The synthesis, properties, and reactivity of Lewis acidic aminoboranes

Article abstract of DOI:10.1039/d1ob00520k

The evolution of frustrated Lewis pair chemistry has led to significant research into the development of new Lewis acidic boranes. Much of this has focused on modifying aryl substituents rather than introducing heteroatoms bound to boron. We recently reported that bis(pentafluorophenyl)phenothiazylborane (1) could be used as a Lewis acid catalyst for the heterolytic dehydrocoupling of stannanes. In this work, we synthesize and characterize a family of Lewis acidic aminoboranes and explored their reactivity with various Lewis bases as well as their efficacy as catalysts for stannane dehydrocoupling and hydrosilylation. Quantum chemical calculations were undertaken to understand the origins of the Lewis acidity and the most Lewis acidic aminoborane (5) was found to be an effective catalyst even in coordinating solvents such as water or acetonitrile, suggesting the amino substituent provides a level of protection against competing donors.

Full text of DOI:10.1039/d1ob00520k

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
The synthesis, properties, and reactivity of Lewis
acidic aminoboranes†
Cite this: Org. Biomol. Chem., 2021,
19, 4796
Jordan N. Bentley,
Christopher B. Caputo
Selvyn A. Simoes, Ekadashi Pradhan, Tao Zeng
*
and
The evolution of frustrated Lewis pair chemistry has led to significant research into the development of
new Lewis acidic boranes. Much of this has focused on modifying aryl substituents rather than introducing
heteroatoms bound to boron. We recently reported that bis(pentafluorophenyl)phenothiazylborane (1)
could be used as a Lewis acid catalyst for the heterolytic dehydrocoupling of stannanes. In this work, we
synthesize and characterize a family of Lewis acidic aminoboranes and explored their reactivity with
various Lewis bases as well as their efficacy as catalysts for stannane dehydrocoupling and hydrosilylation.
Quantum chemical calculations were undertaken to understand the origins of the Lewis acidity and the
most Lewis acidic aminoborane (5) was found to be an effective catalyst even in coordinating solvents
such as water or acetonitrile, suggesting the amino substituent provides a level of protection against com-
peting donors.
Received 17th March 2021,
Accepted 4th May 2021
DOI: 10.1039/d1ob00520k
rsc.li/obc
Nevertheless, Gellrich was able to overcome this through the
use of a non-innocent pyridonate borane complex to reversibly
Introduction
Tri-coordinate boranes have found diverse application as Lewis activate hydrogen.10 This was attributed to a change in the
acidic
reagents
for
stoichiometric
and
catalytic coordination mode from an anionic alkoxypyridine donor to a
transformations.1–4 The advent of frustrated Lewis pair (FLP) neutral pyridinone ligand.10
chemistry has brought the synthesis of novel Lewis acidic
Aminoboranes have also garnered interest due to their iso-
boranes into the spotlight.5–7 The steric and electronic pro- electronic relationship to olefins.11 Specifically, sterically
perties of tri-coordinate boranes can be modified through encumbered aminoboranes have found applications in organic
simple organometallic transformations, opening up a substan- materials
tial library of available Lewis acids.
as they
have
interesting
photochemical
properties.12–16 In these applications, the steric hinderance
Though they are powerful synthetic tools, many electrophi- prevents the aminoborane from oligomerizing and precludes
lic triarylboranes are sensitive to ambient atmospheric con- any Lewis acidity (Yamaguchi, Fig. 1). However, Erker and co-
ditions and require strict air- and moisture-free storage and workers reported the use of pentafluorophenyl substituents on
reaction conditions. Heteroatoms, such as oxygen or nitrogen, boron to increase the Lewis acidity of aminoboranes, synthe-
can be introduced adjacent to the boron centre, however this sizing (N-pyrrolyl)B(C6F5)2 and the saturated (N-pyrrolidinyl)B
often leads to a substantial reduction in Lewis acidic character.
