Introduction

Hydroarylation of alkynes with (hetero)arenes is a straightforward pathway for the construction of C–C bonds1,2,3,4. During the past years, the catalytic enantioselective hydroarylation of alkynes has been developed as an efficient protocol towards diverse chirality elements, such as axial, helical, and planar chirality5,6,7,8. Classified by reaction mechanism, the major strategies include enantioselective addition of arylmetallic intermediates onto alkynes and enantioselective addition of nucleophilic (hetero)arenes onto π-acid-activated alkynes. In the arylmetallic addition strategy, transition metal-catalyzed oxidative addition of aryl halides9,10, transmetallation of aryl boronic acids11, and directing group-oriented C–H activation12,13,14,15 are responsible for the generation of arylmetallic intermediates. Alternatively, the π-acid catalysis16,17,18,19,20,21 approach proceeds through a greener pathway, which does not require pre-functionalized aryl reagents, such as aryl halides, aryl boronic acids, and directing group-containing aryl compounds (Fig. 1a, path 1). In 2011, Tanaka and co-workers reported a palladium-catalyzed enantioselective hydroarylation of propiolamides for the synthesis of axially chiral 4-aryl 2-quinolinones, which represents a breakthrough in catalytic enantioselective intramolecular hydroarylation of alkynes22. The related gold-catalyzed hydroarylation of electron-deficient alkynones could also be applied to construct axial chirality23. In 2014, the Tanaka group further realized the enantioselective construction of S-shaped double azahelicenes through gold-catalyzed sequential intramolecular hydroarylation of diynes24. Alternatively, the Alcarazo group designed several α-cationic phosphonite chiral ancillary ligands to enable a series of gold-catalyzed intramolecular hydroarylation reactions of diynes25,26,27 and diaryl alkynes28,29,30, providing a convenient access towards various helically chiral skeletons. Moreover, the related gold- and platinum-catalyzed enantioselective hydroarylation was extended to the construction of planar chirality by Carreño31, Shibata32, Zhang33, and others. Despite these impressive advances, noble-metal catalysts are required in the π-acid-catalyzed enantioselective hydroarylation of alkynes, and the construction of axial chirality is limited to electron-deficient alkynes (Fig. 1a, path 2). Therefore, the development of non-noble metal catalyzed enantioselective hydroarylation of alkynes with (hetero)arenes is imperative.

Fig. 1: Catalytic enantioselective hydroarylation of alkynes.
Fig. 1: Catalytic enantioselective hydroarylation of alkynes.The alternative text for this image may have been generated using AI.
Full size image

a Major strategies for the catalytic enantioselective hydroarylation of alkynes. b This work: copper-catalyzed intramolecular atroposelective hydroarylation of 1-alkynylindoles with (hetero)arenes.

As a unique type of polarized alkynes, 1-alkynylindoles are useful synthetic intermediates in organic chemistry, and the related catalytic asymmetric transformations have been extensively explored in recent years. In 2021, Li and co-workers disclosed a rhodium-catalyzed asymmetric C–H activation/alkyne insertion cascade reaction of 1-alkynylindoles with arylnitrones for the formation of axially and centrally chiral indenes by directing group strategy34, and this C–H activation methodology was then employed for the synthesis of other atropisomeric heterocycles and alkenes35,36,37,38,39. In addition to C–H activation-initiated alkyne insertion, the enantioselective hydrophosphination40, hydroamination41, hydroarylation11, click cycloaddition42, and [2 + 2 + 2] cycloaddition43,44 were successfully developed, delivering diverse C─N and N─N axially chiral frameworks efficiently. It should be pointed out that noble metal catalysts or directing groups are necessary in the enantioselective transformations of 1-alkynylindoles. To overcome these limitations and expand the repertoire of 1-alkynylindoles in asymmetric synthesis, it is highly desirable to develop a noble-metal-free catalytic system that operates by an atom-economical pathway.

During the past years, our group has been devoted to the catalytic atroposelective transformations of ynamides45,46,47,48,49,50,51,52 for the construction of various axially chiral and planar-chiral skeletons53,54,55,56,57,58,59. In particular, a series of copper-catalyzed atroposelective diyne cyclization reactions could be realized by the remote control of enantioselectivity, such as oxidation and X–H insertion54, formal C–C bond insertion into aldehydes55, dehydro-Diels–Alder reaction56, and C(sp2)–H functionalization of arenes57. Among them, π-acid-catalyzed addition of ynamides was found to be the enantio-determining step for the reaction of sterically hindered diynes54,55. Therefore, we proposed to employ the efficient π-acid catalysis strategy for the enantioselective hydroarylation of alkynes. Herein, we describe the realization of such a copper-catalyzed intramolecular atroposelective hydroarylation of 1-alkynylindoles with (hetero)arenes, leading to the atom-economical synthesis of C─N axially chiral carbazolyl and phenanthryl indoles in excellent yields with good to excellent enantioselectivities (Fig. 1b)60,61,62,63. Further synthetic transformations led to the formation of an axially chiral phosphine, which is potentially useful in asymmetric catalysis. This reaction represents a rare non-noble metal-catalyzed enantioselective hydroarylation of alkynes by π-acid catalysis.

