Synthesis of Electrophiles Derived from Dimeric Aminoboranes and Assessing Their Utility in the Borylation of π Nucleophiles

Dimeric aminoboranes, [H2BNR2]2 (R = Me or CH2CH2) containing B2N2 cores, can be activated by I2, HNTf2 (NTf2 = [N(SO2CF3)2]), or [Ph3C][B(C6F5)4] to form isolable H2B(μ-NR2)2BHX (for X = I or NTf2). For X = [B(C6F5)4]− further reactivity, presumably between [H2B(μ-NMe2)2BH][B(C6F5)4] and aminoborane, forms a B3N3-based monocation containing a three-center two electron B-(μ-H)-B moiety. The structures of H2B(μ-NMe2)2BH(I) and [(μ-NMe2)BH(NTf2)]2 indicated a sterically crowded environment around boron, and this leads to the less common O-bound mode of NTf2 binding. While the iodide congener reacted very slowly with alkynes, the NTf2 analogues were more reactive, with hydroboration of internal alkynes forming (vinyl)2BNR2 species and R2NBH(NTf2) as the major products. Further studies indicated that the B2N2 core is maintained during the first hydroboration, and that it is during subsequent steps that B2N2 dissociation occurs. In the mono-boron systems, for example, iPr2NBH(NTf2), NTf2 is N-bound; thus, they have less steric crowding around boron relative to the B2N2 systems. Notably, the monoboron systems are much less reactive in alkyne hydroboration than the B2N2-based bis-boranes, despite the former being three coordinate at boron while the latter are four coordinate at boron. Finally, these B2N2 electrophiles are much more prone to dissociate into mono-borane species than pyrazabole [H2B(μ-N2C3H3)]2 analogues, making them less useful for the directed diborylation of a single substrate.


General considerations
All reactions were performed under inert conditions using standard Schlenk techniques or in an MBraun Unilab glovebox (<0.1 ppm H2O / O2).
Unless otherwise stated, solvents were degassed with nitrogen, dried over activated aluminium oxide (Solvent Purification System: Inert PureSolv MD5 SPS) and stored over 3 Å molecular sieves in ampules equipped with J. Young's valves. Chlorobenzene, 1,2difluorobenzene and 1,2-dichlorobenzene were dried over calcium hydride, distilled and stored over 3 Å molecular sieves. Deuterated solvents (CDCl3, C6D6 and C6D5Br (99.6% D, Sigma Aldrich)) were dried and stored over 3 Å molecular sieves. All chemicals were, unless stated otherwise, purchased from commercial sources and used as received. BH3·SMe2 was transferred to an ampule fitted with a J. Young's valve prior to use.
[Ph3C][B(C6F5)4] and lithium 2-t Bu-dihydropyridine were synthesized following literature procedure. S1,S2 NMR spectra ( 1 H, 1 H{ 11 B}, 11 B, 11 B{ 1 H}, 13 C{ 1 H} and 19 F) were recorded on Bruker Avance III 400 MHz, Bruker Avance III 500 MHz, Bruker Avance III 600 MHz or Bruker PRO 500 MHz spectrometers. Chemical shifts (δ) are quoted in parts per million (ppm), coupling constants (J) are given in hertz (Hz) to the nearest 0.5 Hz, and as positive vales regardless of their real individual signs. 1 H and 13 C shifts are referenced to the appropriate residual solvent peak while 11 B and 19 F shifts are referenced relative to external BF3·Et2O and C6F6, respectively. Abbreviations used are s (singlet), d (doublet), t (triplet), q (quartet), sep (septet), m (multiplet), br (broad). Background signals in 11 B NMR spectra arise to a significant degree from glass components of the probes used in our spectrometers. Unless otherwise stated, all NMR spectra were recorded at 20 °C.
Mass spectrometry was performed by the Scottish Instrumentation and Resource Centre for Advanced Mass Spectrometry (SIRCAMS) at the University of Edinburgh using electron impact (EI) or electrospray ionisation (ESI) techniques.
CHN Elemental Analyses were carried out by Elemental Microanalysis Ltd.

