Introduction

Preparation and derivatization of aldehydes represent the fundamental transformations in organic synthesis1. While the traditional nucleophilic addition onto aldehydes has been well-studied over the past century, modern strategies that enable the formal insertion into aldehydes have attracted increasing interest recently2,3,4,5,6. Compared with the formal C–H bond insertion of diazo compounds into aldehydes, namely, the Roskamp reaction (path a, Fig. 1a)7,8,9,10,11,12,13, the related formal C–C bond insertion has been much less explored14,15,16,17,18,19,20,21,22,23,24,25,26,27. The reaction starts with the nucleophilic addition of the diazoalkyl carbon onto aldehyde, followed by 1,2-migration of R1 group and the extrusion of N2, thus providing one-carbon homologated aldehyde (path b, Fig. 1a). For instance, Hossain14,15,16,17, Kanemasa18, Kirchner19,20, Maruoka21, Ryu22, and others23,24,25,26,27 independently investigated the C–C bond insertion of diazoacetates into aldehydes for the generation of homologated aldehydes or corresponding enols. In 2013, the only catalytic asymmetric example of formal C−C bond insertion of diazoesters into aldehydes was reported by Ryu group, through the incorporation of a chiral oxazaborolidinium ion catalyst22. Despite the efficacy of these established approaches, the problematic product distribution between C–C and C–H insertion products, as well as the hazardous and explosive nature of diazo compounds should not be ignored, which could significantly limit the application of this strategy. To address these limitations, new catalytic systems operated by non-diazo methodology for the homologation of aldehydes become imperative.

Fig. 1: Catalytic formal C–C bond insertion into aldehydes.
figure 1

a Formal C–C bond insertion into aldehydes via diazo approach. b Our hypothesis: carbonyl ylide enabled formal C–C bond insertion. c This work: formal C–C bond insertion into aldehydes via non-diazo approach.

Alternatively, we suppose that the highly reactive carbonyl ylides28,29,30,31,32,33,34,35 could be potentially applicable to the formal C–C bond insertion into aldehydes. In recent years, special attention were drawn to the formation of carbonyl ylides via the reaction of aldehydes with metal carbenes, which provided efficient access to a variety of 1,3-dioxolanes36,37,38,39,40,41,42,43,44,45,46,47 and epoxides48,49,50. Despite these achievements, the formal aldehyde homologation was not observed in the reaction of aldehydes with metal carbenes. Mechanistically, it originates from the intrinsic requirement of electron-withdrawing substituents on these metal carbenes, which could inhibit the desired carbon migration. Hence, it is vital to figure out a carbene-like intermediate which is not constrained to electron-deficient substituents, such as a donor/donor carbene-like intermediate (Fig. 1b).

To circumvent the inherent challenges in applying carbonyl ylides into the formal C–C bond insertion, we envisaged that vinyl cation could be a prospective option51,52,53, due to the unique carbene-like reactivity. Our laboratory has been engaged in the facile diyne cyclization for the generation of vinyl cations, which were considered as useful donor/donor carbene-like intermediates. The diyne cyclization enabled varieties of asymmetric transformations via remote control of enantioselectivity54,55,56,57,58,59, including intramolecular aromatic C(sp2)–H functionalization54, vinylic C(sp2)–H functionalization55, cyclopropanation54, [1,2]-Stevens-type rearrangement56, intermolecular annulations with styrenes57 and ketones58, as well as oxidation and X–H insertion59. Inspired by these results and our recent study on developing ynamide chemistry for heterocycle synthesis60,61,62,63,64,65, we envisioned that the vinyl cation intermediate generated from diyne cyclization should be a suitable precursor of carbonyl ylide with electron-rich substituents, thus facilitating the formal C–C bond insertion into aldehydes. Moreover, the asymmetric version could be realized through remote-stereocontrol of chiral copper catalyst. Herein, we disclose the development of such an asymmetric formal C–C bond insertion into aldehydes through copper-catalyzed diyne cyclization, which offers a versatile entry to homologated pyrrylaldehydes and axially chiral naphthylpyrroles66 (Fig. 1c). To the best of our knowledge, this protocol represents a rare example of asymmetric formal C–C bond insertion into aldehydes using non-diazo approach.

