Open AccessCCS ChemistryRESEARCH ARTICLE1 May 2021An Unconventional trans-exo-Selective Cyclization of Alkyne-Tethered Cyclohexadienones Initiated by Rhodium(III)-Catalyzed C–H Activation via Insertion Relay Yun-Xuan Tan†, Xing-Yu Liu†, Shuo-Qing Zhang†, Pei-Pei Xie, Xin Wang, Kai-Rui Feng, Shao-Qian Yang, Zhi-Tao He, Xin Hong, Ping Tian and Guo-Qiang Lin Yun-Xuan Tan† The Research Center of Chiral Drugs, Innovation Research Institute of Traditional Chinese Medicine, Shanghai University of Traditional Chinese Medicine, Shanghai 201203 CAS Key Laboratory of Synthetic Chemistry of Natural Substances, Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200032 †Y.-X. Tan, X.-Y. Liu, and S.-Q. Zhang contributed equally to this work.Google Scholar More articles by this author , Xing-Yu Liu† CAS Key Laboratory of Synthetic Chemistry of Natural Substances, Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200032 †Y.-X. Tan, X.-Y. Liu, and S.-Q. Zhang contributed equally to this work.Google Scholar More articles by this author , Shuo-Qing Zhang† Department of Chemistry, Zhejiang University, Hangzhou 310027 †Y.-X. Tan, X.-Y. Liu, and S.-Q. Zhang contributed equally to this work.Google Scholar More articles by this author , Pei-Pei Xie Department of Chemistry, Zhejiang University, Hangzhou 310027 Google Scholar More articles by this author , Xin Wang CAS Key Laboratory of Synthetic Chemistry of Natural Substances, Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200032 Google Scholar More articles by this author , Kai-Rui Feng The Research Center of Chiral Drugs, Innovation Research Institute of Traditional Chinese Medicine, Shanghai University of Traditional Chinese Medicine, Shanghai 201203 Google Scholar More articles by this author , Shao-Qian Yang CAS Key Laboratory of Synthetic Chemistry of Natural Substances, Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200032 Google Scholar More articles by this author , Zhi-Tao He *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] or E-mail Address: [email protected] E-mail Address: [email protected] CAS Key Laboratory of Synthetic Chemistry of Natural Substances, Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200032 Google Scholar More articles by this author , Xin Hong *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] or E-mail Address: [email protected] E-mail Address: [email protected] Department of Chemistry, Zhejiang University, Hangzhou 310027 Google Scholar More articles by this author , Ping Tian *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] or E-mail Address: [email protected] E-mail Address: [email protected] The Research Center of Chiral Drugs, Innovation Research Institute of Traditional Chinese Medicine, Shanghai University of Traditional Chinese Medicine, Shanghai 201203 CAS Key Laboratory of Synthetic Chemistry of Natural Substances, Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200032 Shanghai Key Laboratory for Molecular Engineering of Chiral Drugs, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author and Guo-Qiang Lin *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] or E-mail Address: [email protected] E-mail Address: [email protected] The Research Center of Chiral Drugs, Innovation Research Institute of Traditional Chinese Medicine, Shanghai University of Traditional Chinese Medicine, Shanghai 201203 CAS Key Laboratory of Synthetic Chemistry of Natural Substances, Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200032 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.020.202000339 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Different from the established trans-endo-selective cyclization of alkyne-tethered electrophiles that involve an E/Z isomerization process, herein, the authors present a novel strategy to allow trans-exo-selective arylative cyclization of 1,6-enynes. Through initiation of rhodium(III)-catalyzed C–H activation, a diverse range of N-heterocyclic