Preparation of H2B
HNTf2 (1.18 g, 4.18 mmol, 1.00 equiv.) was dissolved in benzene (10 mL) and slowly added to a solution of [Me2NBH2]2 (0.50 g, 4.39 mmol, 1.05 equiv.) in benzene (10 mL) at room temperature. The solution was stirred at room temperature for 24 hours. The volatiles were removed under vacuum. The product was extracted with pentane (10 mL). Drying under vacuum for 1 hour afforded the product 4 as a colourless oil in 83% yield (1.37 g, 3.48 mmol). 2.5 Preparation of H2B(-pyrrolidine)2BH(OSOCF3NTf) 5 HNTf2 (1.21 g, 4.30 mmol, 1.00 equiv.) was dissolved in benzene (10 mL) and slowly added to a solution of [Pyrrolidine-BH2]2 (0.75 g, 4.52 mmol, 1.05 equiv.) in benzene (10 mL) at room temperature. The solution was stirred at room temperature for 20 hours. The volatiles were removed under vacuum. The product was extracted with pentane (10 mL). Drying under vacuum overnight at 40 °C afforded the product 5 as a colourless oil in 65% yield (1.24 g, 2.79 mmol). Very slow decomposition of 5 was observed over months, even when kept under inert atmosphere.
b Further NMR experiments were carried out for the 1 H NMR characterisation of 7. 1 H NMR experiment at 278 K allowed assignment of the different N(CH3) signals. Hydrogen atoms located on boron centres (BH2, HB(H)BH and HB(H)BH) were identified by 2-D 11 B-1 H HMQC experiment.

Generation of the monomeric borocation ( i Pr)2NBHNTf2 9
A solution of HNTf2 (1.24 g, 4.40 mmol, 1 equiv.) in benzene (10 mL) was added to a solution of ( i Pr)2NBH2 (0.50 g, 4.40 mmol, 1.00 equiv.) in benzene (10 mL) at room temperature. The solution was heated at 70 °C in an unpressurised system (open Schlenk under N2) for 6 days and monitored periodically by NMR spectroscopy. Very slow conversion of the adduct 8 to the product 9 was observed and after 6 days the reaction mixture was filtered, despite not being completed (2.2:1, 8:9). Volatiles were removed under vacuum affording a waxy colourless solid. The solid was dissolved in pentane, cooled down to -78 °C and filtered at this temperature. Removal of pentane fraction afforded the product as a volatile colourless crystalline solid, contaminated with only trace of impurities (however, the impurities in the 19F are sufficient that it is not appropriate to report a yield). Partial decomposition was observed after leaving 9 in solution overnight.
S37 Figure S53: 1 H NMR spectrum (500 MHz, C6D6, 300 K) focusing on the aromatic region of the reaction between diphenylacetylene and 3, with mesitylene as internal standard, after 5 days at 70 °C. The ratio of 10:mesitylene (0.04:1) was calculated by comparison of the integration values between the alkene protons peak from 10 (labelled in red on spectrum) and the aromatic CH peak from mesitylene at 6.72 ppm. H2B(-Me2N)2BH(OSOCF3NTf) 4 (25 mg, 0.06 mmol, 1.00 equiv.), diphenylacetylene (23 mg, 0.12 mmol, 2.00 equiv.) and mesitylene (internal standard) were dissolved in C6D6 (0.5 mL) in an NMR tube and heated at 70 °C for 4.5 days. The in situ conversion to the product 10 was measured to be 95% (conversion was determined by integration of the 1 H NMR spectrum relative to internal standard).
S38 Figure S54: 1 H NMR spectrum (500 MHz, C6D6, 300 K) focussing on the aromatic region of the reaction between diphenylacetylene and 4, with mesitylene as internal standard, before heating. The ratio diphenylacetylene:mesitylene (1.97:1) was calculated by comparison of the integration values between the diphenylacetylene peaks and the aromatic CH peak from mesitylene at 6.72 ppm. Figure S55: 1 H NMR spectrum (500 MHz, C6D6, 300 K) focussing on the aromatic region of the reaction between diphenylacetylene and 4, with mesitylene as internal standard, after 4.5 days at 70 °C. The ratio 10:mesitylene (0.94:1) was calculated by comparison of the integration values between the alkene protons peak from 10 (labelled in red on spectrum) and the aromatic CH peak from mesitylene at 6.72 ppm.

