Iridium-catalyzed dehydrogenative silylation of azulenes based on regioselective C-H bond activation

Article abstract of DOI:10.1021/acs.orglett.5b00575

Use of an iridium catalyst allowed the efficient dehydrogenative functionalization of C-H bonds of azulenes with the production of hydrogen as the sole byproduct. The reaction occurred with excellent chemo- and regioselectivities to provide 2-silylazulene

Full text of DOI:10.1021/acs.orglett.5b00575

Letter
pubs.acs.org/OrgLett
Iridium-Catalyzed Dehydrogenative Silylation of Azulenes Based on
Regioselective C−H Bond Activation
Masahito Murai,*,† Keishi Takami,† Hirotaka Takeshima,† and Kazuhiko Takai*,†,§,#
§
†Division of Chemistry and Biotechnology, Graduate School of Natural Science and Technology, and Research Center of New
Functional Materials for Energy Production, Storage and Transport, Okayama University, 3-1-1 Tsushimanaka, Kita-ku, Okayama
700-8530, Japan
#ACT-C, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan
S
* Supporting Information
ABSTRACT: Use of an iridium catalyst allowed the efficient dehydrogenative
functionalization of C−H bonds of azulenes with the production of hydrogen as
the sole byproduct. The reaction occurred with excellent chemo- and
regioselectivities to provide 2-silylazulenes even without any directing groups.
Effective conjugation through the 2-position of the azulene ring was demonstrated
by the unique stimuli-responsiveness against an acid−base reaction.
zulenes, which show unique photophysical and redox
properties derived from their unusual dipolar and π-
Scheme 1. Direct Functionalization of C−H Bonds at the 2-
A
Position of Azulene Derivatives
electron polarizations, are promising components for the
construction of advanced functional materials.1 The develop-
ment of simple and efficient chemical modification methods
would provide an opportunity to disclose their novel electro- and
physical properties and further extend their utility as new
functional molecules. Since azulene has a dipole moment
between the 6 and 2 positions, selective functionalizations at
these positions are highly desirable to emphasize the dipolar
structure directly.2 However, this is generally difficult due to the
uniquely localized HOMO and LUMO orbitals. The basic
reactivity of azulene can be understood by its resonance structure
and the large HOMO and LUMO coefficients at 1-, 3- (for
HOMO) and 4-, 8-positions (for LUMO), respectively (Figure
1). Therefore, electrophilic functionalization, such as bromina-
catalyzed direct arylation of C(sp2)−H and C(sp3)−H bonds
of guaiazulene with Ar−Br (Scheme 1b).5c,d These methods are
highly useful; however, functionalization occurs at both the 1-
and 2-positions to afford a mixture of several regioisomers with
undesirable substitution patterns. These regioisomers must be
separated for use as building blocks, which decreases the
synthetic utility of these two methods.
Figure 1. Dipolar structure and HOMO/LUMO orbital distributions of
We have developed transition-metal-catalyzed dehydrogen-
ative silylation, borylation, and phosphination of C(sp2)−H or
C(sp3)−H bonds as powerful tools for the construction of
complex heteroaromatic compounds.6 In 2015, we and Hartwig
independently reported iridium-catalyzed intermolecular dehy-
drogenative silylation of aromatic compounds without directing
groups.7,8 The reaction proceeded with opposite chemo- and
site-selectivity compared to that observed for general electro-
philic functionalization of aromatic compounds. Therefore, we
envisioned that the iridium-catalyzed dehydrogenative silylation
of C(sp2)−H bonds of azulenes would allow a straightforward
azulene.
tion and formylation, occurs at the 1- and 3-positions, whereas
nucleophilic addition preferentially occurs at the 4- and 8-
positions.3 Selective functionalization at the 2- and 6-positions
usually requires multistep transformations, which sometimes
includes the construction of the azulene skeleton itself.4
Recently, catalytic direct functionalization of unactivated C−
H bonds has received considerable attention as a practical
preparation method for substituted aromatics. Utilizing this type
of transformation, two methods for functionalization at the 2-
position of the azulene ring have been reported, i.e., iridium-
catalyzed direct borylation of C(sp2)−H bonds of azulenes with
bis (pinacolato)diboron (Scheme 1a)5a,b and palladium-
Received: February 26, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b00575
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
modification of their electronic nature and, hence, of the
properties of the resulting azulene derivatives in a single step.
