Molecular iodine initiates hydroborations of alkenes with N-heterocyclic carbene boranes

Article abstract of DOI:10.1021/ja407678e

The hydroboration of alkenes of diverse structural types by assorted N-heterocyclic carbene boranes can be accomplished by addition of 5-10% diiodine. For example, reaction of 1,3-dimethylimidazol-2-ylidene borane (diMe-Imd-BH₃) with 10% I

Full text of DOI:10.1021/ja407678e

Article
pubs.acs.org/JACS
Molecular Iodine Initiates Hydroborations of Alkenes with
N‑Heterocyclic Carbene Boranes
Xiangcheng Pan, Anne Boussonniere,† and Dennis P. Curran*
̀
Department of Chemistry, University of Pittsburgh, Pittsburgh, Pennsylvania 15260, United States
S
* Supporting Information
ABSTRACT: The hydroboration of alkenes of diverse structural
types by assorted N-heterocyclic carbene boranes can be accom-
plished by addition of 5−10% diiodine. For example, reaction of 1,3-
dimethylimidazol-2-ylidene borane (diMe-Imd-BH3) with 10% I2
followed by addition of 2,3-dimethyl-2-butene provided the
corresponding thexyl NHC-borane (diMe-Imd-BH2thexyl) in 75%
yield. This and related monohydroboration products are stable to
chromatography and storage. The scope of the new reaction is described, and the mechanism is probed by 11B NMR
experiments.
INTRODUCTION
■
The consummate reaction of Lewis base complexes of borane
(LB−BH3) is hydroboration of alkenes.1 Such reactions occur
because most Lewis bases (ethers, sulfides) can undergo rapid
exchange with alkenes. The resulting alkene−borane complexes
rapidly transform to alkyl boranes by hydroboration. In
contrast, complexes of borane with Lewis bases including
phosphines2 and N-heterocyclic carbenes3 are inert to alkene
hydroboration because of their stability. Amine−borane
complexes are in the middle, either leading to hydroboration
or not depending on the structure of the amine and the
reaction temperature.2,4 Hydroboration only occurs when the
amine can decomplex from the borane.
Vedejs and co-workers have spearheaded a new mode of
activation for stable Lewis base complexes of borane including
amine− and phosphine−boranes.5 Activation occurs not by
decomplexation of the Lewis base but instead by formation of a
Figure 1. NHC−borane 1 hydroborate alkenes when triflimide is
added: (a) a typical example and (b) a simplified catalytic cycle
illustrated with free borenium ions.
+
borenium ion (LB−BH2 ) or a species that reacts like a
borenium ion.6 Stabilized borenium ions have been charac-
terized,7 but like free borane, free borenium ions often do not
exist in solution. So called “borenium ion equivalents” are
present, including charged complexes of borenium ions with
solvent, borane, or another Lewis base (called boronium ions8),
or neutral, covalent LB−borane complexes with an excellent
leaving group (LB−BH2X where X is the leaving group).6
In collaboration with Vedejs, we recently reported the first
hydroboration reactions of alkenes with stable N-heterocyclic
carbene boranes (Figure 1).9 This reaction, which otherwise
does not occur even under forcing conditions, can be induced
by adding 5−15% bis(trifluoromethanesulfonimide) (hereafter,
triflimide or Tf2NH). Dihydroborated products quickly form at
rt or below. In a complementary activation approach, Parrain
and co-workers made NHC−boracyclopentanes by rhodium-
catalyzed intramolecular hydroborations.10
BH3) 1 followed by addition of 1-hexene provided dihydrobo-
rated products 2a (with two 1-hexyl groups) and 2b (with one
1-hexyl group and one 2-hexyl group) in a ratio of about 83/17
(Figure 1a). These NHC−dialkylboranes did not survive flash
chromatography, but they were oxidized to provide the
expected 91/9 mixture of 1-hexanol and 2-hexanol in high
yield. A key problem with the triflimide procedure was the
limited scope of alkenes that could be used.9
The catalytic cycle shown in Figure 1b is proposed to
account for these results, with free borenium ions shown for
simplicity. Reaction of Tf2NH with 1 gives the catalyst 3. The
two-step cycle consists of hydroboration of 1-hexene with 3
In a typical example of triflimide activation, addition of 15%
triflimide to 1,3-dimethylimidazol-2-ylidene borane (diMe-Imd-
Received: July 25, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja407678e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(which occurs a second time) followed by hydride transfer from
1 to 4 to regenerate the catalyst 3. This ability of NHC−
boranes and NHC−boreniums to exchange hydride closes a
catalytic cycle that so far has not succeeded with other ligated
boranes (1 equiv of activator is usually needed).5b,g
Table 1. Synthesis of NHC−Alkylborane Complexes Using
Iodine-Catalyzed Hydroboration
Here we expand the scope of borenium ion catalyzed
hydroborations of NHC−boranes in three directions. First, we
report that sensitive and expensive triflimide can be replaced
with easy to handle and inexpensive diiodine (I2) as activator.
