Expression of hyperconjugative stereoelectronic interactions in borazines

Abstract

This paper discusses hyperconjugative stereoelectronic effects in borazines. A series of alkyl-substituted borazines were synthesized and analysed by NMR spectroscopy and X-ray diffraction. Supported by NBO analyses, the significant decreases in ¹J_CH coupling constant for the CH groups adjacent to the boron atoms are consistent with the presence of and interactions. These interactions lower the electrophilicity of boron atoms, enhancing moisture stability and establishing these molecules as valuable scaffolds in synthetic chemistry and materials science.

---

The significance of BN-doped molecular scaffolds has recently surged due to their broad applications in optoelectronic devices,^1–4 thermal management materials,⁵ and catalysis.⁶ Among these, hexasubstituted borazine is an attractive option for BN-based architectures.⁷ The borazine core's susceptibility to hydrolysis in water has hindered its widespread use.⁸ To prevent borazine hydrolytic decomposition, bulky aryl groups can be added to the boron atom, where ortho-substituents create steric hindrance shielding B-atoms from water.⁹ In another architecture, amino groups were used to stabilize the borazine ring within a hybrid cyclomatrix polymer.^9f In search of other approaches preventing hydrolysis, we noticed that B,B′,B′′-trialkyl-N,N′,N′′-triphenyl borazines (alkyl moieties being i- and n-propyl, i- and n-butyl) exhibit remarkable counterintuitive moisture resistance when compared to its aryl-substituted congeners.¹⁰ A comprehensive understanding of the factors affecting the stability of these molecules is still lacking. It is possible that hyperconjugative stereoelectronic interactions occur between the electron-donating alkyl groups and the electrophilic borazine ring (Fig. 1b),¹¹ similar to what is seen in trialkyl boranes.^9h,i With this hypothesis in mind, in this paper, we address the presence of such hyperconjugative effects through NMR spectroscopy and X-ray investigations of a series of B-alkyl borazine derivatives.¹² Theoretical modeling further supported these studies by elucidating the molecular orbitals involved in those interactions.¹³ Three types of alkyl groups to be attached to the boron centres were chosen: –CH₃, –CH₂–, and –CH–. While molecules 1a–b bear methyl substituents, 2a–e and 3a–b feature methylene (as –ⁿBu, –ⁿHex, –ⁿOct, benzyl and (trimethylsilyl)methyl substituents) and methine (as –ⁱPr and cyclohexyl substituents) substitutions on the boron atoms (Fig. 1a), respectively.

Fig. 1 (a) Alkyl borazines studied in this work (*2′,6′-dimethyl-[1,1′-biphenyl]-4-amine used for the synthesis). (b) Stereoelectronic interactions in B-aryl and B-alkyl borazines.

The synthesis for these alkyl borazines have been carried out by adapting known protocols.^9c Intermediate trichloroborazole (BNCl) was prepared by refluxing aniline with BCl₃ in toluene (Scheme S1, ESI†). The reaction of BNCl with the respective organoalkyl-lithium/Grignard nucleophiles in THF gave the final product in good yields (61–74%, Fig. 1a and Scheme S2, ESI†). The structures of all alkyl borazines were characterized by ¹H and ¹³C NMR spectroscopy, and with high-resolution mass spectrometry (Fig. S1–S48, ESI†) and selected structures by single-crystal X-ray diffraction analysis. All B,B′,B′′-trialkyl borazine derivatives exhibited low reactivity towards moisture, allowing them to be handled in the air without precautions and purified through column chromatography. Thermogravimetric analysis (TGA) indicated that all derivatives remain stable above 200 °C (Fig. S49–S51, ESI†).

X-ray diffraction investigations of selected borazine derivatives were first attempted to shed further light on their structural properties (Fig. 2a–f and Tables S1–S6 and Fig. S52–S57, ESI†). In all derivatives, the central borazine ring displays a nearly flat geometry, with average ∠_N–B–N and ∠_(B–N–B) angles of 116.2(4)° and 123.7(4)°, respectively, and with typical B–C₁ bond lengths of 1.575(14) Å (Fig. 2).^11c For molecules 2a, 3a, and 3b, the C₁–C₂ bond lengths (average value ∼1.526(4) Å) fall within the expected range for corresponding single bonds.^11a The average torsional angles ∠_{(N–B–C1–H/C2/Si)}, describing the relative orientation of the B–C–H/C/Si bonds to the borazine ring, are ∼90° for all borazines (Fig. 2). This perpendicular arrangement of the C–H/C/Si bonds is to be expected when in the presence of a hyperconjugation. Moreover, the bond angle ∠_{(B–C1–C2)} for 2a, 3a and 3b is found to be 113(2)°, 116(3)° and 118(6)°, respectively. Similarly, the bond angle ∠_{(B–C1–Si)} is 118(2)° for molecule 2e (Fig. 2). These ∠_{(B–C1–H/C2/Si)} angles are wider than those between C(sp³), again suggesting that a hyperconjugative interaction is at play.^11a

