C–C bond-forming reactions of 2-isocyanobiphenyl·BX₃ adducts: spontaneous construction of polycyclic heteroaromatics

Abstract

The syntheses and reactivities of 2-isocyanobiphenyl·BX₃ adducts (X = I, Br, Cl) are reported. These adducts undergo unexpected C–H bond-functionalizing cyclization upon heating, yielding phenanthridinium-6-trihaloborate zwitterions. Where X = Cl, an unexpected helical polycyclic boronium salt is formed via a competing pathway. Mechanistic details are probed experimentally and computationally.

---

The importance of C–C bond formation to synthetic chemistry has lent prominence to organoboranes and their reactions. Accordingly, new synthetic possibilities arise from new pathways to boron-substituted products. Reactions that increase molecular complexity through concomitant C–C and C–B bond formations, such as borylative cyclizations, hold particular appeal.¹

Isonitrile-BX₃ adducts (X = alkyl, silyl, H, or halide) generally undergo facile insertion of isonitrile into B–X bonds to form (boryl)iminomethanes,² which often dimerize via dual imine-borane coordination. Indeed, this was observed for isonitrile-BX₃ adducts (where X = Br or Cl) by Meller and Batka (Fig. 1a).^2h Nevertheless, alternative reactions of isonitriles and boranes (via (boryl)iminomethanes) have recently been exploited in synthesis. For example, Figueroa and co-workers demonstrated reactivities of sterically encumbered (boryl)iminomethanes that included “frustrated Lewis pair” (FLP) small molecule activations (Fig. 1b).³ Erker and co-workers utilized (boryl)iminomethane intermediates to construct heterocycles such as 1,3-dihydro-1,3-azaborinines via bora-Nazarov reactions,⁴ and to construct various heteroboroles via reactions with carbon–heteroatom double bonds (Fig. 1c).⁵ Reactions of isonitriles with B–B single⁶ and multiple⁷ bonds, bis-diazidoboranes,⁸ metal boryls,⁹ metal borylenes,^9b,10 and N-heterocyclic boryl anions¹¹ have also revealed many unique reaction pathways to complex products.¹²

Fig. 1 Examples of reactivities of boranes and isonitriles via presumed or isolated (boryl)iminomethane intermediates.

Herein, we describe the synthesis and reactivity of 2-isocyanobiphenyl·BX₃ (X = I, Br, Cl) adducts. Instead of anticipated insertion and dimerization, these compounds underwent C–C bond-forming borylative cyclization upon heating, via functionalization of a biphenyl C–H bond, to form phenanthridinium-6-trihaloborate zwitterions. For X = I or Br, the reactions proceeded cleanly without apparent by-products. On the other hand, where X = Cl, the phenanthridinium-6-trichloroborate product was accompanied by an unexpected helical 6,6′-bisphenanthridine-chelated dichloroboronium salt. The latter reaction involves three C–C bond formations, one of which is intermolecular. All isonitrile–borane adducts and cyclization products were isolated, characterized, and studied by single crystal X-ray crystallography. NMR, EPR, and density functional theory (DFT) studies were undertaken to examine reaction mechanisms. These surprising reactivities offer new prospects for C–C bond construction enroute to polycyclic aromatic scaffolds.

Reactions of 2-isocyanobiphenyl and boron trihalides (BI₃, BBr₃, or BCl₃) in dichloromethane at room temperature furnished analytically pure 2-isocyanobiphenyl·BX₃ adducts (1: X = I, 40% yield; 2: X = Br, 32% yield; 3: X = Cl, 36% yield) after recrystallization from n-hexane (Scheme 1). Adducts 1–3 are stable at room temperature under N₂, and each were characterized by ¹H, ¹³C, and ¹¹B NMR spectroscopies, HRMS, and single crystal X-ray crystallography (e.g.Fig. 2a). Nearly identical packing arrangements are adopted by 1, 2, and 3 in the solid state, and distorted linear B–C–N and C–N–C bond geometries are observed for each. B–C bond lengths ascend from 1 (1.562(3) Å), to 2 (1.591(3) Å), to 3 (1.613(2) Å), in line with the relative Lewis acidities of BI₃, BBr₃, and BCl₃, respectively. Isonitrile N–C bonds are 1.145(3) Å (1), 1.141(3) Å (2), and 1.138(2) Å (3), conforming with reported bond lengths between isonitriles and strong boron Lewis acids.¹³

