Introduction

Helicene and helicene-like molecules are intriguing ortho-fused polycyclic chiral frameworks representing typical nonplanar systems1,2,3,4,5,6. Featuring a stereoscopic screw shape, helicene(-like) molecules display a wide variety of topologies and often modified optical and electronic properties7,8,9,10. Consequently, they hold significant promise for applications in various fields, such as liquid crystals11,12, supramolecular chemistry13, chiroptical switches or sensors14, and asymmetric transformations15. Enormous enthusiasm has thus been directed towards the development of helicenes (Fig. 1a, left)16,17,18,19,20,21,22,23,24,25,26,27.

Fig. 1: Research background and our strategy for synthesizing double S-shaped helicene-like molecules.
figure 1

a The research status of catalytic enantioselective synthesis of chemicals with helical chirality. b Current status of enantioselective synthesis of multiple helicenes and helicene-like molecules. The helicene units are indicated with semicircles in light blue and the fused rings are indicated with rectangles in pink. c Synthesis of axially chiral molecules by organocatalyzed [4 + 2] cycloadditions and our vision for the synthesis of multi-helical chiral compounds. d Strategies for the enantioselective synthesis of double S-shaped helicene-like molecule and key issues to be addressed. e Our design blueprint: enantioselective synthesis of double S-shaped helicene-like molecules via organocatalyzed [4 + 2] cycloadditions. E+, electrophilic reagent.

Full size image

As a complement to and a further development of the conventional chemistry of helicenes, the study of multiple helicenes has emerged as a thriving and attractive area of research in recent years28,29,30. Multiple helicenes often exhibit rich stereochemistry, multidimensional intermolecular interactions, and unique physicochemical properties31. However, the accumulation of multiple helicities also presents significant challenges in the stereochemically pure synthesis of multiple helicenes, and the relevant synthetic methods remain very limited (Fig. 1a, right). In 2014, Tanaka et al. pioneered the first enantioselective construction of S-shaped double helicenes32, and in subsequent studies, several more S-shaped double helicene(-like) molecules were presented with intriguing characteristics33,34,35,36. Regarding higher-multiplicity helicenes, which exhibit dramatically increased diastereomers and enantiomers, direct approaches to access enantiopure products are scarcely explored. To the best of our knowledge, only one report on the enantioselective construction of triple helicenes has been disclosed recently37 and the direct catalytic construction of helicene types with four or higher helicity has yet to be achieved (Fig. 1b). Furthermore, all reported enantioselective syntheses of multiple helicenes have involved transition-metal catalysts, and no enantioselective construction of multiple helicenes via organocatalysis has been reported to date. This may be attributed to several factors: (1) unlike building single-helical chiral molecules, controlling multiple-helical chirality is more challenging because the number of isomers increases exponentially with higher helicity (with up to 16 isomers present in quadruple helicenes), and thus the diastereo- and enantioselectivity of the reaction should be strictly controlled; (2) the highly distorted spatial structure and significant steric hindrance caused by the terminal ring complicate the control of configurational stability of multiple helicenes; (3) there are currently no suitable catalytic systems or synthetic methods available. Collectively, the lack of efficient synthetic strategies has greatly hindered the enantioselective synthesis of multiple helicene (-like) molecules via organocatalytic methods. Therefore, it is essential to explore novel organocatalytic synthetic approaches to obtain optically pure multiple helicenes with higher helicity.