Results

To validate our hypothesis, indole-tethered 1-alkynylindole 1a was chosen as the model substrate to evaluate this catalytic atroposelective hydroarylation, and the selected results are summarized in Table 1 (for more conditions, see Supplementary Table 1). At the outset, chiral bisphosphines L1 ((R)-BINAP) and L2 ((R)-SEGPHOS) were employed as ligands in the presence of Cu(CH3CN)4PF6 (10 mol %) and DCE as solvent at 40 °C, affording C─N axially chiral carbazolyl indole 2a in excellent yields with 8–12% ee (entries 1–2). To our delight, the replacement of chiral bisphosphine ligands with bisoxazoline (BOX) ligands L3–L6 resulted in obviously improved enantioselectivities (entries 3–6), and L5, containing 4,5,5-triphenyl-4,5-dihydrooxazole moiety, produced 2a in 97% yield with 44% ee (entry 5). Motivated by these results, Tang’s sidearm-modified bisoxazoline (SaBOX) ligands64 were next explored based on the optimized oxazoline framework. The screening of different SaBOX ligands L7–L14 indicated that the sidearm with a 3,4,5-trimethoxyphenyl group (L14) can further improve the enantioselectivity (entries 7–14), and the expected product 2a was generated in 98% yield with 90% ee (entry 14). However, the screening of other solvents, including CH2Cl2, toluene, and THF, failed to improve the enantiocontrol (entries 15–17). Gratifyingly, an obvious temperature influence was observed, and the carbazolyl indole 2a could be synthesized in 98% yield with 96% ee by lowering the reaction temperature to −10 °C (entries 18 and 19). A slightly lower ee was observed after decreasing the catalyst loading to 5 mol % (entry 20).

Table 1 Optimization of reaction conditions for the atroposelective hydroarylation of 1-alkynylindole 1a
Full size table

After establishing the optimal reaction conditions (Table 1, entry 19), the substrate scope of this copper-catalyzed atroposelective hydroarylation was then evaluated in Fig. 2 (for more details, see Supplementary Fig. 1). Indole-tethered 1-alkynylindoles 1 equipped with different sulfonyl groups on the 2-position of indole ring, including Ts, SO2Ph, SO2(4-F-C6H4), and SO2(4-Cl-C6H4), could undergo the enantioselective transformation smoothly to forge C─N axially chiral carbazolyl indoles 2a–2d in 98–99% yields with 95–96% ee. The switch of sulfonyl group to CO2Ph and P(O)Ph2 delivered 2e and 2f in excellent yields with 91–97% ee, which could be easily diversified or transformed into chiral ligand. The introduction of various electron-donating (Me, and OMe) and -electron-withdrawing (F, Cl, and Br) groups onto the different positions of the indole ring for 1-alkynylindoles 1g–1o led to the formation of desired atropisomeric indoles 2g–2o in 96–99% yields with 93–97% ee. Then, we turned to explore the generality of the nucleophilic indole moiety. The protecting group of the indole moiety could be changed to an easily removable benzyl group, delivering 2p in excellent yield and atroposelectivity. Substrates containing electron-donating (Me, and OMe) and -electron-withdrawing (Br) substituents on the 5- and 6-positions of nucleophilic indole were all compatible, affording the corresponding 2q–2s in 95–98% yields with 95–97% ee. Interestingly, the strategy could be extended to the enantioselective hydroarylation of 1-alkynylpyrroles, and the expected C─N axially chiral carbazolyl pyrroles 2t–2v were generated in 90–98% yields with 92–96% ee. In addition, the enantiomer of 2a could be easily obtained by using ent-L14 as a ligand. The absolute configuration of product 2a was unambiguously determined by single-crystal X-ray crystallographic analysis (for more details, see Supplementary Fig. 6).

Fig. 2: Substrate scope for the synthesis of C─N axially chiral carbazolyl indoles 2.
Fig. 2: Substrate scope for the synthesis of C─N axially chiral carbazolyl indoles 2.The alternative text for this image may have been generated using AI.
Full size image

Reaction conditions: 1 (0.1 mmol), Cu(CH3CN)4PF6 (0.01 mmol), L14 (0.012 mmol), DCE (4 mL), −10 °C, in vials; yields are those for the isolated products; ee values are determined by HPLC analysis. aL10 (0.012 mmol). bL8 (0.012 mmol), 15 °C. cent-L14 (0.012 mmol).