Scheme 1 Synthesis of 2-biphenylisocyanide·BX₃ adducts 1–3 and cyclized products 4–6. (i) BX₃, CH₂Cl₂, r.t. or −78 °C – r.t., 2–18 h, (ii) C₂H₂Cl₄ or C₂D₂Cl₄, 130 °C, 24–48 h.

Fig. 2 Solid-state structures of (a) 2, (b) 5a, (c) the cation of 6b, and (d) 6b (side-on view). C: black, N: blue, B: yellow-green, Cl: green, Br: orange, H: grey. Selected H atoms omitted for clarity.

Transformations of 1 and 2 at elevated temperatures were studied by ¹H and ¹¹B NMR spectroscopy in C₂D₂Cl₄. At 130 °C, each cleanly converted to a single product without observable intermediates (see ESI, Fig. S1–S4†). These products, 4a and 5a, crystallized from solution and were isolated in 75% and 70% yields, respectively. X-ray crystallography revealed their structures as phenanthridinium-6-trihaloborate zwitterions, indicating C–H functionalization of 2-isocyanobiphenyl 2′-positions (Scheme 1). Interestingly, no direct evidence for isonitrile insertion into B–X bonds was observed in either of these reactions. Analytically pure 4a or 5a could also be isolated in 52% (4a) or 67% (5a) yields from reactions of 2-isocyanobiphenyl and BI₃ or BBr₃, without rigorous isolation of intermediates 1 or 2.

Contrary to 1 and 2, heating compound 3 at 130 °C in C₂D₂Cl₄ leads to complex ¹H and ¹¹B NMR spectra as 3 is consumed over 48 h. We postulated that a phenanthridinium 6-trichloroborate zwitterion (6a) was one of the reaction products, based on similarly shifted ¹H NMR peaks to 4a and 5a, and a singlet ¹¹B NMR peak observed at δ = 4.9 ppm. Another product was identified by X-ray analysis of deep red crystals recovered from the reaction mixture as the polycyclic 6,6′-biphenanthridine-chelated dichloroboronium tetrachloroborate salt 6b (Scheme 1 and Fig. 2c, d). This product could be isolated in 16% yield and exhibits ¹¹B NMR singlets at δ = 9.1 and 6.8 ppm attributable to boronium and tetrachloroborate boron atoms. Pure 6b could be isolated from 2-isocyanobiphenyl and BCl₃ in 13% yield without rigorous isolation of 3. In separate experiments, the zwitterionic product 6a could be isolated from this reaction using a purification procedure that involves hydrolysis of 6b, followed by extraction of 6a and recrystallization from chloroform/n-hexane. Some tolerance of 6a towards water is indicated by this purification procedure. Both 6a and 6b were fully characterized by ¹H, ¹¹B, and ¹³C NMR spectroscopies, X-ray crystallography, and HRMS. Ambient stabilities of 4a–6a were assessed by ¹H NMR spectra of solid samples kept under ambient conditions for 24 hours. Spectra for 5a and 6a show no signs of decomposition, while those of 4a show minor impurities.

