Enantioselective construction of polychiral helical compounds via dynamic kinetic resolution

Abstract

Chiral helicenes and their derivatives are valuable across diverse research fields, yet the development of helical chirality has lagged far behind that of point and axial chirality. The synthesis of polychiral or heteroatom-containing helical systems remains especially challenging because of limited starting materials and difficulties in stereochemical control. Here we computationally design a flexible oxa[4]helicenoid-ol scaffold with an optimal rotational barrier (ΔG^‡ = 21.7 kcal mol⁻¹) and enantioselective recognition ability, enabling dynamic kinetic resolution under mild conditions. This platform allows efficient access to helicenes bearing either dual helical chirality or combined helical and axial chirality, with excellent enantio-, diastereo- and E/Z selectivities. The strategy incorporates non-aromatic oxacycles, breaks conventional cyclization patterns and expands the structural diversity of helical chiral scaffolds. Density functional theory calculations elucidate the mechanistic origins of reactivity and stereoselectivity. The resulting helicenoid-quinones are readily derivatized and display promising optical properties for functional materials.

Chiral helicenes are valuable across disciplines, yet strategies to construct helical chirality—particularly in polychiral or heteroatom containing systems—remain limited by limited substrates and challenging stereochemical control. Here, the authors report an enantioselective dynamic kinetic resolution enabling efficient construction of polychiral helical architectures.

Introduction

Helicenes and helicenoids are both ortho-fused polycyclic compounds, featuring π-conjugated skeletons and special helical three-dimensional topological structures, exhibit unique physical and chemical properties^1–4, and have important applications in fields such as asymmetric catalysis^5–10, chiral photoelectric materials^11–18 and other fields^19–24. These unique features have gained significant attention from the chemical community, prompting extensive research into their asymmetric catalytic synthesis of helicenes (helicenoids). Consequently, substantial progress has been made in this field^25–29.

The introduction of multiple stereogenic elements into molecules significantly increases the complexity of their three-dimensional topological structures and enhances molecular specificity. These modifications profoundly influence the physical and chemical properties of the molecules, thereby potentially augmenting their bioactivity, catalytic properties, and material performance^30–37. In recent years, substantial efforts have been directed toward the development of catalytic asymmetric synthesis methods for such compounds, yielding notable achievements in the production of optically pure compounds with various types of chirality, including multiple point chirality, axial and point chirality, planar and point chirality, and multiple axial chirality (Fig. 1A)^38,39.

Fig. 1. Research background and this work.

A Background. B Gaps and opportunities in helicene studies. C This work: Organocatalytic terminal asymmetric substitution via DKR.

Conversely, the catalytic asymmetric synthesis of polychiral helical compounds or helical compounds bearing additional stereogenic elements remains a relatively underexplored area, with only a limited number of reports available in the literature. In 2014, Tanaka’s group reported the first catalytic asymmetric construction of S-type double helical compounds through an asymmetric [2 + 2 + 2] cycloaddition reaction⁴⁰. Subsequently, they conducted in-depth research on the asymmetric synthesis of other S-type double-helical molecules^41–44. The research groups of Kamikawa⁴⁵, Yan³⁵, Liu⁴⁶ and Du⁴⁷ achieved asymmetric synthesis of triple helicenes, quadruple helicene-like molecules and helical polycyclic phenols, respectively. Asymmetric synthetic methods for helical compounds bearing axial or point chirality have also been achieved via alkyne cyclization, cycloaddition and dearomatization by Bonne^48,49, Yan⁵⁰, Yang^51,52 and Wang⁵³ et al. Notably, most of the reported cases rely on catalytic enantioselective ring-expansion strategies, which often require relatively harsh reaction conditions. Thus, the exploration of novel and practically feasible approaches to acquire enantiopure helical compounds remains highly desirable.