Second, we learn that with many types of alkenes
monohydroboration occurs rather than dihydroboration. This
is important because the monohydroborated products are
typically stable to chromatography and can be directly isolated
and characterized. Third, we find that the range of alkene types
that participate in the hydroboration is significantly expanded.
Finally, we also describe experiments that provide insight into
the nature of the reactive intermediates in the catalytic cycle.
RESULTS AND DISCUSSION
Reaction Development. The triflimide activator is thought
to serve as a precatalyst in the reactions in Scheme 1, and we
■
Scheme 1. Results of a Preliminary Hydroboration
Experiment with Diiodide in Place of Triflimide
tested several other activators (TfOH, Br2, NIS) in preliminary
experiments. Among these, diiodine (I2) quickly gave promising
results. In an early experiment patterned after that in Figure 1a,
10 mol % diiodine was added to 1 equiv of 1 in CH2Cl2
(Scheme 1).11 After the bubbling ceased (hydrogen gas), 3
equiv of 1-hexene was added. Then 1 h later, a 11B NMR
spectrum was recorded. The resulting spectrum closely
resembled that from the prior triflimide experiment, with the
major peaks for 2a and 2b detected in about 83/17 ratio. Thus,
a reactive borenium ion equivalent can be access by using
iodine, and this equivalent undergoes both hydroboration and
hydride transfer to close a catalytic cycle.
The two activators Tf2NH and I2 are not as similar as they
look based on this product analysis. Reaction of 1 with a
substoichiometric amount of triflimide in the absence of alkene
provides a borenium ion complexed to a B−H bond of the
starting borane ([NHC−BH2−H−BH2−NHC]+ Tf2N−).9,12 In
contrast, the reaction of 1 with substoichiometric diiodine
produces the covalent boryl iodide NHC−BH2I that is not
complexed to the remaining borane 1.11b The results in Figure
1a and Scheme 1 suggest that these two different species, one
ionic and one covalent, are both capable of setting off
hydroborations by NHC−boranes.
Scope Study. Results of a study of the scope of the iodine-
activation method are summarized in Table 1. We focused on
substrates that were problematic in the original triflimide
procedure.9 This procedure works well for simple mono- and
disubstituted alkenes like 1-, 2-, and 3-hexene, always producing
dialkylboranes (no matter whether the alkene is present in
excess or deficient). However, tri- and tetra-substituted alkenes
and phenyl-substituted alkenes often gave complex mixtures.
a
b
5 mol % iodine was used in a 2 mmol scale reaction. Oxidation of
c
this product provided racemic trans-2-methylcyclohexanol. About
10% of the regioisomer was detected in the 11B NMR spectrum of the
d
crude product. 1 equiv of alkene and 1.4 equiv of 1 were used.
In contrast, addition of 10% I2 to 1 followed by addition of 1
equiv of 2,3-dimethylbutene 5a (tetramethylethylene) provided
NHC−thexylborane adduct 6a as assessed by 11B NMR
spectroscopy (triplet, −24 ppm). No dialkylborane was
B
dx.doi.org/10.1021/ja407678e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
detected (doublet expected at about −20 ppm). Direct solvent
evaporation and flash chromatography provided 6a in 75%
yield (entry 1). We have previously made 6a by complexation
of thexylborane with 1,2-dimethylimidazol-2-ylidene,13 but this
new route is more convenient because 1 is readily available.