Fig. 2 Single-crystal X-ray structures of borazines and relevant structural parameters (a) 1a, C2/c, (b) 1b, C2/c, (c) 2a, Cc, (d) 2e, P2₁/c, (e) 3a, P1̅, and (f) 3b, Pna2₁. Grey: C, pink: B, yellow: Si, and blue: N. The torsion angles are the average of the values measured for C–H/C/Si bonds that display an orientation that is close to that of perpendicularity.

Considering that one-bond NMR spin–spin coupling constants are experimental probes for determining any stereoelectronic interactions (Perlin effect),¹⁴ we tackled the determination of the ¹H–¹³C coupling constants (J_CH) of the C–H bonds connected to the B-atoms by analysis of either the carbon satellite signals in the ¹H NMR spectra or the doublet cross peaks in proton-coupled Heteronuclear single quantum coherence (HSQC) experiments (Fig. 3). For 1a and 1b, the experimental ¹J_CH value for the B–Me groups is 116.4 and 116.6 Hz (Fig. 3a), respectively, which is lower than that of the Me group in the same molecule (²J_CH = 126.6 Hz, Fig. S8 and S13, ESI†). Similarly, the ¹J_CH values for molecules 2a–e are within the 112 Hz to 118 Hz range, which is also lower than the other J_CH values (Fig. 3b and Fig. S18, S23, S28, S33 and S38, ESI†). Further, by examining the ²⁹Si NMR spectrum of molecule 2e, we could determine the coupling constants ¹J_CSi for –CH₂– and –CH₃ groups attached to the Si-atom. The results indicate that ¹J_CSi of –CH₂– exhibits a value of ∼12.3 Hz lower than that measured with –CH₃ (Fig. S38, ESI†). This suggests that the C–Si bond experiences an elongation that one could attribute to a hyperconjugation. For 3a and 3b, the ¹J_CH values are 116.1 and 112.3 Hz, respectively (Fig. 3c and Fig. S43 and S48, ESI†). Notably, these couplings were the lowest among the series. These observations are in line with the X-ray-measured ∠_{(B–C1–C2)} angle values (Fig. 2), confirming strong hyperconjugative interactions between the B(p_z) orbital and the σ_C–H, σ_C–C and σ_C–Si bonds of the relevant substituents.

Fig. 3 Excerpts of the HSQC spectra displaying the relevant ¹J_CH coupling constants for (a) 1a, (b) 2a, and (c) 3a and. (d) 1 ¹J_CH, 2 ¹J_CH, 1 ¹J_CSi and 2 ¹J_CSi values for each alkyl borazines.

DFT calculations were employed to explore hyperconjugation, with bonding analysed via natural bonding orbital (NBO) analyses.^13c The findings revealed weak interactions between the σ_C–H and σ_C–C orbitals and both the and orbitals for all derivatives. The σ_C–H orbital, positioned perpendicularly to the ring, exhibited significant overlap with the π* orbital. For 1a and 1b, the optimised structure contains a C–H lying orthogonal to the borazine plane (Fig. 2a and b) and two that are at the midpoint between the orthogonal and parallel planes ( and in Fig. 4a and Fig. S58, ESI†). Since the overlap between σ_C–H and the given unoccupied orbitals depends on the orientation of the substituent, the stabilisation energy (E₂) associated with these interactions was computationally probed as a function of the angle between the proximal σ_C–H/C/Si orbital and the borazine ring upon rotation about the B–C bond (molecule 1a in Fig. 4). As expected, the interaction is the strongest, 6.5 kcal mol⁻¹, when the C–H bond lies perpendicular to the BN ring as this orientation gives the most significant orbital overlap between the σ_C–H and orbitals. As the Me group is rotated, E₂ decreases, and no interaction is detected with ∠_{(N–B–C1–H)} values <20°. Conversely, the interaction gives the strongest stabilisation, 3.51 kcal mol⁻¹, when the C–H bond lies in the same plane as the BN ring. The effect is not as pronounced as the , but it contributes conferring an additional stabilisation. Similarly, NBO analysis shows both and interactions in the case of –CH₂– and –CH– (3a–b) substitutions (Fig. S59–S67, ESI†). Also, the stabilisation energies were probed for proximal σ_C–C/Si bonds (Fig. S61, ESI†).

