Chiral Sulfide/Phosphoric Acid Cocatalyzed Enantioselective Intermolecular Oxysulfenylation of Alkenes with Phenol and Alcohol O -Nucleophiles

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE3 Oct 2022Chiral Sulfide/Phosphoric Acid Cocatalyzed Enantioselective Intermolecular Oxysulfenylation of Alkenes with Phenol and Alcohol O-Nucleophiles Xiao-Dong Liu†, Yicong Luo†, Xiaohong Huo, Hui-Yun Luo, Ren-Fei Cao and Zhi-Min Chen Xiao-Dong Liu† Shanghai Key Laboratory for Molecular Engineering of Chiral Drugs, School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 †X.-D. Liu and Y. Luo contributed equally to this work.Google Scholar More articles by this author , Yicong Luo† Shanghai Key Laboratory for Molecular Engineering of Chiral Drugs, School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 †X.-D. Liu and Y. Luo contributed equally to this work.Google Scholar More articles by this author , Xiaohong Huo Shanghai Key Laboratory for Molecular Engineering of Chiral Drugs, School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Hui-Yun Luo Shanghai Key Laboratory for Molecular Engineering of Chiral Drugs, School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Ren-Fei Cao Shanghai Key Laboratory for Molecular Engineering of Chiral Drugs, School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author and Zhi-Min Chen *Corresponding author: E-mail Address: [email protected] Shanghai Key Laboratory for Molecular Engineering of Chiral Drugs, School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101590 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Enantioselective intermolecular three-component oxysulfenylations of alkenes with phenols and alcohols as O-nucleophiles were achieved using chiral BINAM-derived sulfide/phosphoric acid as the cocatalyst. Notably, chiral BINAM-derived sulfide/phosphoric acid co-catalysis was first explored to successfully catalyze highly enantioselective transformation. A variety of sulfur-containing benzylic aryl ethers and benzylic alkyl ethers, which contained two contiguous chiral stereocenters, were obtained readily in moderate to excellent yields with high to excellent enantioselectivities. This protocol featured a broad substrate scope with excellent functional group tolerance and high chemo-, regio-, and enantioselectivity. Computational studies were performed to explain the origin of the high selectivity. Additionally, the calculation results indicated that the supposed commonly racemization “olefin-to-olefin” transfer process of thiiranium ion did not occur in this system. Download figure Download PowerPoint Introduction Catalytic asymmetric difunctionalization of internal alkenes is a fundamental and indispensable reaction that provides a facile and noteworthy method for introducing various functional groups into double bonds with the simultaneous construction of two contiguous stereocenters.1–6 Accordingly, several efforts to explore synthetic methodologies in this research area and numerous synthetic strategies have been documented.7–19 For example, catalytic asymmetric intramolecular electrophilic sulfenofunctionalization reactions of alkenes that allow straightforward access to beneficial chiral organosulfur compounds have been well developed recently (Figure 1a).20–25 Pioneering work by Denmark and co-workers26–34 has successfully demonstrated the power of Lewis base catalysis for sulfenium ion transfer to alkenes. The intermediate thiiranium ions have been shown to react with a wide variety of oxygen, nitrogen, and carbon nucleophiles in good yields and with high selectivities. Zhao and co-workers35–39 have also successfully developed a suite of such catalytic enantioselective reactions using chiral bifunctional chalcogenide catalysts. Despite these advances, catalytic asymmetric intermolecular three-component sulfenofunctionalizations of alkenes, which could emerge as a convenient approach to more complex chiral sulfides from simple alkenes, are less investigated. Meanwhile, studies have shown that this system is often less reactive and least favorable in providing high selectivity, thus, currently very challenging.26,40–45 Figure 1 | Design of catalytic enantioselective intermolecular oxysulfenylation of alkenes. Download figure Download PowerPoint Enantiomerically enriched benzylic ethers such as benzylic aryl ethers and benzylic alkyl ethers are valuable compounds because they are widely present in a range of medicinally relevant molecules, as well as common synthetic intermediates (Figure 2).46–51 Therefore, several methods for preparing enantioenriched benzylic ethers have been described.52–59 Although catalytic enantioselective oxyfunctionalizations of readily available 2-substituted styrenes using different phenol derivatives as the intermolecular O-nucleophile provides a modular and complementary approach to a diverse array of chiral benzylic aryl ethers, they remain underdeveloped (Figure 1b).60 We speculated that three main challenges limit the success of the above intermolecular reactions, namely, (1) relatively weak nucleophilic capability of phenol derivatives, which could cause the generation of by-products or decomposition of alkenyl substrates, (2) the indistinguishable C/O chemoselectivity of phenol derivatives and the possible intramolecular rearrangement process of phenolic ethers under strongly acidic conditions that could result in the complexity of the reaction system,45,61,62 and (3) the lack of a suitable and mild catalytic system for the control of high enantioselectivity. Recently, we reported an enantioselective sulfenylation/semipinacol rearrangement of allylic alcohols using chiral selenide/chiral phosphoric acid as the cocatalyst63 and some successful examples of intramolecular sulfenylation of alkenes with phenols as the O-nucleophile under mild reaction conditions.64,65 In this context, we aimed to challenge catalytic enantioselective intermolecular three-component oxysulfenylations of alkenes with phenols and alcohols as O-nucleophiles for the modular synthesis of sulfur-containing benzylic aryl ethers and benzylic alkyl ethers. Herein, we report our preliminary results for these two reactions of 1,2-disubstituted alkenes using chiral 1,1′-binaphthyl-2,2′-diamine (BINAM)-derived sulfide/phosphoric acid as the cocatalyst (Figure 1b). Figure 2 | Selected medicinally relevant molecules with benzylic ether moieties. Download figure Download PowerPoint Experimental Methods Under argon atmosphere, chiral phosphoric acid [(R)- CPA, 3.6 mg, 0.01 mmol, 0.1 equiv], chiral sulfide catalyst 1a (4.7 mg, 0.01 mmol, 0.1 equiv), phenol 3 (0.2 mmol. 2.0 equiv), sulfenylating agent 4 (0.12 mmol, 1.2 equiv), CH2Cl2 (0.8 mL) were charged into an 8 mL vial equipped with a magnetic stirrer bar. The mixture was allowed to stir at 0 °C for 5 min, followed by the addition of a solution of the alkenyl substrate 2 (0.1 mmol, 1.0 equiv) in CH2Cl2 (0.2 mL). The reaction mixture was stirred for another 12–86 h at the corresponding temperature. After quenching the reaction with Et3N, the solution solvent was removed in vacuo, and the crude residue was purified via column chromatography to afford the desired product. Results and Discussion In our initial study, benzyl (E)-4-phenylbut-3-enoate 2a was selected as a model substrate; 4-methoxyphenol 3a and sulfenylating reagent 4a were used as reaction partners. We were pleased to find that the desired product 5aa was obtained in 72% yield with 94% ee when chiral BINAM-derived sulfide (S)- 1a and chiral phosphoric acid (R)- CPA were used as the cocatalyst in CH2Cl2 at 0 °C (Table 1, entry 1). We observed that the chiral Lewis base played a key role in the reactivity and enantioselectivity of this reaction. Replacing 1a with selenide 1b, proven to be the optimal catalyst in several previous works,26–34 led to a reduction in enantioselectivity (entry 2). When 1,1′-spirobiindane-7,7′-diamine (SPINAM)-derived sulfide 1c was used instead of 1a, the desired product 5aa was obtained with a low yield and enantioselectivity (entry 3). An almost racemic product was obtained without 1a (entry 4). Likewise, the acid is also crucial for this reaction (more information, see Supporting Information Table S1). p-Toluenesulfonic (PTS) acid was investigated as an acid catalyst instead of (R)- CPA; however, almost