These heteroatoms can inductively withdraw electron density;
though their lone pairs can donate electron density back into
the empty p-orbital, increasing the bond order and reducing
the Lewis acidity. From an FLP perspective, pentafluorophenyl
esters of borinic, boronic and boronate derivatives have been
shown to exhibit Lewis acidity towards harder bases,8 however
they show limited efficacy in hydrogen activation.9
York University, Department of Chemistry, 4700 Keele St., Toronto, ON, Canada,
M3J 1P3. E-mail: caputo@yorku.ca
†Electronic supplementary information (ESI) available: Stannanes are toxic and
care should be taken when working with them, including careful disposal of
waste. CCDC 2068070–2068072, 2068074 and 2068077. For ESI and crystallo-
Fig. 1 Examples of sterically encumbered aminoboranes (Ar = Mes,
graphic data in CIF or other electronic format see DOI: 10.1039/d1ob00520k
MesF9).
4796 | Org. Biomol. Chem., 2021, 19, 4796–4802
This journal is © The Royal Society of Chemistry 2021

View Article Online
Organic & Biomolecular Chemistry
Paper
(C6F5)2 analogue (Fig. 1).17,18 They illustrated that the pyrrolyl
derivative proved to be Lewis acidic enough to abstract a
methyl group from a zirconocene complex, whereas the satu-
rated pyrrolidinyl aminoborane had limited Lewis acidity due
to the greater BvN character. Mitzel and co-workers explored
the behaviour of the electrophilic aminoboranes (C6F5)2BNMe2
and (CF3)2BNMe2, reporting that the latter will undergo an
insertion with diazomethane whereas the pentafluorophenyl
derivative was unreactive (Fig. 1).19
We previously reported the synthesis and Lewis acidic
behaviour of (N-phenothiazyl)B(C6F5)2 aminoborane (1), that
demonstrated unique reactivity as an intramolecular frustrated
Lewis pair for the catalytic dehydrocoupling of stannanes.20
Given this context, we sought to expand the scope of amino-
boranes studied with electron withdrawing pentafluorophenyl
substituents and herein examine several aminoboranes con-
taining the N–B(C6F5)2 core. We computationally and experi-
mentally investigated their Lewis acidic properties contrasting
our finding with the ubiquitous Lewis acid B(C6F5)3. Further,
we explore the role that the neighbouring nitrogen atom plays
in tempering the Lewis acidity and investigate if this enhances
the stability of these Lewis acids towards donors and exogen-
ous conditions.
Fig. 2 (a) Synthesis and yields of aminoboranes 1–6 via salt metathesis.
(b) Synthesis of 5 and 6 using HB(C6F5)2.
Results and discussion
Synthesis and characterization
We began by evaluating amines that could provide a compre-
hensive scope of both steric and electronic properties. As
noted above, we previously reported the synthesis of the phe-
nothiazyl aminoborane (1)20 and we further synthesized ami-
noboranes containing phenoxazine (2), diphenylamine (3),21
trimethylsilyl (4),22 carbazole (5), and acridane (6) motifs
(Fig. 2). These amines provide a platform to understand the
impact that fused ring systems have on the Lewis acidity.
Similar to the preparation of 1, these new aminoboranes could
be synthesized via salt metathesis, with the addition of the
respective sodium amide precursor to ClB(C6F5)2, yielding
compounds 2–6 in moderate to high yields (26–97%, Fig. 2a).
We have also developed alternative synthetic pathways for 5
and 6 utilizing HB(C6F5)2. Where compound 5 can be prepared
via the dehydrocoupling of HB(C6F5)2 with carbazole and 6 can
be prepared via hydride transfer from the addition of HB
(C6F5)2 to acridine, resulting in a significantly higher yield
(Fig. 2b). To further explore this synthetic route, we attempted
to facilitate the dehydrocoupling with all the respective amines
and HB(C6F5)2 in C6D6. Unfortunately, this process was not
found to be applicable in the synthesis of 1–4 or 6. In all these
cases, no changes were observable by NMR spectroscopy apart
Fig. 3 Solid state structure of the adduct 2, 4, 5 and 6H2. B: pink, C:
grey, F: yellow-green, N: blue. H-atoms omitted for clarity with the
exception of 6H2.
