Enhancement of London Dispersion in Frustrated Lewis Pairs: Towards a Crystalline Encounter Complex

Key words

frustrated Lewis pairs - encounter complex - London dispersion forces - Lewis acid - Lewis base

Figure 1 (a) Reversible association of the encounter complex in archetypical FLP systems. (b) London dispersion facilitated reversible dimerization of all-meta-substituted triphenylmethyl radicals.[8] Red-dashed lines indicate London dispersion interactions.

Since their discovery in 2006,[1] the potential of frustrated Lewis pairs (FLPs) for the metal-free activation of small molecules, such as H₂ and CO₂, has been extensively showcased.[2] However, particularly for substrates like H₂ that only weakly interact with the individual FLP components, the activation mechanism still remains ambiguous. This is due to the weak interactions between the Lewis acid and Lewis base in intermolecular FLPs resulting in highly fluxional structures of low concentrations in solution, which are thus challenging to study and characterize by spectroscopic methods.[3] A number of computational investigations suggested the pre-organization of the FLP components to form an encounter complex, in which the interaction is governed by London dispersion forces between the bulky substituents (Figure [1a]).[4] The first experimental evidence for such association was provided through ¹⁹F,¹H HOESY NMR studies by Rocchigiani and co-workers revealing intermolecular H–F interactions in concentrated solutions of archetypical FLPs R₃P/B(C₆F₅)₃ (R = Mes, tBu).[5] Further evidence of the encounter complex in solution was provided by UV/Vis and transient absorption spectroscopy. The detection of charge-transfer bands of FLPs R₃P/B(C₆F₅)₃ established the equivalence of encounter complexes with electron donor–acceptor (EDA) complexes and allowed for the observation of radical ion pairs resulting from photoinduced single-electron transfer (SET) between the associated Lewis base and Lewis acid by EPR spectroscopy.[6] However, to date, solid-state structures of intermolecular FLPs potentially revealing key information on the FLP pre-organization have not yet been crystallographically characterized.[7]

Schreiner and co-workers demonstrated that the reversible head-to-head association of bulky, all-meta-substituted triphenylmethyl radicals to the corresponding hexaphenylethane derivatives can be attributed to attractive London dispersion forces between the meta substituents and that the extent of stabilization of the dimeric structure can be tuned depending on the quality of the dispersion energy donor (Figure [1b]).[8]

In line with our general interest in the reactivity and reaction mechanisms of FLP systems,[6] [9] we envisioned the utilization of London dispersion stabilization to increase the attraction between the Lewis acid and the Lewis base of an FLP, enhance the pre-organization of the FLP components, which should increase its concentration in solution, ultimately to enable the isolation and (solid-state) characterization of the encounter complex. For this, we computationally investigated the stability of potential encounter complexes containing all-meta-substituted triarylamines/-phosphines and triarylboranes and studied promising combinations experimentally.

Employing an approach analogous to the London dispersion facilitated formation of all-meta-substituted hexaphenylethanes, we included triarylamines N(3,5-tBu₂C₆H₃)₃ and N(3,5-Ph₂C₆H₃)₃ as well as triarylphosphine P(3,5-tBu₂C₆H₃)₃ as the Lewis basic component of potential FLPs in our investigation. For the Lewis acidic counterpart all-meta-substituted triarylboranes B(3,5-tBu₂C₆H₃)₃, B(3,5-Ph₂C₆H₃)₃, and B(3,5-(CF₃)₂C₆H₃)₃ as well as B(C₆F₅)₃ were selected. Moreover, archetypical FLP combinations R₃P/B(C₆F₅)₃ (R = Mes, tBu) were included for comparison. For the computational analysis, initially, the structures of encounter complexes for the different combinations of Lewis acids and Lewis bases were optimized at the ωB97XD/6-311+G(d,p)//ωB97XD/6-31G(d) level of theory, and the formation energies were determined (Table [1]).

