Heptagon-fusion as a molecular design strategy for distorted acenes

Abstract

Distorted acenes have attracted considerable attention owing to their 3-D structures and curvature-dependent properties, which are being used to develop novel organic functional materials. Herein, we report a series of diphenylene-fused acene derivatives incorporating doubly heptagon-fused structures. The introduced heptagons induced two stable conformations, twisted and saddle forms, to the acene cores. Kinetic analyses revealed that the isomerization barrier between the twisted and saddle forms increased with increasing acene length. The twisted isomers exhibited red-shifted absorption spectra compared to the corresponding saddle isomers. Quantum chemical calculations suggest that the narrower band gaps of the twisted isomers primarily originate from the smaller HOMO–LUMO gaps of the twisted acene cores compared with those of the saddle acene cores. These results not only provide a new design strategy for distorted acenes but also provide insights into their unique properties.

---

Introduction

Introducing distortion into planar π-conjugated systems can yield 3-D structures, such as bowl- and helical-shaped molecules, with curvature-dependent properties.^1–10 The development of distorted π-conjugated systems has provided crucial insights into the design and understanding of novel organic functional materials. In particular, distorted acene derivatives have been extensively explored experimentally by (1) introducing steric hindrance at the periphery,^11–18 (2) tethering with a cross-linking chain (Fig. 1a and b),^19,20 and (3) fusion with helically twisted PAH.^21–23 In the former strategy, firstly proposed by Pascal, the incorporation of bulky peripheral substituents onto acenes induces a crowded and twisted structure with pronounced end-to-end torsion angles (up to 184° for tetrabenzohexacene derivative 1).¹⁶ However, this strategy is unsuitable for studying chiroptical properties because of the low racemization barrier and fast racemization owing to rotatable substituents. In the second strategy, the incorporation of cross-linking chains onto anthracene cores along the diagonal direction produced twisted anthracenes 2-Cn (n = 3–6) with end-to-end torsion angles ranging from 23° to 38°, depending on the chain length.¹⁹ However, this method has not yet been applied to larger acenes.

Fig. 1 Structures of (a) highly substituted acene 1, (b) tethered acenes 2-Cn (n = 3–6), (c) fused decapyrrolyl acene 3, (d) twisted-Tet7, (e) saddle-Tet7, and (f) heptagon-fused acene derivatives. For the X-ray structures, the hydrogen and fluorine atoms were omitted for clarity. Distorted acene moieties are highlighted in orange, while the other moieties are shown in transparent form.⁴⁴

Although the scope of current design strategies for distorted acenes remains limited, twisted acenes exhibit chiroptical properties, red-shifted absorption spectra, and enhanced intersystem crossing rates as end-to-end torsion angles increase.^19,24 Theoretical study suggested that the twist of acenes induces diradical character²⁵ and reduces the HOMO–LUMO gap (ΔE_HL), while bending distortions exert only a minor influence on the ΔE_HL.²⁶

The introduction of non-hexagonal rings into π-systems leads to non-planar structures.^{3,5–7,27–43} Recently, Swager reported a fused decapyrrolyl anthracene 3, and Uno reported a fused octapyrrolyl naphthalene derivative (Fig. 1c).^31,42 Both compounds contained multiple heptagons and featured twisted acene moieties, with end-to-end torsion angles of 66° for 3 and 26° for the naphthalene derivative. These compounds display low first oxidation potentials (E^1/2_ox1 = −0.40 V for 3 and −0.52 V for the naphthalene derivative), and DFT calculations revealed that the HOMOs are delocalized primarily over the fused pyrrole moieties rather than the acene moieties. While the pyrrole-fusion strategy generates twisted acenes by introducing heptagons, the pronounced substituent effects make it difficult to investigate the relationship between the distortion of acenes and their physical properties.