Results

Screening of reaction conditions

To examine the feasibility of the proposed formal C–C bond insertion strategy, N-propargyl ynamide 1a was selected as the precursor of vinyl cation to react with 4-methoxybenzaldehyde 2a in the presence of various copper catalysts (Table 1). At the outset, the comparison of Cu(OTf)2 and CuOTf revealed that Cu(I) could be the alternative catalyst for this transformation (entries 1–2). Further evaluation of other Cu(I) catalysts, including Cu(OTf)·1/2C6H6, CuI, and CuTc, resulted in the one-carbon homologated pyrrylaldehyde 3a in low yields (entries 3–5). Gratifyingly, the switch of catalyst to Cu(I) complexes with CH3CN as the ligand, such as Cu(CH3CN)4BF4 and Cu(CH3CN)4PF6 brought significant improvement of yields (44–74%) and reaction rates (entries 6–7). To our delight, the reaction could be further promoted by increasing the equivalent of aldehyde 2a and using higher temperature, which allowed the formation of homologated aldehyde 3a in 80–90% yields (entries 8–9). In addition, the reaction was proved to be less efficient when performed at 80 °C (entry 10) or in other solvents such as DCM and toluene (entries 11–12). The reaction failed to give any conversion in the absence of a copper catalyst (entry 13). It’s notable that the competitive formal C–H bond insertion into aldehyde (ketone byproduct 3a′) was not observed in the above conditions, showing the excellent regioselectivity for the C–C bond cleavage (for more conditions see Supplementary Table 1).

Table 1 Optimization of reaction conditionsa

Reaction scope study

With the optimal reaction conditions in hand (Table 1, entry 9), the substrate scope of the formal C–C bond insertion was then investigated. In general, N-propargyl ynamides 1 containing functional groups with different electronic properties reacted well with arylaldehydes 2 to give the corresponding one-carbon homologated aldehydes 3 (Fig. 2). Diynes 1 with different sulfonyl protecting groups, including Ms (3a), Ts (3b), MBS (3c), SO2Ph (3d) and Bs (3e), reacted smoothly to deliver pyrrylaldehydes in 70−90% yields. A wide array of aryl-substituted N-propargyl ynamides (R1 = Ar) bearing both electron-withdrawing and -donating groups on the benzene ring at para (3f–3j) and meta (3k, 3l) positions, were all compatible with the formal C–C insertion. More challenging alkyl-substituted N-propargyl ynamide containing cyclopropyl substituent was also proved to be a competent substrate (3m). Interestingly, it was found that a range of aryl-substituted N-propargyl ynamides (R2 = Ar) were suitable substrates to generate desired pyrrylaldehydes 3n–3r in 33−79% yields. The substrate scope of arylaldehydes 2 was next investigated under the optimized reaction conditions. Phenyl aldehydes equipped with a series of functional groups, including Et (3s), OTIPS (3t), NMe2 (3u) and OMe (3v), successfully produced the corresponding homologated aldehydes in moderate to good yields. It should be noted that, the transformation also exhibited good tolerance for heterocycles, such as furan (3w) and indole (3x). The structure of product 3b was confirmed by X-ray diffraction. Attempts to extend the reaction to non-substituted benzaldehyde and alkyl aldehydes resulted in low yields (for more details see Supplementary Fig. 1), probably due to the lower nucleophilicity of these aldehydes.

Fig. 2: Substrate scope for the formal C–C bond insertion.
figure 2

Reaction conditions: 1 (0.2 mmol), 2 (0.8 mmol), Cu(CH3CN)4PF6 (0.02 mmol), DCE (4 mL), 60 °C, 0.5–1.5 h, in vials; yields were those for the isolated products. a2b (2 mmol), 50 °C. PG = protecting group, MBS = 4-methoxybenzenesulfonyl, Bs = 4-bromobenzenesulfonyl.