directing groups, including pyridine, pyrazole, imidazo[1,2-a]pyridine, benzoxazole, benzothiazole, and purine, was feasible for the cascade transformation, exhibiting high efficiency (up to 92% yield), broad substrate scope, and excellent functional group compatibility. Moreover, the modification of natural products and pharmaceutical compounds was also demonstrated to showcase its synthetic utility. Based on density functional theory (DFT) calculations, a key three-membered ring intermediate through the insertion relay, rather than the direct E/Z isomerization of alkenyl rhodium species, controlled the stereochemical outcome for this trans-exo-selective cyclization. The subsequent ring-opening protonation of the more favored rotamer led to exclusive trans-exo-selectivity. Download figure Download PowerPoint Introduction Transition metal-catalyzed cascade cyclization reactions have received significant attention in modern organic synthesis and become a reliable, efficient, and step-economical strategy to prepare carbo- and heterocyclic skeletons, which widely exist in natural products, drug molecules, and functional materials.1–3 Substances containing two or more potential electrophilic sites would be suitable substrates for the construction of carbo- and heterocycles through selective insertion and subsequent cyclization of organometallic species. In view of the appealing reactivity of alkynes,4,5 transition metal-catalyzed cascade cyclization reactions of alkyne-tethered electrophiles have been extensively investigated over the past decades.6–20 As shown in Scheme 1a, the elementary cis-insertion of alkyne by organometallic species usually adopts the following two ways: α-insertion or β-insertion. The electronic and sterically unbiased alkynes usually undertake cis-β-insertion to form intermediates I, which can subsequently cyclize in cis-exo-fashion to furnish conventional products II possessing an exocyclic double bond (Scheme 1a, Path A). This cascade cis-β-insertion and cis-exo-cyclization pathway have been well studied for various alkyne-tethered electrophiles (aldehyde, ketone, imine, ester, isocyanate, nitrile, alkene, alkyne, etc.) using rhodium,6–10 palladium,11–13 and other transition metals14–20 as catalysts. On the other hand, alkynes bearing aryl or silyl groups at R1 substituents prefer to choose cis-insertion of the alkynes at the α-position to offer the alkenyl metal intermediates III (Scheme 1a). Obviously, subsequent cis-cyclization of the intermediates III does not occur due to geometric constraints. This issue was first resolved by Zhang et al.21 through a nickel-catalyzed process, in which alkyne-tethered nitriles could react with arylboronic acids to undergo trans-endo-selective arylative cyclization (Scheme 1a, Path B). In other words, intermediates III bypassed the geometric constraints through an E/Z isomerization of alkenyl nickel intermediates.22 Concurrently, an independent study from Clarke et al.23 disclosed an elegant enantioselective nickel-catalyzed trans-endo-selective arylative cyclization of alkyne-tethered cyclic 1,3-diketones and cyclohexadienones with arylboronic acids. They further extended this formal trans-endo-selective strategy to prepare carbo- and heterocyclic compounds using alkyne-tethered allylic phosphates,24 malonate esters,25 and amides.26 Later, Ranjith Kumar et al.27 reported a nickel-catalyzed trans-endo-selective arylative cyclization of alkyne-tethered azides with arylboronic acids. Recently, Petrone et al.28 discovered a palladium-catalyzed trans-endo-selective hydrohalogenation of 1,6-enynes, in which the crucial E-to-Z vinyl–palladium (Pd) isomerization was involved as a key step. Thus, both nickel and palladium could catalyze trans-endo-selective cyclization of alkyne-tethered electrophiles through a cascade process including cis-α-insertion and trans-endo-cyclization (Scheme 1a, Path B). Scheme 1 | Transition