To the best of our knowledge, only one report has described the synthesis of helical compounds bearing other stereogenic elements through the strategy of terminal substitution reaction⁵⁴ of metastable helicenes. This reaction only enabled the formation of helicenes bearing remote axial chirality via kinetic resolution at relatively low temperatures (Fig. 1B)⁵⁵. No terminal functionalization methods have been reported that can construct polychiral helical compounds or helical compounds bearing other stereo-genic elements via dynamic kinetic resolution (DKR). The main solution to the abovementioned issue is rational substrate design to meet the following requirements: (1) an appropriate racemization energy barrier enables the dynamic kinetic resolution via slow racemization of the substrate under the reaction conditions, while preventing racemization of the final helically chiral products; (2) improved stereodiscrimination through an induced large dihedral angle (θ ≈ 50°), generating significant conformational differences between enantiomers to facilitate effective recognition by chiral catalysts⁵⁶; (3) an optimal steric profile imparted by the C1-methyl group, endowing the molecule with a size intermediate between [4]- and [5]helicenes. This strategic molecular design integrates an ideal rotational barrier (21.7 kcal/mol) with rapid tautomerization kinetics under mild reaction conditions, synergistically enabling an efficient DKR process (Fig. 1C). Herein, we achieved the synthesis of a polychiral helical product between the reaction of oxa[4]helicenoids-ols and vinylidene ortho-quinone methides, which features an intricate and fascinating structure. This product was obtained with high enantioselectivity, diastereoselectivity, and E/Z selectivity.

Results

Computationally guided substrate design

To achieve the construction of polychiral helical compounds or helical compounds bearing other stereogenic elements through terminal-ring asymmetric substitution via dynamic kinetic resolution (DKR), we conducted a comprehensive analysis of the rotational energy barriers of a series of polycyclic phenols (1A-1F) using density functional theory (DFT) calculations (Cartesian coordinates are provided in the Source Data file). Subsequently, we investigated their reactivity through substitution reactions, as depicted in Fig. 2. Among these compounds, 1A and 1B exhibit low helical rotational energy barriers (3.9 and 2.2 kcal/mol) and minimal steric hindrance, making them well-suited for DKR processes. However, the construction of configurationally stable helicenes (helicenoids) via terminal-ring substitution remains challenging for these compounds due to their low rotational barriers, which hinder the formation of stable helical structures. In contrast, 1 C and 1D possess higher helical rotational barriers (24.7 and 27.0 kcal/mol), rendering them suitable for terminal-ring substitution strategies. Nevertheless, DKR processes are impeded for these compounds, as the interconversion between enantiomers is significantly hindered when the rotational barrier exceeds 24 kcal/mol (see Table S6 in Supplementary Information for details). Compared to 1A and 1B, 1E and 1F exhibit greater steric hindrance and a significantly elevated rotational energy barrier due to the introduction of methyl substituents (21.3 and 21.7 kcal/mol). These structural modifications not only facilitate the construction of stable helicenes via terminal-ring asymmetric substitution but also maintain rapid interconversion between the compound’s enantiomers. Consequently, 1E and 1F are well-suited for constructing stable helical compounds through terminal-ring asymmetric substitution reactions in DKR processes. Furthermore, DFT calculations revealed dihedral angles of 43° (1E) and 50° (1F) for the two substrates. We propose that the discrepancy between these dihedral angles (43° vs. 50°) and the enhanced skeletal flexibility imparted by the methylene (CH₂) bridge collectively render the fjord regions of the prochiral substrates more conformationally adaptable. This structural feature thereby facilitates selective recognition by the catalyst, leading to improved catalytic activity and stereoselectivity.

Fig. 2. Substrates screening via DFT calculations and experiments.

^aRelative gibbs free energies were calculated via DFT.

We subsequently explored the reactivity of these compounds through terminal-ring asymmetric substitution reactions with vinylidene ortho-quinone methide (VQM) intermediates, catalyzed by chiral phosphoric acid (CPA). VQMs are highly reactive intermediates with multiple active sites and are widely used in the asymmetric synthesis of axially chiral and complex chiral molecules^57–62. To our surprise, the reactions of 1A-1F with 2a yielded not only 3, a helical molecule bearing multi-axial chirality, but also a multi-helical chirality product 4, formed through a further dehydration-condensation reaction from 3. As predicted by DFT calculations, the reaction involving 1A exhibited nearly no enantioselectivity. In contrast, 1B generated the polychiral helical product 4B with moderate enantioselectivity, likely due to its relatively large steric hindrance. Compounds 1C and 1D also produced the corresponding products with moderate enantioselectivities and yields, although the diastereoselectivities of these reactions were not satisfactory. Notably, 12-methyl-[4]helicene-ol (1E), which features an appropriate rotational energy barrier, demonstrated the ability to generate the corresponding product with moderate yield and stereoselectivity. Significantly, when oxa[4]helicenoid-ol (1F) was used as the substrate, a remarkable enhancement in the reaction’s yield, enantioselectivity, and diastereoselectivity was observed.