The current study describes the highly selective dehydrogenative
functionalization enabling the direct use of azulenes as substrates
and the production of hydrogen as the sole byproduct.
Treatment of azulene with 3 equiv of triethylsilane and 3,3-
dimethyl-1-butene as a hydrogen acceptor in the presence of
[Ir(OMe)(cod)]2 as a catalyst and 3,4,7,8-tetramethyl-1,10-
phenanthroline (hereafter referred to as “tmphen”) as a ligand in
cyclohexane (0.20 M) afforded 2-silylazulene 1a as a single
regioisomer in 81% yield (eq 1). In sharp contrast with the
(trifluoromethylphenyl)dimethylsilane as a silane source re-
sulted in a decrease of the corresponding coupling product 1e,
along with diarylmethylsilane 2 in 9% yield (entry 4). This
unexpected coupling product 2 might be formed via the
competitive disproportionation of 3,5-(trifluoromethylphenyl)-
dimethylsilane during the dehydrogenative silylation.11 Benzyl-
dimethylsilane was also a suitable substrate to afford 2-
silylazulene 1f in 79% yield even at 100 °C (entry 5). In this
reaction, slow addition of silane was required to obtain the
coupling product in good yield. 6-Heptylazulene can be used in
place of azulene to provide the corresponding 2-silylazulene 1g in
74% yield.12 Note that 2-silylazulenes were obtained selectively
without forming their regioisomers, including 1-silylazulenes, in
all of the reactions listed in Table 1. 2-(Dimethylsilyl)thiophene
and alkoxysilanes such as silatrane and triethoxysilane were not
applicable as silane sources.
During these investigations, selective dehydrogenative silyla-
tion of azulene was observed in the presence of other aromatic
compounds. For example, competition experiments with
naphthalene in the same flask, which is the structural isomer of
azulene, produced 2-silylazulene selectively with naphthalene
recovered intact (Scheme 2). Similar selective silylation of
azulene was observed in the reactions with pyrene, anthracene,
phenanthrene, and benzene.
previous intermolecular dehydrogenative silylation,9a−h the
reaction proceeded efficiently with small excess amounts of
triethylsilane. The use of [Rh(OMe)(cod)]2 in place of the
iridium catalyst completely shut down the reaction. In contrast to
our previous report on the dehydrogenative silylation of
naphthalene,7 1a was obtained in 19% yield without any
hydrogen acceptors. Other transition-metal complexes, including
rhenium, ruthenium, rhodium, and platinum catalysts, were
totally ineffective with the recovery of most of the azulene used.10
Using the combination of [Ir(OMe)(cod)]2 with tmphen as
the catalyst and ligand, the effect of the substitution pattern of
silanes was investigated (Table 1). Aryldimethylsilanes possess-
ing electron-donating and -withdrawing substituents reacted to
furnish the corresponding 2-silylazulenes 1b−d in moderate
yields (entries 1−3), which implied that variation of the
electronic properties of the silane had little influence on the
reactivity. The use of highly electron-deficient 3,5-
Scheme 2. Selective Dehydrogenative Silylation of Azulene in
the Presence of Naphthalene (Si = SiMe2Bn)
In our previous study on the dehydrogenative silylation of
polycyclic aromatic hydrocarbons, electron-deficient substrates
were more efficiently silylated.7 However, the dehydrogenative
silylation of C(sp2)−H bonds at the electron-deficient seven-
membered ring of azulene was not observed under the present
reaction conditions. This reactivity might be explained by
considering the ease of the π-complexation between an iridium
center and the five-membered ring of azulene. Azulene has a
fused structure of a cyclopentadienyl anion (Cp−) and
cycloheptatrienyl cation, and Cp− is well-known as a good