Based on the efficient reaction of indole-type nucleophiles, we proposed to extend this transformation to the enantioselective hydroarylation of more challenging benzene-type nucleophiles, which could realize the construction of C─N axially chiral phenanthryl indoles. Therefore, 3,5-diethoxy benzene-tethered 1-alkynylindole 3a was employed as a model substrate to conduct the condition optimization, and the desired product 4a was obtained in almost qualitative yield with 91% ee by using Cu(CH3CN)4PF6 as a catalyst and L12 as a ligand in DCE at 40 °C (for more conditions, see Supplementary Table 2). As demonstrated in Fig. 3 (for more details, see Supplementary Fig. 2), 1-alkynylindoles 3 bearing different sulfonyl substituents were evaluated, such as SO2Ph, SO2(4-F-C6H4), SO2(4-Cl-C6H4), Bs, and Ns, and the desired axially chiral phenanthryl indoles 4b–4f could be synthesized in 94–99% yields with 90–92% ee. A range of electron-donating (Me, and OMe) and -withdrawing (F, Cl, and Br) functional groups on the 3-, 4-, 5- and 6-positions of the indole moiety were tolerated well in this reaction, forging the anticipated products 4g–4s in 95–99% yields with 83–95% ee. In addition to 3,5-diethoxy benzene, 3,5-dipropoxy, 3,5-diisopropoxy, and 3,5-dimethoxymethoxy benzene-type nucleophiles could participate in the enantioselective hydroarylation, providing atropisomeric phenanthryl indoles 4t–4v in excellent yields with 84–91% ee. Interestingly, 2-methyl 1-alkynylindole 3w bearing an ester group at the 3-position of indole gave 4w in 97% yield with 56% ee. The absolute configuration of 4h was confirmed by X-ray diffraction (for more details, see Supplementary Fig. 7).

Fig. 3: Substrate scope for the synthesis of C─N axially chiral phenanthryl indoles 4.
Fig. 3: Substrate scope for the synthesis of C─N axially chiral phenanthryl indoles 4.The alternative text for this image may have been generated using AI.
Full size image

Reaction conditions: 3 (0.1 mmol), Cu(CH3CN)4PF6 (0.01 mmol), L12 (0.012 mmol), DCE (4 mL), 40 °C, in vials; yields are those for the isolated products; ee values are determined by HPLC analysis. aL10 (0.012 mmol). bent-L12 (0.012 mmol). Bs = 4-bromobenzenesulfonyl, Ns = 4-nitrobenzenesulfonyl.

To further demonstrate the synthetic potential of this atroposelective hydroarylation, the scale-up reactions of indole-tethered 1-alkynylindoles 1e and 1f were first carried out, resulting in C─N axially chiral carbazolyl indole 2e in 95% yield with 96% ee and 2f in 92% yield with 90% ee (Fig. 4a). The treatment of 2e with KOH produced an axially chiral carboxylic acid intermediate, which could be transformed into Boc-protected amine 5 and urea derivative 6 through one-pot azidation/Curtius rearrangement/nucleophilic addition. The LiAlH4 reduction of the ester moiety on 2e produced atropisomeric alcohol 7 in 92% yield, which could undergo oxidation with MnO2 and azidation with diphenylphosphoryl azide (DPPA) to give aldehyde 8 and azide 9 efficiently. To illustrate the potential utility of constructed chiral skeleton in asymmetric catalysis, the phosphine oxide moiety of product 2f was easily reduced to give phosphine 10 (Fig. 4b). Notably, axially chiral phosphine 10 could be employed as a chiral ligand in palladium-catalyzed enantioselective allylic arylation of rac-1,3-diphenylallyl acetate 11 with indole 12, leading to the arylated product 13 in 60% yield with 65% ee (Fig. 4c). The enantiopurities of all compounds were well retained in these transformations.

Fig. 4: Scale-up reactions and synthetic applications.
Fig. 4: Scale-up reactions and synthetic applications.The alternative text for this image may have been generated using AI.
Full size image

a Scale-up reactions. b Synthetic transformations. c Application in asymmetric catalysis.