Optimal conditions

Based on the aforementioned computational and experimental results, we selected compound 1F (1a) as the substrate for optimizing reaction conditions, as detailed in Table 1. Initially, we screened various chiral phosphoric acids and found that the SPINOL-derived (S)-C2 exhibited significant potential, yielding 4a with 95:5 er, a diastereomeric ratio (dr) of >19:1, and a yield of 18%. The addition of 4 Å molecular sieves (MS) further enhanced the enantioselectivity, although there was no notable increase in yield. Subsequently, we evaluated the impact of catalyst properties and determined that a catalyst with stronger acidity was advantageous for improving the yield of 4a. Gratifyingly, the incorporation of a Lewis acid not only increased the yield of product 4a but also preserved the excellent enantioselectivity and diastereoselectivity⁶³. Further optimization involved investigating the catalyst loading and other reaction conditions. When the reaction was conducted at 40 °C using 10 mol% of the catalyst (S)-C2, 10 mol% of the Lewis acid, and 4 Å MS for 72 h, we achieved a good yield of 67% and excellent enantioselectivity of 98:2 er (see Table S1-3 in Supplementary Information for details).

Table. 1.

Optimization of reaction conditions^a

Entrya	CPA	Additive 1	Additive 2	Time/h	4F(4a)
					Yield(%)b	erc	drd
1e	(S)-A1	---	---	24	79	90:10	>19:1
2e	(S)-A2	---	---	24	38	59:41	>19:1
3e	(S)-A3	---	---	24	41	75:25	16:1
4	(S)-A8	---	---	24	70	81:19	>19:1
5	(S)-B3	---	---	24	70	58:42	13:1
6	(S)-C1	---	---	24	58	22:78	>19:1
7	(S)-C2	---	---	24	18	5:95	>19:1
8	(S)-C3	---	---	24	85	12:88	>19:1
9	(S)-C2	4 Å MS	---	48	33	3:97	>19:1
10	(S)-C2	4 Å MS	B(C6F5)3	48	95	38:62	6:1
11	(S)-C2	4 Å MS	CeCl3	48	43	4:96	>19:1
12	(S)-C2	4 Å MS	ZnCl2	48	47	3:97	>19:1
13	(S)-C2	4 Å MS	LiCl	72	52	2:98	>19:1
14 f	(S)-C2	4 Å MS	LiCl	72	67	2:98	>19:1

^aReaction condition: 1a (0.1 mmol), 2a (0.15 mmol), CPA (5 mol%), Additive 1 (40 mg), Additive 2 (10 mol%) in PhCF₃ (4 mL) at 40 °C for 24 h.

^bIsolated yields.

^cThe ee values were determined by HPLC with a chiral stationary phase.

^dThe dr values were determined by ¹H-NMR.

^eAt RT for 24 h.

^fAdded CPA (10 mol%), LiCl (10 mol%). The E/Z ratio for 4a was >19:1.

Substrate scope investigation

With the optimized conditions established, we investigated the substrate scope of the CPA-catalyzed asymmetric terminal-ring substitution and tandem condensation reaction of metastable helicenoid-ols (helicenols) for the construction of polychiral helical compounds. As shown in Fig. 3, either electron-donating or electron-withdrawing groups at the para-position of the phenyl ring were all compatible with this protocol, affording the corresponding polychiral helical products in good yields and excellent enantioselectivities (4b-4j, 51-72% yield, 94:6-99:1 er).While the substituent was introduced at the meta-position of the benzene ring, the product maintained excellent enantioselectivity and the isolated yields were good (4k-4o, 45-61% yield, 97:3-98:2 er). Changing the substituent to other position of the 1-(phenylethynyl)naphthalen-2-ol had no apparent effect on the reaction results (4u-4x, 51-68% yield, 93:6-98:2 er). Next, we targeted the exploration of substrate scope with respect to oxa[4 or 5]helicenoid-ols. Pleasingly, all substrates underwent smooth conversion, furnishing the desired products in moderate yields with excellent stereoselectivity (5b-5d, 46-65% yield, 90:10-98:2 er). We also investigated the reactivity of different rigid metastable helicenols (1e-1i) under standard conditions. As previously described, metastable helicenols with strong rigidity and a smaller dihedral angle could generate the corresponding products (5e-5i), but exhibited lower enantioselectivity.