1
from the reaction of HB(C6F5)2 with acridane where it was The H NMR spectra of 2–6 agree with the structures and are
evident that a Lewis acid base adduct was formed. We were otherwise unremarkable. Compound 2 showed a doublet of
able to crystallographically characterize the adduct 6H2 (Fig. 3 doublets at 6.83 ppm, and a doublet of triplets at 6.54 ppm as
and ESI, Fig. S65, S67†). All compounds synthesized were fully inequivalent resonances that correspond to the eight protons
characterized by multinuclear NMR spectroscopy and showed on the phenoxazine motif. Compound 3 showed a doublet at
similar features to that of 1, as reported in our earlier work. 6.96, a triplet at 6.80, and a multiplet from 6.78–6.71 ppm,
This journal is © The Royal Society of Chemistry 2021
Org. Biomol. Chem., 2021, 19, 4796–4802 | 4797

View Article Online
Paper
Organic & Biomolecular Chemistry
which corresponds to the two phenyl substituents. Compound energies of those structures, which were used to calculate
4 showed a singlet at 0 ppm which corresponds to the two FIA’s. This selection of methods was inspired by Greb and co-
–SiMe3 groups. Compound 5 showed a doublet 7.99, a triplet workers,29 whose TMS-isodesmic reaction is used to calculate
at 7.40, a multiplet from 7.27–7.21, and a doublet at 7.00 ppm FIA’s here. All calculations in this work except the Natural
integrating to the eight protons of the carbazole. Finally, com- Bond Orbital (NBO) analyses were performed using the ORCA
pound 6 showed a doublet at 7.33 and a multiplet from 4.2 program package.34 The NBO analyses were done using the
7.21–6.98 for the aromatic protons, and a doublet at 3.92 ppm GAMESS-US and NBO 7.0 packages.35 Structurally, the neutral
for the bridgehead CH2 resonances. The chemical shifts in aminoboranes optimized to provide B–N bond lengths of
their respective 11B NMR spectra range from 44.1–37.1 ppm 1.40–1.41 Å (Table 1), which agrees with the experimental crys-
and indicate the pseudo tri-coordinate nature of these species. tallographic observations. In all cases, the N atom is planar,
Most notably the 19F NMR spectra of 2–6 support similar and the sum of the angles are ∼360°. Such a planar structure
degrees of BvN bond character as indicated by their respective may arise from the BvN double bond character and/or the
distance between the para and meta fluorine resonances as π-interaction between N and the two aromatic rings connected
these ranged from Δδpm 8.81 to 10.64 ppm, indicating a to it. NBO analysis was performed on all structures to gauge
pseudo 4-coordinate environment around the boron the level of donation from the N lone pair to the vacant B pz-
centre.23,24 Anomalously, the 19F NMR spectrum for compound orbital.36 The results indicate that all six aminoboranes show
6 presents two resonances for each ortho and meta fluorine similar B–N overlap, ranging from 0.30–0.34 electrons being
atoms at 130.08 & 132.06 ppm and 160.95 & 161.13 ppm. The donated from the N lone pair to B (Table 1). Interestingly, the
appearance of these additional resonances can be attributed to NBO second order perturbation theory analysis indicates a
the steric demand of the acridane functionality restricting the range of energy lowering induced by this donation. In the case
rotation of the C6F5 rings.25
of 1, 2, 3, and 6 this stabilization energy ranges from
Crystals suitable for X-ray diffraction were also obtained for 102–113 kcal mol−1. However, 4 only has a stabilization energy
2, 4, and 5 (Fig. 3) It should be noted the structure of 4 was of 22 kcal mol−1 which can be attributed to the steric hin-
previously elucidated by Green and co-workers22 and our drance between the trimethylsilyl and pentafluorophenyl sub-
values are in agreement with their reports. These show B–N stituents, which twists the B–N bond, increasing the dihedral
bond lengths of 1.395(2) Å, 1.402(3) Å, and 1.405(2) Å respect- angle between the B and N moieties, and reduces the
ively. Furthermore, both the boron and nitrogen centres in 2, π-interaction. Aminoborane 5, bearing a carbazole fragment,
4, and 5 show trigonal planar geometry with the sum of the proved to have a lower than anticipated stabilization of 85 kcal
bond angles around boron being 359.4°, 359.9°, and 360.1°, mol−1. Unlike 4, this can be attributed to the N lone pair being
and around nitrogen being 359.2°, 359.9°, and 359.8°, respect- delocalized into the aromatic carbazole ring system. It is
ively. These data compare nicely to previously reported mono- important to note that the second order perturbation analysis
meric Lewis acidic aminoboranes with trigonal planar geome- is of a qualitative nature. Those overestimated energies reflect
tries around both boron and nitrogen, as well as B–N bond the relative, but not absolute strengths of the BN π bonds.
lengths of ∼1.40 Å (vide supra).
