhu, G. Xu, M. Kazemnejadi, New J. Chem. 45, 11662 (2021)
Q. Hu, X.L. Shi, Y. Chen, F. Wang, Y. Weng, P. Duan, J. Ind. Eng. Chem. 69, 387 (2019)
J.M. Hoover, B.L. Ryland, S.S. Stahl, J. Am. Chem. Soc. 135, 2357 (2013)
H. Sterckx, J. De Houwer, C. Mensch, I. Caretti, K.A. Tehrani, W.A. Herrebout, S. Van Doorslaer, B.U. Maes, Chem. Sci. 7, 346 (2016)
J. Chen, P. Sun, X. Zhou, Y. Zhang, C.H. Huang, Environ. Sci. Technol. 49, 4218 (2015)
Y. Liu, V. Blanchard G., Danoun, Z. Zhang, A. Tlili, W. Zhang, F. Monnier, A. Van Der Lee, J. Mao, M. Taillefer, ChemistrySelect 2, 11599 (2017)
X. Wang, Y. Song, J. Qu, Y. Luo, Organometallics 36, 1042 (2017)
D. Domyati, R. Latifi, L. Tahsini, J. Organomet. Chem. 860, 98 (2018)
X. Zhang, B. Wang, Y. Lu, C. Xia, J. Liu, Mol. Catal. 504, 111452 (2021)
A. Pawar, S. Gajare, A. Jagdale, S. Patil, W. Chandane, G. Rashinkar, S. Patil, Catal. Lett. 152, 1854 (2022)
M. Kazemi Miraki, M. Arefi, A. Salamatmanesh, E. Yazdani, A. Heydari, Catal. Lett. 148, 3378 (2018)
M.T. Chen, B.Y. Hsieh, Y.H. Liu, K.H. Wu, N. Lussari, A.A. Braga, Appl. Organomet. Chem. 34, e5870 (2020)
F. Lazreg, F. Nahra, C.S. Cazin, Coord. Chem. Rev. 293, 48 (2015)
A.M. Thomas, A. Sujatha, G. Anilkumar, RSC Adv. 4, 21688 (2014)
C. Rajalakshmi, S.S. Jibin, R. Sulay, S. Asha, V.I. Thomas, G. Anilkumar, Polyhedron, 193, 114869 (2021)
A.M. Thomas, S. Asha, R. Menon, G. Anilkumar, ChemistrySelect 4, 5544 (2019)
M. Kazemnejadi, S.A. Alavi, Z. Rezazadeh, M.A. Nasseri, A. Allahresani, M. Esmaeilpour, J. Mol. Struct. 1186, 230 (2019)
Y. Hu, M. Ren, M. Kazemnejadi, ChemistryOpen 10, 775 (2021)

No competing interests reported.

SupportingInformation.docx
Tables.docx
Scheme1.png
Scheme 1. (a)-(d) The previously reported protocols for the synthesis of PILs in comparison with (e) this work as a polycondensation on counter ions (PC-CI)
Scheme2.png
Scheme 2. Preparation of poly(TAIm[DCA]) ILs and poly(TAIm[DS]) ILs
Scheme3.png
Scheme 3. Possible interaction of salts and organic molecules with PILs as a suitable ligand and solvent for organic molecules and inorganic salts
Scheme4.png
Scheme 4. An experimental view of the Cu salt and PIL-DS-3 (as a case example) recycling
Scheme5.png
Scheme 5. A plausible reaction mechanism for the Cu-catalyzed C-C Sonogashira coupling in the presence of (poly(TAIm[DCA]) IL) or (poly(TAIm[DS]) IL)
Scheme6.png
Scheme 6. The three-phase test of Wang resin-bounded 4-iodebenzoate and phenylacetylene in the presence of PIL-DS-3.

Download PDF

Version 1

posted

You are reading this latest preprint version

Introduction of versatile and recyclable network poly (ionic liquid)s as an efficient solvent with desired properties for application in C-C cross-coupling reactions

Status:

Version 1

Abstract

Figures

Introduction

Results And Discussion

Conclusion

Experimental

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1