Recently, we reported heptagon-fused tetracene Tet7 with a diphenylene-fused structure in two isolable conformations: C_2h-symmetric saddle-Tet7 and D₂-symmetric twisted-Tet7 (Fig. 1d and e).⁴³ The twisted isomer possesses a twisted tetracene core, whereas the saddle isomer exhibits a saddle-shaped tetracene core. These two conformers displayed curvature-dependent properties; for instance, the twisted form showed a more red-shifted absorption spectrum and greater photo-oxidation stability than the saddle conformer. While our previous study on Tet7 highlighted its promising performance in organic FET and solar cells, a systematic investigation of shorter acene cores is essential to fully elucidate how acene length governs core flexibility, isomerization barriers, and electronic structure. Such fundamental insights provide an indispensable benchmark and design platform for the precise control of next-generation distorted π-electron materials.

Therefore, in this study, we synthesized a series of heptagon-fused naphthalene and anthracene derivatives, Naph7 and Ant7, to generalize the heptagon-fusion strategy as a design approach for introducing in-plane strain into acenes (Fig. 1f). Their physical properties were characterized to examine the impact of acene distortion. It should be noted that a derivative of Naph7 bearing CF₃ substituents has been reported by Leowanawat et al.; its characterization has been limited to absorption and fluorescence properties.³⁴

Results and discussion

Synthesis

The synthesis of heptagon-fused tetracene Tet7 was previously reported,⁴³ which employed a reductive coupling reaction as a key step for constructing incorporated heptagons. The widely used oxidative fusion reaction yielded an undesirable fused product when applied to rubrene. Here, we applied the reductive fusion strategy to synthesize heptagon-fused naphthalene Naph7 and anthracene Ant7 (Scheme 1). Firstly, compound 4 was prepared from 1,4-diiodobenzene in three steps. The Diels–Alder reaction of 5/6 (for Naph7/Ant7) and in situ generated benzyne from 4 afforded the corresponding cycloadducts 7/8. Finally, the intramolecular Yamamoto coupling reaction of 7 and 8 with Ni(cod)₂ furnished Naph7 and Ant7 in 6% and 21% yields, respectively. An excess amount of Ni(cod)₂ was used for the reproducibility of the multi-step fusion process.† We consider the primary bottleneck leading to the low yields to be the reductive cyclization step. MS analysis revealed that the major byproducts were partially fused intermediates and dehalogenated species, although their isolation was precluded by the formation of multiple atropisomers with overlapping chromatographic profiles.

Scheme 1 Synthesis of heptagon-fused naphthalene Naph7 and heptagon-fused anthracene Ant7.

The ¹H NMR spectrum of Naph7 displayed five signals in the aromatic region, consistent with its structure (Fig. S11). On the other hand, Ant7 exhibited two sets of signals in an approximately 1 : 9 ratio, similarly to Tet7, indicating a mixture of two conformers. One minor and two major components were separated by chiral HPLC from the solution of Ant7. The separated major components showed mirror-image CD spectra. In addition, both the major and minor conformers of Ant7 exhibited 11 ¹H NMR signals in the aromatic region, indicating a two-fold symmetry. (Fig. S17 and S22) These results revealed that the major conformer is chiral twisted-Ant7 and the minor conformer is achiral saddle-Ant7. Interconversion between the two conformers of Ant7 occurred slowly at room temperature, as discussed later.

Structural analysis

Recently, our group has reported the single crystal structures of twisted-Tet7 and saddle-Tet7.⁴³ Single crystals of twisted-Naph7 and twisted-Ant7 suitable for X-ray crystallographic analysis were obtained by the vapor diffusion method at room temperature (Fig. 2a and b). The twisted structures of both twisted-Naph7 and twisted-Ant7 with pseudo-D₂ and pseudo-C₂ symmetry were found in the space groups I2/a and P2₁/n, respectively. For twisted-Naph7, The asymmetric unit of the crystal contains two independent but almost identical molecules, and one of which is shown in Fig. 2 (Table S2). The end-to-end torsion angles (ϕ_l) defined by four carbon atoms at the edge of the acene moiety were 32° for twisted-Naph7, 30° for twisted-Ant7, and 26° for twisted-Tet7, which were comparable with those of the tethered anthracene 2-Cn.¹⁷ The average torsion angles between the acene moiety and the diphenylene moiety () were 36° (twisted-Naph7), 41° (twisted-Ant7), and 45° (twisted-Tet7).