Fig. 3. Scope with respect to 1 and 2.

1 (0.1 mmol), 2 (0.15 mmol), (S)-C2 (10 mol%), LiCl (10 mol%), 4 Å MS (40 mg) in PhCF₃ (4 mL) at 40 °C for 72 h. ^a1 (0.1 mmol), 2 (0.15 mmol), (S)-A1 (5 mol%), 4 Å MS (40 mg) in PhCF₃ (4 mL) at RT for 24 h. Isolated yields. Isolated yields. The er values were determined by HPLC with a chiral stationary phase. The E/Z ratio was >19:1.

Subsequently, we probed the reactivity of metastable helicenoid-ols with iminoquinones. To our delight, in the presence of CPA, 1a and 6a underwent an asymmetric terminal-ring substitution reaction by DKR process, affording stable helical compound bearing inner-side axial chirality (7a) with excellent diastereoselectivity (dr >19:1). In particular, under the reactionconditions with (S)-A4 as the catalyst and a mixture of CCl₄ and m-xylene as the solvent, the transformation efficiently constructed 7a with high yield and excellent stereoselectivity (78% yield, er 99:1, dr > 19:1, see Table S4 in Supplementary Information for details). Under the optimized reaction conditions, we systematically investigated the substrate scope of this organocatalyzed DKR reaction. Firstly, a variety of iminoquinones were evaluated. As depicted in Fig. 4, iminoquinones bearing various substituted benzenesulfonyl groups, including electron-donating substituents such as methyl and methoxyl (6a-6e), furnished the corresponding oxa[4]helicenoids 7a-7e with moderate to high yields and moderate to excellent enantioselectivities (up to 99:1 er). Notably, electron-withdrawing substituents (F, Cl, Br, NO₂, CN, CO₂Me) on the benzene ring resulted in the formation of oxa[4]helicenoids 7f-7l with very good yields and enantioselectivities. Furthermore, the reaction with different linear and cyclic alkyl sulfones 6m-6r showed variations in yields but maintained excellent enantioselectivities (up to 99.5:0.5 er). Substrates with other sulfonyl groups, such as 1-naphthyl, styryl, and 2-thienyl (6s-6u), also exhibited remarkable enantioselectivities, with enantiomeric ratios (er) all exceeding 97:3. Further evaluation of substrates 1b and 1c resulted in the formation of the corresponding products (7v and 7w) with enantiomeric ratios of 99.5:0.5 and 85.5:14.5, respectively. In order to further investigate the range of substrates, we assessed the reaction between 1a and quinone methyl ester under standard reaction conditions. Unfortunately, only helicenoid-p-hydroquinone was produced in a yield of 37% with an enantiomeric ratio of 42.5:57.5. Additionally, helicenoid-p-quinone was obtained in 35% yield with the same enantioselectivity. It is evident that quinone ester functioned as an oxidizing agent in this reaction. Therefore, we adjusted the quinone equivalent and conducted a reevaluation of the reaction conditions (see Table S5 in Supplementary Information for details). The optimized conditions resulted in the successful formation of helicenoid-p-quinone 9a in 87% yield with 92:8 er when the reaction was carried out in DCE at 0 °C for 18 hours in the presence of 5 mol% (S)-C4. Subsequently, employing these optimized conditions, we explored the substrate scope of this DKR process by examining various quinone esters and [4]helicenoid-ols, as shown in Fig. 5. Firstly, different quinone esters 8a-8h were examined, which afforded the corresponding helicenoid-p-quinones 9a-9h in good yields (83-89%) with satisfactory enantioselectivities (86:14-96:4 er). Overall, these experimental results indicate that enhanced enantioselectivity was achieved with a bulkier ester group substituent. Furthermore, we investigated the reaction between oxa[4]helicenoid-ols 1b and 1c with quinone isopropyl ester, yielding target products 9i and 9j in good yields with enantiomeric ratios of 97:3 and 93.5:6.5, respectively. It is valuable to note that the quinone ester reacted smoothly with rigid metastable helicenols, resulting in the corresponding products 9k-9n with good yields and moderate enantioselectivities. Clearly, the enantioselectivity of non-aromatic and flexible oxa[4]helicenoid-ols surpass those of aromatic carbo[4]helicenols, as evidenced by the expanded substrate results mentioned above.