Results of catalyzed monohydroborations of an assortment of
other alkenes are summarized in entries 2−10 in Table 1. The
scales were 1 mmol of starting NHC−borane and 10% I2 unless
otherwise indicated. In all cases, the monohydroboration
products were stable, and the reported yields are after
purification by flash chromatography. Hydroboration of 2-
methylbutene (trimethylethylene) 5b provides 6b as a single
regioisomer with the boron atom on the secondary carbon
atom in 59% yield (entry 2). Likewise, hydroboration of 1-
methylcyclohexene 5c by 1 gives a single regio- and
stereoisomer 6c in 68% yield (entry 3).
acid,14 while Stephan and co-workers used the borenium ion
from 1,3-(2,6-diisopropyl-phenyl)imidazol-2-ylidene-9-BBN to
activate dihydrogen at room temperature.15 In both cases, the
starting complexes were made by addition of the appropriate
NHC to 9-BBN.
We made the related 9-BBN complex 8 by direct
hydroboration as shown in Scheme 2. Complex 1 was treated
Scheme 2. Synthesis of Cyclic NHC−Boranes by
Hydroboration
Individual hydroboration of the two enantiomers of α-pinene
(+)-5d and (−)-5d provided enantiomeric pinenyl boranes 6d
and ent-6d as single regio- and stereoisomers in 61% yield
(entries 4 and 5). The results in entries 3−5 show for the first
time that NHC−borenium hydroboration reactions occur with
syn stereochemistry, like borane hydroborations.1
Reactions with 2-methyl-1-phenylpropene 5e and (E)-1-
phenyl-1-propene 5f provided products 6e and 6f in 50% and
59% yield, respectively (entries 6 and 7). In both cases, the
boron atom ended up bonded to the benzylic carbon. The high
regioselectivity with the trisubstituted alkene 5e is perhaps
expected based on substitution, and indeed the only resonance
in the 11B NMR spectrum of the crude product from 5e was
that of 6e. There were two resonances in the spectrum of the
crude product from disubstituted alkene 5f in a ratio of about
90/10. The major resonance at −23.5 ppm belongs to 6f. This
overlaps a small resonance at −23.8 ppm that we suspect
belongs to the regioisomer of 6f (which could not be isolated in
pure form). If so, then the regioselectivity in the hydroboration
of 5e with 1 is considerably higher than free borane (∼3/1) and
on par with pyridine-iodoborane.5b
with 10% I2 as usual, followed by addition of 1,5-cyclo-
octadiene. A slow reaction occurred, and after 2 days, the 11B
NMR spectrum of the mixture showed a large doublet at −16.2
ppm (JBH = 85 Hz). These are values expected for product 8.
The 9-BBN complexes seem to be somewhat more robust than
dialkylborane complexes like 2, and 8 was isolated as a white
solid in 34% yield by flash chromatography. Thus, it is now
possible to make NHC−9-BBN complexes by either hydro-
boration of cyclooctadiene followed by complexation with the
NHC or the reverse.
Next, we looked at two intramolecular hydroborations10 of
complexes 10a and 10b. Such complexes are made by
deprotonation of the corresponding salt 9a,b to generate the
NHC, then addition of BH3·THF.10,16 The terminal alkene is
no competition for the carbene, and alkenyl NHC−boranes
10a,b are formed in good yield. Both compounds are stable and
show no tendency to hydroborate themselves on storage.
The usual reaction of N-allyl analogue 10a did not give a
clean product 11a; however, treatment of N-butenyl analogue
10b with 10% I2 in dichloromethane provided the interesting
fused bicyclic NHC−borane 11b. This was isolated as a white
solid in 25% yield by flash chromatography and characterized
by the usual means. Once isolated, solid 11b is stable. However,
we suspect that material loss occurs during chromatography
because the 11B NMR spectrum of the reaction mixture showed
formation of considerably more than 25% of 11b. These results
are complementary to those of Parrain,10 who reported that
rhodium catalysts promote intramolecular hydroborations to
make boracyclopentanes (including 11a) but not boracyclohex-
anes (like 11b).
