1. Introduction

Axially chiral biaryls are not only key structural units in biologically active compounds^[1-7] but also serve as privileged scaffolds in numerous chiral ligands and organocatalysts^[8-14]. Therefore, the development of efficient and highly enantioselective catalytic processes to approach axially chiral biaryls has been one of the central focuses of organic synthesis over the past decade^[15-37]. Among many established strategies, de novo synthesis of an arene ring provides a unique pathway to access axially chiral biaryls starting from readily available achiral precursors^[25]. Major advances in this category have focused on intramolecular cyclization and cycloaddition^[38-51] as well as intermolecular [2+2+2] alkyne trimerization^[52-56]. In recent years, the use of intermolecular [2+4] annulation to construct an arene ring has attracted increasing attention from organic chemists due to the use of more easily accessible precursors and enhanced structural diversity of the products (Figure 1A). In 2018, Hayashi reported rhodium-catalyzed asymmetric benzannulation of arylalkynes with o-boryl phenylacetonitrile in the atroposelective synthesis of axially chiral 2-aminobiaryls^[57]. A year later, Tong achieved the modular construction of enantioenriched naphthoquinone-based atropisomers via chiral phosphine catalysis, wherein δ-acetoxy allenoates and 2-hydroxyquinone derivatives were employed as the coupling partners^[58]. Furthermore, N-heterocyclic carbene-catalysis has been successfully applied by Zhu and Ye in the de novo construction of an arene ring towards atroposelective synthesis of axially chiral biaryl compounds^[59,60]. Moreover, Cai disclosed copper-catalyzed Diels−Alder/retro-Diels−Alder reactions towards the preparation of axially chiral biaryl esters in a highly enantioselective manner^[61,62]. Very recently, we demonstrated the successful application of reductive [2+4] annulation of β-substituted α-naphthylalkynes with a bis-electrophile of C(sp²)-X type in the atroposelective synthesis of axially chiral biaryls^[63]. Despite these impressive advances, the application of a new pattern of [2+4] annulation in the asymmetric synthesis of axially chiral biaryls is still highly desirable for expanding product structural diversity.

Display Full Size

Download

Figure 1. Synthesis of axially chiral biaryls via intermolecular [2+4] annulation.

Alkynes and o-(pseudo)halostyrenes are both readily available compounds and thus represent desirable starting materials for transition metal-catalyzed annulation reactions involving sequential alkyne insertion and alkene insertion into a carbon-metal bond as the key elementary steps. It is noteworthy that the reactants could undergo either [2+4]^[64,65] or [2+3]^[66-68] annulation depending on the reaction conditions, which results in the construction of a naphthalene or indene framework, respectively (Figure 1B). To the best of our knowledge, [2+4] annulation between alkynes and o-(pseudo)halo styrenes for atroposelective synthesis remains unexplored.

Motivated by the recent advances in nickel-catalyzed synthesis of axially chiral compounds^[69-75] and our ongoing interest in this field^[63,76-80], we envisioned a nickel-catalyzed asymmetric [2+4] annulation, wherein α-naphthylalkynes bearing a bulky ortho substituent and o-iodostyrenes serve as C-2- and a C-4-synthons, respectively. This unprecedented reaction would offer a new method to prepare axially chiral 1,2'-binaphthyls in an enantioselective fashion (Figure 1C). Presumably, this target reaction is initiated by the oxidative addition of o-iodostyrenes to Ni(0). The resulting aryl Ni(II) complex performs subsequently an intermolecular migratory insertion with the C-C triple bond of the α-naphthylalkynes. The issue of regioselectivity is a potential pitfall for this step since the prior successful annulation reactions^[64-68] often require the use of symmetric alkynes. Next, the intermediate A undergoes 6-endo cyclization with its tethered olefinic unit. This key ring closure step for the construction of the naphthalene moiety needs to compete with a 5-exo cyclization, which would lead to the formation of an indene motif. The subsequent β-H elimination from the alkyl Ni(II) species B affords the axially chiral 1,2'-binaphthyls as the aimed product. Finally, the zinc-mediated reduction of Ni(II) to Ni(0) closes the catalytic cycle.

