Fluorine Effects for Tunable C–C and C–S Bond Cleavage in Fluoro-Julia–Kocienski Intermediates

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Jun 2021Fluorine Effects for Tunable C–C and C–S Bond Cleavage in Fluoro-Julia–Kocienski Intermediates Lei Kang†, Yongjia Lin†, Zeng Gao, Jinlong Zhang, Huameng Yang, Jinlong Qian, Qian Peng and Gaoxi Jiang Lei Kang† State Key Laboratory for Oxo Synthesis and Selective Oxidation, Center for Excellence in Molecular Synthesis, Suzhou Research Institute of LICP, Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of Sciences, Lanzhou 730000 State Key Laboratory of Element-Organic Chemistry, College of Chemistry, Nankai University, Tianjin 300071 †L. Kang and Y. Lin contributed equally to this work.Google Scholar More articles by this author , Yongjia Lin† University of Chinese Academy of Sciences, Beijing 100049 †L. Kang and Y. Lin contributed equally to this work.Google Scholar More articles by this author , Zeng Gao State Key Laboratory for Oxo Synthesis and Selective Oxidation, Center for Excellence in Molecular Synthesis, Suzhou Research Institute of LICP, Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of Sciences, Lanzhou 730000 State Key Laboratory of Element-Organic Chemistry, College of Chemistry, Nankai University, Tianjin 300071 Google Scholar More articles by this author , Jinlong Zhang State Key Laboratory for Oxo Synthesis and Selective Oxidation, Center for Excellence in Molecular Synthesis, Suzhou Research Institute of LICP, Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of Sciences, Lanzhou 730000 Google Scholar More articles by this author , Huameng Yang State Key Laboratory for Oxo Synthesis and Selective Oxidation, Center for Excellence in Molecular Synthesis, Suzhou Research Institute of LICP, Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of Sciences, Lanzhou 730000 Google Scholar More articles by this author , Jinlong Qian *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory for Oxo Synthesis and Selective Oxidation, Center for Excellence in Molecular Synthesis, Suzhou Research Institute of LICP, Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of Sciences, Lanzhou 730000 Google Scholar More articles by this author , Qian Peng *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] University of Chinese Academy of Sciences, Beijing 100049 Google Scholar More articles by this author and Gaoxi Jiang *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory for Oxo Synthesis and Selective Oxidation, Center for Excellence in Molecular Synthesis, Suzhou Research Institute of LICP, Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of Sciences, Lanzhou 730000 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.020.202000320 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Classical Julia–Kocienski fluoroolefination represents an indispensable platform for the construction of monofluoroalkenes. Nevertheless, its complex multistep mechanistic manifold along with the unrevealed intrinsic “fluorine effect” in nucleophilic reactions might be responsible for the difficult control of the original stereoselectivity and is thus often ambiguous to predict. Herein, a novel strategy involving the defined fluorine effect and new reaction mechanism was developed for tunable C–C and C–S bond cleavage, providing a versatile avenue for highly stereoselective and easily scalable construction of diverse monofluoroalkenes. Density functional theory (DFT) investigations indicate the fluorine substituents can activate the C–C and C–S bond leading to α-elimination by antiphase orbital interaction. The rate-limiting step were calculated via four-membered transition states with ring strain. Both the sterically eclipsed repulsion and secondary orbital interaction affect the stereoselectivity. Download figure Download PowerPoint Introduction In recent decades, organofluorine chemistry has played an important role in many aspects of pure and applied chemistry since the unique “fluorine effect” can significantly modify the chemical entities’ physical, chemical, and biological properties1–8 and, sometimes, completely change its intrinsic