no desired product was obtained with only the production of the undesired sulfenylation product of 4-methoxyphenol (entry 5). We found that the stereochemistry of phosphoric acid had little impact on the optical purity of the product, but phosphoric acid with 1,1′-Bi-2-naphthol (BINOL) framework produced the product in the best yield and high enantioselectivity (entries 6–9). Almost no reaction (NR) occurred without (R)- CPA (entry 10). Given that (R)- CPA is commercially available and not expensive, we selected (R)- CPA as the acid catalyst. Acetonitrile and tetrahydrofuran (THF) were tested as solvents, but these two solvents were not suitable for this system (entries 11 and 12). The product 5aa was obtained with a lower yield and enantioselectivity when the reaction was performed at room temperature (RT) (entry 13). Table 1 | Reaction Optimizationa Entry Change from the “Standard Conditions” Yield (%) ee (%) 1 None 72 94 2 (S)- 1b instead of (S)- 1a 64 88 3 (S)- 1c instead of (S)- 1a 11 −6 4 Without (S)- 1a 45 2 5 PTS instead of (R)- CPA Trace — 6 (S)-CPA instead of (R)- CPA 65 93 7 rac-PA instead of (R)- CPA 72 92 8 (PhO)2P(O)OH instead of (R)- CPA 58 91 9 (BuO)2P(O)OH instead of (R)- CPA 20 86 10 Without (R)- CPA Trace — 11 CH3CN instead of DCM NR — 12 THF instead of DCM Trace — 13 RT instead of 0 °C 69 90 aStandard reaction conditions: the reaction was conducted with 2a (0.1 mmol), 3a (0.2 mmol), 4a (0.12 mmol), (S)- 1a (0.01 mmol), and (R)- CPA (0.01 mmol) in CH2Cl2 (1.0 mL) at 0 °C under Ar. Isolated yields are shown. The ee was determined by HPLC. With the optimized reaction conditions in hand, the scope of this reaction was first investigated with various substrates bearing different aryl groups (Figure 3). In general, all substrates produced the corresponding benzylic aryl ethers in moderate to good yields with high to excellent enantioselectivities. Compared with the model substrate 2a, the phenyl moiety with electron-withdrawing groups did not affect the ee but obviously decreased the yield ( 5b, 5e, and 5g). The electron-donating group (OMe) at the ortho position of phenyl groups visibly improved the yield and slightly decreased the ee ( 5h). We found that substrates with multiple substituents on the phenyl group were compatible with this transformation ( 5i and 5j). To our delight, the 2-naphthyl-substituted substrate was well-tolerated, delivering the corresponding product 5k in 61% yield and 91% ee. However, the 1-naphthyl-substituted substrate gave lower yield and enantioselectivity because of a steric hindrance ( 5l). Figure 3 | Substrate scope with different aryl groups.aaReaction conditions: unless otherwise noted, the reaction was conducted with 2 (0.1 mmol), 3a (0.2 mmol), 4a (0.12 mmol), (S)-1a (0.01 mmol), and (R)-CPA (0.01 mmol) in CH2Cl2 (1.0 mL) at 0 °C under Ar. Isolated yields are shown. The ee was determined by HPLC analysis. bThe reaction was performed at −10 °C. Download figure Download PowerPoint Next, the scope of this reaction was explored with a series of styrene derivatives bearing different functional groups (Figure 4). In general, the desired products were produced with moderate to excellent yields and high to excellent enantioselectivities in all cases. Different ester groups did not impact the enantioselectivities but reduced the yields ( 5m and 5n). To our delight, this transformation is not limited to allylic esters, and changing the chain length between the ester group and the alkene did not influence the enantioselectivities appreciably ( 5o and 5p). Furthermore, homoallylic protected alcohols and a protected amine were well-tolerated, and the corresponding products were afforded with good yields and high enantioselectivities ( 5q– 5t). Of note, β-methylstyrene and styrene were also amenable to this protocol ( 5u and 5v). The absolute configuration of product 5v was assigned using X-ray crystallographic analysis.a However, unbiased alkyl alkene was not suitable for this reaction. Finally, some different sulfenylating reagents were also evaluated, which is generally well-tolerated ( 5x– 5z). Figure 4 | Substrate scope with various functional