As one of the most efficient methods for constructing ring systems, the organocatalytic [4 + 2] cycloaddition reaction has been widely employed to construct polycyclic systems containing various chiral elements including centers and axes over the past few decades38,39,40,41,42,43. Similarly, the method is particularly attractive for the synthesis of helical chiral molecules. In 2022, our group applied organocatalytic [4 + 2] cycloadditions to the enantioselective synthesis of helical chiral compounds for the first time44, resulting in a variety of highly enantioenriched [4]oxahelicenes. The results demonstrate the feasibility of organocatalytic [4 + 2] cycloadditions for the enantioselective synthesis of helicene(-like) molecules. To further extend the potential of organocatalytic [4 + 2] cycloadditions in the asymmetric synthesis of helical chiral compounds, we aim to apply this method to the synthesis of multiple helicenes with higher helicity. For this purpose, we focus on our earlier work on the atroposelective synthesis of axially chiral heterobiaryls via organocatalytic [4 + 2] cycloaddition reactions41. The simultaneous construction of two rings in a single step facilitates the synthesis of axial chiral aryl-naphthopyran skeletons with excellent yields and enantioselectivity (Fig. 1c, i). This study can inform the direct asymmetric synthesis of multi-helical chiral compounds via organocatalytic [4 + 2] cycloadditions (Fig. 1c, ii). We hypothesize that the addition of a large bulky group to ring A of the original axial chiral skeleton, combined with the extension of fused-ring systems in the terminal ring skeleton, could lead to the formation of an S-shaped double helicene. Moreover, the consecutive cyclizations may yield double S-shaped quadruple helicenes (Fig. 1d). Such scaffold combines both characters of quadruple helicenes and S-shaped helicenes and navigates into an unexplored research area. Notably, all helical units in this structure are 12-position monosubstituted [4]helicenes, whereas reports of conformationally stable 12-position monosubstituted [4]helicenes are rare to date44.

To achieve this quadruple helical chirality, factors such as configurational stability, diastereo- and enantioselectivity during the reaction necessitate the implementation of elaborate strategies. In addition, to attain a rigidly curved shape with sufficient stiffness and configurational stability, the steric crowding of substituents at specific positions is crucial24,45. To realize this objective, we have revisited our previous work on organocatalytic [4 + 2] cycloadditions and designed and synthesized substrate 1 (Fig. 1e). Building on this foundation, an organocatalytic [4 + 2] cycloaddition reaction system involving electrophilic reagents was developed. In this system, a bulky group was introduced at the alkyne end of substrate 1, providing the necessary steric hindrance for the formation of central helical chirality. Meanwhile, a suitable electrophilic reagent (E+) not only participates in the cyclization process in a catalytic helicoselective manner but also delicately introduces steric groups at the 12-position of the terminal helicene-like unit. This strategy can achieve the enantioselective synthesis of double S-shaped helicene-like compounds 2 in one step.

Herein, we report an organocatalysis method to achieve the enantioselective construction of double S-shaped helicene-like systems with quadruple helicity in a single step. It is our understanding that an organocatalytic strategy has never been employed for the enantioselective control of multiple helicenes. Furthermore, to the best of our knowledge, this work represents the enantioselective synthesis of the highest multiplicity of multiple helicenes ever reported. Simultaneously, the synthesized quadruple helices possess the fewest rings of all known configurationally stable quadruple helicenes46,47,48,49.

Results and discussion

Reaction optimization

Since 2a represents 4-fold helicene-like systems composed of a pair of S-shaped helicene-like units and one anthracene core, we speculated that there are ten possible stereoisomers, including four pairs of enantiomers and two meso-isomers. Precise control of isomerism was critically important in the following study. We initiated our investigation by selecting 1a for the primary optimization of the organocatalytic enantioselective [4 + 2] cycloaddition reaction. The optimization survey initially focused on the cinchona alkaloid organocatalysts50,51,52,53,54,55. Their backbones significantly influence the outcome of the reaction. As shown in Fig. 2, neither the monomer cinchona alkaloids with a squaramide (A), thiourea (B), or amide (C) moiety-linked bis(trifluoromethyl)benzene, nor dimeric cinchona alkaloid derivatives (D) or (DHQD)2PHAL (E) could catalyze the reaction (entries 1-5). In sharp contrast, cinchona alkaloid with a squaramide-bridged BINOL56 was found to yield the desired product ((P,P,P,P)-2a) with high diastereoselectivity (dr > 20:1) (entry 6). Further studies indicated that the (R)-BINOL was more efficient than its (S) counterpart with both higher yield (54% versus 37%) and better enantioselectivity (93% versus 70% e.e., entries 6, 7). Other factors, including solvent and temperature, were also tested, but none of them had a positive effect on the reaction. Further refinement of the other parameters led to the validation of the optimal conditions as follows: 20 mol% of catalyst F, 2.0 mL 2-MeTHF, −60 °C, and NBP (entry 8) which were employed in the subsequent work (see Supplementary Section IV for details).