Table 1 Encounter Complex-Formation Energies (kcal mol^–1)^a

Lewis base	Lewis acid	ΔE tot	ΔE Disp
PMes3	B(C6F5)3	–13.90	– 7.95
PtBu3	B(C6F5)3	–13.87	– 5.60
N(3,5-tBu2C6H3)3	B(3,5-tBu2C6H3)3	–39.93	–25.33
	B(3,5-Ph2C6H3)3	–43.73	–27.84
	B(3,5-(CF3)2C6H3)3	–35.01	–20.71
	B(C6F5)3	–20.32	–13.00
N(3,5-Ph2C6H3)3	B(3,5-tBu2C6H3)3	–46.60	–28.17
	B(3,5-Ph2C6H3)3	–54.20	–32.49
	B(3,5-(CF3)2C6H3)3	–40.85	–21.56
	B(C6F5)3	–27.68	–15.66
P(3,5-tBu2C6H3)3	B(3,5-tBu2C6H3)3	–51.58	–26.80
	B(3,5-(CF3)2C6H3)3	–53.80	–21.30
	B(C6F5)3	–33.31	–13.55

^a Calculated at the ωB97xD/6-311+G(d,p) level of theory.

Figure 2 Computed (ωb97xD/6-311+G(d,p)) structure of the encounter complex from N(3,5-tBu₂C₆H₃)₃/B(3,5-tBu₂C₆H₃)₃

Inspection of the obtained formation energies of encounter complexes obtained from all-meta-substituted triarylamines as well as phosphines together with B(C₆F₅)₃ already revealed a significantly increased stability (–20.32 to –33.31 kcal mol^–1) compared to the encounter complexes arising from R₃P/B(C₆F₅)₃ (R = Mes –13.90 kcal mol^–1, tBu –13.87 kcal mol^–1). Substitution of B(C₆F₅)₃ leads to a further increase in stability of the encounter complex, following the order B(C₆F₅)₃ < B(3,5-(CF₃)₂C₆H₃)₃ < B(3,5-tBu₂C₆H₃)₃ < B(3,5-Ph₂C₆H₃)₃. However, it must be noted that the examined P(3,5-tBu₂C₆H₃)₃/BR₃ (R = C₆F₅, 3,5-(CF₃)₂C₆H₃, 3,5-tBu₂C₆H₃) combinations converged to the corresponding classical P–B-bonded Lewis adducts as a result of the longer P–B and P–C bond lengths, effectively reducing the steric bulk of the phosphine. Therefore, the systems containing P(3,5-tBu₂C₆H₃)₃ as the Lewis base are not suitable for further experimental investigations of encounter complexes. In contrast, the optimized structures containing the triarylamines show no considerable amount of N–B interactions as assessed by the essentially planar B centers and the large N–B distances (3.693–4.502 Å). For instance, in case of N(3,5-tBu₂C₆H₃)₃/B(3,5-tBu₂C₆H₃)₃ a N–B separation of 3.778 Å was obtained (Figure [2]), which is in sharp contrast to the weakly bound classical Lewis pair lutidine/B(C₆F₅)₃ that features a N–B bond length of 1.661(2) Å, as determined single-crystal X-ray crystallography.[10]

Having established the increased stability of the encounter complexes with Lewis acids and Lewis bases containing dispersion energy donors, we aimed at a closer inspection and quantification of the London dispersion forces between the Lewis acidic and Lewis basic fragments since the initial computations only account for the total dispersion within the system, i.e., the intermolecular and intramolecular dispersion combined. Therefore, we performed energy decomposition and fragment analyses (ADF; M06-2X-D3/TZ2P) disclosing the interaction energies between the FLP components (Table [2]). Again, the FLPs R₃P/B(C₆F₅)₃ (R = Mes, tBu) were included as reference.