Characterisation of 10
A solution of H2B(-Me2N)2BH(OSOCF3NTf) 4 (0.22 g, 0.56 mmol, 1.00 equiv.) in benzene (1 mL) was added to a solution of diphenylacetylene (0.20 g, 1.12 mmol, 2.00 equiv.) in benzene (1 mL). The resulting solution was heated at 70 °C for 1 week. While heating, the solution turned slowly from colourless to dark orange. The volatiles were removed under vacuum, affording an oil. The oil was extracted with pentane (2 mL) giving a clear orange solution. The solution was cooled down to -35 °C and filtered at this temperature. Removal of volatiles in vacuo afforded the product 10 as an orange oil (0.11 g), contaminated with remaining trace of -NTf2 side-products. a The carbon signal of C F was hidden by the solvent peaks, its assignment was done using a HSQC experiment.    H2B(-Me2N)2BH(OSOCF3NTf) 4 (25 mg, 0.06 mmol, 1.00 equiv.), 1-phenyl-1-propyne (16 µL, 0.12 mmol, 2.00 equiv.) and mesitylene (internal standard) were dissolved in C6D6 (0.5 mL) in an NMR tube and heated at 70 °C for 5 days. The in situ conversion to the product 12a-c was measured to be 81% (conversion was determined by integration of the 1 H NMR relative to internal standard), with a ratio of 29:54:17% (12a:12b:12c). Figure S61: 1 H NMR spectrum (500 MHz, C6D6, 300 K) focussing on the aromatic region of the reaction between 1-phenyl-1-propyne and 4, with mesitylene as internal standard, before heating. The ratio 1-phenyl-1-propyne:mesitylene (1.89:1) was calculated by comparison of the integration values between the 1-phenyl-propyne multiplet at 7.5-7.4 ppm and the aromatic CH peak from mesitylene at 6.72 ppm. Figure S62: 1 H NMR spectrum (500 MHz, C6D6, 300 K) focussing on the aromatic region of the reaction between 1-phenyl-1-propyne and 4, with mesitylene as internal standard, after 5 days at 70 °C. The ratio 12a-c:mesitylene (0.77:1) was calculated by comparison of the integration values between the alkene protons peaks from 12a-c (labelled in red on spectrum) and the aromatic CH peak from mesitylene at 6.72 ppm. Figure S63: 1 H NMR spectrum (500 MHz, C6D6, 300 K) of the reaction between 1-phenyl-1-propyne and 4, with mesitylene as internal standard, after 5 days at 70 °C. Ratio of the different isomers 12a-c was determined by relative integration of the signal from the N(CH3)2 moieties.

Characterisation of isomers of 12a-c
To a solution of H2B(-Me2N)2BH(OSOCF3NTf) 4 (0.20 g, 0.50 mmol, 1.00 equiv.) in benzene (2 mL), was added 1-phenyl-1-propyne (130 µL, 1.00 mmol, 2.00 equiv.) using a micro syringe. The resulting solution was heated at 70 °C for 1 week. Volatiles were removed under vacuum, affording an oil. The oil was extracted with pentane (2 mL) affording a clear orange solution. The solution was cooled down to -35 °C and filtered at this temperature. Removal of volatiles in vacuo afforded the product 12a-c as an orange oil (0.018 g) contaminated with remaining trace of side-products. a Each isomer exhibits a boron signal in similar region of the spectrum, it was not possible to assign the boron signals individually to their respective isomer.
b The C-B and the aromatic carbons signals could not be assigned, even by 2-D NMR experiments, due to the overlapping of the aromatic proton and aromatic carbon signals of the different isomers and impurities.