Furthermore, the photophysical and chiroptical properties of selected C─N axially chiral indoles 2 and 4 were explored (Fig. 5, for more details, see Supplementary Figs. 3-5). The UV/Vis absorption spectra indicated that a panel of carbazolyl indoles (2a, 2j, 2k, 2l, 2n, 2p, and 2u) displayed the maximum absorption peaks around 285 nm, and the maximum absorption peaks of phenanthryl indoles (4a, 4h, 4i, 4l, 4n, 4o, and 4r) ranged from 236 nm to 246 nm (Fig. 5a). Subsequently, the fluorescent properties of these atropisomeric indoles were investigated. Carbazolyl indoles (2a, 2j, 2k, 2l, 2n, and 2p) demonstrated maximum fluorescence emission peaks ranged from 455 nm to 474 nm, and the maximum emission peak of carbazolyl pyrrole 2u was at 365 nm. However, phenanthryl indoles 4 showed maximum emission peak around 370 nm (Fig. 5b). Moreover, the circular dichroism (CD) spectra of the two enantiomers of 2a and 4a exhibited a clear mirror-image relationship (Fig. 5c). The substituent-dependent photophysical and chiroptical properties of C─N axially chiral indoles might be beneficial for material science.

Fig. 5: Photophysical and chiroptical properties of selected C─N axially chiral indoles 2 and 4.
Fig. 5: Photophysical and chiroptical properties of selected C─N axially chiral indoles 2 and 4.The alternative text for this image may have been generated using AI.
Full size image

a UV/Vis absorption spectra of selected compounds 2 and 4. b Fluorescence spectra of selected compounds 2 and 4. c CD spectra of the two enantiomers of 2a and 4a.

Guided by the above experimental results and our previous studies on copper-catalyzed enantioselective reaction of ynamides, a possible reaction mechanism of this copper-catalyzed atroposelective hydroarylation is postulated (Fig. 6). Initially, 1-alkynylindole 1a coordinates with L14-ligated chiral copper complex to generate intermediate A. Then, the coordinated-alkyne moiety undergoes metalation to produce copper-tethered keteniminium intermediate B, which triggers a regioselective cyclization to afford atropisomeric iminium ion intermediate C. Finally, aromatization and protodemetalation take place to deliver the desired C─N axially chiral carbazolyl indole 2a, and regenerate the copper catalyst.

Fig. 6: Plausible reaction mechanism.
Fig. 6: Plausible reaction mechanism.The alternative text for this image may have been generated using AI.
Full size image

Mechanism for the copper-catalyzed atroposelective hydroarylation of 1-alkynylindole 1a.

Discussion

In summary, we have established a copper-catalyzed intramolecular atroposelective hydroarylation of 1-alkynylindoles with (hetero)arenes via a π-acid catalysis pathway, leading to the practical and atom-economical assembly of C─N axially chiral carbazolyl and phenanthryl indoles. Notably, this methodology continues a rare non-noble metal catalyzed enantioselective hydroarylation of alkynes by π-acid catalysis, as well as an important advancement in the construction of axial chirality by π-acid-catalyzed asymmetric reaction of electron-rich alkynes. The utility of this reaction has been demonstrated by scale-up reactions, synthetic transformations, chiral ligand synthesis, application in asymmetric catalysis, and investigation into photophysical and chiroptical properties. The effective enantiocontrol uncovered in this study could have broad implications for the further development of enantioselective π-acid catalysis.

Methods

General

For 1H, 13C, and 19F nuclear magnetic resonance (NMR) spectra of compounds in this manuscript and details of the synthetic procedures as well as more reaction condition screening, see Supplementary Information.

General procedure for the synthesis of C─N axially chiral carbazolyl indoles 2

To an oven-dried vial charged with a stir bar were added Cu(MeCN)4PF6 (3.7 mg, 0.01 mmol), L14 (11.7 mg, 0.012 mmol), and anhydrous DCE (4 mL) sequentially. The mixture was stirred at room temperature for 2 h. After cooling to −10 °C, the 1-alkynylindole 1 (0.1 mmol) was introduced into the reaction, and the resulting mixture was stirred at −10 °C for 6–166 h, and the progress of the reaction was monitored by TLC. Upon completion, the mixture was concentrated under reduced pressure and purified by column chromatography on silica gel (eluent: PE/CH2Cl2) to give the desired C─N axially chiral carbazolyl indole 2.

General procedure for the synthesis of C─N axially chiral phenanthryl indoles 4

To an oven-dried vial charged with a stir bar were added Cu(MeCN)4PF6 (3.7 mg, 0.01 mmol), L12 (11.0 mg, 0.012 mmol), and anhydrous DCE (4 mL) sequentially. The mixture was stirred at room temperature for 2 h. After warming to 40 °C, the 1-alkynylindole 3 (0.1 mmol) was introduced into the reaction. The resulting mixture was stirred at 40 °C for 43–240 h, and the progress of the reaction was monitored by TLC. Upon completion, the mixture was concentrated under reduced pressure and purified by column chromatography on silica gel (eluent: PE/EA) to give the desired C─N axially chiral phenanthryl indole 4.