Table 2 Calculated Interaction Energies (kcal mol^–1) between Lewis Acid and Lewis Base in the Encounter Complexes of Studied FLPs^a

Lewis base	Lewis acid	E int,tot	E int,Disp
PMes3	B(C6F5)3	–13.17	– 3.89
PtBu3	B(C6F5)3	–13.90	– 3.14
N(3,5-tBu2C6H3)3	B(3,5-tBu2C6H3)3	–40.44	–11.45
	B(3,5-Ph2C6H3)3	–40.76	–11.68
	B(3,5-(CF3)2C6H3)3	–32.59	– 8.52
	B(C6F5)3	–20.89	– 6.31
N(3,5-Ph2C6H3)3	B(3,5-tBu2C6H3)3	–41.00	–11.79
	B(3,5-Ph2C6H3)3	–47.32	–11.83
	B(3,5-(CF3)2C6H3)3	–34.82	– 8.19
	B(C6F5)3	–26.79	– 6.42

^a Calculated at the M06-2X-D3/TZ2P level of theory.

Exchanging the phosphine of FLP systems R₃P/B(C₆F₅)₃ (R = Mes –3.89 kcal mol^–1, tBu –3.14 kcal mol^–1) for all-meta-substituted amines N(3,5-tBu₂C₆H₃)₃ (–6.31 kcal mol^–1) and N(3,5-Ph₂C₆H₃)₃ (–6.42 kcal mol^–1) leads to a doubling of the amount of London dispersion stabilization between the FLP components in the encounter complex. Altering the Lewis acid to B(3,5-(CF₃)₂C₆H₃)₃ (–8.52 (tBu), –8.19 (Ph) kcal mol^–1) only shows a marginal increase in dispersion interactions compared to B(C₆F₅)₃, whereas employing all-meta-substituted boranes B(3,5-tBu₂C₆H₃)₃ (–11.45 (tBu), –11.68 (Ph) kcal mol^–1) and B(3,5-Ph₂C₆H₃)₃ (–11.79 (tBu), –11.83 (Ph) kcal mol^–1) effects a further doubling of the amount of London dispersion stabilization. The calculations moreover revealed no significant differences in the dispersion energy for the all-meta-Ph and all-meta-tBu systems making these combinations equally promising candidates for experimental investigations. By installing dispersion energy donors on both Lewis acid and Lewis base, a total increase in the dispersion stabilization of approximately 8 kcal mol^–1 was achieved according to the gas-phase computations. Recent studies demonstrated that dispersion interactions are attenuated in solution to a large extent (about 70%),[11] which results in an expected increased stabilization of the encounter complex comprising all-meta-substituted triarylamines and boranes in solution of approximately 2.5 kcal mol^–1 via interfragment dispersion interactions. This stabilization of the encounter complex is still significant and could lead to a shift in the equilibrium and an increased encounter complex concentration in solution.

Since the interfragment interactions in the N(3,5-R₂C₆H₃)₃/B(3,5-R₂C₆H₃)₃ systems mostly correspond to London dispersion forces with negligible contributions from the N–B interaction, we furthermore calculated the formation energies of the homodimers N(3,5-R₂C₆H₃)₃/N(3,5-R₂C₆H₃)₃ and B(3,5-R₂C₆H₃)₃/B(3,5-R₂C₆H₃)₃ and examined the fragment interactions by energy decomposition analysis (Table [3]).

Table 3 Homodimer Formation^a and Fragment Interaction^b Energies (kcal mol^–1)

R	Complex	ΔE tot	ΔE Disp	E int,tot	E int,Disp
3,5-tBu2C6H3	NR3/BR3	–39.93	–25.33	–40.44	–11.45
	NR3/NR3	–41.03	–26.09	–39.57	–11.55
	BR3/BR3	–37.54	–23.61	–37.99	–11.31
3,5-Ph2C6H3	NR3/BR3	–54.20	–32.49	–47.32	–11.83
	NR3/NR3	–54.20	–31.14	–45.73	–11.91
	BR3/BR3	–54.00	–33.54	–45.60	–11.90

^a Calculated at the ωB97xD/6-311+G(d,p) level of theory.