metal-catalyzed cascade cyclization of alkyne-tethered electrophiles. Download figure Download PowerPoint Despite these notable advances, trans-exo-selective cyclization of alkyne-tethered electrophiles by a transition metal-catalyzed process has never been uncovered to best of our knowledge, probably because cis-exo-cyclization occurs more easily. Therefore, it is quite necessary to develop a new strategy to favor trans-exo-cyclization of alkyne-tethered electrophiles, and as a result, cyclic compounds with a trans-exocyclic double bond can be accordingly obtained, compensating the capacity from currently known methods.21–43 Herein, we present the first rhodium(III)-catalyzed trans-exo-selective arylative cyclization of terminal alkyne-tethered cyclohexadienones (1,6-enynes). This cascade reaction proceeds in the following pathway44: regioselective cis-β-insertion of terminal alkyne with arylrhodium species generates intermediate V, which subsequently undergoes exclusive trans-exo-cyclization rather than conventional cis-exo-cyclization to form trans-exo-selective cyclization products VI (Scheme 1b, Path C). Generally, trans-exo-cyclization remains a more difficult and challenging process. Experimental Section A dried Schlenk flask was charged with (Cp*RhCl2)2 (6.18 mg, 5 mol %), NaBArF4 (42.5 mg, 24 mol %), substrate 2 (0.2 mmol, 1.0 equiv), and 1 or 4 or 6 (0.4 mmol, 2.0 equiv), and then backfilled with argon. Then anhydrous 1,1,1,3,3,3-hexafluoropropan-2-ol (HFIP; 1 mL) was added. After the mixture was stirred at 60 °C for 8 h, the reaction mixture was filtered, washed with ethyl acetate (EtOAc) (10 mL × 3), and concentrated in vacuo. The residue was purified by flash silica gel chromatography to afford desired product 3, 5, or 7. For more details, please see the Supporting Information. Results and Discussion Very recently, we reported a manganese(I)-catalyzed arylative cyclization of terminal alkyne-tethered cyclohexadienones (1,6-enynes) with 2-arylpyridine.45 In general, the cis-exo-selective arylative cyclization product 3aa′ was obtained as the major product together with a small amount of trans-exo-selective cyclization product 3aa [ 3aa:3aa′ = 15∶85, Scheme 2, Eq. (1)]. The formation of trans-exo-selective cyclization product 3aa attracted our special attention and encouraged us to explore the potentially exclusive trans-exo-selectivity for arylative cyclization of 1,6-enynes. Scheme 2 | Preliminary findings. Download figure Download PowerPoint As our continuing interest in cascade reactions initiated by transition metal-catalyzed C–H activation,46–48 we commenced to screen more post-transition metal catalysts for this transformation [Scheme 2, Eq. (2)]. To our delight, the exclusively trans-exo-selective arylative cyclization product 3aa (15% yield) was observed when the reaction was conducted with (Cp*RhCl2)2 as the catalyst and AgSbF6 as the additive ( 3aa:3aa′ = 100∶0). The yield of 3aa could be further improved to 30% by switching the catalyst to [Cp*Rh(MeCN)3](SbF6)2, revealing the importance of the cationic rhodium complex. These exciting preliminary findings prompted us to propose a tentative reaction mechanism for elaborating the formation of 3aa (Scheme 3). First, the five-membered rhodacycle A, generated by C–H activation, underwent cis-β-insertion with 2a to offer the seven-membered intermediate B. Next, a similar reversible E/Z isomerization21–44 between intermediates B and F occurred through a rhodium-carbenoid carbocation intermediate D. During this process, the neighboring pyridine could further stabilize the carbocation in D through the formation of pyridinium intermediate E. Finally, the subsequent intramolecular conjugate addition of intermediate F afforded the formal trans-exo-selective cyclization product 3aa. Although the tentative reversible E/Z isomerization process is the key to the formation of the trans-exo-selective cyclization product, the