Overall, the BN π interactions are weak (∼0.3e donation) and
Finally, 6H2 was found to have two molecules in the unit the two smaller donations of 4 and 5 are consistent with their
cell, with substantially longer B–N bond lengths of 1.683(2) Å less NBO-estimated BN π interaction energies than the others.
and 1.665(2) Å. Furthermore, the B and N atoms are tetra-
The relative Lewis acidities of 1–6 were estimated by the cal-
hedral in nature, as opposed to trigonal planar, strongly culated FIA’s: −385 (1), −389 (2), −360 (3), −353 (4), −413 (5),
suggesting the ammineborane character of 6H2.
and −381 (6) kJ mol−1 (Table 1). Comparatively, B(C6F5)3 has a
reported FIA of −448 kJ mol−1 29
Such a wide range of FIA’s
was unexpected given the small range of N→B π donations and
.
Evaluation of the Lewis acidity of aminoboranes
We attempted to experimentally assess the Lewis acidity of 1–6 the similar BN bond lengths. This indicates that the different
using both the Gutmann–Beckett and the Fluorescent Lewis Lewis acidities of the B centres do not directly arise from the
Adduct method.21,26–28 Unfortunately both of these methods BN π interactions. Instead, the acidities are related to the aro-
provided inconclusive results, which we attribute to the steric
bulk around the boron as well as the pseudo double bond
character of the aminoborane. Therefore, fluoride ion affinity
(FIA) calculations were performed to better elucidate the
Table 1 Computational results for 1–6
impacts that each amino-substituent has on the Lewis acidity
1
2
3
4
5
6
of the boron centre.29 Structures of 1–6 were optimized at the
PBEh-3c/DEF2-mSVP level of theory.30,31 The same method
was employed for hessian calculations. There were no imagin-
ary frequencies at the optimized structures. Single point calcu-
lations at those structures were done at the PW6B95/DEF2-
QZVPP level of theory.32,33 Zero-point energy (ZPE) corrections
were added to those single point energies to give the final
B–N (Å)
1.40
8
0.34
1.40
13
0.34
1.40
10
0.35
1.40
29
0.31
1.41
23
0.30
−85
1.40
8
0.34
−113
B–N dihedral angle (°)
N lp→B pz donation (e)
N lp→B pz donation
−107 −106 −102 −22
energya (kcal mol−1
)
FIA (kJ mol−1
)
−385 −389 −360 −353 −413 −381
a Estimated at the level of NBO 2nd order perturbation theory.
4798 | Org. Biomol. Chem., 2021, 19, 4796–4802
This journal is © The Royal Society of Chemistry 2021

View Article Online
Organic & Biomolecular Chemistry
Paper
maticities of the groups that are connected to N. The Nucleus- vation is in line with the aforementioned dichotomic
Independent Chemical Shifts (NICS(1)zz) calculations were per- Lewis acidity of (N-pyrrolyl)B(C6F5)2 and the saturated
formed for those π cyclic groups,37 and the results are sum- (N-pyrrolidinyl)B(C6F5)2, given that the aromatic amine gener-
marized in Fig. 4. A blue colouration indicates that a ring ates a Lewis acidic aminoborane, whereas the saturated one
becomes more aromatic upon binding a fluoride anion and a shows limited Lewis acidic behaviour.17 Without any π system
red colouration indicates reduced aromaticity. Upon bonding a attached to N to stabilize its pushed-back lone pair, 4 features
F− to B, the N lone pair must be completely pushed back from a low FIA.
the BN π bond and get more delocalized into the π groups
attached to N. We temporarily leave 4 aside since it has no π
groups attached to N. In 3·F−, a phenyl ring rotates to be copla-
nar with the N 2pπ and bears all of the destabilization from
conjugation with the N lone pair. Such a rotation also leads to
steric hindrance between the two phenyls. These two factors
result in a lower FIA. 1, 2, and 6 are similar, as the ligands
attached to N cannot rotate about their CN bonds. The destabi-
lization due to conjugating with the N lone pair is hence atte-
nuated in the two 6-membered side rings. There is no steric
hindrance between the side rings induced by the push-back.