Fig. 4 (a) NBO for and 1a (PBE/def2-TZVP, orbitals with only one B-atom are shown), (b) hyperconjugative energies upon rotation of the B-alkyl group.

Interactions between the σ_C–C and BN ring orbitals are significant when sterically demanding alkyl groups prevent the C–H group from lying perpendicular to the ring, and interactions are present in 2a–e and 3a–b (Fig. S60–S66, ESI†) but offered less of a stabilising effect than those involving σ_C–H (Fig. 4b). Notably, and interactions in 2e displayed stronger stabilization than those with σ_C–C (Fig. S61, ESI†). The maximum E₂ value is observed at the torsion angles ∠_{(N–B–C1–H/C2/Si)} of ∼90° and 0° for and interactions, respectively (Table 1). We evaluate energy changes for 1a using a different basis set and calculation function (Fig. S67, ESI†), confirming the reliability of the data.

Table 1 Stabilisation energies (E₂) as a function of the torsional angle (90° for and 0° for (PBE/def2-TZVP)

Molecule	E2/kcal mol^−1
	1a	1b	2a	2b	2c	2d	2e	3a	3b
	6.53	6.48	6.81	6.40	6.40	5.70	5.28	5.57	4.70
	3.51	3.52	3.98	3.99	3.99	3.96	4.37	4.36	4.39
	—	—	3.83	3.87	3.80	2.66	—	3.85	3.47
	—	—	1.28	1.28	1.28	1.04	—	1.52	1.41
	—	—	—	—	—	—	7.94	—	—
	—	—	—	—	—	—	2.22	—	—

---

Further, we calculated the NMR coupling constant using the GIAO method.¹⁵ The results (Table S7, ESI†) show that the difference ΔJ between ¹J_C1H and ¹J_C2H is 6 to 10 Hz, which is in agreement with the experimental values. Similarly, the calculated ΔJ between ¹J_C1Si and ¹J_C2Si in 2e is 12 Hz, matching the experimental data (Table S8, ESI†). We examined the isodesmic reactions with borazine 3, which has Me and Ph moieties on the N and B atoms (ESI,† Section S7.2.3). Our findings revealed higher values for enthalpy (+18.5 kcal mol⁻¹) and Gibbs free energy (+17.4 kcal mol⁻¹) compared to congener 1a, highlighting the significant impact of hyperconjugation. NBO interactions in Me- and Et-substituted benzene congeners (4_Me and 4_Et) yielded values of 5.1 and 3.4 kcal mol⁻¹. These values indicate that 4_Me and 4_Et exhibit reduced hyperconjugation than their BN counterparts, as evidenced by their higher ¹J_C1H and ¹J_C2H values than those of 1a and 2b, Table S9, ESI†).

As a proof of concept exploiting the hyperconjugation to stabilize borazine derivatives, we successfully prepared cross-linked borazine-based polymers^9f using either 1,6-dilithium hexane or 1,8-dilithium octane as nucleophiles (Fig. 5a). The insoluble nature of the obtained materials in common organic solvents suggests the presence of a crosslinked structure. Also, it exhibits hydrolytic stability. Solid-state ¹³C NMR spectra (Fig. 5b and Fig. S68–S70, ESI†) distinctly displayed aromatic and aliphatic signals corresponding to the phenyl ring and alkyl chain on the borazine structure. Scanning electron microscope (SEM) analyses revealed the morphology of P1 and P2 polymers, showing aggregates and macroscopic porosity (Fig. 5c and Fig. S71 and S72, ESI†). TGA analysis indicated approximately 50% weight loss (up to 500 °C), which matches the alkyl moieties’ loss (Fig. S73, ESI†).