Fig. 4. Scope with respect to 1 and 6.

Reaction condition: 1 (0.1 mmol), 6 (0.11 mmol), (S)-A4 (5 mol%) in a solvent mixture of CCl₄/m-Xylene (2 mL) at room temperature for 12 h. The dr value for all products were >19:1. Isolated yields. The er values were determined by HPLC with a chiral stationary phase.

Fig. 5. Scope with respect to 1 and 8.

1 (0.1 mmol), 8 (0.22 mmol), (S)-C4 (5 mol%) in DCE (2 mL) at 0 °C for 18-30 h. The dr value for all products were >19:1. Isolated yields. The er values were determined by HPLC with a chiral stationary phase.

Scale-up, derivatization and application

To evaluate the practical feasibility of these methodologies and helicenoids, we conducted large-scale reaction using 1 mmol of 1a with 2a’, 6a and 8d, respectively. Both two reactions exhibited the capability to yield the desired products 4u, 7a and 9d in commendable yields with excellent enantioselectivities, demonstrating the viability of implementing these methodologies for large-scale chemical production (Fig. 6). Furthermore, the Pd-catalyzed Suzuki coupling of 4u with arylboronic acid afforded 4ua, while the Sonogashira coupling with phenylacetylene produced the 4ub. Notably, the dioxa-double helicene maintained excellent enantiopurity throughout these transformations, demonstrating remarkable configurational stability despite exposure to transition metal catalysts and elevated reaction temperatures. In addition, enhanced optical purity (99:1 er) can be achieved by further recrystallization of 9d. Subsequently, a series of derivative reactions were performed. Chiral oxa[4]helicenoid 9d could be efficiently reduced to 10a with excellent enantiomeric excess. Furthermore, the reaction between 9d and 4-methoxybenzenethiol in a mixed solution of methanol and toluene at -78 °C produced 10b in 89% yield without any loss of enantiomeric ratio. Additionally, helicenoid 10c can be obtained in high yield and negligible decrease in enantiomeric excess when 9d undergoes a Michael addition process catalyzed by Sc(OTf)₃ with 2-methylindole. Moreover, oxa[4]helicenoid 9d reacted smoothly with diarylphosphine oxides to afforded phosphorus-installing helicenoid 10d in 91% yield with 98:2 er. The absolute configurations of 4m, 7a and 10a were determined through single-crystal X-ray crystallographic analysis (see Figs. S3–6, Tables S7–10 in Supplementary Information for details). To demonstrate the stereochemical stability of these polychiral compounds, we independently determined the racemization barrier of 4a through computational and experimental methods, with the calculated prediction yielding 36.63 kcal/mol and the experimental measurement giving 36.06 kcal/mol (the graph is detailed in Figs. S8–9 in Supplementary Information).

Fig. 6.

Scale-up synthesis and derivatizations of 4u, 7a and 9d.

Due to their fused conjugated structure and unique chiral three-dimensional arrangement, helicenoids have garnered significant interest in the realm of optoelectronic materials, particularly chiral luminescent materials. Therefore, we selected several [4]helicenoids and their derivatives for chiral optical properties as depicted in Fig. 6 (7b, 9d, 10a, 10b, 10c, 10d) and Table S11. The results indicated that 9d exhibited extremely weak fluorescence emission, thus lacking a basis for CPL measurement (Fig. 7b). Meanwhile, 7b showed an extremely weak CPL signal, so that the accurate luminescence dissymmetry factors (g_lum) could not be obtained. However, further derivatization of M-9d facilitates the measurement of CPL, and revealing respectable g_lum value, with maximum g_lum of −2.9 × 10⁻³ for M-10a and −3.6 × 10⁻³ for M-10b. In addition, the CPL direction of oxa[4]helicenoid 10c with indole substitution is reversed compared to others⁶⁴ (Fig. 7).