With growing interest in B-containing helicenes as chiral functional dyes,¹⁴ we were intrigued by the helical structure of the 6b cation revealed by X-ray crystallography. The dihedral angle of 6b defined by C1–C2–C3–C4 (Fig. 2c and d) is 24.8°, considerably less than dihedral angles defined by analogous carbon atoms in recently reported 6,6′-biphenanthridine ruthenium complexes, which lie between 52.3° and 59.5°.¹⁵ The dihedral angle between calculated mean planes of rings A and B is only 12.1°, while that between calculated mean planes of rings C and D is 40.6° (Fig. 2c and d). Accordingly, both phenanthridine ring systems of the helical 6b are warped in comparison to the planar phenanthridine ring systems of 4a–6a. The B–N bond lengths of 6b are 1.563(2) Å and 1.569(2) Å, similar to those observed for 2,2′-bipyridyldichloroboronium chloride.¹⁶

The thermally induced reactions of 1–3 to form 4a–6a stand in apparent contrast to previously reported reactivities of isonitrile boranes. However, borylative cyclization of 2-isocyanobiphenyl to form 6-borylated phenanthridine has been shown by Wang and co-workers,¹⁷ using an N-heterocyclic carbene-borane and a radical initiator. This reactivity was attributed to a radical, Minisci-type¹⁸ mechanism involving imidoyl radicals, recalling oxidative cyclizations of 2-isocyanobiphenyl by Tobisu, Chatani, and co-workers.¹⁹ On the other hand, ionic Friedel–Crafts mechanisms have been proposed by Currie and Tennant for Lewis acid-catalyzed cyclizations of biphenyl-2-isocyanide dihalides to form 6-halophenanthridines,²⁰ as well as by Yu and co-workers for syntheses of trifluoromethylated phenanthridines from 2-isocyanobiphenyl and synthetic equivalents of [CF₃]⁺.²¹

To probe possible radical reaction mechanisms, the reactions of 1–3 in thoroughly degassed C₂H₂Cl₄ were studied by EPR spectroscopy, at room temperature and at 100 °C. No signals were observed for reactions of 2 or 3 at either temperature, although a minor signal was observed for reaction of 1 (Fig. S56–S67†). Nonetheless, thermal reaction rates of 1 assessed by ¹H NMR spectroscopy in C₂D₂Cl₄ were insensitive to freeze–pump–thaw degassing cycles or protection from light (Fig. S50†). Notably, there is no evidence for Büchner-type ring expansion of 1 during reaction under ambient lighting, whereas pure 2-isocyanobiphenyl can undergo this ring expansion under irradiation.²² We interpreted these combined results to favor ionic mechanisms, rather than radical mechanisms, for the thermal reactions of 1–3 to form 4a–6a. Moreover, ionic mechanisms align with the absence of clear radical initiators for these reactions, and the ionic reactivities of related isontrile-BX₃ adducts (where X = Br or Cl) reported by Meller and Batka (Fig. 1a).^2h

To investigate the mechanism underlying the conversion of 2-isocyanobiphenyl and BX₃ (X = I, Br, Cl) to 4a, 5a, and 6a, density functional theory (DFT) calculations were performed. Focusing first on the reaction between 2-isocyanobiphenyl and BCl₃, the DFT results align with experimental observations, indicating the formation of a stable adduct, 3, with a ΔG of −10.7 kcal mol⁻¹ (Fig. 3).

Fig. 3 DFT-calculated Gibbs free energy profile of the formation of 6a. The structures were optimized in CPCM solvation model in C₂H₂Cl₄.

To convert 3 to 6a, both C–H activation and C–C coupling are required. Our results indicate that the reaction begins with C–C coupling, forming a six-membered ring via the pathway 3 → TS1-Cl → Int1-Cl, with activation and reaction free energies (ΔG^‡/ΔG) of 24.1/22.1 kcal mol⁻¹. Subsequently, converting Int1-Cl to 6a requires the migration of a hydrogen atom (marked with blue) from carbon to nitrogen. This migration can occur via three possible mechanisms:

(a) Intramolecular 1,3-H migration: The transition state for this mechanism cannot be located.