After establishing a general and reliable method for this formal C–C bond insertion into aldehydes, we proposed to develop an asymmetric version of this copper-catalyzed aldehyde homologation, for the enantioselective construction of quaternary carbon stereocenters. However, after multiple attempts using an array of chiral ligands, we found the stereocontrol on the quaternary carbon remains unsatisfied (<10% ee). Inspired by our recent work involving axial chirality59,67,68,69, we considered that diynes preinstalled with sterically demanding groups might react with arylaldehydes via atroposelective formal C–C bond insertion to produce axially chiral arylpyrroles (Table 2). Firstly, naphthyl diyne 4a and arylaldehyde 2h were chosen as the model substrates to conduct this atroposelective transformation. The proof-of-concept experiment was performed with catalytic amount of Cu(CH3CN)4PF6/(R)-MeO-BIPHEP (L1) and NaBArF4 in DCE at 30 °C, which delivered the axially chiral naphthylpyrrole 5a in good yield albeit with moderate ee value (entry 1). However, the enantioselectivity couldn’t be further improved after screening of other bisphosphine ligands (for more conditions see Supplementary Table 2). Gratifyingly, a survey into Tang’s sidearm-modified bisoxazoline ligands (SaBOX)70 led to significant improvement in the enantioselectivity. As listed in Table 2, the modification of sidearms and arenes on SaBOX ligands (L2L7) afforded 5a in 75–84% yields with 10–88% ees (entries 2–7). Afterwards, the screening of 2h’s equivalent and solvents failed to promote the reaction (entries 8–10). On further optimization of reaction conditions by varying the reaction temperature, we determined the following optimal conditions to evaluate the substrate generality: 4a (1 equiv) was treated with 2h (2 equiv) in the presence of Cu(CH3CN)4PF6 (10 mol %), L4 (12 mol %) and NaBArF4 (12 mol %) in DCE at 0 °C for 18 h, and the desired axially chiral product 5a was obtained in 88% yield with 92% ee (entry 11).

Table 2 Optimization of reaction conditions for the atroposelective formal C–C bond insertiona

The substrate scope of this atroposelective formal C–C bond insertion was next explored under the optimized reaction conditions (Table 2, entry 11). As depicted in Fig. 3, the reaction proceeded smoothly to deliver the corresponding axially chiral naphthylpyrroles 5 in generally moderate to excellent yields with good to excellent enantioselectivities. First, diynes with different sulfonyl protecting groups reacted well to deliver the homologated aldehydes 5a5e in 70−89% yields with 86−93% ees. The reaction also demonstrated good tolerance towards various substituents (such as alkyl, OMe, TMS) on different positions of naphthalene ring and furnished the desired axially chiral naphthylpyrroles 5f5n in 40−98% yields with 80−99% ees. A variety of N-propargyl ynamides 4 and aldehydes 2 bearing identical aryl substituents (Ar1 = Ar2), were competent substrates to afford the homologated aldehydes in moderate to good yields with 87−94% ees (5o5s). Besides phenyl aldehydes, other heterocyclic aldehydes (Ar1 ≠ Ar2) underwent the regioselective formal C–C bond insertion to provide the corresponding naphthylpyrroles in moderate yields with excellent enantiocontrol (5t5v). It should be mentioned that the diastereocontrol involving the quaternary carbon stereocenter is still challenging at this stage. Therefore, this protocol provides an efficient pathway for the construction of divergent potentially useful axially chiral naphthylpyrroles.

Fig. 3: Substrate scope for the atroposelective formal C–C bond insertion.
figure 3

Reaction conditions: 4 (0.1 mmol), 2 (0.2 mmol), Cu(CH3CN)4PF6 (0.01 mmol), L4 (0.012 mmol), NaBArF4 (0.012 mmol), DCE (2 mL), 0 °C, N2, 10–120 h, in Schlenk tubes; yields were those for the isolated products; ees were determined by HPLC analysis. a20 °C. bL3 was used instead of L4. c25 °C. d−10 °C. Cyp = cyclopentyl.

Synthetic applications

To explore the synthetic utility of this atroposelective formal C–C bond insertion, a gram-scale reaction was first carried out, leading to the desired product 5a in 85% yield with 92% ee (Fig. 4). Furthermore, the axially chiral naphthylpyrrole 5a can readily undergo various transformations to access diverse structural motifs. For example, the sulfonyl-protecting group on pyrrole moiety could be easily removed to provide free pyrrole product 6, which was further diversified to dibrominated pyrrole 7 and Boc-protected pyrrole 8. Alternatively, the aldehyde moiety facilitated a wide range of derivatizations, including 1,2-reduction, reductive amination, condensation, and Wittig olefination, thus resulting in axially chiral naphthylpyrroles 912 efficiently. Note that these transformations all proceeded without erosion of enantiopurity. The absolute configuration of 11 was unambiguously confirmed by X-ray diffraction.