Fig. 2: Optimization of reaction conditions.
figure 2

Reaction conditions: 1a (0.02 mmol, 1.0 equiv), catalyst (0.004 mmol, 20 mol%) in 2-MeTHF (1.0 mL) at corresponding temperature for 1.5 h and then N-bromo-phthalimide (NBP) (0.04 mmol, 2.0 equiv) at corresponding temperature for 2 h. Yield is that of the isolated product and enantiomeric excess (e.e.) value is determined by chiral HPLC analysis. Only one dominant diastereomer was formed under all conditions. n.d., not detected.

Full size image

Substrate scope and synthetic transformations

We then investigated several samples to exemplify the substrate scope of the reaction (Fig. 3a). The replacement of t-Bu with TMS (2b) on the 12-positions of the central helicene-like units resulted in better enantioselectivity (96%), suggesting that bulkier groups are more favorable for enhancing enantioselectivity. For the substituents at the 2-positions of the terminal helicene-like units, unsubstituted phenyl slightly decreased the enantioselectivity (2c, 88% e.e.), while substituents at the o-, m-, or p-positions of the phenyl (2d-f) elevated the e.e. values (94-95%). Alkyls including OMe (2f) and i-Pr (2g) were also tolerated, showing enantioselectivities of 93-96%. However, the linear phenylethynyl group (2 h) impeded the enantioselectivity (81% e.e.). Moreover, the elimination of substituents at this position (2j) dramatically decreased the e.e. value (40%) and yield (39%). We also attempted to expand the methodology for larger products, achieving 2k with one central [5]helicene-like unit, albeit in a low yield (27%) and enantioselectivity (20%) after condition optimization (see Supplementary Section IV for details). In addition, replacing the aminated quinine moiety in catalyst F with aminated quinidine (catalyst H, Fig. 2) yielded quadruple helical chiral products with the opposite configuration, as confirmed by the symmetric circular dichroism (CD) spectra exhibiting opposite Cotton effects of enantiomers (Supplementary Fig. 19d). The catalytic results of catalysts F, G, and H suggest that the chiral cinchona alkaloid component of the catalyst structure plays a pivotal role in regulating helical chirality. In addition, the palladium-catalyzed Sonogashira and Buchwald-Hartwig amination reactions allow the 2-position bromine of (P,P,P,P)-2a to be readily converted to alkynes and morpholine without affecting its stereoselectivity, which further validates the synthetic value of this method (Fig. 3b).

Fig. 3: Substrate scope and synthetic transformations.
figure 3

a Substrate scope of (P,P,P,P)-2 and (M,M,M,M)-2. Reaction conditions: 1 (0.02 mmol, 1.0 equiv), catalyst (0.004 mmol, 20 mol%) in 2-MeTHF (2.0 mL) at corresponding temperature for 1.5 h and then N-bromo-phthalimide (NBP) (0.04 mmol, 2.0 equiv) at −60 °C for 2 h. aThe reaction was conducted at -78 °C for 24 h. bThe reaction was conducted in cyclopentyl methyl ether (CPME) at -78 °C for 24 h. b Synthetic transformations of (P,P,P,P)−2a. t-Bu, tert-butyl. TMS, trimethylsilyl. i-Pr, isopropyl.

Full size image

Mechanistic studies

Given that the reaction involves two cyclization processes, preliminary studies were conducted through controlled experiments to determine whether the reaction proceeds in a concerted manner or via a stepwise mechanism. To facilitate these controlled experiments, an initial attempt was made to isolate the single-cyclized product (P,P)-A. (Fig. 4). It was found that (P,P)-A could be isolated in a lower yield in the presence of a half-equivalent of NBP (optimal condition, Fig. 4a). Subsequently, (P,P)-A was treated with optimal chiral catalyst F and its pseudo-enantiomeric isomer H, respectively, however, neither the desired product (P,P,P,P)-2a nor its isomers could be detected in the reaction system (Fig. 4b). Notably, in several of the aforementioned control experiments, the reaction system was chaotic, with no main product isolated, despite (P,P)-A was completely consumed. These results suggest that the reaction may not occur in a stepwise manner, and the single-cyclized product (P,P)-A may be a by-product rather than an intermediate in this context. This hypothesis is further supported by the differing yields of (P,P)-A and (P,P,P,P)-2a when half-equivalent amounts of NBP were added (Fig. 4a). In addition, control experiments also indicated that the single-cyclized product (P,P)-A was more readily converted to other impurities, resulting in a relatively low yield of the reaction. Based on these observations and our previously reported mechanisms involving the vinylidene ortho-quinone methide (VQM) intermediate57,58,59,60, we speculate that the successive generation of (R)-VQM-1 and (R,R)-VQM-2 from the substrate in the presence of the catalyst and NBP, followed by sequential [4 + 2] cycloaddition reactions to generate the final product via (R,P,P)-VQM-3 represents the most plausible conversion pathway (Fig. 4c).