transformation behavior.9–14 Fluorine substitution on carbanion centers will significantly decrease the carbanion’s nucleophilicity, or its “negative fluorine effect,” which has been successfully used in organic reactions.15,16 Increasing fluorine substitution will lead to lowered thermal stability of carbanions caused by α-elimination of a fluoride ion that can be difficult to control. Among these, the fluorine effect on the reactivity of adjacent carbon atoms has yet been undisclosed (Scheme 1a). Therefore, further investigation and utilization of the fluorine effect to exploit new reactions, especially to realize the tunable activation of contiguous C–C/C–X bonds, should streamline the synthesis of fluorine-containing compounds that are difficult to achieve with traditional methods. As the stable bioisostere of amide bonds, monofluoroalkenes are vitally important for the exploitation of pharmacologically active candidates and fluorine-containing compounds for material science applications.17 As a part of the “olefination toolbox,” classical fluoro-Julia–Kocienski olefinations provide a powerful platform to nonselective Z/E monofluoroalkenes by simultaneous cleavage of two C–S bonds, in which a special heteroaryl (benzothiazal-2-yl, pyridin-2-yl, 1-phenyl-1H-tetrazol-5-yl, etc.) or electron-deficient aromatic sulfone should be implanted into the starting material in advance and the assistant unit will be released with sulfur dioxide (Scheme 1b).18–20 Highly stereoselective construction of monofluoroalkenes, especially trisubstituted ones, still remains a formidable challenge due to its larger steric hindrance and minimal energy difference between the two inseparable Z/E stereoisomeric forms.21 Besides transition metal catalyzed, site-selective defluorinative coupling of gem-difluoroalkenes, which are scarce and currently limited to 1-aryl-2,2-difluoroalkenes,22–29 a metal-free breakthrough was achieved by the Hu group.30,31 They realized a smart stepwise extraction and independent workup procedure from a one-pot Julia–Kocienski-type fluoroolefination reaction mixture to separate Z and E monofluoroisomers. Basically, due to the differences in bond energy between the C–C (334 kJ/mol) and C–S bond (276 kJ/mol),32 for the adjacent C–C and C–S bonds, the activation and cleavage of the C–S bond is more favorable. In this issue, to the best of our knowledge, highly atom-economical survival of the sulfonyl group through Julia–Kocienski fluoroolefination to produce useful (α-fluoro)vinyl sulfones33–36 via preferentially selective conversion of the contiguous C–C bond is unprecedented. Herein, we employ the defined fluorine effect to control α-elimination for the tunable C–C and C–S bond cleavage of fluoro-Julia–Kocienski intermediates, providing a versatile avenue for highly stereoselective and easily scalable construction of diverse monofluoroalkenes (Scheme 1c). This method features wide substrate scope, independence of conventional assistant groups, and exclusive stereoselectivity, which is a powerful complement to traditional desulfonated fluoroolefinations. Density functional theory (DFT) investigations indicate the fluorine substituent can activate the C–C and C–S bond leading to α-elimination by antiphase orbital interaction. The four-membered transition states with ring strain were calculated to be the rate-determining steps of the reaction. Both the steric eclipsed repulsion and secondary orbital interaction affect the stereoselectivity. Scheme 1 | (a) Fluorine effect for carbanions, (b) classical Julia–Kocienski fluoroolefination, and (c) the current strategy. Download figure Download PowerPoint Experimental Method General procedure for the reactions: a dry 5 mL screw cap tube equipped with a Teflon-coated magnetic stirring bar was charged with anhydrous Na2SO4 (100 mg), α-fluoro-sulfonylones (1.0 equiv, 0.1 mmol), 1,8-Diazabicyclo[5.4.0]undec-7-ene (DBU) (1.2 equiv, 0.12 mmol or Cs2CO3 as the base), and aldehyde (1.0 equiv, 0.1 mmol) in 0.5 mL anhydrous MeCN. Then the reaction mixture was stirred and heated at 60–80 °C for 12 h. After completion of the reaction monitored by thin-layer chromatography (TLC), until of the starting materials were disappeared, the solvent was removed in vacuum. The crude product was purified by flash column chromatography on silica gel with petroleum ether (PE)/ethyl acetate (EA) = 15∶1 to provide a series of monofluoroalkene compounds. Experimental details and characterization methods are available in the Supporting Information. Results and Discussion Synthesis of sulfones via C–C bond cleavage After careful optimization of the reaction conditions (see Supporting Information), we proved that a wide range of (α-fluoro)vinyl sulfones can readily be furnished in high yields with excellent (E)-stereoselectivity (all E/Z ratio > 50∶1) by the reaction of 2-fluoro-1-phenyl-2-(phenylsulfonyl)ethan-1-one 1a with different kinds of aldehydes 2 (Table 1). Aromatic aldehydes 2a– p were highly compatible with the transformation to deliver the desired products 3aa– ap in good yields regardless of substituted electrical properties in the aromatic rings. Aliphatic aldehydes 2q– t, even the polyformaldehyde 2u, were also applicable to this reaction, assembling the corresponding (E)-monofluoroalkenes 3aq– au with satisfactory outcomes. Gratifyingly, substrates with various heterocyclic groups, such as pyridine ( 2v), benzo[b]thiophene ( 2w), furan ( 2x), chiral pyrrolidine ( 2y), and imidazole ( 2z), were well tolerated and smoothly provided 3av– az in 70–92% yields. It is noted that natural products (R)-citronellal ( 2za) and myrac aldehyde ( 2zb), as well as complex substrates with furanose ( 2zc), can also be readily converted into the relevant (α-fluoro)vinyl sulfones in 80–95% yields. Olefinic isomerization was observed with 2r and 2z to provide conjugated (1-fluoroallyl)sulfonyl products iso- 3ar and 3az. The reaction can be amplified up to gram scale without appreciable decreases in product yields. Accordingly, both 3am and 3aq can be isolated at 91% yields with 0.5 and 0.95 g, respectively. Table 1 | Substrate Generality of Aldehydesa aUnless indicated otherwise, the reaction was carried out on 0.1 mmol. 1 (1.0 equiv, 0.1 mmol), aldehyde 2 (1.0 equiv), and DBU (1.2 equiv, 0.12 mmol) in MeCN (0.5 mL) at 80 °C for 12 h. bUse of Cs2CO3 (2.0 equiv) instead of DBU at 60 °C for 24 h. cat room temperature for 12 h. We then investigated the scope of α-fluoro-β-keto-sulfones 1. As shown in Table 2, a series of α-fluoro-β-phenylacetyl sulfones bearing aromatic ( 1b–n), ortho-pyridyl ( 1o), 2-methyl-furanyl ( 1p), aliphatically substituted benzyl ( 1q), methyl ( 1r), cyclohexyl ( 1s), and even bulky tertiary butyl groups ( 1t), reacted smoothly with 2m, generating 3bm–tm in good to excellent yields and high E-selectivity. Other kinds of leaving acyl groups were also amenable to the reaction. Based on benzenesulfonyl substitution, electron-deficient phenylacetyl groups were more reactive ( 1a, 1u, and 1v). The yield of 3am could be increased from 78% to 95% when para-OMe ( 1u) was replaced by fluorine ( 1v) at the phenylacetyl group. Pivaloyl ( 1w) and acetyl ( 1x) α-fluoro-sulfones were also effective for the reaction, giving 3am in 57% and 86% yields, respectively. Table 2 | Substrate Scope of α-Fluoro-β-Keto-Sulfonesa aUnless indicated otherwise, the reaction was carried out on 0.1 mmol. 1 (1.0 equiv, 0.1 mmol), aldehyde 2 (1.0 equiv), and DBU (1.2 equiv, 0.12 mmol) in MeCN (0.5 mL) at 80 °C for 12 h. Synthesis of classical Julia–Kocienski fluoroolefination via C–S bond cleavage As demonstrated in Table 3, to be an interesting complementary and contradistinctive protocol to the above decarbonylative fluoro-olefinations, we found that desulfonation was carried out when 2-fluoro-1-phenyl-2-((trifluoromethyl)sulfonyl)ethan-1-one 1y was employed to react with 2m and afforded 4a in higher yield (89%) than that of classical Julia–Kocienski olefination process ( 1z and 1za). With the treatment of 1y with a variety of aldehydes, adducts 4b–f were furnished in 80–91% yields. (R)-citronellal ( 2za), myrac aldehyde ( 2zb), and furanose ( 2zc) were also suitable to the desulfonation process, providing 4g–i in good yields. Table 3 | Aldehyde Scope for Desulfonation Processa aUnless indicated otherwise, the