Fig. 2 (a–c) and (f) X-ray structures of (a) twisted-Naph7, (b) twisted-Ant7, (c) twisted-Tet7,⁴³ and (f) saddle-Tet7;⁴³ (d) and (e) optimized structures of (d) saddle-Naph7 and (e) saddle-Ant7 calculated at the B3LYP/6-31G(d) level. The average torsion angle φ, end-to-end dihedral angle ϕ_l, and angles θ_acd θ_deb are shown. Hydrogen atoms are omitted for clarity. In (a, b, c and f), thermal ellipsoids are shown at the 50% probability. One of the two almost identical molecules in the asymmetric unit is shown for twisted-Naph7 and saddle-Tet7.

The structures of the twisted and saddle isomers of the fused acenes were further investigated by DFT calculations. The energy-minimized geometries of all twisted isomers and saddle-Tet7 closely matched with the corresponding X-ray structures (Table S2, Fig. S30 and S31). The calculated saddle-shaped conformers of Naph7 and Ant7 are shown in Fig. 2d and e, exhibiting pseudo-C_2h (saddle-Naph7) and pseudo-C_s (saddle-Ant7) symmetry. The calculated end-to-end angles (ϕ_l) of all saddle-conformers were 0°, and the dihedral angles between the acene core and diphenylene moiety () were compatible with those of the corresponding twisted conformers (35° for saddle-Naph7, 38° for saddle-Ant7, and 42° for saddle-Tet7 (X-ray: 43°)). The acene moiety of the saddle-isomers possesses a doubly bended structure with bend angles (θ_acd, θ_deb) of (164°, 164°) for saddle-Naph7, (166°, 159°) for saddle-Ant7, and (161°, 161°) for saddle-Tet7 (X-ray: 161°, 161°).

Kinetics of the isomerization pathway between the conformers

The thermodynamics and kinetics of the isomerization between the saddle and twisted conformers of Naph7 and Ant7 were investigated by ¹H NMR spectroscopy. While the saddle and twisted conformers of Ant7 could be separated by HPLC, HPLC analysis on Naph7 showed a single peak, and the ¹H NMR spectra of Naph7 remained unchanged down to −80 °C in CD₂Cl₂ (Fig. S32). These results indicate the presence of only one conformer in solutions or very rapid interconversion between the conformers of Naph7. For Ant7, interconversion between the conformers was observed without detectable intermediates or byproducts (Fig. 3a, Fig. S33 and S34). Based on the equilibrium ratio of Ant7 determined by the ¹H NMR spectra (saddle/twisted = 0.131/1 at 25 °C, 0.125/1 at 20 °C, 0.123/1 at 15 °C), the isomerization enthalpy (ΔH) and entropy (ΔS) were estimated to be 4.5 kJ mol^–1 and −1.8 J mol^–1 K^–1, respectively (Fig. 3b). The rate constants for the saddle-to-twisted (k_{twist→saddle}) and the twisted-to-saddle (k_{saddle→twist}) processes were obtained by fitting the data to a reversible first-order interconversion model (See the SI for details). Eyring plots using these rate constants yielded activation enthalpy (ΔH^‡) and entropy (ΔS^‡) of 63.7 kJ mol^–1 and −0.114 kJ mol^–1 K^–1 for the saddle-to-twisted isomerization, and 68.2 kJ mol^–1 and −0.119 kJ mol^–1 K^–1 for the twisted-to-saddle isomerization, respectively (Fig. S36). Both ΔH^‡ were smaller than those of Tet7 (128.4 kJ mol^–1 for twisted-to-saddle isomerization, 143.8 kJ mol^–1 for saddle-to-twisted isomerization),⁴³ which was attributed to reduced repulsion in the transition state.