^b Calculated at the M06-2X-D3/TZ2P level of theory.

The computational analysis revealed similar formation and dispersion interaction energies for the amine–amine (–11.55 (tBu), –11.91 (Ph) kcal mol^–1) and borane–borane (–11.31 (tBu), –11.90 (Ph) kcal mol^–1) dimers as for the amine–borane systems, suggesting that amine–borane encounter complex and homodimer formation are equally favored.

As a result of the computational analysis, we investigated combinations consisting of all-meta-tBu-substituted triarylamine N(3,5-tBu₂C₆H₃)₃ as the Lewis base and triarylboranes B(3,5-(CF₃)₂C₆H₃)₃ and B(3,5-tBu₂C₆H₃)₃ as Lewis acids in the subsequent experimental investigation.[12] Amine N(3,5-tBu₂C₆H₃)₃ was synthesized according to a literature procedure.[13] For purposes of comparison, single crystals of N(3,5-tBu₂C₆H₃)₃ were grown from pentane at –30 °C and the solid-state structure was determined by single-crystal X-ray structure analysis. The crystal structure of the current pentane solvate is isostructural with the toluene solvate of the corresponding hydrocarbon.[14] The central nitrogen atom and one phenyl ring are located on a twofold axis. The nitrogen geometry is planar with an angle sum of 360.0(3)° and N–C distances of 1.430(3) Å and 1.424(2) Å. The aryl groups are arranged propeller-like with C–N–C–C torsion angles of –39.37(15)° and –35.3(2)°. Two of the tBu groups were refined with a disorder model (Figure [3]). The crystal structure contains no intermolecular N···N distances shorter than 10 Å.

Figure 3 Displacement ellipsoid plot of N(3,5-tBu₂C₆H₃)₃ in the crystal (50% probability level). Only the major disorder component is shown. Hydrogen atoms and pentane solvent molecule are omitted for clarity. Symmetry code i: -x, y, ½-z.

B(3,5-(CF₃)₂C₆H₃)₃ was obtained following a literature procedure,[15] whereas B(3,5-tBu₂C₆H₃)₃ could not be obtained in an analogous manner due to the formation of complex product mixtures from which the borane could not be isolated. However, the synthesis of B(3,5-tBu₂C₆H₃)₃ was achieved via a modified synthetic route for BPh₃ developed by Lammertsma et al.[16] First, Na[B(3,5-tBu₂C₆H₃)₄] (¹¹B NMR: δ = –4.72) was synthesized by reaction of NaBF₄ with four equivalents of Li(3,5-tBu₂C₆H₃). Subsequently, [Me₃NH][B(3,5-tBu₂C₆H₃)₄] was generated in situ by cation exchange using [Me₃NH]Cl, which spontaneously eliminates NMe₃ and 1,3-tBu₂C₆H₄ with formation of B(3,5-tBu₂C₆H₃)₃ upon treatment with THF.[17] It was found, though, that B(3,5-tBu₂C₆H₃)₃ is highly susceptible towards decomposition upon workup of the reaction mixture. Isolation of B(3,5-tBu₂C₆H₃)₃ was achieved by addition of 2,2,6,6-tetramethylpiperidine to the THF solution to scavenge any residual protons, removal of volatiles, and extraction into Et₂O, from which colorless X-ray quality single crystals were obtained after storage at –30 °C (Scheme [1]).[18]

Borane B(3,5-tBu₂C₆H₃)₃ features the expected signals for the aryl groups in its ¹H and ¹³C{¹H} NMR spectra. The ¹¹B NMR spectrum exhibits a broad resonance at 68.4 ppm, comparable to that of BPh₃ (67.8 ppm).[16] The molecular structure of B(3,5-tBu₂C₆H₃)₃ determined by X-ray crystal-structure determination features the expected trigonal-planar coordination geometry at B with an angle sum of 360.1(4)°. Here, the boron atom is on a general position without symmetry. The propeller-like arrangement of the aryl substituents can be seen in the C–B–C–C torsion angles of –31.4(4)°, –32.6(4), and –34.7(4)°. The B–C bond lengths are in the range of 1.558(4)–1.570(4) Å (Figure [4]). The overall geometry is highly similar to that of N(3,5-tBu₂C₆H₃)₃. Again, there are no intermolecular B···B distances shorter than 10 Å.