Fig. 4: Gram-scale reaction and further transformations.
figure 4

Reagents and conditions: (i) KOH (3 equiv), EtOH/THF = 1/1, 50 °C, 0.5 h. (ii) NBS (2.1 equiv), THF, 0 °C, 0.5 h. (iii) Boc2O (4 equiv), 4-DMAP (0.5 equiv), MeCN, 40 °C, 1 h. (iv) NaBH4 (2 equiv), MeOH/THF = 1/1, rt, 0.5 h. (v) BnNH2 (4 equiv), NaBH3CN (4 equiv), AcOH (3.3 equiv), MeOH/THF = 1/1, rt, 2 h. (vi) DNPH (5 equiv), EtOH/DCE/AcOH = 1/1/1, 80 °C, 1.5 h. (vii) Ph3PMeBr (2 equiv), nBuLi (2 equiv), THF, −78 °C–rt, 2 h.

Moreover, an axially chiral phosphine ligand based on enantioenriched naphthylpyrrole 12 was successfully prepared to prove the practicability of this methodology (Fig. 5a). Naphthylpyrrole 12 could undergo deprotection and bromination to give monobrominated free pyrrole 13 in good yield. Subsequent methylation and base-promoted phosphination furnished the desired axially chiral phosphine ligand 14. Notably, phosphine 14 could be used as a suitable ligand to realize the silver-catalyzed enantioselective [3 + 2] cycloaddition of azomethine ylide 15 and maleimide 16, as well as the palladium-catalyzed enantioselective allylic alkylation of 1,3-diphenylallyl acetate 18 and dimethyl malonate 19 (Fig. 5b). These results demonstrate that the constructed enantioenriched axially chiral skeleton is capable of inducing the chirality in asymmetric synthesis.

Fig. 5: Synthetic applications.
figure 5

a Chiral ligand synthesis. b Applications in asymmetric catalysis. Reagents and conditions: (i) KOH (3 equiv), EtOH/THF = 1/1, 50 °C, 1 h. (ii) DBMDH (0.6 equiv), THF, −78 °C, 10 min. (iii) MeI (4 equiv), NaH (6 equiv), DMF, 40 °C, 1 h. (iv) ClPPh2 (5 equiv), nBuLi (5 equiv), THF, −78 °C, 40 min.

Mechanistic investigations

To elucidate the mechanistic details, several control experiments were conducted (Fig. 6). The reaction of selected diyne 1s with aldehyde 2e under optimized reaction conditions at 30 °C allowed the formation of desired product 3y as well as 1,3-dioxolane 3y′36,37,38,39,40,41,42,43,44,45,46,47. Further treatment of 3y′ with Cu(CH3CN)4PF6 at 60 °C delivered the desired product 3y with the exclusion of aldehyde 2e in similar yields. These results indicated that the 1,3-dioxolane 3y′ should be the key intermediate and subsequent ring-opening step is presumably involved in the transformation.

Fig. 6: Control experiments.
figure 6

a Isolation of 1,3-dioxolane intermediate. b Reaction of 1,3-dioxolane intermediate in standard conditions.

On the basis of the above experimental results, density functional theory (DFT) calculations were performed to illustrate the mechanism for the formal C–C bond insertion. As displayed in Fig. 7a, the reaction begins with the coordination of BOX-ligated copper species to diyne 4f to form intermediate A, followed by intramolecular atroposelective cyclization to give vinyl cation intermediate B via transition state TSA with a free energy barrier of 8.6 kcal/mol. Nucleophilic addition of aldehyde onto vinyl cation intermediate B affords vinyl copper intermediate C which is considered as the resonance form of carbonyl ylide. In line with the developed reactions of aldehydes with metal carbenes36,37,38,39,40,41,42,43,44,45,46,47, the second nucleophilic addition of aldehyde onto intermediate C delivers copper carbene intermediate D bearing a 1,3-dioxolane moiety. Subsequent Lewis base-assisted 1,4-proton transfer and demetallation54,55,56,57,58,59 take place to afford 1,3-dioxolane intermediate E, which has been isolated in the above control experiments. Then, a Lewis-acid promoted ring-opening of copper-coordinated 1,3-dioxolane intermediate F leads to carbocation intermediate G with a free energy barrier of 20.7 kcal/mol, which is the rate-determining step. Finally, the desired naphthylpyrrole (R)-5f forms via 1,2-aryl migration. In another alternative pathway (blue line), the coordination of copper with another oxygen atom on 1,3-dioxolane unit could result in intermediate F′, which undergoes further ring-opening to give the ketone product 5f′. A comparison of the intermediates F and F′ indicates that the copper-coordinated intermediate F′ has much higher free energy compared with intermediate F (−50.1 kcal/mol versus −59.9 kcal/mol), which might due to the difficult coordination of copper onto sterically hindered oxygen. Therefore, the generation of ketone product 5f′ is less likely and the regioselective formation of homologated aldehyde (R)-5f is thermodynamically preferred. In addition, the detailed reaction mechanism for the formation of the enantiomer (S)-5f was also calculated (for more details see Supplementary Figure 5).