Fig. 3 (a) ¹H NMR spectra of twisted-Ant7 and saddle-Ant7 during the progress of saddle-to-twisted isomerization in dichloromethane-d₂ at 25 °C after 0 h (start), 6 h, and 21 h (equilibrium), (b) thermodynamics parameters of Naph7, Ant7, and Tet7,⁴³ and (c) isomerization pathways between the twisted and saddle isomers for Naph7, Ant7, and Tet7⁴³ calculated at the B3LYP/6-311G(2d,p)//B3LYP/6-31G(d) level.

The observed isomerization dynamics between the conformers were further investigated using quantum chemical calculations at the B3LYP/6-311G(2d,p)//B3LYP/6-31G(d) level (Fig. 3c). In contrast to Tet7,⁴³ the calculations predicted that the twisted isomers of Naph7 and Ant7 are the global minima. The saddle conformations were calculated to be 16.3 kJ mol^–1 (Naph7) and 2.4 kJ mol^–1 (Ant7) higher than the twisted form at 25 °C. For Naph7, the calculated equilibrium ratio (0.999/0.001 = twist/saddle at 25 °C) suggested that Naph7 predominantly existed as a twisted isomer in solution. The most plausible isomerization pathways are shown in Fig. 3c. The isomerization of Naph7 proceeds via a transition state, TS1, associated with the inversion of the diphenylene moiety, with an activation barrier of 80.3 kJ mol^–1. In the case of Ant7, the isomerization occurs through two transition states, TS1 (89.0 kJ mol^–1) and TS2 (100.0 kJ mol^–1). The structure and energy of TS1 are similar between Naph7 and Ant7. In the case of Tet7, isomerization proceeds via TS2 and TS3, and the TS2 of Ant7 and Tet7 were also structurally and energetically similar. The highest isomerization barrier of Tet7 was TS3 at 146.1 kJ mol^–1. These theoretical activation barriers are in good agreement with the experimental results (103.6 kJ mol^–1 at 25 °C for Ant7, 136.2 kJ mol^–1 at 25 °C for Tet7⁴³).

Photophysical properties

To clarify the impact of molecular conformation on the electronic characteristics, we measured the absorption and fluorescence spectra of Naph7, saddle-Ant7, and twisted-Ant7 in toluene (Fig. 4a and b). The spectra of saddle-Tet7 and twisted-Tet7 in toluene were reported in our previous work.⁴³ Naph7 showed the lowest energy absorption maximum at 415 nm (ε = 1.31 × 10⁴ cm^–1 M^–1) and 0–0 emission band at 435 nm (Φ_f = 0.20). For Ant7 and Tet7, twisted isomers exhibited red-shifted absorption spectra relative to the corresponding saddle isomers; 508 nm (ε = 1.07 × 10⁴ cm^–1 M^–1) for twisted-Ant7, 483 nm (ε = 1.50 × 10⁴ cm^–1 M^–1) for saddle-Ant7, 651 nm (ε = 1.13 × 10⁴ cm^–1 M^–1) for twisted-Tet7, and 591 nm (ε = 1.68 × 104 cm^–1 M^–1) for saddle-Tet7. The emission spectra of saddle-Ant7 and twisted-Ant7 are quite similar (λ_em = 539 nm for saddle-Ant7 and 541 nm for twisted-Ant7). The excitation spectra matched their absorption spectra, confirming that the similar emission arises from distinct conformations (Fig. S39 and S40). In the case of Tet7, the twisted isomer exhibited red-shifted fluorescence spectra compared to the saddle isomer (λ_em = 611 nm for saddle-Tet7, and λ_em = 663 nm for twisted-Tet7, Fig. 4a). The saddle and twisted isomers possess compatible dihedral angles between the acene core and the fused diphenylene moiety. Therefore, the observed trend in the optical band gaps is primarily attributed to structural distortion of the acene core, rather than to variations in the π-extension of the diphenylene units. The fluorescence quantum yield of saddle-Ant7 (Φ_f = 0.49) is smaller than that of twisted-Ant7 (Φ_f = 0.62) due to the twice faster non-radiative decay of saddle-Ant7 (k_nr = 0.11 ns^–1) compared to that of twisted-Ant7 (k_nr = 0.05 ns^–1). The small quantum yield of Naph7 (Φ_f = 0.20) is also attributed to the high k_nr (0.82 ns^–1). The fluorescence and excitation spectra of Naph7 are independent of the excitation and monitoring wavelengths, supporting the conclusion that Naph7 predominantly exists as a twisted isomer in solution (Fig. S38). The enantiomers of twisted-Ant7 and twisted-Tet7 showed mirror-image circular dichroism (CD) spectra (Δε = 36 M⁻¹ cm⁻¹, |g_CD| = 3.7 × 10⁻³ at 508 nm for Ant7, and Δε = 56 M⁻¹ cm⁻¹, |g_CD| = 4.9 × 10⁻³ at 651 nm for Tet7⁴³), respectively (Fig. 4a).