The three compounds thus show similar FIA’s, which are
greater than that of 3. In 5·F−, the N lone pair is pushed to a
triple-ring 14 electron π system. The number of electrons is
consistent with the 4n + 2 Hückel’s rule, and the system is aro-
matic, highlighted by the blue colour across all three rings.
The enhancement of aromaticity is most significant in the
central ring that contains N, with its NICS(1)zz changing from
−11 to −24 ppm upon F− adduction. This enhancement of aro-
maticity explains the largest FIA of 5. This result is reminiscent
of the allene derived carbon Lewis acids reported by Alcarazo
and co-workers,38 which showed fluorene-substituted allenes
can delocalize negative charge into the rings, increasing the
partial positive character of the allenyl carbon. This obser-
Behaviour of aminoboranes as Lewis acids
The calculations predict that these aminoboranes will have
varying degrees of Lewis acidity thus, we began exploring reac-
tivity with exogenous Lewis bases as it is generally understood
that Lewis acids have poor tolerance for donor solvents and
Lewis basic functional groups. This reactivity often narrows
substrate scopes when they are applied as catalysts. Contrary
to conventional Lewis acids, these Lewis acidic aminoboranes
appear more tolerant of donor solvents. Exposure of aminobor-
anes 1–4 and 6 to acetonitrile did not appear to show any
changes in the multinuclear NMR spectra, indicating that
adduct formation is not favourable at room temperature.
However, the more Lewis acidic 5 appears to form an adduct
with acetonitrile when dissolved in CD3CN as observed in the
11B NMR spectrum with a sharper upfield resonance at
−3.2 ppm (ESI, Fig. S50†).
Increasing the Lewis base donor abilities, we explored the
reactivity of the aminoboranes with 4-dimethylaminopyridine
(DMAP). However, again only 5 formed an isolable adduct (7),
which supports the FIA calculations that is has the greatest
Lewis acidity. The evidence for adduct formation was clear by
multinuclear NMR analysis, the 11B NMR spectrum has a reso-
nance at δ: −0.60 ppm, indicating a tetracoordinate boron
centre. Further, this is supported by the 19F NMR spectrum,
with resonances at δ: −130.38, −156.62 and −162.95 ppm.
Crystals suitable for X-ray diffraction were grown from a con-
centrated benzene solution (Fig. 5). The B–N bond length for
the carbazole substituent was found to be 1.548(3) Å, which is
significantly longer than the neutral species (1.405(2) Å) and
Fig. 4 Top: Computed reaction for FIA values; bottom: NICS values for
the neutral and fluoride bound aminoboranes (in ppm).
Fig. 5 Solid state structure of the adduct 7. B: pink, C: grey, F: yellow-
green, N: blue. H-atoms omitted for clarity.
This journal is © The Royal Society of Chemistry 2021
Org. Biomol. Chem., 2021, 19, 4796–4802 | 4799

View Article Online
Paper
Organic & Biomolecular Chemistry
consistent with the 1.55 Å of 5·F− obtained in calculation, indi- vides additional stability to the Lewis acidic boron centre. This
cating that the nitrogen lone pair is no longer being donated finding is reminiscent of work by Ingleson where they report
to the boron centre. The B–N bond length between the amino- that weakly Lewis acidic boranes can facilitate FLP reductions
borane and DMAP was found to be slightly longer at 1.607(4) in wet solvents.41
Å, supporting the dative bond formation.
Finally, to further explore the hypothesis that the BvN
As noted above, the adduct of HB(C6F)2 and acridane (6H2) double bond provides a level of “protection” to the Lewis
could be synthesized, however 6 could not be generated from acidic borane, we computationally explored the sensitivity of 5,
this product via H2 loss, even when heated. The fact that only the most Lewis acidic aminoborane, with water. Structural
5 could be successfully prepared via a dehydrocoupling optimizations, hessian, and energy calculations were all per-
approach led us to explore if 5 could reversibly activate H2 in formed at the CAM-B3LYP/def2-QZVPP/D3/CPCM(Toluene)
an intramolecular FLP fashion. Exposure of 5 to an atmo- level and there is no imaginary frequency.42–44 The subsequent
sphere of H2 in a J-Young NMR tube produced new resonances NBO analysis was performed at the CAM-B3LYP/cc-pVTZ level.