Hydroborations of (E)- and (Z)-1,2-diphenylethene (stil-
bene) Z-5g and E-5g gave the same product 6g in about the
same yield (66% and 64%, entries 8 and 9). Hydroboration of
2,5-dimethylhexa-2,4-diene 5h gave substituted allyl borane 6h
in 37% yield (entry 10).
A couple of highly conjugated alkenes did not give clean
hydroboration products with this procedure. Specifically,
triphenylethene (Ph2CCHPh) and (E,E)-1,4-diphenylbuta-
1,3-diene (PhCHCH−CHCHPh) did not appear to react
under the standard conditions. Instead, gradual decomposition
of the starting NHC−borane was observed over two days.
We also looked briefly at two modifications of the standard
procedure. Reasoning that formation of the NHC−BH2I
catalyst would leave the amount of 1 deficient for product
formation with a 1/1 stoichiometry of 1 and alkene, we
increased the amount of 1 to 1.4 equiv in two experiments. In
entry 7, the yield of 6f increased significantly from 59% to 80%,
while in entry 8 the yield of 6g increased modestly from 66% to
73%. More practical of course is to simply decrease the amount
of iodine added. We probed this on a larger scale (2 mmol) by
adding only 5% iodine to 1 and 5b. This time, the yield of 6b
increased from 59% to 75%.
To close the scope study, we surveyed the reactions of six
typical NHC−boranes with 2-methyl-2-butene 5b. The stand-
ard procedure was used (10% I2, CH2Cl2, rt), and the structures
and isolated yields of the hydroboration products are shown in
Figure 2. Reaction of triazol-2-ylidene and benzimidazol-2-
ylidene NHC−boranes bearing (N,N)-dimethyl groups pro-
vided hydroboration products 12 and 13 in 70% and 71% yield.
Recently, NHC−9-BBN complexes (9-BBN is 9-
borabicyclo[3.3.1]nonane) have attracted attention in several
respects. Lindsay and co-workers made a stable borenium by
reaction of 1,3-dimesitylimidazol-2-ylidene-9-BBN with triflic
C
dx.doi.org/10.1021/ja407678e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
The catalytic cycle has steps of hydroboration (2) and
hydride transfer (3). Hydroboration with 18 could occur by
displacement of the iodide by the alkene 5b to give a species
like 19 (either an intermediate or a transition state), which
would rapidly hydroborate itself, then collapse by readdition of
iodide to 20. The formation of 19 can be considered a Lewis
base exchange or a nucleophilic substitution at boron.
Nucleophilic substitutions with anionic nucleophiles and
NHC−boryl halides and triflates are well precedented.11a
In the hydride transfer step (3), the boryl iodide 20 again
reacts by Lewis base exchange (or substitution at B) of the
iodide with a BH bond of the precursor 1 to provide a bridging
species 21,9,12 which then completes the hydride transfer to
give product 6b and return the catalyst 18. The bridging species
21 could be a transient intermediate rather than a transition
state. Related species have been observed in the absence of
good nucleophiles,9,12 though we did not observe 21 during
reactions.
Figure 2. Structures of products and yields in a scope study of NHC−
boranes with 2-methyl-2-butene.
Imidazol-2-ylidene NHC−boranes bearing medium-sized N-
alkyl groups (isopropyl or cyclohexyl) also reacted smoothly to
provide 14 and 15 in 57% and 50% yield. However, products
16 and 17 were not isolated from reactions of imidazol-2-
ylidene NHC−boranes bearing large N-aryl groups (mesityl
and 2,6-diisopropylphenyl). We know by 11B NMR spectros-
copy that iodine reacts with the starting boranes NHC−BH3 to
give the expected NHC−BH2I. In the last case, the
intermediate dipp-Imd-BH2I is not consumed when the alkene
is added, so apparently the hydroboration step fails due to the
large N-aryl substituents.
Several 11B NMR spectroscopic experiments supported key
aspects of this catalytic cycle. A stoichiometric reaction between
NHC−BH2I 18 and 2-methyl-2-butene 5b provided a single
major product with a broad resonance at −12 ppm. This is a
reasonable chemical shift for the intermediate boryl iodide 20.