Fig. 4 (a) Absorption, fluorescence, and CD spectra, (b) photophysical parameters of Naph7, saddle-Ant7, twisted-Ant7, saddle-Tet7,⁴³ and twisted-Tet7⁴³ in toluene, and (c) molecular orbital diagram of saddle-Naph7, twisted-Naph7, saddle-Ant7, twisted-Ant7, saddle-Tet7,⁴³ and twisted-Tet7⁴³ calculated at the (TD-)B3LYP/6-311G(2d,p)//B3LYP/6-31G(d) level.

The electronic transition characters were disclosed by TD-DFT calculations at the TD-B3LYP/6-311G(2d,p)//B3LYP/6-31G(d) level (Fig. 4c). The observed lowest-energy absorption bands were assigned to the HOMO–LUMO transitions. The calculated frontier orbitals were delocalized over the entire π-electronic system, despite the highly curved structures. The calculated excitation wavelengths of the HOMO–LUMO transitions for twisted isomers (λ = 415 nm for twisted-Naph7; λ = 527 nm for twisted-Ant7; λ = 705 nm for twisted-Tet7⁴³) were longer than those of the corresponding saddle isomers (λ = 404 nm for saddle-Naph7; λ = 507 nm for saddle-Ant7; λ = 642 nm for saddle-Tet7⁴³), which is consistent with the experimental results. We also calculated the energy-minimized geometry at the S₁ state (Fig. S46). The calculation results showed that the difference in emission wavelength between the saddle and twisted isomers increases with increasing acene length (Naph7 < Ant7 < Tet7).

Distortion effect on the optical band gap of acenes

The optical band gaps of Naph7/Ant7/Tet7 are considered as the band gap of the pristine acene being reduced by the structural distortion on the acene core (ΔE_dist) and the π-extension of the diphenylene units (ΔE_π-ext); i.e. ΔE_HL(saddle-Naph7) = ΔE_HL(naphthalene) + ΔE_dist + ΔE_π-ext (Fig. 5c). To quantify ΔE_dist and ΔE_π-ext for Naph7/Ant7/Tet7, we calculated the corresponding unsubstituted acene cores extracted from their energy-minimized structures (denoted as Naph7-H/Ant7-H/Tet7-H, see Fig. 5a and b and Fig. S44). The difference in optical band gaps between Naph7-H/Ant7-H/Tet7-H and the corresponding pristine acenes was interpreted as the effect of distortion at the acene cores (ΔE_dist). Similarly, the difference in optical band gaps between Naph7-H/Ant7-H/Tet7-H and Naph7/Ant7/Tet7 was interpreted to be the effect of π-extension (ΔE_π-ext). Accordingly, ΔE_dist and ΔE_π-ext for Naph7/Ant7/Tet7 were calculated and summarized in Fig. 5c and Table 1. The results revealed that both π-extension and structural distortion contribute comparably to ΔE_HL. Furthermore, the |ΔE_dist| values of twisted isomers were larger than those of the corresponding saddle isomers. Conversely, the |ΔE_π-ext| values of twisted isomers were smaller than those of the corresponding saddle isomers. Therefore, the observed narrower optical band gaps of twisted isomers were ascribed to the larger distortion effect on the acene cores. The difference in ΔE_dist between the twisted and saddle conformations is consistent with the trend observed in a previous theoretical study, showing that the twisting acenes narrows the ΔE_HL, while bending distortions have little effect on ΔE_HL.²⁶