Fig. 5 (a) Synthesis of cross-linked polymers P1 and P2, (b) solid-state ¹H → ¹³C cross-polarization and magic angle spinning (CPMAS) NMR spectrum of P1, and (c) SEM images of P1 (see CPMAS NMR spectrum for reference borazine 2a in Fig. S68, ESI†).

In conclusion, our studies into alkyl borazines using NMR spectroscopy have revealed key insights into stereoelectronic interactions. The findings confirm the existence of interactions between the C₁–H, C₁–C₂, and C₁–Si bonds of the B-alkyl group with the orbital of the BN ring. Computational analyses provide a nuanced understanding of the interplay between the σ_C–H/C/Si orbitals and both the and orbitals. As proof of principle, we have developed stable crosslinked borazine polymers that demonstrate moisture resistance. Our findings enhance the knowledge of borazines and expand their applications, offering exciting opportunities for using inorganic benzene in new materials and advanced chemistry.

D. B. acknowledges the EU (MSCA-ITN-ETN, STiBNite, no. 956923 & HORIZON-IA, DecoChrom, no. 760973) and the University of Vienna for funding; we acknowledge access to the HPC service (ARCCA) at Cardiff University, we thank Dr D. Romito (Vienna) and Dr P. K. Mondal (Elettra Trieste) for help in some X-ray analyses.