Fig. 7: Plausible reaction mechanism.
figure 7

a Plausible reaction pathway. b The geometries and relative free energies of the transition states TSA-(R) and TSA-(S) with the chiral ligand. Relative free energies (ΔG, in kcal/mol) of all the transition states and intermediates were computed at the SMD(dichloroethane)-B3LYP-D3/def2TZVP//SMD(dichloroethane)-B3LYP-D3/6-31 G(d)/LANL2DZ level of theory.

The origin of enantioselectivity was also theoretically explored by using SaBOX coordinated Cu(I) complex in the enantioselectivity-determining intramolecular cyclization step (A → B). As shown in Fig. 7b, the free energy of transition state TSA-(R) (leading to the major enantiomer) is predicted to be 2.2 kcal/mol lower than that of TSA-(S) (leading to the minor enantiomer) and the theoretically predicted enantioselectivity matches well with the experimental ee value (96.6% versus 94%). Inspection of the structures of transition states shows that TSA-(S) has a shorter C···C distance than TSA-(R) for the bond-forming position (2.07 Å versus 2.09 Å), and the ligand L4 on TSA-(S) is closer to naphthyl group than TSA-(R) (closest distance: 2.41 Å versus 2.47 Å). These results suggest that TSA-(S) has stronger steric repulsion and lower stability. Therefore, the observed enantioselectivity is dominated by steric effects.

Discussion

In summary, an asymmetric formal C–C bond insertion into aldehydes has been developed through a copper-catalyzed diyne cyclization strategy, which provides a convenient entry for the construction of axially chiral naphthylpyrroles. This methodology has significant advantages over the existing diazo approach, including the avoidance of hazardous and explosive reagents, as well as high regioselectivity. Importantly, the reaction represents a rare example of asymmetric formal C–C bond insertion into aldehydes using a non-diazo approach. The scalability and further derivatizations lead to the divergent synthesis of axially chiral skeletons, which were proven to be potentially applicable as chiral ligands to asymmetric catalysis. Furthermore, experimental and computational mechanistic studies have been carried out to understand the reaction mechanism, origin of regioselectivity and stereoselectivity. Although still in its infancy, we believe the findings shown here will shed lights on the development of aldehyde homologation. Efforts to expand the aldehyde scope and the discovery of more efficient catalytic systems are the topics of the ongoing investigation.

Methods

General procedure for the synthesis of pyrrylaldehydes 3

To the solution of N-propargyl ynamide 1 (0.2 mmol) and arylaldehyde 2 (0.8 mmol) in DCE (4 mL) was added Cu(CH3CN)4PF6 (7.6 mg, 0.02 mmol). The resulting mixture was then stirred at 60 °C for 0.5–1.5 h and the progress of the reaction was monitored by TLC. Upon completion, the resulting mixture was concentrated under reduced pressure and purified by column chromatography on silica gel (eluent: hexanes/ethyl acetate) to afford the desired pyrrylaldehyde 3.

General procedure for the synthesis of axially chiral naphthylpyrroles 5

To a dry 10 mL Schlenk tube charged with a stir bar were added Cu(CH3CN)4PF6 (3.8 mg, 0.01 mmol), L4 (7.5 mg, 0.012 mmol), NaBArF4 (10.6 mg, 0.012 mmol) and DCE (1 mL) sequentially under N2 atmosphere. The solution was stirred at room temperature for 2 h. After cooling to 0 °C, the solution of N-propargyl ynamide 4 (0.1 mmol) and arylaldehyde 2 (0.2 mmol) in DCE (1 mL) was added into the reaction dropwise. The resulting mixture was stirred at 0 °C for 10–120 h and the progress of the reaction was monitored by TLC. Upon completion, the resulting mixture was concentrated under reduced pressure and purified by column chromatography on silica gel (eluent: hexanes/ethyl acetate) to afford the desired axially chiral naphthylpyrrole 5.