Fig. 5 (a) Structure (top and side views) of saddle-Naph-H, (b) molecular orbital diagram of naphthalene, twisted-Naph7-H, and saddle-Naph7-H calculated at the B3LYP/6-311G(2d,p) level, (c) scheme of deconvolution analysis of ΔΔE_HL into distortion and π-extension effects (ΔE_dist and ΔE_π-ext) for saddle-Naph7 as an example and summary of the deconvolution.

Table 1 Obtained distortion and π-extension effect on the optical band gaps of Naph7-H/Ant7-H/Tet7-H

Cmpd.	Conformation	ΔEHL	ΔΔEHL	ΔEdist	ΔEπ-ext
Unit: eV. ΔΔEHL = ΔEdist + ΔEπ-ext
Naph7	Twisted	3.42	−1.34	−0.40	−0.94
	Saddle	3.48	−1.28	−0.33	−0.95
Ant7	Twisted	2.75	−0.80	−0.32	−0.48
	Saddle	2.77	−0.77	−0.26	−0.51
Tet7	Twisted	2.22	−0.63	−0.34	−0.29
	Saddle	2.11	−0.52	−0.22	−0.30

---

Conclusions

In summary, this work demonstrated that the incorporation of heptagons generally induces twisted and saddle-shaped structures into a series of acenes by in-plane strain. Structural and kinetic analyses revealed distinct conformational preferences, where the activation barriers for twisted-to-saddle isomerization increased with acene length (103.6 kJ mol^–1 for Ant7 and 136.2 kJ mol^–1 for Tet7 at 25 °C). Optical measurements and (TD-)DFT calculations showed that twisted isomers exhibited narrower HOMO–LUMO gaps compared to the saddle isomers (e.g., 2.75 eV for twisted-Ant7 and 2.77 eV for saddle-Ant7). Further investigations separately examined the effects of structural distortion and π-extension on the HOMO–LUMO gaps, indicating that the narrower band gaps of the twisted isomer can be ascribed to the induced twist within the acene cores. While the photophysical and chiroptical properties of Naph7 and Ant7 (e.g., Φ_f = 0.20 for Naph7 and |g_CD| = 3.7 × 10⁻³ for twisted-Ant7) are moderate compared to those of larger acenes, the systematic correlation between acene length and isomerization dynamics unveiled here offers crucial guidelines for material optimization. A current limitation of this methodology is the relatively low synthetic yield, which stems from the complex multi-step cyclization process. Future work will focus on optimizing reaction conditions to enhance the scalability and applications of these materials in high-performance organic electronics.

Author contributions

M. H. and D. S. conceived the idea. M. H. performed all experimental studies and wrote the original draft. All authors discussed the results and interpretations. The manuscript was written through the contributions of all authors.

Conflicts of interest

There are no conflicts to declare.

Data availability

Supplementary information (SI): experimental details, spectroscopic data, crystallographic data, electrochemical analysis, and theoretical calculations (PDF). See DOI: https://doi.org/10.1039/d6tc01255h.

CCDC 2534033–2534035 contain the supplementary crystallographic data for this paper.^45a–c

Acknowledgements

This work was supported by a Grant-in-Aid for Transformative Research Areas (A) “Condensed Conjugation” (JSPS KAKENHI Grant Numbers JP20H05866 and JP20H05868) from MEXT, Japan, and JST FOREST program (Grant Number JPMJFR2427). This work was also supported by a Grant-in-Aid for Scientific Research (B) (JSPS KAKENHI Grant Number 23K26641), a Grant-in-Aid for JSPS Fellows (JSPS KAKENHI Grant Number JP23KJ1377) from JSPS, Japan. The computational resource was provided by the SuperComputer System, Institute for Chemical Research, Kyoto University.