conventional cis-cyclization process cannot be excluded to generate 3aa′ (Scheme 3, Path A). Scheme 3 | A proposed and tentative reaction mechanism. Download figure Download PowerPoint With the above considerations in mind, we conducted further optimization of this trans-exo-selective arylative cyclization reaction, and the selected results are summarized in Table 1. First, other aprotic solvents, including tetrahydrofuran (THF), dimethylformamide (DMF), toluene, and 1,2-dichloroethane (DCE) exclusively gave the trans-exo-selective arylative cyclization product 3aa albeit in low yield (Table 1, Entries 1−6), except for CH3CN (Table 1, Entry 4, 3aa: 3aa′ = 2.5∶1). Screening of several protic solvents, such as MeOH, 2,2,2-trifluoroethan-1-ol (TFE), and HFIP, led to enhanced yield of 3aa (Table 1, Entries 7−11), with HFIP solvent giving the reaction yield up to 60% (Table 1, Entry 10), which is probably attributed to the fact that HFIP possesses high dielectric constant and low nucleophilicity, which strongly affect the stabilization of the rhodium cation intermediates.49 Moreover, the use of cosolvent, 1,4-dioxane and water (1∶1), also led to satisfactory yields (Table 1, Entry 7). Lowering the reaction temperature did not impair the yield (Table 1, Entry 12). Considering the important role of counter anions in Cp*Rh(III)-catalyzed reactions,50 we replaced [Cp*Rh(MeCN)3](SbF6)2 with a combination of (Cp*RhCl2)2 and NaBArF4. To our delight, the yield of 3aa could be further increased to 82% (Table 1, Entry 13). Reducing the loading of (Cp*RhCl2)2 or 1a led to an erosion of the yield (Table 1, Entries 14 and 15). Under optimized conditions, we reinvestigated the Cp*Co(III)- and Cp*Ir(III)-complexes, and no cyclization product was observed (Table 1, Entries 16 and 17), which is consistent with the preliminary studies shown in Scheme 2. Table 1 | Reaction Condition Optimizationa Entry Catalyst Additive Solvent Temperature (°C) Yield of 3aa (%)b 1 [Cp*Rh(MeCN)3](SbF6)2 / Dioxane 100 30 2 [Cp*Rh(MeCN)3](SbF6)2 / THF 100 24 3 [Cp*Rh(MeCN)3](SbF6)2 / DMF 100 15 4 [Cp*Rh(MeCN)3](SbF6)2 / CH3CN 100 28 ( 3aa: 3aa′ = 2.5∶1) 5 [Cp*Rh(MeCN)3](SbF6)2 / Toluene 100 18 6 [Cp*Rh(MeCN)3](SbF6)2 / DCE 100 18 7 [Cp*Rh(MeCN)3](SbF6)2 / Dioxane/H2O = 1∶1 100 52 8 [Cp*Rh(MeCN)3](SbF6)2 / MeOH 100 14 9 [Cp*Rh(MeCN)3](SbF6)2 / TFEc 100 38 10 [Cp*Rh(MeCN)3](SbF6)2 / HFIPd 100 60 11 [Cp*Rh(MeCN)3](SbF6)2 / HFIP/H2O = 1∶1 100 35 12 [Cp*Rh(MeCN)3](SbF6)2 / HFIP 60 59 13 (Cp*RhCl2)2e NaBArF4f HFIP 60 82(77)g 14 (Cp*RhCl2)2h NaBArF4 HFIP 60 45 15i (Cp*RhCl2)2 NaBArF4 HFIP 60 63 16 (Cp*IrCl2)2j NaBArF4 HFIP 60 0 17 Cp*Co(CO)I2k NaBArF4 HFIP 60 0 aReactions were carried out with 1a (0.4 mmol), 2a (0.2 mmol), [Cp*Rh(MeCN)3](SbF6)2 (10 mol %), and solvent (1.0 mL) under an Ar atmosphere, 8 h. bDetermined by 1H NMR analysis of unpurified mixtures with CH2Br2 as an internal standard. cTFE = 2,2,2-trifluoroethan-1-ol. dHFIP = 1,1,1,3,3,3-hexafluoropropan-2-ol. e(Cp*RhCl2)2 (5 mol %) was used instead of [Cp*Rh(MeCN)3](SbF6)2. fNaBArF4 = sodium tetrakis[3,5-bis(trifluoromethyl)phenyl]borate, NaBArF4 (24 mol %) was used. gYield of isolated product 3aa. h(Cp*RhCl2)2 (2.5 mol %) and NaBArF4 (12 mol %) were used. i 1a (0.2 mmol) and 2a (0.2 mmol) were used. j(Cp*IrCl2)2 (5 mol %) was used instead of [Cp*Rh(MeCN)3](SbF6)2. kCp*Co(CO)I2 (10 mol %) was used instead of [Cp*Rh(MeCN)3](SbF6)2. With the optimal reaction conditions in hand, we began to explore the scope of this rhodium-catalyzed trans-exo-selective arylative cyclization of terminal alkyne-tethered cyclohexadienones (1,6-enynes). First, various 1,6-enynes were screened and the results are summarized in Table 2. With R1 substituents as alkyl, benzyl, and phenyl groups, the reactions proceeded smoothly with moderate to high yields (33−77%), exclusively affording trans-exo-selective arylative cyclization products (Table 2, 3aa–3ae). It is noteworthy that steric hindrance has a