Scheme 1 Synthesis of B(3,5-tBu₂C₆H₃)₃

Figure 4 Displacement ellipsoid plot of B(3,5-tBu₂C₆H₃)₃ in the crystal (50% probability level). Only the major disorder component is shown. Hydrogen atoms and diethyl ether solvent molecule are omitted for clarity.

With the Lewis acids and Lewis base in hand, encounter complex formation was investigated. Mixing solutions of N(3,5-tBu₂C₆H₃)₃ and B(3,5-(CF₃)₂C₆H₃)₃ in toluene or dichloromethane initially yielded a pale yellow solution which gradually turned dark blue over the course of several hours. As the dark blue color is characteristic of triarylamine radical cations, EPR analysis was conducted. The X-band EPR spectrum (Figure [5]) of this solution revealed a signal featuring a multitude of hyperfine coupling interactions characteristic for an organic radical in a doublet spin system, indicating formation of ^+•N(3,5-tBu₂C₆H₃)₃.

Figure 5 Simulated (red) and experimental (black) X-band EPR spectrum of ^+•N(3,5-tBu₂C₆H₃)₃ in dichloromethane (capillary) at room temperature. Experimental parameters; microwave frequency 9.388019 GHz, power 0.7962 mW, modulation amplitude; 1.000 G. Simulation parameters; S = ½, g _iso = 2.0042, A ^14N _iso = 21.5777 MHz, 6 × A ^1H-ortho _iso = –6.0593, 3 × A ^1H-para _iso = –15.4771, lw = 0.067165.

Satisfactory simulation of the experimental spectrum was achieved with g _iso = 2.0042 and inclusion of hyperfine coupling interactions with nitrogen (A ^14N _iso = 21.58 MHz), six equivalent protons (ortho-protons on the aryl moieties of the amine; A ^1H-ortho _iso = –6.06 MHz), and another three equivalent protons (para protons on the aryl moieties of the amine; A ^1H-para _iso = –15.48 MHz) consistent with the presence of ^+•N(3,5-tBu₂C₆H₃)₃. Moreover, the calculated EPR parameters for ^+•N(3,5-tBu₂C₆H₃)₃ by DFT (g _iso = 2.0030, A ^14N _iso = 21.20 MHz, A ^1H-ortho _iso = –6.18 MHz, A ^1H-para _iso = –9.18 MHz) are in reasonable agreement with the simulated data. Further support was obtained by independent generation of ^+•N(3,5-tBu₂C₆H₃)₃ via the oxidation of N(3,5-tBu₂C₆H₃)₃ with Cu(ClO₄)₂ [19] and simulation of the obtained spectra (Figures S9, S10), which can be achieved with the same hyperfine coupling constants and a g _iso value of 2.0028. The formation of ^+•N(3,5-tBu₂C₆H₃)₃ in the reaction of N(3,5-tBu₂C₆H₃)₃ and B(3,5-(CF₃)₂C₆H₃)₃ is in line with previous reports on the one-electron oxidation of triarylamines by B(C₆F₅)₃ [20] and our investigation on photoinduced SET in FLP systems,[6] driven by rapid decomposition of the corresponding triarylborane radical anion via solvolytic pathways.[21] This photolability is expected to complicate characterization and isolation of the corresponding encounter complex. Crystallization of an encounter complex with exclusion of light proved unsuccessful. Therefore, we focused on B(3,5-tBu₂C₆H₃)₃ as Lewis acid. The decreased Lewis acidity and electron affinity of B(3,5-tBu₂C₆H₃)₃ leads to a large energy gap for the excitation of the charge-transfer complex to the corresponding radicals (ΔE = 4.26 eV (291 nm); Table S1) preventing visible-light-induced SET.