groups.aaReaction conditions: unless otherwise noted, the reaction was conducted with 2 (0.1 mmol), 3a (0.2 mmol), 4a (0.12 mmol), (S)-1a (0.01 mmol), and (R)-CPA (0.01 mmol) in CH2Cl2 (1.0 mL) at 0 °C under Ar. Isolated yields are shown. The ee was determined by HPLC analysis. bThe reaction was performed at −10 °C. cThe reaction was performed at −5 °C. Download figure Download PowerPoint Subsequently, we evaluated the scope of phenol derivatives (Figure 5). To our satisfaction, we found that the electronic nature of substituents at the para position of the phenyl group had little influence on the enantioselectivities and the yields ( 5aa– 5ag). However, the iodine and fluorine groups at the meta or ortho position of the phenyl group always resulted in lower yields with maintained enantioselectivities ( 5ah– 5aj). Finally, two phenol derivatives with multiple substituents were suitable coupling partners in this reaction, providing the corresponding products in moderate yields with high enantioselectivities ( 5ak and 5al). Figure 5 | Substrate scope with different phenols.aaReaction conditions: unless otherwise noted, the reaction was conducted with 2a (0.1 mmol), 3 (0.2 mmol), 4a (0.12 mmol), (S)-1a (0.01 mmol), and (R)-CPA (0.01 mmol) in CH2Cl2 (1.0 mL) at 0 °C under Ar. Isolated yields are shown. The ee was determined by HPLC analysis. bThe reaction was performed at −5 °C. Download figure Download PowerPoint To expand the capability of this catalytic system further, we turned our attention to studying a three-component reaction using methanol instead of phenol as the nucleophile, considering the importance of benzylic alkyl ethers. It is worth mentioning that studies on catalytic asymmetric intermolecular sulfenylation/methoxylation of alkenes are also limiting.26 To our delight, using 2a as the model substrate, the desired product 6a was obtained with 88% yield and 96% ee when this reaction was conducted in methanol at −5 °C. Next, the scope of this catalytic system was surveyed (Figure 6). The electron-rich groups at the para position of the phenyl moiety increased the yield, whereas electron-poor groups decreased the yield ( 6d and 6e). The different substituents at ortho and meta positions were all suitable, providing the desired benzylic methyl ethers with high efficiency and high selectivities ( 6f– 6i). Notably, multiple substituents and heteroaromatic-substituted substrates were well-tolerated, yielding products with high levels of enantioselectivities ( 6j– 6m). We found that unbiased alkyl alkenes could also be used as substrates. The use of symmetric alkene gave the product 6p in 88% yield with 95% ee, whereas unsymmetric alkenes formed products with excellent levels of enantioselectivities but with modest regioselectivities ( 6n and 6o). To our delight, homoallylic ester, protected alcohols, and amine substrates were also amenable to this system, and the products were always obtained with good yields and high ee. Similarly, β-methylstyrene and styrene were also suitable for this sulfenylation/methoxylation reaction ( 6u and 6v). Simple sulfenylating reagent 4b was also employed as an electrophile ( 6w). In particular, deuterated product 6x was observed in 96% yield and 93% ee when methanol-d was used as the solvent. Finally, the absolute configuration of the benzylic methyl ether product was assigned using X-ray crystallographic analysis of the amide derivative of 6y, which was produced using free allylic alcohol as the substrate. Figure 6 | Substrate scope of methoxylation reaction.aaReaction conditions: unless otherwise noted, the reaction was conducted with 2 (0.1 mmol), 4a (0.12 mmol), (S)-1a (0.01 mmol), and (R)-CPA (0.01 mmol) in MeOH (1.0 mL) at −5 °C under Ar. Isolated yields are shown. The ee was determined using HPLC. bThe reaction was performed at −20 °C. cMeOH (1.5 equiv) was added in CH2Cl2 (1.0 mL). dThe reaction was run at −40 °C. Download figure Download PowerPoint To further examine other aliphatic alcohols as O-nucleophiles and consider other aliphatic alcohols are not widely used as solvents, we attempted using 1.5 equiv of methanol as reactant reacted with three different