Fig. 7. Photophysical and chiroptical properties of selected products (7b, 9d, 10a, 10b, 10c, 10d).

a Absorption spectra for selected products measured at 2×10⁻⁵M in DCM. b Emission spectra for selected products measured at 2 × 10⁻⁵M in DCM. c Circular dichroism (CD) spectra of selected products measured at 2 × 10⁻⁵M in DCM. d g_abs spectra of selected products. e Circularly Polarized Luminescence (CPL) of selected products measured at 1 × 10^-4M in DCM. f g_lum spectra of selected products.

Mechanism exploration

Under the reaction conditions detailed in the manuscript (Table 1), compounds 1a and 2 not only convert to bisphenol 3 but also undergo an unusual dehydration to form product 4. Based on these observations, we propose the following reaction pathway: the reaction likely proceeds via an isomerization of 1-(phenylethynyl)naphthalen-2-ol, followed by nucleophilic addition and aromatization to yield product 3 (Fig. 8). The hydrogen bonds formed between the two hydroxyl groups in 3 serve to stabilize the product and facilitate rapid interconversion between 3 and its keto tautomer 3-keto. Upon activation by the catalyst, 3-keto undergoes intramolecular nucleophilic attack by the hydroxyl group, forming a cyclized intermediate. Subsequent elimination of a water molecule then leads to the formation of the final product 4⁶⁵. Next, we performed DFT calculations^66–69 to investigate the proposed reaction mechanism above and the origin of stereoselectivity. Geometry optimizations were performed in the gas phase using the M06-2X functional with the 6-31 G(d) basis set. Frequency calculations were conducted at the same level of theory to verify the nature of stationary points and acquire thermal corrections. High-accuracy single-point energies of transition states were separately calculated in the gas phase at the M06-2X-D3/def2-TZVPP level. Solvation energies were determined as follows with SMD model: single-point energies of the optimized geometries were first computed at the M05-2X/6-31 G(d) level in both the gas phase and (trifluoromethyl)benzene solution; the solvation energy was defined as the difference between these two energies. Final Gibbs free energies were obtained by summing the high-accuracy single-point energy, thermal corrections from frequency calculations, and solvation energy. The calculated free energy profile of the reaction is shown in Fig. 8.

Fig. 8.

Calculated free energy profile of the model reaction.

Initially, the bifunctional phosphoric acid catalyst acts as a proton shuttle, facilitating the conversion of substrate 2a into the vinylidene ortho-quinone methide (VQM) via a 1,5-hydrogen shift (transition state TS1). A closer examination of the transition states reveals that in the favored TS-1A, the bond distance between the benzylic carbon and the acidic proton of the phosphoric acid (1.25 Å) is shorter than that in TS-1B (1.27 Å), suggesting a higher degree of protonation at the benzylic carbon. Additionally, the P = O···H distance in TS-1A (1.59 Å) is shorter than that in TS-1B (1.63 Å). These factors collectively lower the energy of TS-1A by 2.9 kcal/mol compared to TS-1B, indicating that R-VQM is the dominant intermediate. The higher energy barrier associated with TS-1, as compared to that of any other step in the reaction sequence, not only designates it as the rate-determining step but also serves as the key factor governing the determination of the absolute configuration of the axially chirality. Since the racemization barrier of 1a is only 21.7 kcal/mol (Fig. 2), P-1a can rapidly convert to M-1a, enabling dynamic kinetic resolution. Thus, CP-3 could be smoothly formed with M-1a and CP-2a through hydrogen bonding.

Subsequently, the ortho-carbon of the phenolic hydroxyl group in 1a attacks VQM to form a C-C bond via TS-2, yielding intermediate CP-4. The energy barrier for this C-C bond formation step (TS-2) is higher than that of the subsequent cyclization step (TS-3), indicating that the C-C bond formation determines the relative configuration of the helical chirality. In this regard, four possible TSs were located to account for the formation of both the major and minor diastereomers (Fig. 8 and Fig. S11). As expected, TS-2A, with the lowest relative energy among the four calculated TSs, led to the major product. Specifically, the relative energy of TS-2A is 0.82 kcal/mol lower than that of TS-2C, associated with the minor diastereomer. The calculated diastereomeric ratio (dr) is 89:11, which agrees with the experimentally observed dr.