Notes and references

1. M. Ball, Y. Zhong, Y. Wu, C. Schenck, F. Ng, M. Steigerwald, S. Xiao and C. Nuckolls, Acc. Chem. Res., 2015, 48, 267–276.
2. Y. Segawa, H. Ito and K. Itami, Nat. Rev. Mater., 2016, 1, 15002.
3. M. Saito, H. Shinokubo and H. Sakurai, Mater. Chem. Front., 2018, 2, 635–661.
4. D. Wu, W. Cheng, X. Ban and J. Xia, Asian J. Org. Chem., 2018, 7, 2161–2181.
5. S. H. Pun and Q. Miao, Acc. Chem. Res., 2018, 51, 1630–1642.
6. I. R. Márquez, S. Castro-Fernández, A. Millán and A. G. Campaña, Chem. Commun., 2018, 54, 6705–6718.
7. M. A. Majewski and M. Stępień, Angew. Chem., Int. Ed., 2019, 58, 86–116.
8. Y. Liang, M. Tang and Z. Liu, Chem. Lett., 2020, 49, 1329–1336.
9. Z. Hassan, Adv. Funct. Mater., 2024, 34, 2311828.
10. R. Kumar, H. Aggarwal and A. Srivastava, Chem. – Eur. J., 2020, 26, 10653–10675.
11. R. A. Pascal Jr, Chem. Rev., 2006, 106, 4809–4819.
12. M. A. Dobrowolski, M. K. Cyrański and Z. Wróbel, Phys. Chem. Chem. Phys., 2016, 18, 11813–11820.
13. Y. Yano, H. Ito, Y. Segawa and K. Itami, Synlett, 2016, 2081–2084.
14. T. Fujikawa, Y. Segawa and K. Itami, J. Am. Chem. Soc., 2016, 138, 3587–3595.
15. W. Fan, T. Winands, N. L. Doltsinis, Y. Li and Z. Wang, Angew. Chem. Int. Ed., 2017, 56, 15373–15377.
16. R. G. Clevenger, B. Kumar, E. M. Menuey and K. V. Kilway, Chem. – Eur. J., 2018, 24, 3113–3116.
17. W. Matsuoka, K. P. Kawahara, H. Ito, D. Sarlah and K. Itami, J. Am. Chem. Soc., 2023, 145, 658–666.
18. T. Matsunaga, K. Oki, S. Mori, T. Nishiuchi, M. Higashi, T. Kubo, T. Okujima, H. Uno and M. Takase, Org. Lett., 2025, 27, 11005–11010.
19. A. Bedi, L. J. W. Shimon and O. Gidron, J. Am. Chem. Soc., 2018, 140, 8086–8090.
20. I. Shioukhi, A. Agrawal, Y. Deree, B. Bogoslavsky and O. Gidron, Org. Lett., 2025, 27, 10642–10646.
21. J. A. Weber, E. L. Clennan and N. Arulsamy, European J. Org. Chem., 2022, e202101533.
22. A. Swain, K. Radacki, H. Braunschweig and P. Ravat, Chem. Sci., 2024, 15, 11737–11747.
23. A. R. Agrawal, I. Shioukhi, Y. Deree, B. Bogoslavsky, O. Shalev, R. Hoffman and O. Gidron, Angew. Chem. Int. Ed., 2025, 64, e202510423.
24. P. Malakar, V. Borin, A. Bedi, I. Schapiro, O. Gidron and S. Ruhman, Phys. Chem. Chem. Phys., 2022, 24, 2357–2362.
25. J. E. Norton and K. N. Houk, J. Am. Chem. Soc., 2005, 127, 4162–4163.
26. A. M. Armon, A. Bedi, V. Borin, I. Schapiro and O. Gidron, Eur. J. Org. Chem., 2021, 5424–5429.
27. W. E. Barth and R. G. Lawton, J. Am. Chem. Soc., 1966, 88, 380–381.