substrates in dichloromethane (DCM), respectively. Fortunately, we found that products 6e, 6u, and 6v were obtained in moderate yields and good enantioselectivities (Figure 6). Encouraged by the above results, β-methylstyrene was selected as an alkene substrate, and seven different alcohols were tested under the above reaction conditions (Figure 7). We found that ethanol, n-butyl alcohol, and benzyl alcohol gave products in moderate yields and high levels of enantioselectivity, whereas alcohols with greater steric hindrance, such as isopropanol and tert-butanol, produced the corresponding products in lower yield with high ee. To our delight, but-2-yn-1-ol was also amenable to this system. Interestingly, using (E)-3-phenylprop-2-en-1-ol as O-nucleophile, product 6ag was obtained in 50% yield with 95% ee. Simultaneously, product 6ah was obtained in this system. The reaction was also performed with (E)-3-phenylprop-2-en-1-ol, which acted as alkenyl substrate and O-nucleophile at the same time, and sulfenylating reagent 4a at −40 °C, producing the product 6ah in 69% yield with 66% ee (the detail, see Supporting Information). Finally, we tested the system using water as the O-nucleophile and obtained the product 6ai in 68% yield and 94% ee. Figure 7 | The scope of aliphatic alcohols.aaReaction conditions: unless otherwise noted, the reaction was conducted with β-methylstyrene (0.1 mmol), 4a (0.12 mmol), aliphatic alcohol (0.15 mmol), (S)-1a (0.01 mmol), and (R)-CPA (0.01 mmol) in CH2Cl2 (1.0 mL) at −5 °C under Ar. Isolated yields are shown. The ee was determined by HPLC analysis. b2.0 equiv of H2O was used. c5.0 equiv of tBuOH was used. dThe additional reaction was performed at −40 °C under Ar, and 2.5 equiv of (E)-3-phenylprop-2-en-1-ol and 1.0 equiv of 4a were used. Download figure Download PowerPoint To evaluate the synthetic utility of this catalytic asymmetric reaction, a gram-scale reaction of 2a with 3a was performed under the standard conditions. As illustrated in Figure 8a, the desired product 5aa was obtained in 81% yield and 95% ee. Next, the product 5aa could be converted rapidly into other functionalized derivatives (Figure 8b). Treatment of 5aa with an excess amount of m-chloroperbenzoic acid afforded the sulfone 7aa in 92% yield and 94% ee. 5aa could also be transformed readily into amine 9aa in 85% total yield and 94% ee, an analog of drug molecules. Subsequently, compound 6s was subjected to a reduction reaction to deliver desulfurization product 10s in 80% yield with 91% ee. Figure 8 | Gram-scale reaction and derivatization. Download figure Download PowerPoint With the above experimental results in hand, density functional theory (DFT) calculations were then used to study the origin of stereoselectivity and regioselectivity (Figure 9). Based on the previous work,30 BINAM-derived sulfide (S)- 1a, P=S moiety of which plays as Lewis base site (more detail, see Supporting Information Tables S4 and S5), first reacted with sulfenylating reagent 4a in the presence of CPA to generate the catalytically active sulfenium ion intermediate IM-1 (the detail process, please see Supporting Information Scheme S1 ). Then, the activated sulfenium ion attacked the double bond of the substrate from two possible sides, releasing the catalyst (S)- 1a and generating a thiiranium ion intermediate. This step was the stereodetermining step based on our DFT calculations. The chiral catalyst (S)- 1a dominated the configuration with ΔΔG = 5.31 kcal/mol, which led to over 99% ee. In this case, the experimental 92% ee could be attributed to the background reaction (see Table 1, entry 4). Once the intermediate IM-2a was formed, at least three pathways for the SN2 reaction with nucleophile 3a were possible. Within all three pathways, the nucleophilic substitution reaction and the hydrogen transfer process were synergistic. In the pathway with the lowest activation energy (the path through TS-2a), the O atom in 3a attacked the carbon near the phenyl group in IM-2a, which led to product TM-1, matching the experimental product 5m. As for the pathway involving TS-2b, the C atom on the ortho position of 3a attacked the same carbon as