(b) Two consecutive intramolecular 1,2-H migrations: The first step (Int1-Cl → TS3-Cl → Int4-Cl) has an activation barrier of ΔG^‡ = 15.3 kcal mol⁻¹ and leads to BCl₃ dissociation.

(c) Intermolecular hydrogen transfer: In this pathway, two Int1-Cl molecules form a van der Waals complex (Int1-Cl → Int2-Cl, ΔG = −11.8 kcal mol⁻¹), and each molecule transfers its hydrogen atom to the nitrogen of the other. The first hydrogen transfer (Int2-Cl → TS2-Cl → Int3-Cl) has a Gibbs free energy barrier of ΔG^‡/ΔG = 3.5/−21.2 kcal mol⁻¹, while the second one (Int3-Cl → ½-(6a)₂) is barrierless with ΔG = −35.1 kcal mol⁻¹. The dissociation of ½-(6a)₂ to 6a is slightly endergonic, with a ΔG of 0.5 kcal mol⁻¹. Recomputing this energy using the M06 functional, a widely used alternative, we find that ΔG becomes slightly exergonic (ΔG = −2.2 kcal mol⁻¹). These results suggest that the intermolecular interaction between the monomers is weak, facilitating dimer dissociation. Among these mechanisms, the intermolecular process is the most energetically favorable. Overall, the rate-determining step for the conversion of 2-isocyanobiphenyl and BCl₃ to 6a is the C–C bond formation, with ΔG^‡ = 24.1 kcal mol⁻¹.

After identifying the most favorable pathway for forming 6a from 2-isocyanobiphenyl + BCl₃, we computed the Gibbs free energy surfaces for 2-isocyanobiphenyl + BI₃ → 4a and 2-isocyanobiphenyl + BBr₃ → 5a. Similar to 2-isocyanobiphenyl + BCl₃ → 6a, the rate-determining step for both 2-isocyanobiphenyl + BI₃ → 4a and 2-isocyanobiphenyl + BBr₃ → 5a is also C–C bond formation, with comparable ΔG^‡ values of 25.0 and 27.1 kcal mol⁻¹, respectively (Fig. S75 and S76†).

The formation of 6b involves C–H functionalizing cyclization of biphenyl, similar to that observed for the formation of 4a–6a. However, an additional intermolecular C–C bond is formed between precursors, and a series of steps lead to altered bonding environments for boron atoms. Peaks corresponding to 6a and 6b appear to grow simultaneously in our NMR studies (Fig. S5†). We also note that no significant changes are observed by NMR spectroscopy after heating 6a for 3 hours at 130 °C in C₂D₂Cl₄, with or without BCl₃. Furthermore, treatment of 6a with 2-isocyanobiphenyl at 130 °C for 24 hours does not yield compound 6b (Fig. S53†). Thus, we suspect that 6a is not an intermediate enroute to 6b, but rather that 6b is formed via an independent pathway. Nevertheless, the mechanism of formation of 6b is still unclear based on preliminary studies (see ESI†). Further investigations of mechanistic details are ongoing in our laboratory.

Conclusions

Adducts between 2-isocyanobiphenyl and BX₃ (X = I, Br, Cl) have been prepared. These compounds undergo C–C bond-forming borylative cyclizations at elevated temperatures to give phenanthridinium-6-trihaloborate zwitterions, contrasting with reactivities of other reported isonitrile-borane adducts. Reaction of 2-isocyanobiphenyl·BCl₃ forms an additional, unexpected helical 6,6′-bisphenanthridine-chelated dichloroboronium salt via three C–C bond formations. Based on NMR, EPR, and DFT studies, we propose ionic mechanisms for reactions leading to phenanthridinium-6-trihaloborates. The unique reactivities described herein offer new possibilities for isonitrile-boranes in synthesis.

Author contributions

Y.-T. K. performed the experimental work. C.-Y. Y. performed DFT calculations. Y.-H L. performed X-ray crystallography. M.-J. C. directed DFT studies. J. M. F. supervised the research and wrote the manuscript. All authors edited the manuscript.