ligand for various transition-metal complexes. Although C−H
bond activation without using a heteroatom-containing directing
group is generally difficult,13 the above interaction might fix the
iridium center near the C−H bond on the five-membered ring
and promote this energetically unfavorable step (see inter-
mediate B in Scheme 3).14
Taking these observations into consideration, a tentative
reaction mechanism is proposed in Scheme 3 (tmphen ligands
are omitted for clarity). The Ir−OMe complex is initially
converted to an Ir−H species via oxidative addition to
hydrosilane followed by the reductive elimination of Si−
OMe.15 This species is then oxidatively added to hydrosilane,
inserted onto 3,3-dimethyl-1-butene, and reductively eliminated
to furnish the Ir−Si species A.16 Oxidative addition of A to the
C(sp2)−H bond at the 2-position of azulene, which can be
promoted by the facile interaction of the iridium center with the
Cp− moiety of azulene, was followed by reductive elimination to
afford the corresponding 2-silylazulenes. Chemoselective
Table 1. Iridium-Catalyzed Regioselective Dehydrogenative
Silylation of Azulenes with Hydrosilanes
temp
(°C)
yield
product (%)
a
entry method
R1
SiR2
3
b
1
2
A
A
125
125
H
H
SiMe2Ph
1b
1c
67
66
b
SiMe2(4-
MeOC6H4)
3
4
A
A
125
150
H
H
H
SiMe2(4-
1d
1e
61
41
CF3C6H4)
c
SiMe2(3,5-
(CF3)2C6H3)
b
d
5
6
B
B
100
100
SiMe2Bn
nC7H15 SiEt3
1f
79
74
1g
a
Method A: Norbornene (hydrogen acceptor) in octane. Method B:
b
3,3-Dimethyl-1-butene (hydrogen acceptor) in cyclohexane. For 10 h.
c
d
Compound 2 was obtained as a byproduct in 9%. For 4 h.
B
DOI: 10.1021/acs.orglett.5b00575
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
The resulting 2-silylazulene can be utilized for further
derivatization of the azulene ring. Treatment of 2-silylazulene
1a with bis(pinacolato)diboron in the presence of an iridium
catalyst furnished 6-boryl-2-silyl- and 5-boryl-2-silylazulenes 5a
and 5b in 52% and 18% yields, respectively (Scheme 6a).5a,b The
Scheme 3. Plausible Reaction Mechanism (Si = SiR3)
Scheme 6. Synthetic Derivatization of 2-Silylazulenes 1a and
1e (Si and B = SiEt3 and Bpin, Respectively)
silylation of azulene in the presence of other polycyclic aromatic
hydrocarbons observed in Scheme 2 can be explained by
predominant formation of the Cp−Ir intermediate B.17
Furthermore, steric factors dominated the site-selectivity of the
reaction to provide 2-silylazulenes exclusively via regioselective
C−H bond activation at the least sterically hindered 2-position.
In the course of these studies, unexpected formation of 2,6-
disilyl- and 2,5-disilylazulenes 3a and 3b was observed in the
reaction of azulene with triethylsilane in a reduced amount of
cyclohexane (1.0 M). These disilylated azulenes were likely
formed via two sequential dehydrogenative silylations on the
five- and seven-membered rings of azulene. In fact, treatment of
2-silylazulene 1a with triethylsilane in the presence of the iridium
complex furnished disilylazulenes 3a and 3b in 39% and 12%
yields, respectively (Scheme 4).
borylation occurred preferentially at the 6-position and followed
similar selectivity as was observed in the silylation of the 7-
membered ring (see Scheme 4). Chlorination of 1a using CuCl2
afforded 1-chloro-2-silylazulene 6 in 46% yield (Scheme 6b).
Note that this chlorination without affecting the silyl group
functionality cannot be attained by the typical chlorination
schemes using chlorine or N-chlorosuccinimide. 2-Arylazulenes
7a and 7b were obtained by the palladium-catalyzed Hiyama
cross-coupling reaction of 2-silylazulene 1e with aryl iodides,19,20
showing the potential utility of the present reaction toward the
effective expansion of azulene’s π-conjugated system (Scheme
6c).
Scheme 4. Iridium-Catalyzed Dehydrogenative Silylation of
C(sp2)−H Bonds on the Seven-Membered Ring of Azulene
Effective conjugation through the 2-position of the azulene
ring was confirmed by the longest absorption maximum peak for
7b (385 nm), which was red-shifted by 15 nm compared with
that of the regioisomer 1-(4-(trifluoromethyl)phenyl)azulene
7b′ (Figure 2).21 Moreover, the unique stimuli-responsiveness
was demonstrated by the reversible color change against acid−
base reaction of 7b.22 A significant change in the absorption was
found upon addition of TFA to a solution of 7b in CH2Cl2. An
absorption at 427 nm appeared along with an instant color
Unfortunately, dihydrosilanes, including Et2SiH2, Ph2SiH2,
and PhMeSiH2, were not applicable, and the scope of the silane
was limited to monohydrosilanes under the current reaction
conditions. Taking advantage of this reactivity difference, a
cascade reaction initiated by hydrosilylation with an olefin
followed by the dehydrogenative silylation of azulene was carried
out. First, dihydrosilane Et2SiH2 was treated with 1-hexene in the
presence of Wilkinson catalyst, RhCl(PPh3)3, at 25 °C for 8 h.18
Subsequently, the dehydrogenative silylation step was performed
under the current iridium-catalyzed reaction conditions to
furnish the expected three-component coupling product, 2-
silylazulene 4, in 84% yield (Scheme 5).
Scheme 5. Three-Component Coupling Reaction Leading to
2-Silylazulene 4
Figure 2. UV−vis spectra in CH2Cl2 of 7b (red) and 7b′ (blue) for
neutral state (solid line) and upon the addition of CF3CO2H (dashed
line). Inset: photographs of the neutral state (left) and after protonation
by CF3CO2H (right) for each compound.
C
DOI: 10.1021/acs.orglett.5b00575
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Seiki, T.; Takai, K. Angew. Chem., Int. Ed. 2013, 52, 1520. (d) Kuninobu,
Y.; Iwanaga, T.; Omura, T.; Takai, K. Angew. Chem., Int. Ed. 2013, 52,
4431. (e) Murai, M.; Matsumoto, K.; Okada, R.; Takai, K. Org. Lett.
2014, 16, 6492.
change of the solution from blue to yellow. In contrast, the
regioisomer 1-arylazulene 7b′ did not show such behavior, which
clearly implies the utility of the current regioselective
functionalization at the 2-position of the azulene ring. These
differences can be explained by the perturbation of the π-
conjugation upon the formation of the azulenium cation. It is
known that protonation of azulene derivatives occurs at the 1-
and/or 3-position on the electron-rich five-membered ring.
Therefore, the structure of 1-arylazulene 7b′ might not be
effective for the conjugation since it cannot retain the planar
structure after the protonation.
In conclusion, a novel straightforward functionalization of C−
H bonds at the 2-position of azulene based on regioselective
dehydrogenative silylation has been described. The current
method provides a practical route for the construction of
azulene’s uniquely polarized π-system, which is difficult to access
by conventional approaches.
(7) Murai, M.; Takami, K.; Takai, K. Chem.Eur. J. 2015, 21, 4566.
(8) Cheng, C.; Harwig, J. F. J. Am. Chem. Soc. 2015, 137, 592.
(9) For intermolecular dehydrogenative silylation without any
directing groups, see: (a) Gustavson, W. A.; Epstein, P. S.; Curtis, M.
D. Organometallics 1982, 1, 884. (b) Sakakura, T.; Tokunaga, Y.;
Sodeyama, Y.; Tanaka, M. Chem. Lett. 1987, 2375. (c) Ishikawa, M.;
Okazaki, S.; Naka, A.; Sakamoto, H. Organometallics 1992, 11, 4135.
(d) Uchimaru, Y.; El Sayed, A. M. M.; Tanaka, M. Organometallics 1993,
12, 2065. (e) Djurovich, P. I.; Dolich, A. R.; Berry, D. H. J. Chem. Soc.,
Chem. Commun. 1994, 1897. (f) Ezbiansky, K.; Djurovich, P. I.;
LaForest, M.; Sinning, D. J.; Zayes, R.; Berry, D. H. Organometallics
1998, 17, 1455. (g) Tsukada, N.; Hartwig, J. F. J. Am. Chem. Soc. 2005,
127, 5022. (h) Murata, M.; Fukuyama, N.; Wada, J.-i.; Watanabe, S.;
Masuda, Y. Chem. Lett. 2007, 36, 910. (i) Cheng, C.; Hartwig, J. F.
Science 2014, 343, 853. (j) Cheng, C.; Hartwig, J. F. J. Am. Chem. Soc.
2014, 136, 12064.
(10) Other transition-metal complexes, including RhCl(PPh3)3,
[Rh(OAc)2]2, Re2(CO)10, Ru3(CO)12, and Ir4(CO)12, did not promote
the desired reaction, and most of azulene was recovered.
(11) Disproportionation of 1e to 2 via the exchange of the methyl and
3,5-di(trifluoromethyl)phenyl groups was not observed under the
current conditions, which implies that the disproportionation occurred
before dehydrogenative silylation of the C−H bond.
(12) The reaction of guaiazulene with triethylsilane under the reaction
conditions noted in Table 1, entry 5, afforded the corresponding 2-
silylazulene as a single regioisomer in 22% yield.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, spectroscopic data for all new
1
compounds, and copies of H and 13C NMR spectra. This
material is available free of charge via the Internet at http://pubs.
acs.org.
AUTHOR INFORMATION
Corresponding Authors
■
(13) For a review, see: Kuhl, N.; Hopkinson, M. N.; Wencel-Delord, J.;
Glorius, F. Angew. Chem., Int. Ed. 2012, 51, 10236.
(14) (a) Sibbald, P. A.; Rosewall, C. F.; Swartz, R. D.; Michael, F. E. J.
Am. Chem. Soc. 2009, 131, 15945. (b) Hickman, A. J.; Sanford, M. S. ACS
Catal. 2011, 1, 170.
*E-mail: masahito.murai@cc.okayama-u.ac.jp.
*E-mail: ktakai@cc.okayama-u.ac.jp.
Notes
The authors declare no competing financial interest.
(15) The reaction of Ir−Cl with H−SiR3 to generate Ir−H species via
oxidative addition followed by reductive elimination of Cl−SiR3 has
ACKNOWLEDGMENTS
■
been reported. See: Esteruelas, M. A.; Olivan
2013, 52, 12108.
́ ́
, M.; Velez, A. Inorg. Chem.
This work was financially supported by a Grant-in-Aid (No.
26248030) from MEXT, Japan, and the MEXT program for
promoting the enhancement of research universities. We
gratefully thank Mr. Shinji Iba and Takahisa Sakae (Okayama
University) for HRMS measurements.
(16) For theoretical studies on the mechanism of iridium-catalyzed
intramolecular dehydrogenative silylation of C−H bonds, see: (a) Parija,
A.; Sunoj, R. B. Org. Lett. 2013, 15, 4066. The alternative mechanism via
oxidative addition or σ-bond metathesis of Ir−(SiEt3)3 with Ar−H,
which is generally proposed for the iridium-catalyzed borylation of the
C−H bond, cannot be ruled out. See: (b) Mkhalid, I. A. I.; Barnard, J. H.;
Marder, T. B.; Murphy, J. M.; Hartwig, J. F. Chem. Rev. 2010, 110, 890.
(c) Hartwig, J. F. Acc. Chem. Res. 2012, 45, 864.
(17) Kunkely, H.; Vogler, A. Inorg. Chem. Commun. 2006, 9, 830.
(18) Nielsen, L.; Skrydstrup, T. J. Am. Chem. Soc. 2008, 130, 13145.
(19) For representative works, see: (a) Hatanaka, Y.; Hiyama, T.
Synlett 1991, 845. (b) Hiyama, T.; Hatanaka, Y. Pure Appl. Chem. 1994,
66, 1471. (c) Katayama, H.; Taniguchi, K.; Kobayashi, M.; Sagawa, T.;
Minami, T.; Ozawa, F. J. Organomet. Chem. 2002, 645, 192. (d) Hiyama,
T. J. Organomet. Chem. 2002, 653, 58. (e) Nakao, Y.; Imanaka, H.;
Sahoo, A. K.; Yada, A.; Hiyama, T. J. Am. Chem. Soc. 2005, 127, 6952.
(f) Sore, H. F.; Galloway, W. R. J. D.; Spring, D. R. Chem. Soc. Rev. 2012,
41, 1845.
REFERENCES
(1) Nozoe, T. Pure Appl. Chem. 1972, 28, 239.
■
(2) (a) Noll, G.; Daub, J.; Lutz, M.; Rurack, K. J. Org. Chem. 2011, 76,
̈
4859. (b) Koch, M.; Blacque, O.; Venkatesan, K. Org. Lett. 2012, 14,
1580. (c) Yamaguchi, Y.; Maruya, Y.; Katagiri, H.; Nakayama, K.; Ohba,
Y. Org. Lett. 2012, 14, 2316. (d) Xia, J.; Capozzi, B.; Wei, S.; Strange, M.;
Batra, A.; Moreno, J. R.; Amir, R. J.; Amir, E.; Solomon, G. C.;
Venkataraman, L.; Campos, L. M. Nano Lett. 2014, 14, 2941.
(3) (a) Yokoyama, R.; Ito, S.; Watanabe, M.; Harada, N.; Kabuto, C.;
Morita, N. J. Chem. Soc., Perkin Trans. 1 2001, 2257. (b) Ito, S.; Nomura,
A.; Morita, N.; Kabuto, C.; Kobayashi, H.; Maejima, S.; Fujimori, K.;
Yasunami, M. J. Org. Chem. 2002, 67, 7295. (c) Zhang, J.; Petoud, S.
Chem.Eur. J. 2008, 14, 1264.
(20) 2-Silylazulene 2 can be also utilized as a coupling partner to afford
7a in 66% yield by the reaction with iodobenzene.
(4) (a) Nozoe, T.; Seto, S.; Matsumura, S.; Murase, Y. Bull. Chem. Soc.
Jpn. 1962, 35, 1179. (b) McDonald, R. N.; Richmond, J. M.; Curtis, J. R.;
Petty, H. E.; Hoskins, T. L. J. Org. Chem. 1976, 41, 1811.
(21) For 7b: λmax = 297 nm (ε = 2.7 × 104), 306 nm (ε = 2.7 × 104), 367
nm (ε = 0.48 × 104), 385 nm (ε = 0.57 × 104). For 7b′: λmax = 302 (ε =
1.7 × 104), 370 (ε = 0.48 × 104).
(5) (a) Kurotobi, K.; Miyauchi, M.; Takakura, K.; Murafuji, T.;
Sugihara, Y. Eur. J. Org. Chem. 2003, 3663. (b) Fujinaga, M.; Murafuji,
T.; Kurotobi, K.; Sugihara, Y. Tetrahedron 2009, 65, 7115. (c) Zhao, L.;
Bruneau, C.; Doucet, H. Chem. Commun. 2013, 49, 5598. (d) Dyker, G.;
(22) For recent examples, see: (a) Murai, M.; Ku, S.-Y.; Treat, N. D.;
Robb, M. J.; Chabinyc, M. L.; Hawker, C. J. Chem. Sci. 2014, 5, 3753.
(b) Amir, E.; Murai, M.; Amir, R. J.; Cowart, J. S., Jr; Chabinyc, M. L.;
Hawker, C. J. Chem. Sci. 2014, 5, 4483.
Borowski, S.; Heiermann, J.; Korning, J.; Opwis, K.; Henkel, G.;
̈
Kockerling, M. J. Organomet. Chem. 2000, 606, 108.
̈
(6) (a) Ureshino, T.; Yoshida, T.; Kuninobu, Y.; Takai, K. J. Am. Chem.
Soc. 2010, 132, 14324. (b) Kuninobu, Y.; Yoshida, T.; Takai, K. J. Org.
Chem. 2011, 76, 7370. (c) Kuninobu, Y.; Yamauchi, K.; Tamura, N.;
D
DOI: 10.1021/acs.orglett.5b00575
Org. Lett. XXXX, XXX, XXX−XXX