mentioned above, and for the pathway of TS-2c, the O atom in 3a attacked the other carbon of the three-membered ring. However, these two pathways were extremely unfavorable due to high ΔΔG‡ values of 3.74 and 10.52 kcal/mol, respectively. Figure 9 | Gibbs energy profile of overall reactions, calculated at 273.15 K. Download figure Download PowerPoint To study the high stereoselectivity in the formation of the thiiranium ion intermediate, an Atoms in Molecules (AIM) analysis was used to quantify the secondary interactions in TS-1a and TS-1b. For TS-1a, eight interactions were found between substrate and catalyst. It is worth noting that the interaction between the C–H group at the benzyl position of 2m and the naphthyl group of (S) -1a (“f” in Figure 10) could be classified as an unconventional hydrogen bond ( C sp 2 –H…π) with a bond energy of −1.11 kcal/mol. In addition, there were other interactions in TS-1a than in TS-1b, leading to high ee (Figure 10). Meanwhile, we also studied if a thiiranium ion “olefin-to-olefin” transfer process was involved.24 It turned out that “olefin-to-olefin” transfer was unlikely to occur in our established standard conditions (more detail, see Supporting Information Figures S1 and S2). Figure 10 | The result of AIM analysis of TS-1a and TS-1b with all secondary interactions found between substrate and catalyst. The number before the brackets was the distance of secondary interaction (in Å), and the number in the brackets was the electronic density of the bond critical point (in ρ × 10−2 a.u.). Visualization of IM-2a and 3a with a nucleophilic index or electrophilic index based on CDFT. Download figure Download PowerPoint To study the regioselectivity of this reaction further, conceptual density functional theory (CDFT) was used to quantify multiple indices related to chemical activity. For the electrophile IM-2a, the S–C1 bond was somewhat longer than the S–C2 bond. As a result, the condensed local electrophilicity of C1 (0.42) was larger than that of C2 (0.11), indicating that C1 was more electrophilic (Figure 10). For nucleophile 3a, the condensed local nucleophilicity of atom O was larger than that of ortho-position carbons, suggesting that the nucleophilicity of atom O was higher. Combined with the results above, product TM-1 ( 5m) was naturally the major product in our weakly acidic reaction conditions (more detail, see Supporting Information Figure S3). Conclusion We have successfully developed chiral sulfide/phosphoric acid cocatalyzed enantioselective intermolecular oxysulfenylations of alkenes with phenols and alcohols as O-nucleophiles. These two reactions could produce a variety of sulfur-containing benzylic aryl ethers and benzylic alkyl ethers in moderate to excellent yields with high to excellent enantioselectivities. Moreover, two vicinal tertiary carbon stereocenters were facilely constructed. The origin of high chemo-, regio-, and enantioselectivity was revealed using computational studies. More catalytic enantioselective intermolecular sulfenofunctionalizations of alkenes are ongoing in our laboratory. Footnote a CCDC 2094621 ( 5v) and CCDC 2094620 ( 6z) contain the supplementary crystallographic data for this paper. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via the link www.ccdc.cam.ac.uk/structures Supporting Information Supporting Information is available and includes experimental details, X-ray crystallography structures of compounds 5v ( Table S2) and 6z ( Table S3), analytical data for new compounds, nuclear magnetic resonance (NMR) spectra of new compounds, and high-performance liquid chromatography (HPLC) spectra. Conflict of Interest There is no conflict of interest to report. Funding Information We thank the National Natural Science Foundation of China (NSFC) (nos. 21871178, 22071149, 21901158, 21702135) and the STCSM (no. 19JC1430100) for their financial support. This research was also supported by The Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning. Acknowledgments The authors gratefully thank Prof. Yong-Qiang Tu (Shanghai Jiao Tong University & Lanzhou University) for helpful suggestions and comments on this manuscript.