Data availability

All experimental details and data supporting this article are available in the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We are grateful for the financial support of the National Science and Technology Council of Taiwan (grant no. 113-2113-M-002-016-MY3) and the Yushan Young Scholar Program of the Ministry of Education of Taiwan (grant no. MOE-110-YSFMS-0003-003-P1). We acknowledge the mass spectrometry technical research services from Consortia of Key Technologies, National Taiwan University and the Instrumentation Center of National Taiwan Normal University. We would like to thank the Instrumentation Center at National Tsing Hua University for EPR measurements.

References

1. (a) E. Buñuel and D. J. Cárdenas, Eur. J. Org. Chem., 2016, 5446–5464; (b) K. Kubota, H. Iwamoto and H. Ito, Org. Biomol. Chem., 2017, 15, 285–300; (c) J. R. Lawson and R. L. Melen, Inorg. Chem., 2017, 56, 8627–8643; (d) G. Kehr and G. Erker, Chem. Sci., 2016, 7, 56–65; (e) S. Kirschner, K. Yuan and M. J. Ingleson, New J. Chem., 2021, 45, 14855–14868.
2. (a) G. Hesse and H. Witte, Angew. Chem., Int. Ed. Engl., 1963, 2, 617–617; (b) G. Hesse, H. Witte and W. Gulden, Tetrahedron Lett., 1966, 7, 2707–2710; (c) J. Casanova and R. E. Schuster, Tetrahedron Lett., 1964, 5, 405–409; (d) J. Casanova Jr. and H. R. Kiefer, J. Org. Chem., 1969, 34, 2579–2583; (e) S. Bresadola, G. Carraro, C. Pecile and A. Turco, Tetrahedron Lett., 1964, 5, 3185–3188; (f) S. Bresadola, F. Rossetto and G. Puosi, Tetrahedron Lett., 1965, 6, 4775–4778; (g) J. Tanaka and J. C. Carter, Tetrahedron Lett., 1965, 6, 329–332; (h) A. Meller and H. Batka, Monatsh. Chem., 1969, 100, 1823–1828; (i) K. Škoch, C. G. Daniliuc, G. Kehr and G. Erker, Chem. Commun., 2020, 56, 12178–12181; (j) M. Suginome, T. Fukuda and Y. Ito, J. Organomet. Chem., 2002, 643–644, 508–511; (k) M. Suginome, T. Fukuda, H. Nakamura and Y. Ito, Organometallics, 2000, 19, 719–721.
3. (a) B. R. Barnett, C. E. Moore, A. L. Rheingold and J. S. Figueroa, Chem. Commun., 2015, 51, 541–544; (b) B. R. Barnett, C. E. Moore, A. L. Rheingold and J. S. Figueroa, J. Am. Chem. Soc., 2014, 136, 10262–10265.
4. J. Li, C. G. Daniliuc, C. Mück-Lichtenfeld, G. Kehr and G. Erker, Angew. Chem., Int. Ed., 2019, 58, 15377–15380.
5. J. Li, C. G. Daniliuc, G. Kehr and G. Erker, Angew. Chem., Int. Ed., 2021, 60, 27053–27061.
6. (a) H. T. W. Shere, H.-Y. Liu, S. E. Neale, M. S. Hill, M. F. Mahon and C. L. McMullin, Chem. Sci., 2024, 15, 7999–8007; (b) H. Asakawa, K.-H. Lee, Z. Lin and M. Yamashita, Nat. Commun., 2014, 5, 4245; (c) N. Arnold, H. Braunschweig, W. C. Ewing, T. Kupfer, K. Radacki, T. Thiess and A. Trumpp, Chem. – Eur. J., 2016, 22, 11441–11449; (d) Y. Katsuma, H. Asakawa and M. Yamashita, Chem. Sci., 2018, 9, 1301–1310; (e) Y. Katsuma, H. Asakawa, K.-H. Lee, Z. Lin and M. Yamashita, Organometallics, 2016, 35, 2563–2566.
7. J. Böhnke, H. Braunschweig, T. Dellermann, W. C. Ewing, T. Kramer, I. Krummenacher and A. Vargas, Angew. Chem., Int. Ed., 2015, 54, 4469–4473.
8. T.-T. Liu and Y.-S. Cui, Chem. – Eur. J., 2023, 29, e202302683.
9. (a) M. Tobisu, H. Fujihara, K. Koh and N. Chatani, J. Org. Chem., 2010, 75, 4841–4847; (b) H. Braunschweig, W. C. Ewing, K. Ferkinghoff, A. Hermann, T. Kramer, R. Shang, E. Siedler and C. Werner, Chem. Commun., 2015, 51, 13032–13035.
10. (a) H. Braunschweig, R. D. Dewhurst, F. Hupp, M. Nutz, K. Radacki, C. W. Tate, A. Vargas and Q. Ye, Nature, 2015, 522, 327–330; (b) H. Braunschweig, T. Herbst, K. Radacki, C. W. Tate and A. Vargas, Chem. Commun., 2013, 49, 1702–1704; (c) H. Braunschweig, K. Radacki, R. Shang and C. W. Tate, Angew. Chem., Int. Ed., 2013, 52, 729–733; (d) T. J. Hadlington, T. Szilvási and M. Driess, Chem. Sci., 2018, 9, 2595–2600; (e) M. Nutz, B. Borthakur, C. Pranckevicius, R. D. Dewhurst, M. Schäfer, T. Dellermann, F. Glaab, M. Thaler, A. K. Phukan and H. Braunschweig, Chem. – Eur. J., 2018, 24, 6843–6847.
11. W. Lu, H. Hu, Y. Li, R. Ganguly and R. Kinjo, J. Am. Chem. Soc., 2016, 138, 6650–6661.
12. See also, reactivity of an isonitrile-silaborane made via BH₃ insertion: T. Kajiwara, N. Takeda, T. Sasamori and N. Tokitoh, Organometallics, 2004, 23, 4723–4734.
13. H. Jacobsen, H. Berke, S. Döring, G. Kehr, G. Erker, R. Fröhlich and O. Meyer, Organometallics, 1999, 18, 1724–1735.
14. A. Nowak-Król, P. T. Geppert and K. R. Naveen, Chem. Sci., 2024, 15, 7408–7440.
15. D. B. Nemez, I. B. Lozada, J. D. Braun, J. A. G. Williams and D. E. Herbert, Inorg. Chem., 2022, 61, 13386–13398.
16. S. M. Mansell, C. J. Adams, G. Bramham, M. F. Haddow, W. Kaim, N. C. Norman, J. E. McGrady, C. A. Russell and S. J. Udeen, Chem. Commun., 2010, 46, 5070–5072.
17. Y. Liu, J.-L. Li, X.-G. Liu, J.-Q. Wu, Z.-S. Huang, Q. Li and H. Wang, Org. Lett., 2021, 23, 1891–1897.
18. D. Nanni, P. Pareschi, C. Rizzoli, P. Sgarabotto and A. Tundo, Tetrahedron, 1995, 51, 9045–9062.
19. M. Tobisu, K. Koh, T. Furukawa and N. Chatani, Angew. Chem., Int. Ed., 2012, 51, 11363–11366.
20. K. S. Currie and G. Tennant, J. Chem. Soc., Chem. Commun., 1995, 2295–2296.
21. Y. Cheng, H. Jiang, Y. Zhang and S. Yu, Org. Lett., 2013, 15, 5520–5523.
22. J. H. Boyer and J. De Jong, J. Am. Chem. Soc., 1969, 91, 5929–5930.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2418957–2418963. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5dt00210a

---