reaction was carried out on 0.1 mmol. 1 (1.0 equiv, 0.1 mmol), aldehyde 2 (1.0 equiv), and DBU (1.2 equiv, 0.12 mmol) in MeCN (0.5 mL) at 80 °C for 12 h. bUse of Cs2CO3 (2.0 equiv) instead of DBU at 60 °C for 24 h. DFT calculations for the reaction mechanism As mentioned earlier, to gain further insight into this fluoroolefination reaction, we propose a plausible mechanism based on experiments and DFT calculations (Scheme 2). Substrate 1 abstracted hydrogen by the base DBU to obtain intermediate IN1. Oxygen anion intermediate IN2 can be produced via nucleophilic addition of IN1 with 2-fluorobenzaldehyde 2m. When the R group is trifluoromethyl, the reaction proceeds through β-O elimination via b-TS2, which is the rate-determining step. Then trifluoromethylsulfonate is removed and produces the vinylcarbonyl product 4a. When the R group is methyl, the reaction proceeds through benzoyl 1,3-migration via a-TS2 to form a-IN3, where a-TS2 is the rate-determining step. Compared with trifluoromethylsulfonate, it is more difficult to release the benzoyloxy unit. Therefore, a-IN3 subsequently undergoes olefination via a-TS3 with only a 7.6 kcal/mol energy barrier, generating the vinylsulphonyl product 3rm. As shown in Figures 1 and 2, DFT calculations at the SMD-M06-2X/6–31G(d,p)//6–311+G(d,p) level of theory were introduced to understand the mechanism for the competed C–C/C–S bond cleavage and Z/E stereoselectivity of olefins dependent on −CH3 or −CF3 substituted sulfones (for computational details, see Supporting Information).37–40 More functionals B3LYP-D3 and ω-B97XD gave comparable results based on selected key transition states (see Supporting Information Table 9-S2). Calculated intermediates from different C-C/C-S pathways could be converted to each other via C-C bond rotations that were a facile step supporting by the energy scan of a-IN2 in Supporting Information Figure S2. IN2 and the other isomer IN2-1 can be obtained via the nucleophilic addition of 2-fluorobenzaldehyde 2m with IN1. Figure 1 | Energy profile on the pathway of intermediates with methyl group a-IN1. The low barrier isomerization step by the Cα–Cβ bond rotation was omitted for clarity, supported by the energy scan of a-IN2 in Supporting Information Figure S2. Download figure Download PowerPoint While R group is methyl, the transition state of C–C bond cleavage a-TS2, leading to E-olefin, was 8.7 kcal/mol more stable than a-TS2-3 via C–S bond cleavage. In addition, the energy barrier of a-TS2-1 and a-TS2-2 to form the other Z-isomer illustrated in the dashed line was over 4 kcal/mol higher than a-TS2, suggesting the final vinylsulphonyl product (E)- 3rm via a-TS2 is both kinetically and thermodynamically favorable and highly stereospecific. While the R group was trifluoromethyl, the transition state of C–S bond cleavage b-TS2 was 4.7 kcal/mol more favorable than b-TS2-2 via C–C bond cleavage for Z-olefins. The pathway to form favorable vinylcarbonyl product (Z)- 4a is kinetically irreversible and thermodynamically stable. Additional transition states for different substitutions on carbonyl and sulfonyl groups were calculated, and all of them gave comparable results with our experiment, see Supporting Information Table S3 for details. Our calculated mechanism predicts the C–C/C–S bond cleavage and Z/E stereoselectivity controlled by a slight change, via −CH3 and −CF3 substitutions. Scheme 2 | Plausible mechanism based on DFT calculations. Download figure Download PowerPoint Selectivity control based on the reaction mechanism Computational analysis was conducted to reveal the original control of this reaction and then classified into four aspects as shown in Figure 3 and Supporting Information Schemes S3–S5 for C–C and C–S bond cleavage, respectively. Frontier orbital analysis was used to study how the fluorine atom promotes the reaction (Figure 3).41,42 Yu and co-workers43 successfully used frontier molecular orbital (FMO) energy to describe electrophilicity and nucleophilicity issues that can be applied to discuss many mechanistic issues. Figure 3 | Computational analysis. (a) HOMO orbital analysis of key transition states a-TS2 