The free energy difference between TS-2A and TS-2B is 1.57 kcal/mol, leading to an er value of 92:8, which agrees well with the experimentally observed er value of 88:12. To elucidate the origin of stereoselectivity, we employed the Interaction Region Indicator (IRI) method^70–72 developed by Lu et al. to analyze non-covalent interactions in TS-2 (Figs. 9 and S10). Multiple non-covalent interactions, including O-H···O, C-H···O, C-H···π, and π···π interactions, are present in TS-2A and TS-2B. The O-H···O hydrogen bond is stronger in TS-2A, where the O-H distance (1.03 Å) is shorter than the corresponding O-H distance (1.04 Å) in TS-2B, suggesting more effective activation of VQM by catalyst S-C3 in TS-2A. Similarly, the P = O···H distance in TS-2A (1.54 Å) is notably shorter than the P = O···H distance in TS-2B (1.62 Å), enhancing the nucleophilicity of 1a in TS-2A. Furthermore, the non-covalent analysis reveals that 1F interacts with the catalyst via a C-H···O hydrogen bond, while VQM forms a C-H···O hydrogen bond and a C-H···π hydrogen bond with the catalyst in TS-2A (Fig. 9). These interactions are absent from TS-2B, affording TS-2A with additional stability. Thus, the observed stereoselectivity is attributed to the strength difference of O-H···O hydrogen bonds and the synergistic effect of multiple C-H···O and C-H···π hydrogen bonds. Energy decomposition analysis (EDA) of transition states TS-2A and TS-2B was also performed via the sobEDAw method⁷³ developed by Lu et al. to further support the observed stereoselectivity. The anionic moiety of the chiral phosphate acid catalyst was designated as one independent fragment, while the remaining constituents of the transition-state complex constituted the second fragment. The calculations were conducted at the B3LYP-D3(BJ)/6-311 + G(2 d,p) w.CB level in the gas phase. All computed EDA results are compiled in Table 2. The total interaction energy of TS-2A ( − 121.72 kcal/mol) is 0.79 kcal/mol more negative than that of TS-2B ( − 120.93 kcal/mol), which exhibits good consistency with the 1.57 kcal/mol relative energy difference predicted by density functional theory (DFT) calculations. More importantly, distinct differences are observed in the individual energy components between the two transition states: compared with TS-2B, TS-2A possesses a more negative electrostatic energy (ΔΔE = −6.81 kcal/mol) and dispersion energy (ΔΔE = −5.03 kcal/mol), which contribute to the enhanced stabilization; however, this stabilization is partially offset by the increased exchange-repulsion energy (ΔΔE = +7.98 kcal/mol) and weakened orbital interaction energy (ΔΔE = +3.07 kcal/mol) in TS-2A. The ΔE_disp/ΔE_els ratios of TS-2A and TS-2B are approximately 0.45 and 0.43, respectively, confirming that electrostatic interactions dominate the intermolecular interaction in both systems. Nevertheless, the dispersion energy contributes substantially to the total stabilization (accounting for ~45% of the electrostatic stabilization), indicating that dispersion effects are non-negligible in the interaction mechanism (Fig. S10). The more negative ΔE_int of TS-2A is presumably attributed to the more favorable synergistic hydrogen-bonding configuration between the chiral phosphate anion and the substrate complex, which enhances the electrostatic attraction. Collectively, these energy decomposition features strongly corroborate the hypothesis that synergistic hydrogen-bonding interactions are the core driving force for the observed enantioselectivity. Besides, an interesting observation is that CP-4 serves as a branching point for the formation of products 4 and 3. Dissociation of CP-4a from the catalyst followed by aromatization yields 3, while product 4 can be obtained in two paths. The first path is that tautomerization of 3 into its keto form (3-keto) and ultimately undergo cyclization via TS-3A to form 4. The alternative path is that CP-4a readily undergoes cyclization via TS-3A to form CP-5a, which subsequently dehydrates to yield 4 (Cartesian coordinates are provided in the Source Data file).