28. K. Yamamoto, T. Harada, M. Nakazaki, T. Naka, Y. Kai, S. Harada and N. Kasai, J. Am. Chem. Soc., 1983, 105, 7171–7172.
29. K. Kawasumi, Q. Zhang, Y. Segawa, L. T. Scott and K. Itami, Nat. Chem., 2013, 5, 739–744.
30. N. Fukui, T. Kim, D. Kim and A. Osuka, J. Am. Chem. Soc., 2017, 139, 9075–9088.
31. K. Oki, M. Takase, S. Mori, A. Shiotari, Y. Sugimoto, K. Ohara, T. Okujima and H. Uno, J. Am. Chem. Soc., 2018, 140, 10430–10434.
32. J. M. Farrell, V. Grande, D. Schmidt and F. Würthner, Angew. Chem. Int. Ed., 2019, 58, 16504–16507.
33. S. H. Pun, Y. Wang, M. Chu, C. K. Chan, Y. Li, Z. Liu and Q. Miao, J. Am. Chem. Soc., 2019, 141, 9680–9686.
34. K. Kantarod, T. Worakul, D. Soorukram, C. Kuhakarn, V. Reutrakul, P. Surawatanawong, W. Wattanathana and P. Leowanawat, Org. Chem. Front., 2021, 8, 522–530.
35. M. Hisada, D. Shimizu and K. Matsuda, Org. Lett., 2022, 24, 3707–3711.
36. M. Hisada, D. Shimizu and K. Matsuda, J. Org. Chem., 2022, 87, 9034–9043.
37. K. M. Cheung, Y. Xiong, S. H. Pun, X. Zhuo, Q. Gong, X. Zeng, S. Su and Q. Miao, Chem, 2023, 9, 2855–2868.
38. M. Krzeszewski, Ł. Dobrzycki, A. L. Sobolewski, M. K. Cyrański and D. T. Gryko, Saddle-Shaped Aza-Nanographene with Multiple Odd-Membered Rings, Chem. Sci., 2023, 14, 2353–2360.
39. M. Hisada, D. Shimizu and K. Matsuda, Chem. Lett., 2024, 53, upae021.
40. S. M. Elbert, O. T. A. Paine, T. Kirschbaum, M. P. Schuldt, L. Weber, F. Rominger and M. Mastalerz, J. Am. Chem. Soc., 2024, 146, 27324–27334.
41. B. Borrisov, G. M. Beneventi, Y. Fu, Z.-L. Qiu, H. Komber, Q.-S. Deng, P. M. Greißel, A. Cadranel, D. M. Guldi, J. Ma and X. Feng, J. Am. Chem. Soc., 2024, 146, 27335–27344.
42. A. A. Suleymanov, Q. He, P. Müller and T. M. Swager, Org. Lett., 2024, 26, 5227–5231.
43. M. Hisada, K. Bulgarevich, Y. Tsutsui, K. Miyata, D. Shimizu, S. Seki, K. Takimiya and K. Matsuda, J. Am. Chem. Soc., 2026, 148, 6716–6726.
44. E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch, D. M. Greenblatt, E. C. Meng and T. E. Ferrin, J. Comput. Chem., 2004, 25, 1605–1612.
45. (a) CCDC 2534033: Experimental Crystal Structure Determination, 2026 DOI:10.5517/ccdc.csd.cc2r1w0q; (b) CCDC 2534034: Experimental Crystal Structure Determination, 2026 DOI:10.5517/ccdc.csd.cc2r1w1r; (c) CCDC 2534035: Experimental Crystal Structure Determination, 2026 DOI:10.5517/ccdc.csd.cc2r1w2s.

---

Footnote

† We confirmed that precursors 7 and 8 were completely consumed after the reductive reaction.

---