and a-TS2-H. (b) Distortion–interaction analysis of key transition states a-TS2 and a-TS2-3. (c) Molecular geometry of key transition states a-TS2 and a-TS2-1. Dihedral angle values are given in degrees, and for the Z-isomer, the angle values are shown by adding 180°. (d) NPA charge analysis and molecular geometry of a-TS2 and b-TS2-3. Energy values are given in kcal/mol. Bond distance values are given in Å. Download figure Download PowerPoint Fluorine substitution effect As a control, we replaced the fluorine atom in a-TS2 with a hydrogen atom to study the effects of fluorine atoms on the transition state of C–C bond cleavage (Figure 3a). The relative energy barrier of the rate-determining step via a-TS2 is more favorable for 9.4 kcal/mol than that via a-TS2-H. According to these two highest occupied molecular orbital (HOMO) orbitals, we can find that the p orbital of the fluorine atom interacts with the reaction site of C–C bond cleavage in an antiphase, resulting in activation of this bond. A lower orbital energy gap between HOMO and lowest unoccupied molecular orbital (LUMO) for a-TS2 rather than a-TS2-H (6.03 vs. 6.50 eV) promotes a-TS2 as a favorable transition state of the reaction. The relative orbital energy difference of 0.47 eV (10.8 kcal/mol) also gives a comparable energy level with our calculated relative energy barriers of 9.4 kcal/mol, implying FMO contributions. While the substituent group is a hydrogen atom, there is slight in-phase orbital interactions with the neighboring carbon atom to stabilize the HOMO orbital. Therefore, the fluorine substituent facilitates C–C bond cleavage by the activation effect of the antiphase p orbital interaction, which is consistent with bond dissociation energy (BDE) analysis of the C–C bond for a-IN1 and a-IN1-H.44,45,a Further experiments using the nonfluorinated counterparts of 1x and 1y (Scheme 3) afford low reactivity and selectivity that reasonably support the fluorine effect. Scheme 3 | Controlled experiment to understand the fluorine effect. Download figure Download PowerPoint C–C cleavage Versus C–S cleavage For the selectivity of C–C bond or C–S bond cleavage, the oxygen anionic moiety from a-IN2 nucleophilically attacks the benzoyl sulfonyl groups via a-TS2 or a-TS2-3 transition states, respectively (Figures 1 and 3b). The relative LUMO energy of a-TS2 and a-TS2-3 indicates trends to accept oxygen anions for the favorable C–C cleavage reaction. Distortion–interaction analysis was used to discuss the regioselectivity of oxygen anionic moieties when the R group is a methyl group (Figure 3b). The key transition state is decomposed into two parts: the original substrate aldehyde and carbon anion intermediate a-IN1. There are distortion and interaction energies of both parts in corresponding transition states. It can be concluded that the activation energy of the C–S bond cleavage for transition state a-TS2-3 is 8.0 kcal/mol higher than the C–C bond cleavage for transition state a-TS2, which is mainly derived from the interaction energy between the aldehyde and a-IN1, about 30.7 kcal/mol. According to the molecular geometry of a-TS2 and a-TS2-3, forming a C4–O1 bond or S–O1 bond is vital in the interaction between the part of the aldehyde and a-IN1 anion. The bond distance of the C4–O1 bond in a-TS2 is about 1.42 Å shorter than S–O1 bond of a-TS2-3. Considering the 0.11 Å radius difference between carbon and sulfur atoms, the C4–O1 distance of a-TS2 is still 0.17 Å shorter than another one.46,47 It is suggested that the interaction of C4 and O1 in a-TS2 is more favorable than that of S and O1 in a-TS2-3, leading to the product of C–C bond cleavage. Figure 2 | Energy profile on the pathway of intermediate with trifluoromethyl group b-IN1. Download figure Download PowerPoint Stereoselectivity for Z- or E-olefins Further inspection of the key transition states affording stereoisomer (Z)- or (E)-olefins (Figure 3c) shows that the dihedral angle ∠C3–C2–C1–F of the E-isomer transition state a-TS2 is 15.1° closer to the coplanar of the forming olefin, which is 21.3° smaller than the Z-isomer a-TS2-1. This is likely because of the eclipsed repulsion between