in the 19F NMR spectrum, indicating the formation of a 4-coor- 5·H2O has its rBO = 1.64 Å, larger than the typical 1.34 Å
dinate boron species (ESI Fig. S52–S54†). However, this was a covalent bond length. For the 5·H2O adduct, the NBO analysis
minor product, and most of the starting material remained assigned a Lewis structure with a coordinate bond between the
intact. Isolation of this product was not successful as once the O lone pair and the B vacant orbital, which features a s0.11p0.89
H2 was released from the system, the only observable product hybridization. The second order perturbation theory analysis
was 5. Nevertheless, this suggests that this aminoborane may indicates a 201 kcal mol−1 large interaction. Although it is well
be able to reversibly activate H2. Attempts to utilize 5 as a known that the perturbation theory overestimates interactions
hydrogenation catalyst were thus far unsuccessful.
between orbitals, such a large magnitude clearly points to a
We previously reported that 1 was able to catalyze the dehy- coordinate (dative) bond. The occupancies of the B vacant and
drocoupling of stannanes.20 This was unexpected because it the O lone pair orbitals are 0.35 and 1.71e, respectively. The
was the first time an intramolecular FLP was shown to hetero- occupancy of the B vacant orbital does not solely arise from
lytically facilitate this reaction (there has been subsequent the O lone pair. The perturbation analysis shows that the N
work illuminating the heterolytic nature of this transform- lone pair also interacts with the B vacant orbital, with a 23 kcal
ation).39 To further understand this reaction, we explored the mol−1 magnitude. Note that the B and N moieties has been
ability of 2–6 to act as catalysts for the dehydrocoupling of distorted to nonplanar by the H2O attachment to B, with a 43
Ph3SnH under similar conditions to our prior report (5 mol%, degrees dihedral angle. The N-to-B π donation is hence largely
50 °C in C6D6). Only 2 promoted this reaction to an appreci- reduced. Overall, the DFT calculation showed that the H2O-
able extent as evidenced by the increase in the Ph3SnSnPh3 addition is favorable by −5.0 kcal mol−1 energy with ZPE
resonance at δ: −141 ppm in the 119Sn NMR spectrum correction. Though we emphasize, again, the overestimating
(Fig. S28†). These results make sense given the structural and nature of the perturbation theory analysis. The 5.0 kcal mol−1
electronic similarities between phenothiazine and phenoxa- here does not contradict the 201 kcal mol−1 mentioned
zine. Trace dehydrocoupled product could be observed when 6 above, as they are obtained with respect to different references.
is used as a catalyst, however this reaction does not proceed in The orbitals discussed above can be seen in Fig. S66 in the
appreciable yields.
ESI.†
These computations were supported by experimental evi-
The FIA calculations indicate that the aminoboranes 1–6
are somewhat electrophilic (vide supra). To experimentally dence, 5 was dissolved in undried CDCl3 and left for 7 days in
verify these values, we explored the ability of these Lewis acids solution open to ambient atmosphere, as well as a solid
to catalyze the hydrosilylation of acetophenone at 5 mol% cata- sample of 5 that was also left out on the bench. In both cases,
lyst loading.40 We monitored these reactions via 1H NMR spec- an apparent weak water adduct was observed in the 1H, 11B,
troscopy for up to 6 hours, however it was observed that most and 19F NMR spectra, however the change in the 11B chemical
of these reductions proceeded to >90% conversion within shift from the parent species to the water adduct was less than
1 hour (ESI Fig. S41–S46†). Predictably, aminoborane 4, which 1 ppm (40.5 ppm vs. 39.9 ppm, see ESI Fig. S58–S60†). Similar
bears a bis(trimethylsilylamido) functionality, was unable to observations can be seen in the 19F NMR spectrum, supporting
facilitate this reaction given the steric constraints of the tri- that the H2O adduct is quite weak, consistent with the calcu-
methylsilyl substituents. These results speak to the ability of lated 1.63 Å long BO bond and relative −5.0 kcal mol−1 energy
the aminoboranes to act as Lewis acid catalysts. Excitingly, we of the adduct. Encouragingly, decomposition of the aminobor-
were able to perform this same hydrosilylation reaction with 5 ane was not observed. Furthermore, addition of acetophenone
(5 mol%) in CD3CN at room temperature resulting in near and triethylsilane to a solution of 5·H2O in CDCl3 (5 mol%)
complete conversion in under 6 h (ESI Fig. S49–S51†). The 11B led to the near complete reduction (∼89% conversion) of the
and 19F NMR spectra indicate that some catalyst decompo- acetophenone after 48 h at 60 °C (ESI Fig. S47 and S48†).