Mixing solutions of N(3,5-tBu₂C₆H₃)₃ and B(3,5-tBu₂C₆H₃)₃ in toluene produced a colorless solution and ¹H NMR spectroscopic analysis only showed the presence of the starting materials as typically observed for FLP systems.[22] Attempts to observe a potential pre-organization of the FLP in solution by ¹H,¹H NOESY NMR studies revealed no through-space correlations between the protons of the Lewis acid and the Lewis base. This could be a result of the limited solubility of the FLP and its components precluding analysis of highly concentrated solutions as used by Rocchigiani and co-workers.[5] Moreover, homodimer association could further reduce the effective concentration of Lewis acid–Lewis base couples in solution. However, due the limited solubility, crystallization of colorless crystals was achieved from both toluene and n-pentane solutions.[23] The X-ray crystal-structure determination of the N(3,5-tBu₂C₆H₃)₃/B(3,5-tBu₂C₆H₃)₃ crystals showed that these crystals are isostructural to the pentane solvate of N(3,5-tBu₂C₆H₃)₃ described above (see the Supporting Information). In addition to sharp Bragg reflections these crystals showed diffuse peaks for reflections with ℓ = odd (Figure S15). Due to similar X-ray scattering factors of nitrogen and boron, it was not possible to assign these centers unambiguously and thus resolve substitutional disorder of the central atom, which is why the occupancy was set to ½ for both elements. The formation of a co-crystal is supported by NMR spectroscopic analysis of the isolated crystals showing a 1:1 mixture of N(3,5-tBu₂C₆H₃)₃ and B(3,5-tBu₂C₆H₃)₃ (Figures S11, S12), which proves the presence of both components in the crystal lattice. Moreover, the IR spectrum of the isolated crystals (Figure S13) exhibits features found in the IR spectra of both separate components (Figures S1, S8), while the melting point (224 °C) lies between those of N(3,5-tBu₂C₆H₃)₃ (201 °C) and B(3,5-tBu₂C₆H₃)₃ (255 °C). The random stacking of both components in the crystal lattice yielding an overall 1:1 ratio of Lewis acid and Lewis base most likely results from the almost identical shapes of N(3,5-tBu₂C₆H₃)₃ and B(3,5-tBu₂C₆H₃)₃ and the computed similar hetero- and homofragment dispersion interaction energies between the components of the FLP (Table [3]) resulting in no significant driving force towards an alternating pattern. Still, this result emphasizes the importance of London dispersion forces in FLP chemistry and the possibility of crystallizing the encounter complex.

On the basis of computational analyses, which showed that the London dispersion forces between the components of FLP systems can be significantly enhanced by judicious choice of the substituents at the Lewis acid and Lewis base, we investigated the formation of encounter complexes in two amine–borane FLP systems experimentally. In case of N(3,5-tBu₂C₆H₃)₃/B(3,5-(CF₃)₂C₆H₃)₃ it was shown that photoinduced SET can be generally observed in FLPs with matching electron affinities and ionization potentials. For the N(3,5-tBu₂C₆H₃)₃/B(3,5-tBu₂C₆H₃)₃ FLP system a co-crystal containing both components was obtained, which showed positional disorder of the B and N centers due to random arrangement of the individual components in the crystal. This moreover showed the potential for homodimer formation in dispersion interaction-governed systems. This work represents an important contribution to the structural characterization of an intermolecular FLP and the structural verification of the concept of the encounter complex in FLP chemistry. Future investigations will focus on the use of dissimilar Lewis acids and Lewis bases to favor heterofragment interactions and ultimately allow the unambiguous identification of an encounter complex in the crystalline state.

