Reversible 1,1-hydroboration: Boryl insertion into a C-N bond and competitive elimination of HBR2 or R-H

Article abstract of DOI:10.1002/anie.201500487

Boranes with the general formula of HBR₂ have been found to undergo a facile 1,1-hydroboration reaction with pyrido[1,2-a]isoindole (A), resulting in insertion of a BR₂ unit into a C-N bond and the formation of a variety of BN heterocycles. Investigation on the thermal reactivity of the BN heterocycles revealed that these molecules have two distinct and competitive thermal elimination pathways: HBR₂ elimination (or retro-hydroboration) versus R-H elimination, depending on the R group on the B atom and the chelate backbone. Mechanistic aspects of these highly unusual reactions have been established from both experimental and computational evidence. Adduct formation between HBR₂ and A was found to be the key intermediate in 1,1-hydroboration of A.

Full text of DOI:10.1002/anie.201500487

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201500487
B,N Heterocycles
German Edition:
DOI: 10.1002/ange.201500487
À
Reversible 1,1-Hydroboration: Boryl Insertion into a C N Bond and
Competitive Elimination of HBR2 or R H **
À
Deng-Tao Yang, Soren K. Mellerup, Xiang Wang, Jia-Sheng Lu, and Suning Wang*
Abstract: Boranes with the general formula of HBR2 have
been found to undergo a facile 1,1-hydroboration reaction with
pyrido[1,2-a]isoindole (A), resulting in insertion of a BR2 unit
ing reagents are known,[13] the straightforward 1,1-hydro-
boration of alkenes or alkynes by common boranes, such as
HBR2, remains unknown. Herein we disclose the first
observation of a facile 1,1-hydroboration involving the
À
into a C N bond and the formation of a variety of BN
À
heterocycles. Investigation on the thermal reactivity of the BN
heterocycles revealed that these molecules have two distinct
and competitive thermal elimination pathways: HBR2 elimi-
addition of a variety of HBR2 into a C N bond of pyrido-
[1,2-a]isoindole (Scheme 1, A), ultimately generating both
known[14] and novel BN heterocycles. Furthermore, during the
À
nation (or retro-hydroboration) versus R H elimination,
depending on the R group on the B atom and the chelate
backbone. Mechanistic aspects of these highly unusual reac-
tions have been established from both experimental and
computational evidence. Adduct formation between HBR2
and A was found to be the key intermediate in 1,1-hydro-
boration of A.
Scheme 1. Reversible borane insertion (A!B) and photo/thermal
elimination pathways available to B.
O
rganoboron compounds are an important class of mole-
cules, which can be found in a number of important fields[1]
including hydrogen storage,[2] optoelectronic devices,[3–4] sen-
sors,[5] and organic synthesis.[6] Of the many organoboron
species, boranes (particularly those of the form HBR2) have
long been studied for their use in the functionalization of
alkenes and alkynes by 1,2-hydroboration which was origi-
nally pioneered by Herbert C. Brown.[7] Despite the large
number of examples/uses[8] of 1,2-hydroboration, the related
1,1-hydroboration reactions have, for the most part, eluded
researchers. Building upon the seminal work of Wrack-
meyer,[9] Erker and co-workers recently reported a 1,1-
investigation of the BN heterocyclic compounds B, we
discovered that depending on the R group of the borane
and the ligand backbone in B, the 1,1-hydroboration product
can either thermally revert back to the borane and A, or
À
eliminate an R H group to form the BN phenanthrene C.
This unprecedented competing thermal elimination phenom-
enon is also presented.