big influence on reaction efficiency (Table 2, 3ab vs 3ac). Furthermore, diverse functional groups, such as aryl-bromide, aryl-cyanide, ester, alkyl-cyanide, silyl ether, halogens (Cl and Br), imide, and amine, were tolerated well, and the corresponding trans-exo-selective arylative cyclization products could be obtained with yields up to 78% (Table 2, 3af–3ao). In addition, the easily available 1,6-enynes 2p, 2q, and 2r derived from coumarin, estrone, and dehydrocholic acid, were applied to this reaction, successfully providing cyclization products in high yields (Table 2, 3ap–3ar). Table 2 | Substrate Scope of 1,6-Enynes and Pyridinesa a Condition A: Reactions were performed using 1 (0.4 mmol, 2.0 equiv), 2 (0.2 mmol, 1.0 equiv), (Cp*RhCl2)2 (5 mol %), NaBArF4 (24 mol %), and HFIP (1.0 mL) under Ar atmosphere, 60 °C, 8 h. bYield of isolated product 3. cNPhth = N-phthaloyl. d Condition B: Reactions were performed using 1 (0.4 mmol, 2.0 equiv), 2 (0.2 mmol, 1.0 equiv), [Cp*Rh(MeCN)3](SbF6)2 (10 mol %), dioxane (0.5 mL), and H2O (0.5 mL) under Ar atmosphere, 60 °C, 8 h. The reactions of different 2-arylpyridines 1 with methyl-substituted 1,6-enynes 2a were then examined (Table 2). Various functional groups, including halogen (F, Cl, and Br), aldehyde, and cyanide in the para-position of phenyl rings, were well compatible (Table 2, 3ba–3ia). The relative configuration of 3ha was undoubtedly confirmed by X-ray crystallography. With an ester group at the meta-position of the phenyl ring, C–H activation selectively occurred at the less sterically hindered position, and only a single cyclization product was observed with good yield (Table 2, 3ja). However, substrate 1k with a methoxyl group at the meta-position of phenyl ring provided both regioisomers (Table 2, 3ka). As for the substrates with para-substitutions at the pyridine rings, all reactions could afford the desired products, and obvious electronic effects were observed, with electron-withdrawing groups (cyano and nitro) leading to lower yields (Table 2, 3la–3pa). The 2-phenyl 5-substituted pyridine 1q also obtained the corresponding trans-exo-selective arylative cyclization product with high yield (Table 2, 3qa). Notably, the C–H activation of thiophene was compatible in this reaction to furnish the desired trans-exo-selective arylative cyclization product 3ra and its relative configuration was also confirmed by X-ray crystallography. As an overall trend for 2-arylpyridines, both neutral and electron-donating groups at either aryls or pyridines, gave higher yields of the trans-exo-selective arylative cyclization products than those with an electron-withdrawing group. These results fully support our proposed tentative catalytic cycle, in which these neutral or electron-donating groups can stabilize benzyl-carbocation or pyridinium ion intermediates (Scheme 3). Subsequently, we started to evaluate other directing groups on the reactions with 1,6-enynes 2a (Table 3). With the directing groups as pyrazolyl, imidazo[1,2-a]pyridyl, benzoxazolyl, benzothiazolyl, and purinyl, all reactions proceeded smoothly with acceptable to good yields (33–76%), and only trans-exo-selective arylative cyclization products were observed (Table 3, 5aa– 5la). Notably, both cyclopropyl and allyl groups survived in this reaction (Table 3, 5ka and 5la). The relative stereochemistry of cyclization product 5ea was unambiguously established by X-ray crystallography. Encouraged by the broad substrate scope and good functional group compatibility of this rhodium-catalyzed trans-exo-selective arylative cyclization reaction, we next focused our attention on the modification of functional molecules. Both reactions of estrone- and phenylalanine-derived pyridine substrates 6a and 6b were performed smoothly with good to excellent yields (Table 3, 7aa and 7ba). Similarly, fenofibrate and clofibrate derivatives were also amenable to this transformation (Table 3, 7ca and 