• References and Notes

• 1 Welch GC, Juan RR. S, Masuda JD, Stephan DW. Science 2006; 314: 1124
• 2a Stephan DW, Erker G. Angew. Chem. Int. Ed. 2010; 49: 46
• 2b Stephan DW. Acc. Chem. Res. 2015; 48: 306
• 2c Stephan DW. J. Am. Chem. Soc. 2015; 137: 10018
• 2d Stephan DW. Science 2016; 354: aaf7229
• 2e Paradies J. Coord. Chem. Rev. 2019; 380: 170
• 2f Jupp AR, Stephan DW. Trends Chem. 2019; 1: 35
• 3a Paradies J. Eur. J. Inorg. Chem. 2019; 283
• 3b Jupp AR. Dalton Trans. 2022; 51: 10681
• 4a Rokob TA, Hamza A, Stirling A, Soós T, Pápai I. Angew. Chem. Int. Ed. 2008; 47: 2435
• 4b Bistoni G, Auer AA, Neese F. Chem. Eur. J. 2017; 23: 865
• 4c Grimme S, Kruse H, Goerigk L, Erker G. Angew. Chem. Int. Ed. 2010; 49: 1402
• 4d Skara G, Pinter B, Top J, Geerlings P, De Proft F, De Vleeschouwer F. Chem. Eur. J. 2015; 21: 5510
• 4e Zeonjuk LL, Vankova N, Mavrandonakis A, Heine T, Röschenthaler GV, Eicher J. Chem. Eur. J. 2013; 19: 17413
• 5 Rocchigiani L, Ciancaleoni G, Zuccaccia C, Macchioni A. J. Am. Chem. Soc. 2014; 136: 112
• 6 Holtrop F, Jupp AR, van Leest NP, Paradiz Dominguez M, Williams RM, Brouwer AM, de Bruin B, Ehlers AW, Slootweg JC. Chem. Eur. J. 2020; 26: 9005
• 7 A π–π-stacked co-crystal of 1-Br-4-(NMe₂)C₆H₄ and B(C₆F₅)₃ was reported: Aramaki Y, Imaizumi N, Hatta M, Kumagai J, Ooi T. Chem. Sci. 2020; 11: 4305
• 8a Rösel S, Balestrieri C, Schreiner PR. Chem. Sci. 2017; 8: 405
• 8b Rösel S, Becker J, Allen WD, Schreiner PR. J. Am. Chem. Soc. 2018; 140: 14421
• 9 Holtrop F, Jupp AR, Kooij BJ, van Leest NP, de Bruin B, Slootweg JC. Angew. Chem. Int. Ed. 2020; 59: 22210
• 10 Geier SJ, Stephan DW. J. Am. Chem. Soc. 2009; 131: 3476
• 11a Pollice R, Bot M, Kobylianskii IJ, Shenderovich I, Chen P. J. Am. Chem. Soc. 2017; 139: 13126
• 11b Schümann JM, Wagner JP, Eckhardt AK, Quanz H, Schreiner PR. J. Am. Chem. Soc. 2021; 143: 41
• 12 The syntheses of all-meta-Ph-substituted triarylamine and triarylborane were attempted. However, only in common organic solvents insoluble products were obtained precluding purification, characterization, and further investigations.
• 13 Bedford RB, Betham M. Tetrahedron Lett. 2007; 48: 8947
• 14 Kahr B, Van Engen D, Gopalan P. Chem. Mater. 1993; 5: 729
• 15 Kolychev EL, Bannenberg T, Freytag M, Daniliuc CG, Jones PG, Tamm M. Chem. Eur. J. 2012; 18: 16938
• 16 Borger JE, Ehlers AW, Lutz M, Slootweg JC, Lammertsma K. Angew. Chem. Int. Ed. 2015; 55: 613