Fig. 4: Preliminary mechanistic study.
figure 4

a Isolation of single-cyclized product. b Control experiments. c Proposed mechanistic pathway.

Full size image

To further understand the reaction process for the enantioselective synthesis of [4]helicene-like molecules, density functional theory (DFT) calculations were performed to investigate: (1) the origin of the enantioselectivity in the asymmetric addition step; (2) the origin of the diastereoselectivity in the cyclization step? The calculations indicate that catalyst F brings the substrate close to the NBP through multiple hydrogen bonding interactions and locks the relative orientation of the two substrates via hydrogen bonding interactions with the O-H of the binaphthol, the N-H of the squaramide and the alkylammonium ion in the catalyst. In the most favorable transition state, TS-2, the bifunctional catalyst stabilizes substrate 1 using the N-H group of the squaramide and the alkylammonium ion, while the binaphthol activates the NBP through hydrogen bonding. In the present catalytic system, the formation of (S)-VQM-1 is relatively delayed by 4.6 kcal·mol−1 compared to the formation of (R)-VQM-1, primarily due to the absence of hydrogen bonding interactions between the NBP and binaphthol of catalyst F in the formation of (S)-VQM-1 (Fig. 5a, TS-2-minor). The formation of (R,R)-VQM-2 is similar to that of (R)-VQM-1, attributed to the lack of hydrogen bonding between the NBP and the binaphthol of catalyst F, resulting in a relative delay of 4.6 kcal·mol−1 in the formation of (R,S)-VQM-2 compared to (R,R)-VQM-2 (Fig. 5a, TS-5 vs TS-5-minor). The independent gradient model based on Hirshfeld partition analysis (IGMH)61,62 of TS-2 and TS-5 further demonstrates a strong interaction between NBP and the binaphthol of catalyst F (Fig. 5a). The enantioselective generation of (R)-VQM-1 and (R,R)-VQM-2 described above represents the enantioselective determining step of the reaction. Once the chiral intermediate (R,R)-VQM-2 is formed, two consecutive intramolecular [4 + 2] cycloadditions between the VQM moiety and the tert-butyl alkynyl group take place to produce the final product 2. This process is thermodynamically controlled and exothermic at 118.0 kcal·mol−1 (see Supplementary Section VII for details, cartesian coordinates of all optimized structures are shown in Supplementary Data 1).

Fig. 5: DFT calculations and proposed mechanism.
figure 5

a DFT-computed Gibbs free energies of key transition states and the independent gradient model based on Hirshfeld partition (IGMH) analyses. b The catalytic cycle commences with substrate 1 and NBP being activated by the hydrogen bonding of catalyst F. It proceeds through stereocontrolled two successive electrophilically brominated and rapid proton transfer and concludes with the release of the chiral VQM intermediate and catalyst regeneration. The generated axially chiral VQM intermediate then undergoes two consecutive [4 + 2] cycloaddition reactions to give the final chiral product via axial to helical chirality transfer.