In addition, the same resonance was generated by adding a
stoichiometric amount of diiodine (50 mol %) to the product
6b.17 As expected for a boryl iodide, intermediate 20 did not
survive flash chromatography. To simulate the catalytic
conditions for the hydride transfer step (3), we generated 20
in situ by adding iodine to 6b, then adding 10 equiv of starting
NHC−borane 1. The major new resonances in the 11B NMR
spectrum belonged to hydride transfer products 6b and 18.
This hydride transfer reaction might be reversible, but further
hydroboration of boryl iodide 18 drives the process forward.
Mechanistic Experiments. Several 11B NMR experiments
also probed what intermediates might be formed and how they
are consumed in these reactions. Figure 3 shows plausible steps
of a catalytic cycle that is more sophisticated than the cycle in
Figure 1. In the catalyst generation step (1), reaction of 2 equiv
of 1 with I2 provides 2 equiv of covalent boryl iodide 18.11b Left
alone, this boryl iodide is stable for long periods in solution.
However, it is sensitive to neutral nucleophiles like water,
methanol, and presumably alkenes.
CONCLUSIONS
■
The new procedure for activation of NHC−boranes toward
hydroboration by diiodine is both more convenient and less
expensive than the original triflimide activation procedure.
Most importantly, the scope of the hydroboration has been
dramatically expanded. Indeed, with the exception of 1-hexene
(Scheme 1), all of the hydroboration substrates used in this
paper either did fail or would be expected to fail with the
original triflimide method. Monohydroboration occurs with di-
and tri- and tetra-substituted alkenes, and these products can be
isolated by flash chromatography and are stable on storage.
11B NMR experiments have also provided helpful informa-
tion about the likely intermediates and the mechanism.
Reactions of NHC−BH3 with triflimide and diiodine provided
two structurally different intermediates. Even though the first
was ionic and the second was covalent, both reacted like a
borenium ion in the subsequent steps of the catalytic cycle.
These steps are hydroboration and hydride transfer. This and
the prior results of Vedejs5 with amine- and phosphine-boranes
suggest that it is possible to evince borenium ion chemistry of
N-heterocyclic carbene boranes from several different kinds of
intermediates. On the basis of the results of the scope studies,
the intermediates in the triflimide procedure seem more
aggressive, while those in the iodine procedure seem gentler. All
this bodes well for expansion of the chemistry of N-heterocyclic
carbene boranes through the use of borenium ions of their
reactive equivalents.
Figure 3. Plausible steps in a catalytic cycle for hydroboration
illustrated with 1 (NHC is 1,3-dimethylimidazol-2-ylidene) and 2-
methyl-2-butene 5b.
D
dx.doi.org/10.1021/ja407678e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(14) McArthur, D.; Butts, C. P.; Lindsay, D. M. Chem. Commun.
2011, 47, 6650−6652.
ASSOCIATED CONTENT
■
S
* Supporting Information
(15) Farrell, J. M.; Hatnean, J. A.; Stephan, D. W. J. Am. Chem. Soc.
2012, 134, 15728−15731.
Complete details of experiments and compound character-
1
ization and copies of H, 11B, and 13C NMR spectra of all
(16) Solovyev, A.; Ueng, S.-H.; Monot, J.; Fensterbank, L.; Malacria,
products. This material is available free of charge via the
Internet at http://pubs.acs.org.
̂
M.; Lacote, E.; Curran, D. P. Org. Lett. 2010, 12, 2998−3001.
(17) Alkyl-subsituted boryl iodides like 20 are less stable than boryl
iodide 18. Gradual decomposition generates a resonance at 33.7 ppm
in the 11B NMR spectrum. This shift is consistent with a borinic acid,
RB(OH)2.
AUTHOR INFORMATION
■
Corresponding Author
curran@pitt.edu
Present Address
†Institut des Molec
́
ules et Mater
́
iaux du Mans (IMMM) - UMR
6283, Universite
́
du Maine, Le Mans, France.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Science Foundation for funding, and we
thank Prof. Edwin Vedejs, University of Michigan, and Dr.
̂ ́
Emmanuel Lacote, CNRS, Universite de Lyon, for helpful
discussions. We dedicate this paper to Prof. Edwin Vedejs for
his inspirational contributions to organoboron chemistry.