Compound A was initially obtained from the thermal
reaction of B1, with the intention of converting it to our
reported highly luminescent compound C1 (Scheme 2)[14]
carboboration[10] involving the simple activation of C C
À
bonds through the treatment of internal alkynes with several
highly electrophilic boranes (e.g. B(C6F5)3). This unique
reactivity has been extended to the activation/insertion of
[11]
=
a BR2 unit into a C C bond as well as the formation of
borole derivatives,[12] showcasing the potential usefulness of
such a chemical transformation. Although a few examples of
formal alkyne 1,1-hydroboration involving the use of activat-
[*] D.-T. Yang, S. K. Mellerup, X. Wang, Prof. Dr. S. Wang
Department of Chemistry
Queen’s University
Kingston, Ontario, K7L 3N6 (Canada)
E-mail: suning.wang@chem.queensu.ca
Scheme 2. 1,1-Hydroboration of A with HBMes2 and 9-BBN, and the
thermal and photoreactivity of B1 and B2.
Dr. J.-S. Lu, Prof. Dr. S. Wang
Beijing Key Laboratory of Photoelectronic/Electrophotonic Conver-
sion Materials, School of Chemistry, Beijing Institute of Technology
5 South Zhongguancun Street, Beijing (P.R. China)
using heat instead of light. Upon heating B1 in the presence
of a small amount of mineral oil at 2808C for over 1 h, bright
green–yellow fluorescent vapors were observed which readily
deposited as a green fluorescent solid film at the top of the
reaction vessel. This solid was isolated and identified as
[**] We thank the Natural Sciences and Engineering Research Council
(NSERC) of Canada for financial support. S.K.M. thanks NSERC for
the Canada Graduate Schloarship (CGS-M). S.W. thanks the Canada
Council for the Art for the Killam research fellowship.
1
pyrido[1,2-a]isoindole (or isocarbazole) by H and 13C NMR
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201500487.
spectroscopy, mass spectrometry, and X-ray diffraction anal-
5498
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 5498 –5501

Angewandte
Chemie
yses,[15] and can be readily prepared on a gram scale by
established procedures.[16] NMR spectroscopic data indicated
that compound A along with HBMes2 formed quantitatively
from the thermal reaction of B1, and compound C1 was not
observed at all.[15] Compound C1 is not an intermediate in the
conversion of B1 into A as it remains unchanged when heated
to 2808C in the presence of mesitylene under the same
conditions used for the thermal reaction of B1. Thus, the
transformation of B1 into A could be formally described as
a retro-1,1-hydroboration. To obtain mechanistic insights into
this highly unusual elimination reaction, we set out to
examine the reverse reaction, namely the 1,1-hydroboration
of A with HBMes2.
photoreactivity of B2 was investigated. Upon irradiation
with UV light (300 nm) in C6D6, B2 was found to convert into
the BN-phenanthrene C2 quantitatively by an intramolecular
À
R H elimination process (Scheme 2). Similar to C1, molecule
C2 has a low-energy absorption band with well resolved
vibrational features at 400–500 nm and is brightly fluorescent
with an emission peak at lmax = 520 nm.[15]
Additionally, A was found to readily react with highly
Lewis acidic boranes, such as HB(C6F5)2 and B(C6F5)3. As
shown by the 1H NMR spectra in Figure 2, the reaction of A
Much to our surprise, A reacts readily with HBMes2 in
THFat 808C, generating B1 in approximately 95% yield after
3 days (Scheme 2). The fact that 1,1-hydroboration of A
occurs at a much lower temperature than the reverse retro-
hydroboration of B1 suggests that either the activation barrier
from A + HBMes2 to B1 is lower than the retro-1,1-
hydroboration of B1, or, there is an accessible intermediate
from A + HBMes2 to B1. Although no intermediates were
observed in the 1H NMR spectra tracking the 1,1-hydro-
boration of A, below we will provide evidence which supports
that an intermediate is involved in the transformation. The
apparent paradox that A and HBMes2 can be obtained
quantitatively from B1 at a temperature much higher than the
temperature required to form B1 from A + HBMes2 can be
explained in that A is continuously sublimed from the
reaction medium and deposited on the cooler upper surface
of the reaction vessel.
Figure 2. Bottom: The 1H NMR spectra showing the clean formation
of D3 upon the mixing of A and HB(C6F5)2 in [D8]toluene at ambient
temperature, and the subsequent clean transformation of D3 to B3
upon heating at 1108C. Top: Scheme showing the structure of
intermediate D3 and its conversion into B3.