• 17 Osi A, Mahaut D, Tumanov N, Fusaro L, Wouters J, Champagne B, Chardon A, Berionni G. Angew. Chem. Int. Ed. 2022; 61: e202112342
• 18 Synthesis of B(3,5-tBu₂C₆H₃)₃ [Me₃NH]Cl (50 mg, 0.55 mmol) was added to a solution of Na[B(3,5-tBu₂C₆H₃)₄] (396 mg, 0.5 mmol) in MeCN (10 mL) at ambient temperature. The white suspension was stirred for 3 h and subsequently filtered over Celite. Volatiles were removed from the filtrate in vacuo. The resulting white solid was dissolved in THF (40 mL), and 2,2,6,6-tetramethylpiperidine (78 mg, 0.55 mmol, 0.093 mL) was added at ambient temperature. The mixture was stirred for 1 h, after which volatiles were removed in vacuo. The residue was extracted into diethyl ether (30 mL), and the filtrate was concentrated until incipient crystallization and stored at –30 °C to yield colorless analytically pure crystals of B(3,5-tBu₂C₆H₃)₃. Yield: 260 mg (0.45 mmol, 90%); mp 255 °C. HRMS (LIFDI): m/z calcd for C₄₂H₆₃B: 578.5030; found: 578.5028. IR (neat): ν = 2952, 2902, 2866, 1584, 1475, 1460, 1412, 1390, 1361, 1309, 1247, 1220, 993, 895, 885, 822, 725, 671, 539 cm^–1. ¹H NMR (500 MHz, CDCl₃): δ = 7.60 (s, 3 H, p-CH), 7.51 (s, 6 H, o-CH), 1.35 (s, 54 H, tBu). ¹³C{¹H} NMR (125.6 MHz, CDCl₃): δ = 148.9 (m-C ₆H₃ tBu₂), 133.7 (o-C ₆H₃ tBu₂), 125.0 (p-C ₆H₃ tBu₂), 35.0 (C(CH₃)₃), 31.8 (C(CH₃)₃) (ipso-C ₆H₃ tBu₂ not observed). ¹¹B NMR (160.5 MHz, CD₃CN): δ = 68.4 (very br).
• 19 Sreenath K, Suneesh CV, Gopidas KR, Flowers RA. II. J. Phys. Chem. A 2009; 113: 6477
• 20 Zheng X, Wang X, Qiu Y, Li Y, Zhou C, Sui Y, Li Y, Ma J, Wang X. J. Am. Chem. Soc. 2013; 135: 14912
• 21 Lawrence EJ, Oganesyan VS, Wildgoose GG, Aschley AE. Dalton Trans. 2013; 42: 782
• 22 Welch GC, Stephan DW. J. Am. Chem. Soc. 2007; 129: 1880
• 23 Synthesis of N(3,5-tBu₂C₆H₃)₃/B(3,5-tBu₂C₆H₃)₃ N(3,5-tBu₂C₆H₃)₃ (10 mg, 0.017 mmol) and B(3,5-tBu₂C₆H₃)₃ (10 mg, 0.017 mmol) were dissolved in toluene-d ₈ (0.5 mL). Storage of the solution at –30 °C overnight afforded colorless crystals of N(3,5-tBu₂C₆H₃)₃/B(3,5-tBu₂C₆H₃)₃. Yield: 16 mg (0.014 mmol, 80%); mp 224 °C. IR (neat): ν = 2951, 2902, 2866, 1582, 1495, 1476, 1461, 1447, 1426, 1413, 1391, 1361, 1307, 1246, 1221, 1204, 1130, 1038, 1008, 993, 924, 896, 886, 865, 822, 727, 720, 711, 694, 671, 539, 463 cm^–1. ¹H NMR (500 MHz, C₆D₆): δ 8.03 (s, 3 H, p-CH), 7.86 (s, 6 H, o-CH), 7.46 (s, 3 H, p-CH), 7.27 (s, 6 H, o-CH), 1.32 (s, 54 H, tBu), 1.23 (s, 54 H, tBu).