Full size image

Based on the control experiments and DFT calculation results, a plausible reaction pathway is proposed in Fig. 5b. First, the quinuclidine moiety of catalyst F acted as a Brønsted base to deprotonate the naphthalenol moiety of substrate 1. Subsequently, the alkynyl group was electrophilically brominated by NBP to yield the axial chiral VQM intermediate (R)-VQM-1. Afterward, the remaining succinimide anion participates in a rapid proton transfer completing the first catalytic cycle. After the formation of (R)-VQM-1 intermediate, it undergoes a second deprotonation-bromination process with catalyst F and NBP to synthesize the (R,R)-VQM-2 intermediate. Subsequently, the (R,R)-VQM-2 intermediate undergoes sequential [4 + 2] cyclization and yield (P,P,P,P)-2 with axial to helicene chirality transfer60. In this process, the chirality of the brominated VQM intermediate plays a crucial role in determining both the diastereoselectivity and enantioselectivity of chiral quadruple helicene molecules.

Configurational stability

To gain further insight into the configurational stability of quadruple helicene-like molecules, we have calculated the relative Gibbs free energies of (P,P,P,P)-2a and its possible diastereomers (Fig. 6a). The result demonstrates that the meso-isomer (M,M,P,P)-2a is the most thermodynamically stable isomer, whereas (P,P,P,P)-2a is the second most stable isomer (with a relative energy of 3.1 kcal·mol−1 with respect to (M,M,P,P)-2a). Based on this, we further investigated the interconversion pathways between the ten isomers of 2a identifying eight transition states with face-to-face oriented terminal groups of helical substructures (see Supplementary Section VIII for calculation details, cartesian coordinates of all optimized structures are shown in Supplementary Data 1). By the comparison of the energies of each transition state, the most plausible isomerization process is the path: (P,P,P,P)-2a → (P,M,P,P)-2a→(M,M,P,P)-2a (meso-isomer). Along this conversion, the free energy barrier of transition state TSA (ΔG‡ = 33.8 kcal·mol−1) is significantly higher compared to that of transition state TSB (ΔG‡ = 13.0 kcal·mol−1). Thus, the isomerization of the single helicene with tert-butyl terminal group can be considered the rate-determining step (RDS). The isomerization barrier from (P,P,P,P)-2a to (M,M,P,P)-2a is confirmed to be 33.8 kcal·mol−1 (Fig. 6b).

Fig. 6: Configurational stability.
figure 6

a Stereoisomers and calculated energies of 2a. b The most plausible diastereomerization pathway from (P,P,P,P)−2a to (M,M,P,P)−2a. Relative Gibbs free energy profile for interconversion among diastereoisomers of 2a calculated at the B3LYP/6-31 G(d)//M06−2x/6-311 + G(d,p) level69,70,71,72. The energy diagram of the process from (P,P,P,P)−2a to (P,P,M,M)−2a is equal to that of the process from (M,M,M,M)−2a to (M,M,P,P)-2a. Helical substructures that invert in transition states are highlighted in green. c Isomerization of (P,P,P,P)-2a to (M,M,P,P)-2a. d 1H-NMR traces for time course of isomerization of (P,P,P,P)-2a to (P,P,M,M)-2a in toluene-d8 at 100 °C. e Plot of the decreasing integration of (P,P,P,P)-2a in the 1H-NMR spectra in toluene-d8 upon heating at 100 °C, where [(P,P,P,P)-2a]0 and [(P,P,P,P)-2a]t are the ratios of the integration of (P,P,P,P)-2a to the sum of the integration of (P,P,P,P)-2a and (P,P,M,M,)-2a at the initial stage and at a certain time t (s) during the conversion, respectively. The integration of the signal peak at 8.31 ppm for (P,P,P,P)-2a was used as a reference. RDS, rate-determining step.

Full size image

Following this, we further verified the configurational stability of the products through thermal stability experiments. The kinetics of the isomerization of (P,P,P,P)-2a were studied experimentally by heating in toluene-d8 at 100 °C (Fig. 6c). The 1H-NMR spectra showed that (P,P,P,P)-2a can be gradually converted to (M,M,P,P)-2a (Fig. 6d). The isomerization between (P,P,P,P)-2a and (M,M,P,P)-2a can be described as a reversible first order kinetics (see Supplementary Section VIII for details). According to the plot of time (t) against ln([(P,P,P,P)-2a]t/[(P,P,P,P)-2a]0), the energy barrier for the isomerization of (P,P,P,P)-2a at 373 K is determined to be 31.5 kcal·mol-1, corresponding to a half-life of 434.7 years at room temperature, indicating excellent configurational stability (Fig. 6e). Notably, this value is much higher than the inversion energy barrier of 1-methyl substituted [4]helicene (ΔG‡ = 21.2 kcal·mol−1)45, which may be related to the increase in spatial site resistance of the substituent groups in the molecular structure as well as the increase in the overall repulsive force due to the condensation of multiple helical units.