2. Experimental Section

2.1 Material

Unless otherwise noted, all reagents and starting materials were purchased from commercial vendors and used as received without further purification. Flash chromatography was performed using 300-400 mesh silica gel with the indicated solvent system. All air- or moisture-sensitive reactions were protected with a nitrogen atmosphere. All reactions were monitored through thin layer chromatography [Merck 60 F254 precoated silica gel plate (0.2 mm thickness)]. Subsequent to elution, spots were visualized using UV radiation (254 nm) on Spectroline Model ZF-7 254 nm. Other visualization methods include staining with a basic solution of potassium permanganate or acidic solution of ceric ammonium molybdate, followed by heating.

2.2 General procedure for nickel-catalyzed asymmetric [2+4] annulation

NiBr₂•diglyme (5.3 mg, 0.015 mmol, 15 mol%), (4S,5R)-4,5-diphenyl-2-(pyridin-2-yl)-4,5-dihydrooxazole (L9) (6.0 mg, 0.02 mmol, 20 mol%), Zn (9.7 mg, 0.15 mmol, 1.5 equiv), the olefins 1 (0.1 mmol, 1.0 equiv), and the alkynes 2 (0.2 mmol, 2.0 equiv) were placed in a Schlenk tube equipped with a stir bar. Subsequently, the Schlenk tube was evacuated and filled with nitrogen (three cycles). To this mixture, 0.5 mL of anhydrous DMA was added subsequently under a nitrogen atmosphere. The reaction mixture was stirred at 40 °C for 24 hours, before it was quenched by successive addition of water (3 mL) and a saturated aqueous solution of NH₄Cl (3 mL). The aqueous phase was extracted with ethyl acetate (3 × 4 mL), and the combined organic phases were dried over Na₂SO₄, filtered, and concentrated in vacuo. The crude material was purified through column chromatography on silica gel (petroleum ether/ethyl acetate) to afford the products 3.

3. Results and Discussion

3.1 Optimization of the reaction conditions

For optimization of the reaction conditions, α-methyl o-iodostyrene (1a) and the α-naphthylalkyne 2a bearing a bulky mesitylmethoxy as the ortho-substituent were utilized as the standard substrates (Table 1). Initially, a series of chiral ligands were investigated using NiBr₂•glyme as the precatalyst, zinc as the reductant, and DMA as the solvent at 40 °C for 24 h. In the cases of the BiOX ligand L1, the BOX ligand L2, and the PyBOX ligand L3, the target product 3aa was obtained in low yields and poor enantiocontrol (entries 1−3). No desired reaction occurred in the case of the PHOX ligand L4 and the BINAP L5 (entries 4 and 5). Various PyrOX ligands L6−9 turned out to be able to promote the anticipated asymmetric annulation reaction (entries 6−9), and the best outcome in terms of both efficiency and enantioselectivity was achieved in the case of ligand L9 (entry 9). The introduction of a methyl group to the pyridine ring of the PyrOX ligand (L10) gave rise to a slightly lower enantiomeric excess (entry 10), while the use of the strong electron-withdrawing CF₃-substituted PyrOX L11 as the ligand attenuated the enantioselectivity of this process significantly (entry 11). When the trans-configured ligand L12 was employed as the ligand, the enantioselectivity was reversed but still poor (entry 12). Tuning the ligand structure by replacing pyridine with pyrimidine (L13) led to a decrease in both yield and enantiocontrol (entry 13). Brief screening of nickel-precatalysts was undertaken (entries 14−17), revealing that the best outcome could be achieved in the case of NiBr₂•diglyme. The target reaction proceeded also in other polar aprotic solvents like DMF and NMP but less efficiently (entries 18 and 19). Because one equivalent of HI is released during the reaction, we wondered if basic additives could have a positive effect. However, in the presence of K₂CO₃ or NEt₃, the [2+4] annulation reaction was almost shut down (entries 20 and 21). Moreover, the reaction yield diminished dramatically when manganese was used as the reducing agent instead of zinc (entry 22). Furthermore, performing the reaction with a catalyst loading of 10 mol% resulted in a lower yield (entry 23).

Table 1. Optimization of the reaction conditions^a.

Display Full Size

entry	ligand	Ni-precat.	yield (%)b	ee (%)c
1	L1	NiBr2•glyme	32	8
2	L2	NiBr2•glyme	11	3
3	L3	NiBr2•glyme	10	34
4	L4	NiBr2•glyme	0	-
5	L5	NiBr2•glyme	0	-
6	L6	NiBr2•glyme	41	25
7	L7	NiBr2•glyme	70	11
8	L8	NiBr2•glyme	76	75
9	L9	NiBr2•glyme	86	91
10	L10	NiBr2•glyme	88	90
11	L11	NiBr2•glyme	55	63
12	L12	NiBr2•glyme	55	‒40
13	L13	NiBr2•glyme	27	82
14	L9	NiCl2•glyme	85	88
15	L9	NiI2	70	90
16	L9	Ni(COD)2	86	91
17	L9	NiBr2•diglyme	90	91
18d	L9	NiBr2•diglyme	57	88
19e	L9	NiBr2•diglyme	79	90
20f	L9	NiBr2•diglyme	trace	n.d.
21g	L9	NiBr2•diglyme	trace	n.d.
22h	L9	NiBr2•diglyme	10	80
23i	L9	NiBr2•diglyme	53	91

^a: Unless otherwise specified, the reactions were performed on a 0.1 mmol scale of α-methyl o-iodostyrene (1a) using 2 equiv of the α-naphthylalkyne 2a, 15 mol% Ni-precatalyst, 20 mol% ligand L, and 1.5 equiv of Zn in 0.5 mL DMA under N₂ atmosphere for 24 h at 40 °C; ^b: Yield of the isolated product; ^c: Determined by HPLC analysis on a chiral stationary phase; ^d: The reaction was performed in DMF; ^e: The reaction was performed in NMP; ^f: The reaction was performed with 2 equiv of K₂CO₃; ^g: The reaction was performed with 2 equiv of NEt₃; ^h: Mn was used as the reductant instead of Zn; ⁱ: The reaction was performed with 10 mol% NiBr₂•diglyme and 15 mol% L9. DMA: N,N-dimethylacetamide; HPLC: high-performance liquid chromatography; DMF: N,N-dimethylformamide; NMP: N-methyl-2-pyrrolidone.

3.2 Evaluation of the substrate scope

After identifying the optimized reaction conditions, we commenced evaluating the substrate scope of this nickel-catalyzed annulation. First, the generality of o-halostyrenes was investigated and the results were summarized in Figure 2. The use of the α-methyl o-bromostyrene (1a') instead of its iodo analog 1a also afforded the annulation product 3aa in high enantioselectivity but less efficiently. In the case of ethyl (1b) or n-propyl (1c) as the geminal substituent of the styrenes, the corresponding products 3ba and 3ca were obtained in good yields and high enantioselectivities. Although bulkier isopropyl (1d) and cyclopentyl (1e) turned out to have a detrimental effect on the reaction efficiency, the asymmetric induction remained at a high level (3da and 3ea). In contrast, the reaction employing α-phenyl o-iodostyrene (1f) proceeded with somewhat lower enantiocontrol (3fa). The influence of substituents on the phenyl ring of the styrene substrates was subsequently examined. An array of functionalities including phenyl (1g and 1m), furyl (1h), thienyl (1i), chloro (1j and 1k), methyl (1l), methoxy (1n), and trifluoromethyl (1o) on either the meta or para position of the iodo group were interrogated, and the corresponding products 3ga−oa were formed in moderate to excellent yields and moderate to high atroposelectivities. In general, relatively low enantiocontrol was observed for the alkenes bearing a substituent on the meta position of the iodide. In addition, the reaction for the synthesis of compound 3aa on a 2 mmol scale afforded a similar result (85% yield, 90% ee).

Display Full Size

Download

Figure 2. Evaluation of the scope of o-iodostyrenes^[a-c]. ^a: Unless otherwise specified, the reactions were performed on a 0.1 mmol scale of o-iodostyrenes 1a−o using 2 equiv of the α-naphthylalkyne 2a, 15 mol% NiBr₂•diglyme, 20 mol% ligand L9, and 1.5 equiv of Zn in 0.5 mL DMA under N₂ atmosphere for 24 h at 40 °C; ^b: Yield of the isolated products; ^c: Determined by HPLC analysis on a chiral stationary phase; ^d: α-Methyl o-bromostyrene (1a') was used instead of 1a; ^e: The reaction was performed on a 2 mmol scale of 1a; ^f: Reaction time: 36 h. DMA: N,N-dimethylacetamide; HPLC: high-performance liquid chromatography.

Next, we moved to assess the scope of β-substituted α-naphthylalkynes as the coupling partner (Figure 3). Permutation of the substituents on the C-7, C-6, or C-5 position of the naphthyl ring was first carried out. It turned out that both electron-donating and electron-withdrawing groups were well tolerated, and the corresponding products 3ab−at were provided in good to high enantiomeric excesses. Moreover, in the case of the methyl, benzyl, or TBS-protected propargylic alcohols (2u−w), the homopropargylic alcohol 2x, and the simple internal alkyne 2y without a pendant functional group, good enantiocontrol could also be achieved for the corresponding products 3au−ay, confirming that the hydroxyl moiety on the propargylic position is not requisite for this asymmetric annulation reaction. The atroposelectivity of the studied reaction depended significantly on the bulkiness of the ortho alkoxy group of the naphthalene backbone. In the case of methoxy (2z), the enantiomeric excess of the product 3az decreased dramatically to 57%, while good results in terms of enantiocontrol could be obtained in the case of sterically more demanding benzyloxy (3aaa) and β-naphthylmethoxy (3aab) groups. Furthermore, the reaction using ortho-phenyl substituted naphthyl propargylic alcohol 2ac as the precursor also proceeded enantioselectively albeit with relatively poor asymmetric induction (3aac). All the alkynes that underwent the desired annulation are electronically and sterically biased, bearing one aryl and one alkyl substituent, and the results obtained with compounds 3ay and 3aac are inconsistent with a chelation-directed regiocontrol model. Therefore, the consistently observed regioselectivity across these substrates is more likely governed by electronic and/or steric differentiation between the two alkyne termini. Notably, we did not observe the formation of a [2+3] annulation by-product in all the cases demonstrated in Table 2 and Table 3. In addition, the absolute configuration of compound 3ae was unambiguously determined to be R through X-ray crystallography (CCDC number: 2247891).

Display Full Size

Download

Figure 3. Evaluation of the scope of α-naphthylalkynes^[a-c]. ^a: Unless otherwise specified, the reactions were performed on a 0.1 mmol scale of α-methyl o-iodostyrene (1a) using 2 equiv of the α-naphthylalkynes 2b−ac, 15 mol% NiBr₂•diglyme, 20 mol% ligand L9, and 1.5 equiv of Zn in 0.5 mL DMA under N₂ atmosphere for 24 h at 40 °C; ^b: Yield of the isolated products; ^c: Determined by HPLC analysis on a chiral stationary phase; ^d: Reaction time: 36 h. DMA: N,N-dimethylacetamide; HPLC: high-performance liquid chromatography.

3.3 Synthetic transformations

To demonstrate the utility of this method, various derivatizations of the annulation product 3aa were conducted toward the synthesis of potential organocatalysts or ligands with axial chirality (Figure 4). First, Dess Martin oxidation of 3aa afforded an aldehyde 4 in 83% yield, which was subsequently subjected to the Pinick oxidation, furnishing an axially chiral carboxylic acid 5 in 77% yield. Furthermore, the sequential reactions consisting of mesylation and nucleophilic substitution by NaN₃ converted compound 3aa into an azide 6 in 76% yield over two steps, starting from which a primary amine 7 bearing a chiral 1,2'-binaphthyl scaffold was synthesized in 64% yield by means of the Staudinger reaction. Upon treatment with a thioisocyanate, the amine 7 was transformed into a Schreiner’s thiourea 8 with axial chirality in 67% yield. In the presence of BBr₃, compound 3aa underwent not only the removal of the mesitylmethyl group but also the simultaneous bromination of the alcohol moiety, affording a binaphthyl 9 containing both phenol and bromide moieties in a nearly quantitative yield. The use of the pendant bromide of 9 as the substrate in the nucleophilic substitution by morpholine provided a chiral bifunctional 1,2'-binaphthyl 10 in 48% yield, which incorporates a phenol as a Brønsted acid and a tertiary amine as a Brønsted base. Notably, no significant erosion of the enantioselectivity was observed in the reactions aforementioned.