7da). More importantly, commercially available drug substances containing 1-phenyl-1H-pyrazole or 2-phenylpyridine substructure, such as pyraclostrobin (fungicide), GSK1292263 (GPR119 agonist), and atazanavir (HIV-1 protease inhibitor), were practicable in this rhodium-catalyzed cascade reaction without requiring further pretreatments (Table 3, 7ea– 7ga). Heterocycle, amine, sulfone, hydrazine, amide, and even exposed hydroxyl groups were well tolerated in this reaction, demonstrating remarkable functional group compatibility of this cascade reaction once again. Table 3 | Substrate Scope of Other Directing Groups and Late-Stage Modification of Functional Moleculesa a Condition C: Reactions were performed using 4 or 6 (0.4 mmol, 2.0 equiv), 2a (0.2 mmol, 1.0 equiv), [Cp*Rh(MeCN)3](SbF6)2 (10 mol %), and HFIP (1.0 mL) under Ar atmosphere, 60 °C, 8 h. b Condition A: Reactions were performed using 4 or 6 (0.4 mmol, 2.0 equiv), 2a (0.2 mmol, 1.0 equiv), (Cp*RhCl2)2 (5 mol %), NaBArF4 (24 mol %), and HFIP (1.0 mL) under Ar atmosphere, 60 °C, 8 h. cYield of isolated product 5 or 7. dDr value was determined by NMR. eIsolated yield for gram-scale experiment. To gain further insight into this uncommon rhodium-catalyzed trans-exo-selective arylative cyclization reaction, a series of mechanistic experiments were conducted (Scheme 4). First, deuterium-labeling experiments were carried out. When pentadeuterated 2-phenylpyridine ( [D5]-1a) itself was subjected to standard conditions, obvious deuterium loss at the ortho-positions of [D5]-1a was detected. However, when [D5]-1a was treated with 2a under standard conditions for 5 min, minor hydrogen/deuterium (H/D) exchange in the recycled [D5]-1a and cyclization product [D4]-3aa was observed (Scheme 4a). These results indicated that the C–H bond activation step might be reversible and the subsequent arylative cyclization of the rhodacycle intermediate was probably faster than the reversed H/D exchange. The intermolecular kinetic isotope effect (KIE) experiment was then performed under standard conditions for 5 min, and the KIE value of 1.38 implied that the cleavage of the C–H bond was not involved in the rate-determining step (Scheme 4b).51 Intermolecular competition experiments of 1c and 1h were conducted by reacting both with 2a. As a result, the more electron-rich substrate 1c was preferentially transformed, suggesting that arene-coordination is essential in the C–H activation process (Scheme 4c).52,53 Next, cyclometalated Cp*Rh(III) complexes C1 and C2 were prepared and then applied to catalyze the reactions of 1a and 2a. Rhodacycle complex C1 gave relatively low yields of product 3aa (Scheme 4d, Entry 1). However, when the use of C1 was combined with NaBArF4 or cationic catalyst C2 was employed, the yield of product 3aa was significantly improved (Scheme 4d, Entries 2 and 3). The crucial role of NaBArF4 in this transformation strongly suggested that the reaction could go through a cyclorhodium cation complex. Scheme 4 | Mechanistic experiments. Download figure Download PowerPoint Afterwards, several control experiments were performed to further understand the unusual trans-selectivity of this reaction. When (Z)-product 3aa and (E)-product 3ff were individually subjected to standard reaction conditions, no E/Z isomerization was observed for the recovered material, demonstrating the stability of (Z)- and (E)-isomers under standard conditions (Scheme 4e). Several parallel reactions of 1a and 2a were conducted under different temperatures and only the trans-exo-selective cyclization product 3aa was observed in all cases (Scheme 4f). The supply of radical inhibitors, including 2,2,6,6-tetramethyl-1-piperinedinyloxy (TEMPO) and 2,6-bis(1,1-dimethylethyl)-4-methyl-phenol (BHT), into the reaction mixture did not have apparent effects on the yield and E/Z ratio of the product (Scheme 4g).54–59 Finally, upon treatment of the C-linked substrate 2s with 