Formation of 6
To a solution of H2B(-Me2N)2BH(OSOCF3NTf) 4 (0.20 g, 0.50 mmol, 1.00 equiv.) in benzene (2 mL), was added 1-phenyl-1-propyne (130 µL, 1.00 mmol, 2.00 equiv.) using micro syringe. The resulting solution was heated at 70 °C for 1 week. Volatiles were removed under vacuum, affording an oil. The oil was extracted with pentane (2 mL) affording a clear orange solution. The solution was cooled down to 0 °C. Colourless needles started to crystallise out of the solution. After one hour at 0 °C, the crystals were isolated by filtration, affording the by-product 11 in its dimeric form 6 as colourless crystals in extremely small quantity, contaminated with other impurities.

S58
6 Crystallographic data 6.1 Crystal structure of 3 CCDC Deposition Number: 2192971 Experimental: Single colourless plate-shaped crystals of 3 recrystallised by sublimation. A suitable crystal with dimensions 0.20 × 0.09 × 0.05 mm 3 was selected and mounted on a MITIGEN holder in Paratone oil on a Rigaku Oxford Diffraction Excalibur diffractometer equipped with Eos CCD detector. The crystal was kept at a steady T = 120.0(2) K during data collection. The structure was solved with the ShelXS S8 (Sheldrick, 2008) solution program using dual methods and by using Olex2 1.5-beta S9 as the graphical interface. The model was refined with ShelXL 2018/3 S10 using full matrix least squares minimisation on F 2 . Experimental: Single colourless plate-shaped crystals of 6 recrystallised from pentane by slow cooling. A suitable crystal with dimensions 0.50 × 0.23 × 0.12 mm 3 was selected and mounted on a MITIGEN holder in Paratone oil on a Rigaku Oxford Diffraction XCalibur diffractometer. The crystal was kept at a steady T = 120.00(10) K during data collection.
The structure was solved with the ShelXT 2018/2 S11 solution program using dual methods and by using Olex2 1.5-beta S9 as the graphical interface. The model was refined with ShelXL 2018/3 S10 using full matrix least squares minimisation on F 2 . Experimental: Single colourless crystals of 7 were obtained by layering C6H5Cl with pentane. A suitable crystal with dimensions 0.162 x 0.133 x 0.09 mm 3 was selected and mounted on a MITIGEN holder in Paratone oil on a Rigaku Oxford Diffraction SuperNova diffractometer. The crystal was kept at a steady T = 120.01(10) K during data collection.
The structure was solved with the ShelXT S11 solution program using dual methods and by using Olex2 1.5-beta S9 as the graphical interface. The model was refined with ShelXL 2018/3 S10 using full matrix least squares minimisation on F 2 . Experimental: Single colourless plate-shaped crystals of 9 recrystallised by sublimation. A suitable crystal with dimensions 0.30 × 0.15 × 0.05 mm 3 was selected and mounted on a MITIGEN holder in Paratone oil on a Rigaku Oxford Diffraction SuperNova diffractometer. The crystal was kept at a steady T = 100.0 K during data collection. The structure was solved with the ShelXT S11 solution program using dual methods and by using Olex2 1.5-beta S9 as the graphical interface. The model was refined with ShelXL 2018/3 S10 using full matrix least squares minimisation on F 2 . recrystallised from hexane by slow evaporation. A suitable crystal with dimensions 0.08 × 0.06 × 0.03 mm 3 was selected and mounted on a MITIGEN holder in Paratone oil on a Rigaku Oxford Diffraction SuperNova diffractometer. The crystal was kept at a steady T = 120.01(10) K during data collection. The structure was solved with the ShelXT 2014/5 S11 solution program using dual methods and by using Olex2 1.5-beta S9 as the graphical interface. The model was refined with ShelXL 2018/3 S10 using full matrix least squares minimisation on F 2 .
All calculations were performed using the Gaussian09 programme. S12 Geometries optimisation were completed with the DFT method using the M06-2X functional S13 and the 6-311G(d,p) as a basis set. All geometry optimizations were full, with no restrictions. Stationary points located in the potential energy surface were characterized as minima by vibrational analysis. Solvent effects of the dichloromethane were introduced using the self consistent field approach, by means of the integral equation formalism polarizable continuum model (IEFPCM). S14