Fig. 9.

Origin of stereoselectivity in the model reaction.

Table. 2.

sobEDAw Results for TS-2A and TS-2B^a

sobEDAw B3LYP-D3(BJ)/6-311 + G(2 d,p) w.CBb
	ΔEint	ΔEels	ΔExrep	ΔEorb	ΔEdisp	ΔEdisp/ΔEels
TS-2A	−121.72	−132.49	128.45	−58.54	−59.14	0.45
TS-2B	−120.93	−125.68	120.47	−61.61	−54.11	0.43
Difference	−0.79	−6.81	7.98	3.07	−5.03	-

^aEnergies are in kcal/mol. int: interaction energy, els: electrostatic, xrep: exchange-repulsion, orb: orbital, disp: dispersion.

^bw. CB means using complex basis functions in monomer calculations.

Discussion

In summary, we have successfully developed an efficient strategy for the enantioselective construction of polychiral helical compounds (polychiral helical compounds and oxa[4]helicenoids bearing inner-side axial chirality) via the CPA-catalyzed dynamic kinetic resolution process, which based on the screened out flexible metastable oxa[4]helicenoid-ol substrates by experiments and calculations, which exhibit enhanced chiral recognition and appropriate racemic barrier. The reaction enables the convenient synthesis of various substituted stable polychiral helical compounds and [4]helicenoids, achieving good yields and excellent stereoselectivity (up to 89% yield, up to 99:1 enantiomeric ratio, diastereomeric ratio >19:1, and E/Z ratio >19:1). DFT calculations reveal that the observed stereoselectivity is attributed to various hydrogen bonding interactions, including C-H···O and C-H···π hydrogen bonds. The synthetic utility of these compounds is demonstrated through large-scale reactions and subsequent transformations. Furthermore, both [4]helicenoid-quinone and its derivatives exhibit favorable optical properties, indicating their potential applications in various related fields.

Methods

Procedure for the synthesis of 4(5)

To a solution of 1 (0.1 mmol), (S)-C2 (0.01 mmol, 10 mol%) and 4 Å MS (40 mg) in PhCF₃ (4 mL), 2 (0.15 mmol) was added in one portion at 40 °C. After stirring for 72 h at 40 °C, the reaction mixture was concentrated and purified by silica gel column chromatography using petroleum ether/ethyl acetate to afford pure product 4(5).

Procedure for the synthesis of 7

To a solution of 1 (0.1 mmol) in a 1:1 mixture of CCl₄ and m-Xylene (2 mL) was added 6 (0.11 mmol) and (S)-A4 (0.005 mmol, 5 mol%). The reaction was stirred at room temperature for 12 h. Purification of the crude product by column chromatography on silica gel using petroleum ether/ethyl acetate afforded product 7.

Procedure for the synthesis of 9

To a 0 °C mixture of 1 (0.1 mmol), 8 (0.22 mmol) and (S)-C4 (0.005 mmol, 5 mol%) was added cooled DCE (2 mL). Then stirred at 0 °C for 18-30 h. The reaction mixture was concentrated and purified by silica gel column chromatography using petroleum ether/ethyl acetate to afford pure product 9.

Author contributions

X.L., C.L. and X.-Q.L. conceived and designed the experiments. C.L., X.-Q.L. and Z.Z. performed the experiments. C.Y. and Y.-B.S. performed the DFT calculations. C.L. and X.-Q.L. wrote the original manuscript. X.-Q.L. and X.L. reviewed and edited the manuscript.

Peer review

Peer review information

Nature Communications thanks Hanmin Huang, Junzhi Liu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Funding

The authors disclose support for the research of this work from the National Natural Science Foundation of China (grant Nos. 22371219 and 22193011) and the Haihe Laboratory of Sustainable Chemical Transformations.

Data availability

The authors declare that the data supporting the findings of this study are available within the article and its Supplementary Information files. Additional data are available from the author upon request. Crystallographic data for the structures reported in this article have been deposited at the Cambridge Crystallographic Data Center, under deposition numbers CCDC 2432700 (4a), 2464622 (5 h), 2315554 (7 m) and 2325667 (10a). Copies of the data can be obtained free of charge via https://www.ccdc.cam.ac.uk/structures/. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-73870-y.