1a under standard conditions, no cyclization product was observed. Instead, only cis-protonation took place to produce the E-olefin 8as (Scheme 4h). Different from the fact that E/Z mixtures was observed in previous report,23 the exclusive trans-exo-selectivity was achieved in this rhodium-catalyzed cyclization reaction, suggesting the alkenyl-metal species could bypass the potential E/Z isomerization. We next explored the reaction mechanism with density functional theory (DFT) calculations (for more details about computational studies, please see the Supporting Information).60–65 The DFT-computed free-energy profiles for the generation of 3aa are shown in Figure 1, and the optimized structures of key transition states are shown in Figure 2. From the di-cationic intermediate int1, substrate 1a acts as a Brønsted base to deprotonate the arene C–H bond via TS2, which must overcome a 15.7 kcal mol−1 barrier. The resulting cationic five-membered rhodacycle intermediate int3 coordinates with 2a to give int5, which subsequently undertakes facile alkyne insertion via TS6 to irreversibly generate the seven-membered rhodacycle intermediate int7. Int7 isomerizes to int8 with the coordination of enone moiety, followed by a facile enone insertion via TS9 to generate the bicyclic intermediate int10. This enone insertion is more efficient than the possible E/Z isomerization of int7 ( Supporting Information Figure S1), which corroborates the exclusive Z-product formation in the experiments. From int10, the insertion of the styrene moiety only requires a barrier of 13.5 kcal mol−1, favoring the generation of the three-membered ring intermediate int12. Int12 then isomerizes to the more stable intermediate int14 with the pyridine coordination. Due to the sequential insertions, the original π-bonds of alkyne are fully converted to σ-bonds in int14, which allows for facile rotation and corresponding equilibrium between int14 and int16 through TS15. Owing to the higher stability of int16 than int14, the subsequent stereochemistry-determining protonation via TS17 opens the three-membered ring and leads to the product-coordinated complex int18, in which the stereochemistry of olefin is defined as Z-form. Int18 then undergoes ligand exchange to liberate the desired Z-product 3aa and regenerate the catalytically active species int1. Based on the DFT-computed free-energy changes of the whole catalytic cycle, the three-membered ring intermediate int16 is the resting state and the subsequent protonation is the rate-determining step, which must overcome a 24.9 kcal mol−1 barrier via TS17. Figure 1 | DFT-computed Gibbs free-energy diagrams of proposed reaction pathways for the generation of 3aa. Relative configurations are shown only for clarification. Download figure Download PowerPoint Figure 2 | Optimized structures of key transition states. Trivial hydrogens are omitted for clarity. Download figure Download PowerPoint The trans/cis selectivity of cyclization is controlled by the competitive protonation of the equilibrated three-membered ring intermediates int14 and int16 (Figure 3). The protonation of the more stable intermediate int16 via TS17 leads to int18, which eventually produces the Z-product 3aa. On the other hand, the alternative protonation of the less stable intermediate int14 via TS19 leads to int20, which eventually produces the E-product 3aa′. Our computational calculations reveal a 3.9 kcal mol−1 preference for the protonation via TS17, which supports the exclusive formation of Z-product. This high E/Z-selectivity could be related to the intrinsic stabilities of the two three-membered ring intermediates ( int14 and int16), since consistent energy differences were discovered between the three-membered ring intermediates ( int14 and int16) and the determining transition states ( TS17 and TS19). Figure 3 | Competition between the formation of 3aa (Z-product) and 3aa′ (E-product). Trivial hydrogen