Crystal structural analysis

Single crystals of (P,P,P,P)-2a and (P,P,M,M)-2a were obtained by slow evaporation of their solutions in dichloromethane/petroleum ether mixtures, respectively. X-ray crystallography determines the absolute configurations of (P,P,P,P)-2a and (P,P,M,M)-2a unambiguously. The single crystals from different viewings clearly show the highly twisted structure of the multi-ring system ((P,P,P,P)-2a colored in green and (P,P,M,M)-2a colored in yellow in Fig. 7). There are obvious differences in the structures of the two isomers (Fig. 7). The four helical units of (P,P,P,P)-2a have the same configuration, forming a twisted molecule with a propeller-shaped geometry, while (P,P,M,M)-2a has a symmetry plane, and the helicene-like units on the terminal and central sides are mirrored along the plane perpendicular to the anthracene core plane, showing a saddle-shaped structure (Fig. 7a). For (P,P,P,P)-2a, the two tert-butyl groups were respectively located on the upper and lower sides of the anthracene core plane and the two terminal naphthalene rings are orientated in opposite directions. The terminal Br atoms are 6.027 Å away from the anthracene plane and this distance for tert-butyl quaternary carbon atoms is 2.484 Å (Fig. 7a, left). However, the two tert-butyl groups and the two terminal naphthalene rings of (P,P,M,M)-2a have the same orientation, which are located on the upper and lower sides of the central plane of the anthracene core, respectively. The distance between the helical ends and the central anthracene core is similar to that of (P,P,P,P)-2a, which is 5.824 Å and 2.448 Å, respectively (Fig. 7a, right). This results in the torsion angles of each helical unit of the two isomers very close to each other, with the two terminal helical units having torsion angles of 47.289° ((P,P,P,P)-2a), 46.651° ((P,P,M,M)-2a), and the two central helical units having torsion angles of 52.559° ((P,P,P,P)-2a) and 50.302° ((P,P,M,M)-2a), respectively (Fig. 7b).

Fig. 7: Crystal structures of (P,P,P,P)-2a and (P,P,M,M)-2a.
figure 7

ORTEP drawings are shown showing 50% probability of thermal ellipsoids. Hydrogen atoms are omitted for clarity. a Different viewings of (P,P,P,P)-2a and (P,P,M,M)-2a in the crystals. b Torsion angles of (P,P,P,P)-2a and (P,P,M,M)-2a. c Twisting angles relative to the central anthracyclic core. d Packing structures of (P,P,P,P)-2a and (P,P,M,M)-2a.

Full size image

It is worth noting that the most notable structural difference between (P,P,P,P)-2a and (P,P,M,M)-2a lies in the degree of distortion of the anthracene core skeleton. The anthracyclic skeleton of (P,P,P,P)-2a experiences repulsive forces from the tert-butyl groups in opposite directions, causing both ring A and ring C to twist by different degrees, with the twist angles of 6.619° and 8.493° respectively, but the ring B is not twisted, which helps maintain a certain planarity of the anthracyclic core. However, (P,P,M,M)-2a is completely opposite. Because the tert-butyl group with large steric hindrance is located on the same side of the anthracycline plane, the substantial steric repulsion in the same direction results in a pronounced distortion of ring B with a twist angle of 15.054°, while rings A and C are not affected. This significant distortion reduces the planarity of the entire anthracene core, resulting in a saddle-shaped structure (Fig. 7c). In addition, the packing structures of the two isomers differ. (P,P,P,P)-2a is packed along the c-axis, forming a columnar packing structure. Due to its highly distorted structure, there is no intermolecular π-π interaction in this packing model (Fig. 7d, left). In (P,P,M,M)-2a, two naphthalene rings located at the terminal of the molecular have π-π interaction with the naphthalene rings at the terminal of the other two molecules, respectively, leading to the formation of a layered packing structure along the c-axis direction in (P,P,M,M)-2a (Fig. 7d, right, alse see Supplementary Fig. 18).