Display Full Size

Download

Figure 4. Derivatizations of the [2+4] annulation product.

Subsequently, the axially chiral amino naphthol 10 was investigated as the ligand in the enantioselective ethylation of benzaldehyde with diethyl zinc^[81,82]. As depicted in Figure 5, a chiral alcohol 11 was obtained in 77% yield and 95% ee. Notably, the enantiomeric excess of the product 11 was slightly higher than the ligand 10 (89% ee), suggesting possible ee amplification in the catalytic system; however, further systematic studies would be required to establish a definitive nonlinear effect^[83].

Display Full Size

Download

Figure 5. Asymmetric ethylation of benzaldehyde using the axially chiral compound 10 as the ligand.

3.4 Mechanistic studies

To gain further insight into the mechanism of the studied reaction, several experiments were conducted (Figure 6). First, the reaction between α-methyl o-iodostyrene (1a) and the α-naphthylalkyne 2a using 1.0 equiv of Ni(COD)₂ as the promoter in the absence of Zn delivered the annulation product 3aa in 68% yield and 80% ee, supporting the feasibility of a Ni(0)/Ni(II) catalytic cycle that does not require reduction of a nickel intermediate by zinc (Figure 6A). Moreover, the addition of 1.0 equiv of BHT (2,6-di-tert-butyl-4-methylphenol) as a radical scavenger to the reaction mixture had a minimal effect on the reaction outcome, arguing against a radical-mediated pathway (Figure 6B). Next, experiments were conducted to compare the initial conversion rates at 10 min for aryl iodide 1a, aryl bromide 1a', and aryl chloride 1a'' (Figure 6C). Comparable rates for these reactions would be inconsistent with oxidative addition of the aryl halide to a low-valent nickel species being rate-limiting. However, aryl iodide 1a reacted much faster than aryl bromide 1a', whereas aryl chloride 1a'' remained intact under the reaction conditions. These reactions alone could not establish that oxidative addition is rate-limiting in the reactions engaging aryl iodides. Therefore, more detailed kinetic studies were performed by measuring the initial reaction rates at 10 min under different concentrations of catalyst or reactants. As shown in Figure 6D, the results reveal zero-order dependence on α-methyl o-iodostyrene (1a) and the α-naphthylalkyne 2a, and first-order dependence on the nickel catalyst. Therefore, oxidative addition of 1a to Ni(0) and the intermolecular alkyne insertion are unlikely to be the turnover-limiting step, narrowing down the possible ones to the unimolecular elementary steps, which are intramolecular alkene insertion and β-H elimination. Furthermore, the intermolecular competition between 1a and its dideuterated analog d₂-1a gave an inverse secondary kinetic isotope effect, consistent with a change in hybridization at the olefinic carbon during 6-endo cyclization. This result supports intramolecular alkene insertion as the turnover-limiting step (Figure 6E).

Display Full Size

Download

Figure 6. Mechanistic studies.

4. Conclusion

In conclusion, we have developed a nickel-catalyzed asymmetric [2+4] annulation of β-substituted α-naphthylalkynes with o-iodostyrenes, providing new and highly enantioselective access to axially chiral 1,2'-binaphthyls via the de novo construction of a benzene ring. The synthetic utility is demonstrated by a variety of derivatizations of the annulation product towards the synthesis of potential organocatalysts with axial chirality.

Supplementary material

The supplementary material for this article is available at: Supplementary materials.

Authors contribution

Hu W, Mao Y: Investigation, data curation, formal analysis, writing-review & editing.

Wang C: Conceptualization, methodology, project administration, writing-original draft, writing-review & editing.

Conflicts of interest

The authors declare no competing interests.

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent to publication

Not applicable.

Availability of data and materials

The data reported in this paper are available in the main text or supplementary material, including methods, NMR data, HRMS data, HPLC spectra Crystal data and NMR spectra. Crystallographic data for the structure reported in this article have been deposited at the Cambridge Crystallographic Data Centre, under deposition numbers CCDC 2247891 (3ae).

Funding

This work was supported by the National Natural Science Foundation of China (Grant Nos. 22271267 and 22471256).

Copyright

© The Author(s) 2026.