sition occurs, however at the end of the reaction some catalyst Similar to the reaction done in CD3CN, some catalyst
remains. This reaction would not be possible with the more decomposition occurred, nevertheless, these results are
electrophilic Lewis acid, B(C6F5)3, and supports our hypothesis encouraging with respect to aminoborane Lewis acidic
that the lone pair donation from the amino substituent pro- capabilities.
4800 | Org. Biomol. Chem., 2021, 19, 4796–4802
This journal is © The Royal Society of Chemistry 2021

View Article Online
Organic & Biomolecular Chemistry
Paper
8 G. J. P. Britovsek, J. Ugolotti and A. J. P. White,
Organometallics, 2005, 24, 1685–1691.
Conclusions
We demonstrate the synthesis, characterization of Lewis acidic
aminoboranes. Through analysis of the Fluoride Ion Affinity
9 R. C. Neu, E. Y. Ouyang, S. J. Geier, X. Zhao, A. Ramos and
D. W. Stephan, Dalton Trans., 2010, 39, 4285–4294.
we surmised that the Lewis acidity of these aminoboranes is 10 U. Gellrich, Angew. Chem., Int. Ed., 2018, 57, 4779–4782.
not immediately related to B–N bond length, but instead the 11 A. J. V. Marwitz, M. H. Matus, L. N. Zakharov, D. A. Dixon
resulting change in aromaticity of the N-contained rings upon
adduct formation. All of the tested aminoboranes, with the 12 M. E. Glogowski, P. J. Grisdale, J. L. R. Williams and
exception of 4, showed catalytic activity in the hydrosilylation L. Costa, J. Organomet. Chem., 1974, 74, 175–183.
of acetophenone, though only 2 proved effective in the dehy- 13 J. Wang, Y. Wang, T. Taniguchi, S. Yamaguchi and S. Irle,
drocoupling of stannanes – owing to the similarity of phenoxa- J. Phys. Chem. A, 2012, 116, 1151–1158.
zine to phenothiazine. Finally, the carbazole substituted ami- 14 T. Taniguchi, J. Wang, S. Irle and S. Yamaguchi, Dalton
noborane 5 was shown to effect catalytic transformations even Trans., 2013, 42, 620–624.
in the presence of Lewis basic donors and donor solvents. This 15 N. M. D. Brown, F. Davidson and J. W. Wilson,
activity, despite the presence of water and donor solvent could J. Organomet. Chem., 1980, 192, 133–138.
open the doors to new, kinetically bench-stable Lewis acids for 16 C. Arivazhagan, A. Maity, K. Bakthavachalam, A. Jana,
and S.-Y. Liu, Angew. Chem., Int. Ed., 2009, 48, 973–977.
frustrated Lewis pair transformations. We have begun investi-
gating this exciting prospect and will report our findings in
due course.
S. K. Panigrahi, E. Suresh, A. Das and S. Ghosh, Chem. –
Eur. J., 2017, 23, 7046–7051.
17 G. Kehr, R. Fröhlich, B. Wibbeling and G. Erker, Chem. –
Eur. J., 2000, 6, 258–266.
18 J. W. Strauch, J. L. Fauré, S. Bredeau, C. Wang, G. Kehr,
R. Fröhlich, H. Luftmann and G. Erker, J. Am. Chem. Soc.,
2004, 126, 2089–2104.
Author contributions
J. N. B., S. S., and E. P. were responsible for data 19 D. Winkelhaus, Y. V. Vishnevskiy, R. J. F. Berger,
collection. C. B. C., and T. Z. were responsibly for
supervision. J. N. B., S. S., E. P., T. Z., and C. B. C. were respon-
sible for writing, reviewing, and editing the manuscript.
H.-G. Stammler, B. Neumann and N. W. Mitzel, Z. Anorg.
Allg. Chem., 2013, 639, 2086–2095.
20 J. N. Bentley, E. Pradhan, T. Zeng and C. B. Caputo, Dalton
Trans., 2020, 49, 16054–16058.
21 J. N. Bentley, S. A. Elgadi, J. R. Gaffen, P. Demay-Drouhard,
T. Baumgartner and C. B. Caputo, Organometallics, 2020,
39, 3645–3655.
Conflicts of interest
There are no conflicts to declare.
22 J. R. Galsworthy, M. L. H. Green, V. C. Williams and
A. N. Chernega, Polyhedron, 1998, 17, 119–124.
23 A. G. Massey and A. J. Park, J. Organomet. Chem., 1966, 5,
218–225.
24 M. G. Hogben and W. A. G. Graham, J. Am. Chem. Soc.,
1969, 91, 283–291.
Acknowledgements
We are grateful to York University, the Natural Sciences and
Engineering Research Council of Canada, the Canada 25 F. Blank, H. Scherer, J. Ruiz, V. Rodríguez and C. Janiak,
Foundation for Innovation, and the Canada Research Chairs
program.
Dalton Trans., 2010, 39, 3609–3619.
26 U. Mayer, V. Gutmann and W. Gerger, Monatsh. Chem.,
1975, 106, 1235–1257.
27 M. A. Beckett, G. C. Strickland, J. R. Holland and
K. S. Varma, Polymer, 1996, 37, 4629–4631.
28 J. R. Gaffen, J. N. Bentley, L. C. Torres, C. Chu,
T. Baumgartner and C. B. Caputo, Chem, 2019, 5, 1567–1583.
Notes and references
1 J. W. B. Fyfe and A. J. B. Watson, Chem, 2017, 3, 31–55.
2 A. Dasgupta, R. Babaahmadi, B. Slater, B. F. Yates, 29 P. Erdmann, J. Leitner, J. Schwarz and L. Greb,
A. Ariafard and R. L. Melen, Chem, 2020, 6, 2364–2381. ChemPhysChem, 2020, 21, 987–994.
3 H. C. Brown, K. W. Kim, M. Srebnik and S. Bakthan, 30 S. Grimme, J. G. Brandenburg, C. Bannwarth and
Tetrahedron, 1987, 43, 4071–4078. A. Hansen, J. Chem. Phys., 2015, 143, 054107.
4 Y. Soltani, L. C. Wilkins and R. L. Melen, Angew. Chem., Int. 31 R. Sure, J. G. Brandenburg and S. Grimme, ChemistryOpen,
Ed., 2017, 56, 11995–11999. 2016, 5, 94–109.
5 W. Meng, X. Feng and H. Du, Acc. Chem. Res., 2018, 51, 32 Y. Zhao and D. G. Truhlar, J. Phys. Chem. A, 2005, 109,
191–201.
5656–5667.
6 D. W. Stephan, Science, 2016, 354, aaf7229.
7 J. R. Lawson and R. L. Melen, Inorg. Chem., 2017, 56, 8627–
8643.
33 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005,
7, 3297–3305.
34 F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2018, 8, 4–9.
This journal is © The Royal Society of Chemistry 2021
Org. Biomol. Chem., 2021, 19, 4796–4802 | 4801

View Article Online
Paper
Organic & Biomolecular Chemistry
35 E. D. Glendening, C. R. Landis and F. Weinhold, J. Comput. 40 D. J. Parks and W. E. Piers, J. Am. Chem. Soc., 1996, 118,
Chem., 2019, 40, 2234–2241. 9440–9441.
36 A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 41 V. Fasano and M. J. Ingleson, Chem. – Eur. J., 2017, 23,
88, 899–926. 2217–2224.
37 Z. Chen, C. S. Wannere, C. Corminboeuf, R. Puchta and 42 T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett.,
P. von Ragué Schleyer, Chem. Rev., 2005, 105, 3842–3888. 2004, 393, 51–57.
38 B. Inés, S. Holle, R. Goddard and M. Alcarazo, Angew. 43 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem.,
Chem., Int. Ed., 2010, 49, 8389–8391. 2011, 32, 1456–1465.
39 F. Holtrop, A. R. Jupp, B. J. Kooij, N. P. Leest, B. Bruin and 44 J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005,
J. C. Slootweg, Angew. Chem., Int. Ed., 2020, 59, 22210–22216. 105, 2999–3094.
4802 | Org. Biomol. Chem., 2021, 19, 4796–4802
This journal is © The Royal Society of Chemistry 2021