For further discussion of structural information of (P,P,P,P)-2a, including the nucleus-independent chemical shift (NICS)63, harmonic-oscillator model of aromaticity (HOMA)64, multicenter delocalization energies (MCI)65, electron density of delocalized bonds (EBBD)66, and the molecular electrostatic potential (MEP), please refer to Section XIII of the Supplementary Information.

Photophysical Properties

UV-Vis absorption spectroscopy, fluorescence spectroscopy, time-resolved fluorescence decay, and circular dichroism (CD) spectra were measured to explore the photophysical properties of compounds 2a-c, and the results showed that 2a-c exhibited similar photophysical properties (Fig. 8a–d, See Section X of the Supplementary Information for a detailed discussion). Notably, compounds 2a-c have strong UV absorption bands at longer wavelengths (>550 nm) with maximum absorption wavelengths of about 593-606 nm, which is rare in helicene(-like) structures. The intense absorption bands at longer wavelengths can be attributed to the HOMO → LUMO transition (f = 0.5975, λcalc = 639.06 nm), according to the results of time-dependent density functional theory (TD-DFT) calculations of (P,P,P,P)-2a at the B3LYP/6-31G(d) level (see Supplementary Section XII for details). Since the λmax of compounds 2a-c at long wavelengths were very close to the sodium D line (589 nm) and exhibit strong UV absorption at 589 nm (ε= 1.63 × 104 M-1 cm-1, 1.33 × 104 M-1 cm-1, 1.70 × 104 M-1 cm-1, respectively), the strong dispersion enhancement leads to the fact that all these compounds showing large specific rotations (\({\left[{{{\rm{\alpha }}}}\right]}_{{{{\rm{D}}}}}^{25}\) up to +36881 for (P,P,P,P)-2g). To the best of our knowledge, the specific rotations of such compounds are the highest reported to date67,68 (see Supplementary Section X for details).

Fig. 8: Photophysical properties of 2a-c.
figure 8

a UV/Vis spectra of 2a-c (2.0 × 10−5 M in DCM). b Fluorescence spectra of 2a-c (Excitation wavelength = 375 nm, 2.0 × 10−4 M in DCM). c Fluorescence lifetime decay curves of 2a-c. d CD spectra of 2a-c (2.0 × 10−5 M in DCM).

Full size image

In summary, we successfully developed a direct asymmetric strategy to prepare chiral double S-shaped helicene-like molecules with quadruple helicity in one step via an organocatalyzed enantioselective [4 + 2] cycloaddition reaction. This is the first report of enantioselective control of multiple helicenes through an organocatalytic strategy. Moreover, this work represents the enantioselective synthesis of the highest multiplicity of multiple helicenes ever reported. The configurations of these quadruple helicene-like molecules can be precisely controlled to be either (P,P,P,P) or (M,M,M,M) through catalyst regulation. Several samples were prepared to exemplify the universality of the synthetic method, and amenable physicochemical properties were disclosed through peripheral modifications. Experimental tests and DFT calculations have also been performed to illustrate the mechanistic essence of the enantioselective reaction and the configurational stability of the products. The single crystal structures of (P,P,P,P)-2a and (P,P,M,M)-2a revealed unique deformation features. Overall, this work explores new double S-shaped helicene-like molecules and may provide new insights into the research field of multiple helicenes with higher helicity. The investigation into the properties and potential applications of these structurally novel quadruple helical chiral compounds are ongoing.

Methods

General procedure for synthesis of (P,P,P,P)-2 via organocatalytic annulation

A solution of 1 (0.02 mmol) and catalyst-F (20 mol%) in 2-methyltetrahydrofuran (2.0 mL) was stirred at -60 °C or -78 °C for 1.5 h, then N-bromo-phthalimide (0.04 mmol) was added. After stirring at -60 °C for 2 h or at -78 °C for 24 h, the reaction mixture was concentrated under reduced pressure. The resulting residue was purified by silica gel flash column chromatography using petroleum ether/ethyl acetate eluent (100:1 to 50:1) to afford the compound (P,P,P,P)-2.