the aryl substituent and sulfonate group at the Z-isomer. In addition, the fluorine atom in E-isomer a-TS2 showed two secondary orbital interactions with carbonyl and aryl substituent to stabilize the transition state as our illustrated HOMO orbital analysis. The relative LUMO energy of a-TS2 and a-TS2-1 gives a consistent trend for the favorable (E)-isomer, and the corresponding shorter atom distances of F···O and F···Ar in E-isomer were able to provide further support. Substitution effect for −CH3 and −CF3 Comparing transition states a-TS2 and b-TS2-3, the effects of −CH3 and −CF3 groups on C–C cleavage were investigated (Figure 3d). Due to the different electron withdrawing capacities for −CH3 and −CF3, the negative charge of the C2 atom at a-TS2 is accumulated more than that at b-TS2-3 according to the nature population analysis (NPA) charge. Also, the C2 atom at a-TS2 with a more negative charge has strong nucleophilic ability to interact with the C4 atom of the carbonyl group forming the late transition state and leading to more stable a-IN3. Therefore, a-TS2 is stabilized by 2.5 kcal/mol free energy compared with b-TS2-3. In addition, a-TS2 with −SO2CH3 has lower LUMO energy than b-TS2-3 with−SO2CH3 that lead to a favorable transition state for a-TS2, suggesting the reactivity of the substrate with −SO2CH3 is more feasible for C2–C4 bond cleavage. However, in C–S cleavage, the sulfur group is a leaving group and b-TS2 containing −SO2CF3 has a lower LUMO than a-TS2-2 with −SO2CH3, indicating that LUMOs contribute to the regioselectivity of this reaction ( Supporting Information Scheme S4). Conclusions We have developed a new strategy for highly stereoselective and easily scalable construction of trisubstituted monofluoroalkenes via controlled C–C or C–S bond cleavage, providing a versatile avenue for the synthesis of diverse (E)-(α-fluoro)vinyl sulfones and (Z)-α-fluoroenones. According to the DFT results, C–C or C–S bond cleavage via constrained four-membered transition states is the rate-determining step of the reaction. We reveal that the fluorine substituent plays a key promoting effect by fluorine antiphase p orbitals in both C–C and C–S bond cleavage. In activation modes by the fluorine, the chemoselectivity of C–C/C–S bond cleavage and the stereoselectivity for Z/E-olefin were able to be controlled by slightly changing the substitutions of −CH3 and −CF3 on sulfones. Oxygen anion interactions, sterically eclipsed repulsion, and secondary orbital interactions were uncovered to explain and rationalize the corresponding selectivity. It is anticipated that the highly stereoselective construction of multisubstituted monofluoroalkenes via accurate control of adjacent C–X bond activation and cleavage will be a powerful complement to the classical Julia–Kocienski fluoroolefinations. Further understanding of the conversion of fluorosulfones may facilitate the development of new reactions and provide new insights for the fluorine effect. Supporting Information Supporting Information is available. Conflict of Interest The authors declare no competing financial interests. Acknowledgments The authors gratefully acknowledge the National Natural Science Foundation of China (nos. 21602231, 21890722, 21702109, and 11811530637), Chinese Academy of Sciences “Light of West China” Program, and the Natural Science Foundation of Jiangsu Province (nos. BK20160396 and BK20191197), the Natural Science Foundation of Tianjin Municipality (nos. 18JCYBJC21400 and 19JCJQJC62300), Tianjin Research Innovation Project for Postgraduate Students (no. 2019YJSB081) and the Fundamental Research Funds for Central Universities [Nankai University (nos.63191515 and 63196021)] for generous financial support. Footnote a Regarding the heterolytic cleavage type, when fluorine atom is replaced by hydrogen atom, the C–C bond cleavage of a-IN1 is favorable about 6.0 kcal/mol compared to aIN1-H, indicating that the fluorine atom plays an important role in the substrate. And the BDE of the C–C bond cleavage is lower than that of C–S bond that imply the relative weak C–C bond readily for the reaction, see Supporting Information Scheme S2.