To examine the scope of this unprecedented 1,1-hydro-
boration, the reaction of A with the commercially available 9-
borabicyclo[3.3.1]nonane (9-BBN) was also studied. After
10 h in toluene at 1108C, molecule A reacts completely with
9-BBN generating a new BN heterocycle B2 in 90% yield.
Again, no intermediate was detected in the 1H NMR spectra
while monitoring the transformation.[15] The crystal structure
of B2 was determined by X-ray diffraction analysis and is
shown in Figure 1. Compound B2 is a unique member of the
BN heterocyclic compounds (Scheme 2) with a chelating
cyclooctyl group bound to the boron atom. To examine if the
cyclooctyl group prohibits the previously demonstrated
photoelimination of related molecules, such as B1, the
with HB(C6F5)2 occurred instantaneously at ambient temper-
ature. Rather than forming the BN heterocycle B3, a new
species D3 was obtained quantitatively. Based on the NMR
spectroscopy data, the structure of D3 was determined to be
an adduct of A and HB(C6F5)2 as shown in Figure 2. Most
significant is the observation that upon heating (808C in
benzene or 1108C in toluene), D3 converts quantitatively into
compound B3, a new BN heterocycle, which was isolated in
98% yield and fully characterized by X-ray diffraction
analysis.[15] Formally, the conversion of D3 into B3 may be
À
described as hydride migration and boryl insertion into a C N
bond. This establishes that the adduct D3 is a key intermedi-
ate involved in the 1,1-hydroboration of A with HB(C6F5)2.
From this result, we expect that the analogous adducts of A
with HBMes2 and 9-BBN are likely to be involved in the
conversion into B1 and B2, respectively.
Not surprisingly, the reaction of A with B(C6F5)3 pro-
ceeded in the same manner as that of HB(C6F5)2, forming the
air-sensitive and colorless adduct D4 quantitatively. As shown
in Figure 1, D4 possesses a crystallographically imposed
mirror plane. Consequently, the C(2) and N(1) atoms in D4
À
are disordered. The B(1) C(1) bond (1.718(3) Š) is quite
long, owing to steric congestion around the B atom. The ph-py
unit in D4 has p-stacking interactions with a C6F5 ring (C(8)
ring) as indicated by some of the short distances (2.92–
3.21 Š). The structures of D3 and D4 establish that pyrido-
[1,2-a]isoindole is a good nucleophile, capable of binding to
a trivalent boron center through its C(1) atom. Heating the
solution of D4 in THF or toluene causes the solution color
Figure 1. The crystal structures of B2 (left) and D4 (right). H atoms
are omitted for clarity. Key bond lengths [Š] and angles [8] for B2:
B(1)–C(1) 1.635(2), B(1)–N(1) 1.651(1), B(1)–C(13) 1.637(2), B(1)–
C(17) 1.632(2); C(1)-B(1)-N(1) 100.33(8), C(13)-B(1)-C(17) 104.94(9);
for D4: B(1)–C(1) 1.718(3), B(1)–C(8) 1.662(3), B(1)–C(12) 1.653(2),
C(1)–C(2) 1.506(9), C(1)–N(1) 1.504(7); C(1)-B(1)-C(8) 102.8(2), C(1)-
B(1)-C(12) 112.9(1).
Angew. Chem. Int. Ed. 2015, 54, 5498 –5501
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 5499

.
Angewandte
Communications
change to deep blue, accompanied by a complex NMR
spectral change which is being resolved and will be reported
in due course. For less Lewis acidic and sterically encumbered
boranes, such as BPh3, no reaction was observed with A.
The above examples establish that the 1,1-hydroboration
of A is a general phenomenon for HBR2 boranes. Syntheti-
cally this is a very useful and convenient reaction for the
synthesis of BN-heterocyclic compounds which are either
difficult or not possible to obtain by other methods.[14]
Having established that the species D is a key intermedi-
ate, the transformation of A +HBR2 to B is proposed to
involve the following sequence (Scheme 3): 1) formation of
the substituent groups on ph-py and boron clearly dictate
which is preferred.
To explain the two distinct thermal elimination pathways
of the BN heterocycles, computational studies using DFT
methods were carried out. Prior to modeling the reaction
pathways, the nucleophilic nature of Aꢀs C(1) atom was
confirmed by Mulliken population analysis where it was
found to be negatively charged and contributing significantly
to the HOMO level.[15] The thermal reaction pathways of B1
are depicted in Figure 3, while those of the other BN
Scheme 3. The proposed mechanism of the 1,1-hydroboration reaction.
the four-coordinated borane adduct D, 2) a hydride transfer
from the boron atom to the carbon atom and cleavage of the
À
C N bond in D resulting in the formation of intermediate E,
which is likely facilitated by the negatively charged borane
and the positively charged nitrogen in D, 3) coordination of
the nitrogen atom to the electron-deficient boron atom in E,
forming the chelate product B. Mechanistically, the 1,1-
hydroboration of A with HBR2 described bears some
resemblance to the 1,1-carboboration of an aminohydropen-
talene derivative with B(C6F5)2R reported by Erker and co-
workers.[10] The fact that the borane adduct of Awith HBMes2
or 9-BBN was not observed at ambient temperature could be
explained by the low Lewis acidity of HBMes2 and 9-BBN
that prevents adduct formation at ambient temperature. At
elevated temperatures, the corresponding adducts could form,
but they would be short-lived and undergo a rapid hydride
transfer because of the low Lewis acidity of the boron center.
To determine if the reversible 1,1-hydroboration is gen-
eral for other BN heterocycles, we investigated the thermal
reaction of the B2, B3, B5, and B6 (Scheme 4) under the same
Figure 3. Relative ground-state Gibbs free energies and structures
(B3LYP/6-31g*) of the reactant, intermediates, transition states, and
products involved in the two thermal elimination pathways of B1.
heterocycles can be found in the Supporting Information.
The calculated mechanistic pathway is consistent with our
proposed mechanism (Scheme 3), in which the calculated
activation energy for HBR2 elimination (Ea1) is approxi-
mately 100 kJmolÀ1 less than the R H elimination barrier
À
(Ea2, Figure 3). Not only does this suggest that HBR2
elimination is favored for B1, but it also helps explain why
the 1,1-hydroboration of A with HBMes2 requires less energy
to proceed than the reverse, as the intermediate D1 resides
close in energy and geometry to the transition-state structure
E1*. That this intermediate is not observed experimentally is
not surprising given the proximity of the D1 and E1* energy
levels. It is interesting to note that there is a large difference
between the two product energies of each thermal elimination
À
pathway, where the R H elimination pathway produces the
thermodynamically favored product. As argued before, we
suspect that sublimation of A away from the reaction mixture
as it forms allows for isolation of this kinetic product.
Although we attempted to locate the transition states
between A +HBMes2!D1 and B1!E1, they could not be
found most likely because they reside very close in energy to
the product of each transformation.
Based on the calculated pathway data of each B species
(Table 1), different reactivities would be expected based on
the R groups at boron and ph-py. In all the other systems, the
calculated relative energies of the HBR2 +A (or A’, 2,9-
dimethylpyrido-[1,2-a]isoindole) molecules were found to be
significantly higher than that of HBMes2 +A. In the case of
B5 and B6, this is most likely the reason that the HBR2
elimination of B5 and B6 is not observed as the products are
Scheme 4. The thermal elimination of previously reported B5 and B6.
conditions as B1. Molecules B2 and B3 do not undergo any
chemical change below 3608C, while above 3608C they
decompose producing unidentifiable species. Oddly enough,
heating B5 and B6 at 3108C and 3608C, respectively, did not
result in retro-hydroboration, but instead produced the BN-
phenanthrene products C5 and C6, respectively
(Scheme 4).[15] Based on these observations it is evident that
there are two competing thermal elimination pathways for the
À
BN heterocycles, namely, R H or HBR2 elimination, where
5500
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 5498 –5501

Angewandte
Chemie
Table 1: Summary of activation barriers and pertinent ground-state
energies (kJmolÀ1) relative to the calculated energies of B1–B6.
[1] Boron Science New Technologies and Applications (Ed.: N. S.
Hosmane), CRC, Taylor & Francis, Boca Raton, 2012.
[2] a) W. Luo, P. G. Campbell, L. N. Zakharov, S.-Y. Liu, J. Am.
Chem. Soc. 2011, 133, 19326 – 19329; b) C. W. Hamilton, R. T.
Baker, A. Staubitz, I. Manners, Chem. Soc. Rev. 2009, 38, 279 –
293.
[3] a) C. D. Entwistle, T. B. Marder, Angew. Chem. Int. Ed. 2002, 41,
2927 – 2931; Angew. Chem. 2002, 114, 3051 – 3056; b) C. D.
Entwistle, T. B. Marder, Chem. Mater. 2004, 16, 4574 – 4585;
c) A. Wakamiya, T. Taniguchi, S. Yamaguchi, Angew. Chem. Int.
Ed. 2006, 45, 3170 – 3173; Angew. Chem. 2006, 118, 3242 – 3245;
d) P. Chen, R. A. Lalancette, F. Jäkle, Angew. Chem. Int. Ed.
2012, 51, 7994 – 7998; Angew. Chem. 2012, 124, 8118 – 8122; e) A.
Lorbach, A. Hübner, M. Wagner, Dalton Trans. 2012, 41, 6048 –
6063.
Cmpd
A(A’) + HBR2
Ea1
EaÀ1
Ea2
D
C + RH
B1
B2
B3
B5
B6
8
95
139
54
154
205
206
154
227
146
110
107
100
109
250
287
277
256
288
89
137
99
92
145
À140
1
7
À142
118
À29
thermodynamically unstable. It is important to note that A’
has not been synthesized to date, which further supports this
claim. Given that these reactions were carried out at elevated
temperatures, the excess kinetic energy of the system could
allow the Ea2 barrier to be crossed for B5 and B6, leading the
[4] a) Y.-L. Rao, H. Amarne, S.-B. Zhao, T. M. McCormick, S.
´
Martic, Y. Sun, R.-Y. Wang, S. Wang, J. Am. Chem. Soc. 2008,
À
reaction to the “thermodynamic minimum” that is the R H
130, 12898 – 12900; b) K. Nagura, S. Saito, R. Frçhlich, F.
Glorius, S. Yamaguchi, Angew. Chem. Int. Ed. 2012, 51, 7762 –
7766; Angew. Chem. 2012, 124, 7882 – 7886; c) Y.-L. Rao, H.
Amarne, S. Wang, Coord. Chem. Rev. 2012, 256, 759 – 770.
[5] C. R. Wade, A. E. J. Broomsgrove, S. Aldridge, F. P. Gabbäi,
Chem. Rev. 2010, 110, 3958 – 3984.
[6] a) H. C. Brown, Tetrahedron 1961, 12, 117 – 138; b) R. M.
Mikhailov, Organoboron Compounds in Organic Synthesis,
Harwood Academic, 1984; c) H. C. Brown, Boranes in Organic
Chemistry, Cornell University Press, Ithaca, 1972.
elimination products. With respect to B2 and B3, the
predicted EaÀ1 barrier is approximately 40 kJmolÀ1 less than
that for B1, which correlates nicely with the varying reaction
times/temperatures of these three hydroboration reactions. In
both cases, neither competing elimination pathway results in
the formation of a thermodynamically stable product, which is
most likely why heating of B2 and B3 does not induce either
of these reactivities. Lastly, it is interesting to note that the
NMR spectroscopic observation of D3 is also in agreement
with the calculation results, as the adduct D3 was found to be
more thermodynamically stable than the mixture of A and
HB(C6F5)2.[15] Subsequent heating of D3 gives the reaction
[7] a) H. C. Brown, Hydroboration, W. A. Benjamin, New York,
1962; b) H. C. Brown, C. F. Lane, J. Am. Chem. Soc. 1970, 92,
6660 – 6661; c) J. V. B. Kanth, H. C. Brown, J. Org. Chem. 2001,
66, 5359 – 5365.
[8] a) M. Scheideman, G. Wang, E. Vedejs, J. Am. Chem. Soc. 2008,
130, 8669 – 8676; b) H. Braunschweig, R. D. Dewhurst, C. Hçrl,
A. K. Phukan, F. Pinzner, S. Ullrich, Angew. Chem. Int. Ed. 2014,
53, 3241 – 3244; Angew. Chem. 2014, 126, 3305 – 3308; c) H.
Braunschweig, C. Hçrl, Chem. Commun. 2014, 50, 10983 –
10985; d) C. D. Entwistle, A. S. Batsanov, J. A. Howard, M. A.
Fox, T. B. Marder, Chem. Commun. 2004, 702 – 703.
[9] a) B. Wrackmeyer, Coord. Chem. Rev. 1995, 145, 125 – 156, and
references therein; b) B. Wrackmeyer, Heteroat. Chem. 2006, 17,
188 – 208, and references therein.
[10] C. Chen, G. Kehr, R. Frçhlich, G. Erker, J. Am. Chem. Soc. 2010,
132, 13594 – 13595.
[11] B. H. Xu, G. Kehr, R. Frçhlich, S. Grimme, G. Erker, J. Am.
Chem. Soc. 2011, 133, 3480 – 3491.
[12] F. Ge, G. Kehr, C. G. Daniliuc, G. Erker, J. Am. Chem. Soc. 2014,
136, 68 – 71.
[13] a) J. Ugolotti, G. Dierker, G. Kehr, R. Frçhlich, S. Grimme, G.
Erker, Angew. Chem. Int. Ed. 2008, 47, 2622 – 2625; Angew.
Chem. 2008, 120, 2662 – 2665; b) A. Boussonni›re, X. Pan, S. J.
Geib, D. P. Curran, Organometallics 2013, 32, 7445 – 7450.
[14] J.-S. Lu, S. B. Ko, N. R. Walters, Y. Kang, F. Sauriol, S. Wang,
Angew. Chem. Int. Ed. 2013, 52, 4544 – 4548; Angew. Chem.
2013, 125, 4642 – 4646.
mixture enough energy to traverse the activation barrier EaÀ1
ultimately yielding B3.
,
In summary, the unusual 1,1-hydroboration reaction
involving different boranes and molecule A has been
established, which provides an efficient synthetic method
for a variety of BN heterocyclic compounds. It is conceivable
that this reactivity could be extended to systems that have
similar structural features to those of A, and which we are
now beginning to examine as a route to new BN heterocycles.
Two competitive thermal elimination pathways of BN heter-
ocyclic compounds have been established, one of which could
be used for the bulk synthesis of certain BN-phenanthrene
compounds through a convenient thermal elimination reac-
tion. Compound B2 was found to undergo an intramolecular
À
R H photoelimination, converting the 1,5-cyclooctyl into a 1-
cyclooctyl ligand and generating a new BN-phenanthrene C2.
Mechanistic insight on 1,1-hydroboration of A and the two
competing thermal elimination pathways of B was obtained
from experimental work and computational studies.
[15] See Supporting Information for details.
[16] S. Kajigaeshi, S. Mori, S. Fujisaki, S. Kanemasa, Bull. Chem. Soc.
Jpn. 1985, 58, 3547 – 3551.
Keywords: 1,1-hydroboration · BN heterocycles ·
boryl insertion · thermal elimination
How to cite: Angew. Chem. Int. Ed. 2015, 54, 5498–5501
Angew. Chem. 2015, 127, 5588–5592
Received: January 18, 2015
Published online: March 20, 2015
Angew. Chem. Int. Ed. 2015, 54, 5498 –5501
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5501