Data availability

The data supporting this article are in the ESI,† while the original data are available from the authors upon request.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. (a) I. H. T. Sham, C. C. Kwok, C. M. Che and N. Zhu, Chem. Commun., 2005, 3547–3549; (b) D. Bonifazi, F. Fasano, M. M. Lorenzo-Garcia, D. Marinelli, H. Oubaha and J. Tasseroul, Chem. Commun., 2015, 51, 15222–15236; (c) D. Marchionni, S. Basak, A. N. Khodadadi, A. Marrocchi and L. Vaccaro, Adv. Funct. Mater., 2023, 33, 2303635; (d) I. Neogi and A. M. Szpilman, Synthesis, 2022, 1877–1907.
2. (a) Z. Huang, S. Wang, R. D. Dewhurst, N. V. Ignat‘ev, M. Finze and H. Braunschweig, Angew. Chem., Int. Ed., 2020, 59, 8800–8816; (b) S. Madayanad Suresh, D. Hall, D. Beljonne, Y. Olivier and E. Zysman-Colman, Adv. Funct. Mater., 2020, 30, 1908677.
3. (a) J. Dosso, J. Tasseroul, F. Fasano, D. Marinelli, N. Biot, A. Fermi and D. Bonifazi, Angew. Chem., Int. Ed., 2017, 56, 4483–4487; (b) S. Oda and T. Hatakeyama, Bull. Chem. Soc. Jpn., 2021, 94, 950–960; (c) J. Kashida, Y. Shoji, Y. Ikabata, H. Taka, H. Sakai, T. Hasobe, H. Nakai and T. Fukushima, Angew. Chem., Int. Ed., 2021, 60, 23812–23818.
4. (a) S. Wang, Coord. Chem. Rev., 2001, 215, 79–98; (b) Y.-L. Rao and S. Wang, Inorg. Chem., 2011, 50, 12263–12274.
5. (a) M. R. H. Mazumder, L. D. Mathews, S. Mateti, N. V. Salim, J. Parameswaranpillai, P. Govindaraj and N. Hameed, Appl. Mater. Today, 2022, 29, 101672; (b) Y. Luo, L. Zou, J. Qiao, J. Zhang, K. Liu, H. Wu, P. Lin and Y. Chen, ACS Appl. Energy Mater., 2022, 5, 11591–11603.
6. A. K. Thakur, K. Kurtyka, M. Majumder, X. Yang, H. Q. Ta, A. Bachmatiuk, L. Liu, B. Trzebicka and M. H. Rummeli, Adv. Mater. Interfaces, 2022, 9, 2101964.
7. (a) A. Wakamiya, T. Ide and S. Yamaguchi, J. Am. Chem. Soc., 2005, 127, 14859–14866; (b) S. Kervyn, N. Kalashnyk, M. Riello, B. Moreton, J. Tasseroul, J. Wouters, T. S. Jones, A. De Vita, G. Costantini and D. Bonifazi, Angew. Chem., Int. Ed., 2013, 52, 7410–7414; (c) J. Dosso, T. Battisti, B. D. Ward, N. Demitri, C. Hughes, A. P. Williams, K. D. M. Harris and D. Bonifazi, Chem. – Eur. J., 2020, 26, 6608–6621.
8. T. Yoshizaki, H. Watanabe and T. Nakagawa, Inorg. Chem., 1968, 7, 422–429.
9. (a) K. Nagasawa, Inorg. Chem., 1966, 5, 442–445; (b) S. Kervyn, O. Fenwick, F. Di Stasio, Y. S. Shin, J. Wouters, G. Accorsi, S. Osella, D. Beljonne, F. Cacialli and D. Bonifazi, Chem. – Eur. J., 2013, 19, 7771–7779; (c) D. Marinelli, F. Fasano, B. Najjari, N. Demitri and D. Bonifazi, J. Am. Chem. Soc., 2017, 139, 5503–5519; (d) N. Kalashnyk, P. Ganesh Nagaswaran, S. Kervyn, M. Riello, B. Moreton, T. S. Jones, A. De Vita, D. Bonifazi and G. Costantini, Chem. – Eur. J., 2014, 20, 11856–11862; (e) J. Dosso, D. Marinelli, N. Demitri and D. Bonifazi, ACS Omega, 2019, 4, 9343–9351; (f) N. A. Riensch, A. Deniz, S. Kühl, L. Müller, A. Adams, A. Pich and H. Helten, Polym. Chem., 2017, 8, 5264; (g) J. R. Sanzone, C. T. Hu and K. A. Woerpel, J. Am. Chem. Soc., 2017, 139, 8404–8407; (h) A. R. Cowley, A. J. Downs, S. Marchant, V. A. Macrae, R. A. Taylor and S. Parsons, Organometallics, 2005, 24, 5702–5709; (i) B. Wrackmeyer and O. L. Tok, Z. Naturforsch. B, 2005, 60, 259–264.
10. (a) S. J. Groszos and S. F. Stafiej, J. Am. Chem. Soc., 1958, 80, 1357–1360; (b) J. H. Smalley and S. F. Stafiej, J. Am. Chem. Soc., 1959, 81, 582–586; (c) R. J. Brotherton and A. L. McCloskey, Adv. Chem., 1964, 42, 131–138; (d) I. B. Atkinson, D. C. Blundell and D. B. Clapp, J. Inorg. Nucl. Chem., 1972, 34, 3037–3041.
11. (a) A. H. Maulitz, P. Stellberg and R. Boese, J. Mol. Struct.: THEOCHEM, 1995, 338, 331–340; (b) I. V. Alabugin, Stereoelectronic Effects: A Bridge Between Structure and Reactivity, Wiley, Hoboken, 2016; (c) R. Boese, A. H. Maulitz and P. Stellberg, Chem. Ber., 1994, 127, 1887–1889.
12. (a) G. Cuevas and E. Juaristi, J. Am. Chem. Soc., 2002, 124, 13088–13096; (b) R. Boese, D. Blaser, N. Niederprum, M. Nusse, W. A. Brett, P. V. R. Schleyer, M. Buhl and N. J. R. V. E. Hommes, Angew. Chem., Int. Ed. Engl., 1992, 31, 314–316.
13. (a) I. V. Alabugin, J. Org. Chem., 2000, 65, 3910–3919; (b) I. V. Alabugin and T. A. Zeidan, J. Am. Chem. Soc., 2002, 124, 3175–3185; (c) I. V. Alabugin, G. P. Gomes and M. Abdo, Comput. Mol. Sci., 2018, 9, e1389; (d) Y. Mo, H. Jiao and P. V. R. Schleyer, J. Org. Chem., 2004, 69, 3493–3499; (e) F. Weinhold and C. R. Landis, Valency and bonding: a natural bond orbital donor-acceptor perspective, Cambridge University Press, Cambridge, UK, 2005.
14. (a) A. S. Perlin and B. Casu, Tetrahedron Lett., 1969, 10, 2921–2924; (b) S. Wolfe, B. M. Pinto, V. Varma and R. Y. N. Leung, Can. J. Chem., 1990, 68, 1051–1062; (c) E. Juaristi and G. Cuevas, Acc. Chem. Res., 2007, 40, 961–970.
15. X. S. Monfort and O. Eisenstein, Polyhedron, 2006, 25, 339–348.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2261150, 2261151, 2280127, 2280128, 2284380 and 2307378. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4cc05188b

---