REFERENCES
■
(1) (a) Zaidlewicz, M. In Comprehensive Organometallic Chemistry;
Wilkinson, G., Stone, F. G. A., Abel, E., Eds.; Pergamon: Oxford, 1982;
Vol. 7, pp 143−254; (b) Brown, H. C. Hydroboration; W. A. Benjamin:
Reading, MA, 1980. (c) Brown, H. C.; Zaidlewicz, M. Organic
Syntheses Via Boranes; Aldrich Chemical Co.: Milwaukee, WI, 2001;
Vol. 2.
(2) Staubitz, A.; Robertson, A. P. M.; Sloan, M. E.; Manners, I. Chem.
Rev. 2010, 110, 4023−4078.
(3) Curran, D. P.; Solovyev, A.; Makhlouf Brahmi, M.; Fensterbank,
L.; Malacria, M.; Laco
10317.
̂
te, E. Angew. Chem., Int. Ed. 2011, 50, 10294−
(4) Kanth, J. V. B. Aldrichim. Acta 2002, 35, 57−66.
(5) (a) Shapland, P.; Vedejs, E. J. Org. Chem. 2006, 71, 6666−6669.
(b) Karatjas, A. G.; Vedejs, E. J. Org. Chem. 2008, 73, 9508−9510.
(c) De Vries, T. S.; Prokofjevs, A.; Harvey, J. N.; Vedejs, E. J. Am.
Chem. Soc. 2009, 131, 14679−14687. (d) Prokofjevs, A.; Vedejs, E. J.
Am. Chem. Soc. 2011, 133, 20056−20059. (e) Prokofjevs, A.; Kampf, J.
W.; Vedejs, E. Angew. Chem., Int. Ed. 2011, 50, 2098−2101. (f) Cazorla,
C.; De Vries, T. S.; Vedejs, E. Org. Lett. 2013, 15, 984−987. (g)
Hydroboration of pinene with DMAP−borane can also be catalyzed
by iodine. See reference 5a.
(6) De Vries, T. S.; Prokofjevs, A.; Vedejs, E. Chem. Rev. 2012, 112,
4246−4282.
(7) Solovyev, A.; Geib, S. J.; Laco
2012, 31, 54−56.
(8) Kolle, P.; Noth, H. Chem. Rev. 1985, 85, 399−418.
̂
te, E.; Curran, D. P. Organometallics
̈
̈
̀ ̂
(9) Prokofjevs, A.; Boussonniere, A.; Li, L.; Bonin, H.; Lacote, E.;
Curran, D. P.; Vedejs, E. J. Am. Chem. Soc. 2012, 134, 12281−12288.
(10) Toure, M.; Chuzel, O.; Parrain, J. L. J. Am. Chem. Soc. 2012, 134,
17892−17895.
(11) (a) Solovyev, A.; Chu, Q.; Geib, S. J.; Fensterbank, L.; Malacria,
M.; Laco
15080. (b) Monot, J.; Solovyev, A.; Bonin-Dubarle, H.; Derat, E.;
Curran, D. P.; Robert, M.; Fensterbank, L.; Malacria, M.; Lacote, E.
Angew. Chem., Int. Ed. 2010, 49, 9166−9169.
(12) Ines, B.; Patil, M.; Carreras, J.; Goddard, R.; Thiel, W.; Alcarazo,
M. Angew. Chem., Int. Ed. 2011, 50, 8400−8403.
(13) (a) Pan, X.; Lacote, E.; Lalevee, J.; Curran, D. P. J. Am. Chem.
Soc. 2012, 134, 5669−5675. (b) Walton, J. C.; Makhlouf Brahmi, M.;
Fensterbank, L.; Lacote, E.; Malacria, M.; Chu, Q.; Ueng, S.-H.;
Solovyev, A.; Curran, D. P. J. Am. Chem. Soc. 2010, 132, 2350−2358.
̂
te, E.; Curran, D. P. J. Am. Chem. Soc. 2010, 132, 15072−
́
̂
́
̂
́
̂
E
dx.doi.org/10.1021/ja407678e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX