Rhodium-catalyzed intramolecular C-H silylation by silacyclobutanes

Article abstract of DOI:10.1002/anie.201602376

Silacyclobutane was discovered to be an efficient C-H bond silylation reagent. Under the catalysis of Rh^I/TMS-segphos, silacyclobutane undergoes sequential C-Si/C-H bond activations, affording a series of π-conjugated siloles in high yields and regioselectivities. The catalytic cycle was proposed to involve a rarely documented endocyclic β-hydride elimination of five-membered metallacycles, which after reductive elimination gave rise to a Si-Rh^I species that is capable of C-H activation. Old reagent, new reactivity: Silacyclobutane was discovered to be an efficient C-H bond silylation reagent under the catalysis of Rh/TMS-segophos. This new reactivity was attributed to a key Si-Rh^I intermediate formed through a Si-C activation, endocyclic β-H elimination and reductive elimination cascade. A wide array of siloles was obtained in high yields and excellent regioselectivities.

Full text of DOI:10.1002/anie.201602376

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201602376
German Edition: DOI: 10.1002/ange.201602376
À
C H Activation
À
Rhodium-Catalyzed Intramolecular C H Silylation by
Silacyclobutanes
Qing-Wei Zhang+, Kun An+, Li-Chuan Liu, Shuangxi Guo, Chenran Jiang, Huifang Guo, and
Wei He*
Abstract: Silacyclobutane was discovered to be an efficient
I
À
C H bond silylation reagent. Under the catalysis of Rh /TMS-
À
À
segphos, silacyclobutane undergoes sequential C Si/C H
bond activations, affording a series of p-conjugated siloles in
high yields and regioselectivities. The catalytic cycle was
proposed to involve a rarely documented endocyclic b-hydride
elimination of five-membered metallacycles, which after
I
À
reductive elimination gave rise to a Si Rh species that is
À
capable of C H activation.
S
ilacyclobutanes (SCB) were first reported by Sommer and
Baum over half a century ago.[1] Compared with other
organosilicons, SCB is unique due to their high ring strain
energy, as evidenced in ring opening polymerization,[2] aldol
reactions of SCB[3] and retro-[2+2] reaction.[4] Also because
of the high ring strain, low valent transition metals can facilely
insert into the SCB to form 5-membered silametallacycles.[5]
A signature reactivity of the resulting silametallacycles
À
Figure 1. Silacyclobutanes as C H silylation reagent.
to RhI). However, these silametallacycles were proposed to
be stable against b-H elimination.[12] In general, endocyclic b-
H elimination of related metallacycles (especially that of 3- to
6-membered rings) are conceived to be extremely challeng-
ing.[13] A classic example is the alkene–alkyne coupling
reaction reported by Trost, in which acyclic b-H elimination
was shown to prevail.[14] Although DFT calculations sug-
gested that endocyclic b-H elimination might be possible,[15]
only a few examples were reported very recently.[16] Herein
À
involves the migratory insertion of p systems into the M Si
bonds. Based on this reactivity, various fascinating formal
cycloaddition reactions of SCB have been developed.[6] For
example, reactions of alkynes or alkenes with SCB resulted in
formal [s2+p2] cycloaddition[7] (Figure 1a). Similarly, enones
were shown to give formal [s2+p4] products.[8] Most recently,
Murakami and co-workers discovered that electron-rich
silapalladacycle is capable of oxidative insertion into an
adjacent cyclobutanone, cultivated in an intriguing formal
s bond metathesis reaction.[9] Despite these advancements,
À
we reported that SCB is an efficient reagent for C H bond
silylation[17] under Rh catalysis in the presence of a hindered
ligand, possibly enabled by endocyclic b-H elimination of Rh
silametallacycles (Figure 1b). This reaction produces various
siloles in high yield with excellent regioselectivity.
À
SCB is not shown to react with C H bonds.
In our continuing research in the Rh catalyzed construc-
tion of Si C bonds,[10] we became curious if SCB could act as
À
À
a C H silylation reagent. If a 5-membered Rh silametalla-
Our investigation started with the intramolecular silyla-
tion reaction of SCB 1a. Since our previous work showed that
the steric hindrance of diphosphine ligands has a profound
cycle is to be assumed, converting this RhIII to catalytically
I
À
active Si Rh would be required, since only the latter is
[11]
I
influence[10b] on the C H bond silylation reaction, we
À
À
À
known to perform C H activation. Such a Si Rh might be
achieved by a two-step sequence: b-hydride elimination of the
Rh silametallacycle followed with reductive elimination (RhIII
screened a large panel of ligands (Table 1). No reaction
took place in the absence of a ligand (entry 1). Similar to our
previous observations,[10b] ligands with different dihedral
angles showed dramatically different reactivity. The less
hindered ligand spirophos and BINAP were ineffective
(entries 2–3). DTBM-MeO-Biphep, which possesses a wider
dihedral angle and more hindered 3, 5-di-t-Bu-4-MeO-phenyl
substituent, gave the silole product[18] 2a in 15% yield
(entry 4). A systematic screening of ligands with segphos
backbone that possess wider dihedral angles[19] was then
conducted (entries 5–8). Among these ligands, TMS-seg-
phos[20] gave the best result (52% yield, entry 7). Further
increasing the bulkiness, either by using larger 3, 5-di-
triethylsilylphenyl (TES)[21] or by employing C1-tunephos[22]
[*] Dr. Q.-W. Zhang,[+] K. An,[+] L.-C. Liu, S. Guo, C. Jiang, Dr. H. Guo,
Prof. Dr. W. He
School of Pharmaceutical Sciences, Tsinghua University
Beijing, 100084 (China)
E-mail: whe@mail.tsinghua.edu.cn
Homepage: http://www.med.tsinghua.edu.cn/index.php/en/
research/faculty/item/he-wei
[+] These authors contributed equally to this work.
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
1002/anie.201602376.
Angew. Chem. Int. Ed. 2016, 55, 6319 –6323
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6319

Angewandte
Communications
Chemie
Table 1: Screening of the reaction parameters.[a]
Table 2: Substrate scope.[a]
Entry
Substrate
Product
Yield[b]
1
2
3
4
5
6
R=H, 1a
2a
2b
2c
2d
2e
2 f
84
89[d]
90
93
84
86
78
83
Entry
Ligand
Yield[b]
R=3-Me, 1b
R=4-Me, 1c
R=4-t-Bu, 1d
R=4-Cl, 1e
R=4-F, 1 f
R=4-C(O)(CH2)3CH3, 1g
R=4-OMe, 1h
1
2
3
4
5
6
7
8
–
n.r.
n.r.
spirophos
BINAP
DTBM-MeO-Biphep
segphos
DTBM-segphos
TMS-segphos
TES-segphos
DTBM-C1-tunephos
TMS-segphos
TMS-segphos
TMS-segphos
trace
15%
n.r
10%
52%
trace
trace
trace
trace
84%
7[c]
8
2g
2h
9
10[c]
11[d]
12[e]
9
1i
1j
2i
2j
90
78
[a] Conditions: 0.1 mmol substrate, catalyst (10 mol%), ligand
(10 mol%) and toluene (1 mL) were dissolved in the glove box.
[b] Isolated yield. [c] Catalyst is [Rh(cod)2]BF4. [d] Catalyst is [Rh(cod)]-
(acac). [e] 48 h.
10
with narrower dihedral angle backbone, resulted in loss of
activity (entries 8 and 9). This nonlinear dependence of
catalyst performance on the steric hindrance of the ligands is
[10b]
À
in line with our previous C H silylation study.
Catalysts
11
12
1k
1l
2k
2l
85
with other counterions, such as [Rh(cod)2]BF4 and [Rh(cod)]-
(acac), were also inactive (entries 10 and 11). Gratifyingly,
when the reaction time was extended to 48 h, the desired
product 2a was isolated in 84% yield (entry 8 vs. 12).
À
This new C H silylation reaction is applicable to a wide
84[d]
scope of SCB substrates (Table 2). First, substituents on aryl
ring A were examined. Methyl substituent at either 3-position
(1b) or 4-position (1c) gave comparable yields to that of 1a
(entries 2 and 3 vs. entry 1). Only one of the two possible
regioisomeric products was obtained from substrate 1b: the
silylation occurred exclusively at the less hindered 6-position
(entry 2). Alkyl substitution at the 4-position (1c and 1d) did
not interfere with the reaction. With the bulky 4-t-Butyl group
(1d), the reaction yield increased to 93% (entry 4). Electron
withdrawing groups such as Cl, F, and keto (1e, 1 f and 1g) at
the 4 position were well tolerated and the desired products
(2e, 2 f and 2g) were isolated in 84%, 86% and 78% yields,
respectively (entries 5–7). It is worth noting that for substrate
13
1m
1n
2m
2n
91
57
14[c]
[a] Conditions: 0.1 mmol substrate, [Rh(cod)Cl]2 (5 mol%), TMS-seg-
phos (10 mol%) and toluene (1 mL) were dissolved in the glove box.
[b] Isolated yield. [c] 72 h. [d] other possible regioisomeric products were
not detected.
=
1g, possible cycloaddition to the C O bond was not observed,
neither was the transfer hydrogenation of the C O bond. This
might suggest that the C H activation is kinetically favored
=
À
and the Rh was not scramble to other parts of the reactant
during the reaction. Substrate 1e with the electron-rich
methoxyl group was also amenable to the protocol, giving the
desired product 2e in 83% yield (entry 8).
Other types of aryl ring A were also tested (entries 9–12).
Both 2-naphthyl (1i) and 1-naphthyl (1j) substrates afforded
p-extended conjugated silole products 2i, 2j in respectively
90% and 78% yield (entries 9 and 10). In the case of 1i, the
materials,[18] but their syntheses are more challenging than
siloles. Using this SCB method, both benzofuran and
thiophene fused siloles 2k and 2l were rapidly obtained in
85% and 84% yields, respectively (entries 11 and 12). Again,
the C H bond silylation of 1l occurred exclusively at the
ortho position of the sulfur atom (entry 12). Double C H
À
À
À
C H silylation occurred exclusively at the less hindered
position. Heterocyclic siloles are promising optoelectronic
silylation is possible: the fused bis-silole product 2m was
obtained in an excellent yield (91%) (entry 13). Although the
6320 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6319 –6323

Angewandte
Communications
Chemie
2
À
À
stereo center generated from the first C H silylation is far
away from the second silicon atom, excellent stereoselectivity
high yields (Figure 3a). When there is a competing C(sp ) H
À
bond such as in substrate 1u, silyation of both types of C H
bonds were observed and the C(sp ) H bond silylation
2
À
À
of the second C H silylation was observed, affording only one
diastereomeric product. Substrate 1n with ethyl substituent
on the silicon atom gave product 2n in a lower yield
(entry 14), confirming that the reaction is very sensitive to
steric effect.
product is slightly favoured (Figure 3b).
A series of experiments were carried out to explore the
reaction mechanism (Scheme 1). First, the fully deuterated
The effect of substituents on aryl ring B was then
investigated (Figure 2). Overall, this class of substrates is
less reactive, such that longer reaction time (72 h) was
Figure 2. Substituents on aryl ring B
Scheme 1. Deuterium experiments.
required for complete conversion. The reaction yields range
from moderate to good. 3-Isopropyl substituted substrate 1o
gave a 58% yield. When this same position was substituted
with a tert-butyl group (1q), the yield dropped to 21%. This
result again suggested that the steric hindrance, even seem-
ingly remote from the Si center, has a dramatic influence on
the reaction (c.f. 1m). In line with this observation, less
hindered aryl ring B is beneficial. For example, 3-methoxyl
substituted 2q and 3,4-dimethoxyl substituted substrate 2r
substrate 1a–d5 was subjected to the reaction conditions
(Scheme 1a). The desired product 2a–d5 was isolated in
a comparable yield (80%) to that of 1a (84%). The
deuterium atom was observed to transfer to both C1 and C2
positions of the propyl group (the existence of D atom in C1
and C2 was confirmed by both 1H and 2H NMR, see the
Supporting Information). This deuterium incorporation indi-
cates that the SCB is the internal hydrogen acceptor and the
re-hydrogenation step proceeded in a non-regioselective
fashion. Moreover, while reaction of mono-deuterated sub-
strate 1a-d1 revealed a moderate kinetic isotope effect (KIE)
(Scheme 1b), the parallel reactions between substrates 1a
and 1a-d5 however did not show a similar effect (Scheme 1c).
furnished the desired products in good yields.
3
À
During the screening, we discovered that C(sp ) H bonds
could also be silylated (Figure 3).[23] 2,6-Dimethyl substituted
substrates 1s and 1t reacted smoothly under the standard
conditions to afford the six-member silacycles 3a and 3b in
À
This pair of experiments suggested that the C H bond
activation is unlikely the rate limiting step.[24]
Based on our preliminary mechanistic study (Supporting
Information) we proposed a plausible mechanism (Figure 4)
using 1a as an example. First, five-membered silametallacycle
I is formed via reversible oxidative insertion of RhI complex
into the SCB.[5,25] Subsequent b-H elimination of I, possibly
enabled by the sterically hindered TMS-segphos ligand,
would give the key hydrido-rhodium(III) II, in which the
terminal alkene probably still coordinates to the Rh. Since
only RhI is known to activate C H bonds, reductive
À
elimination of intermediate II is assumed to give RhI
I
À
intermediated III along with HCl. The Si Rh then undergoes
2
3
À
À
À
Figure 3. C(sp ) H vs. C(sp ) H silylation.
reversible C H bond activation,[11] affording RhIII intermedi-
Angew. Chem. Int. Ed. 2016, 55, 6319 –6323
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6321

Angewandte
Communications
Chemie
À
Keywords: C H activation · metallocycle · silole · silylation ·
b-hydrogen elimination
How to cite: Angew. Chem. Int. Ed. 2016, 55, 6319–6323
Angew. Chem. 2016, 128, 6427–6431
[1] L. H. Sommer, G. A. Baum, J. Am. Chem. Soc. 1954, 76, 5002.
[2] N. V. Ushakov, E. S. Finkelshtein, Russ. Chem. Rev. 2013, 82, 205.
[3] a) A. G. Myers, S. E. Kephart, H. Chen, J. Am. Chem. Soc. 1992,
114, 7922; b) K. Matsumoto, K. Oshima, K. Utimoto, J. Org.
Chem. 1994, 59, 7152; c) S. E. Denmark, J. Org. Chem. 2009, 74,
2915, and references therein.
[4] R. T. Conlin, R. S. Gill, J. Am. Chem. Soc. 1983, 105, 618. For
review see: A. G. Brook, M. A. Brook, Adv. Organomet. Chem.
1996, 39, 71.
[5] Y. Tanaka, H. Yamashita, S. Shimada, M. Tanaka, Organo-
metallics 1997, 16, 3246.
[6] For reviews see, a) L. Li, Y. Zhang, L. Gao, Z. Song, Tetrahedron
Lett. 2015, 56, 1466; b) K. Hirano, H. Yorimitsu, K. Oshima,
Chem. Commun. 2008, 3234.
Figure 4. Proposed mechanism.
[7] a) H. Sakurai, T. Imai, Chem. Lett. 1975, 891; b) Y. Takeyama, K.
Nozaki, K. Matsumoto, K. Oshima, K. Utimoto, Bull. Chem.
Soc. Jpn. 1991, 64, 1461; c) J. Liu, X. Sun, M. Miyazaki, L. Liu, C.
Wang, Z. Xi, J. Org. Chem. 2007, 72, 3137; d) K. Hirano, H.
Yorimitsu, K. Oshima, J. Am. Chem. Soc. 2007, 129, 6094; e) R.
Shintani, K. Moriya, T. Hayashi, J. Am. Chem. Soc. 2011, 133,
16440; f) R. Shintani, K. Moriya, T. Hayashi, Org. Lett. 2012, 14,
2902.
[8] K. Hirano, H. Yorimitsu, K. Oshima, Org. Lett. 2008, 10, 2199.
[9] N. Ishida, W. Ikemoto, M. Murakami, J. Am. Chem. Soc. 2014,
136, 5912.
[10] a) Q. W. Zhang, K. An, W. He, Angew. Chem. Int. Ed. 2014, 53,
5667; Angew. Chem. 2014, 126, 5773; b) Q. W. Zhang, K. An,
L. C. Liu, Y. Yue, W. He, Angew. Chem. Int. Ed. 2015, 54, 6918;
Angew. Chem. 2015, 127, 7022.
ate IV. A second reductive elimination would generate the
silole precursor V. Hydrometallation of the terminal alkene
and subsequent protodemetallation with HCl would produce
the final product 2a and regenerate Rh complex VI. In
essence, the endocyclic b-H elimination not only leads to the
I
À
formation of catalytic active Si Rh , but also generates
a pendant alkene as the internal hydrogen acceptor. Such
a catalytic cycle is in good agreement with current exper-
imental observations and is under active investigation.
In conclusion, we discovered that silacyclobutane is a new
I
À
C H silylation reagent under the catalysis of Rh , providing
efficient and regioselective syntheses of functionalized siloles.
This new reactivity of SCB can be attributed to a rarely
documented endocyclic b-H elimination of five-membered
metallacycles. We expect SCB to find applications in other
[11] C. Cheng, J. F. Hartwig, J. Am. Chem. Soc. 2014, 136, 12064.
[12] For review see, F. Zheng, A. Sivaramakrishna, J. R. Moss, Coord.
Chem. Rev. 2007, 251, 2056.
[13] J. X. McDermott, J. F. White, G. M. Whitesides, J. Am. Chem.
Soc. 1976, 98, 6521.
À
C H silylation reactions. Related studies along this line, as
well as full mechanistic studies are being pursued and will be
reported in due course.
[14] B. M. Trost, A. F. Indolese, T. J. J. Mueller, B. Treptow, J. Am.
Chem. Soc. 1995, 117, 615.
[15] a) T. J. M. de Bruin, L. Magna, P. Raybaud, H. Toulhoat,
Organometallics 2003, 22, 3404; b) Z.-X. Yu, K. N. Houk,
Angew. Chem. Int. Ed. 2003, 42, 808; Angew. Chem. 2003, 115,
832; c) X. Huang, J. Zhu, Z. Lin, Organometallics 2004, 23, 4154.
[16] a) J. W. Park, K. G. Kou, D. K. Kim, V. M. Dong, Chem. Sci.
2015, 6, 4479; b) X. Zhou, G. Dong, J. Am. Chem. Soc. 2015, 137,
13715; c) S. Y. Yip, C. Aissa, Angew. Chem. Int. Ed. 2015, 54,
6870; Angew. Chem. 2015, 127, 6974.
[17] For an up-to-date review, see a) C. Cheng, J. F. Hartwig, Chem.
Rev. 2015, 115, 8946. For selected examples, see: b) B. Lu, J. R.
Falck, Angew. Chem. Int. Ed. 2008, 47, 7508; Angew. Chem. 2008,
120, 7618; c) T. Ureshino, T. Yoshida, Y. Kuninobu, K. Takai, J.
Am. Chem. Soc. 2010, 132, 14324 – 14326; d) C. Cheng, J. F.
Hartwig, Science 2014, 343, 853; e) A. A. Toutov, W. B. Liu, K. N.
Betz, A. Fedorov, B. M. Stoltz, R. H. Grubbs, Nature 2015, 518,
80; f) L. Omann, M. Oestreich, Angew. Chem. Int. Ed. 2015, 54,
10276; Angew. Chem. 2015, 127, 10414.
Experimental Section
To a screw caped tube were added 1 (0.1 mmol), [Rh(cod)Cl]2
(5 mol%, 2.5 mg), TMS-segphos (10 mol%, 12.0 mg) and degassed
toluene (1 mL) in a N2 flushed glove box. The tube was capped,
removed from the glove box. The system was stirred at indicated
temperature and concentrated under vacuum after completion. The
residue was purified by silica gel chromatography to afford the
desired product.
Acknowledgements
We acknowledge financial support from the National Key
Basic Research Program of China (2012CB224802), NSF of
China (21371107, 21521091), Tsinghua-Peking Joint Centers
for Life Sciences, and the China Postdoctoral Science
Foundation (2013M540083). We also thank Dr. Haifang Li
for his assistance in mass spectrometry.
[18] For reviews on catalytic synthesis of siloles see: a) T. Meng, K.
Ouyang, Z. Xi, RSC Adv. 2013, 3, 14273; b) L. Wang, Z. Duan,
Chin. Sci. Bull. 2012, 58, 307.
[19] M.-N. Birkholz (nØe Gensow), Z. Freixa, P. W. N. M. van Leeu-
wen, Chem. Soc. Rev. 2009, 38, 1099.
[20] C. S. Sevov, J. F. Hartwig, J. Am. Chem. Soc. 2014, 136, 10625.
[21] a) Z. Zhang, H. Qian, J. Longmire, X. Zhang, J. Org. Chem.
2000, 65, 6223; b) F. Liu, D. Qian, L. Li, X. Zhao, J. Zhang,
6322
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6319 –6323

Angewandte
Communications
Chemie
Angew. Chem. Int. Ed. 2010, 49, 6669; Angew. Chem. 2010, 122,
Soc. 2014, 136, 6586; i) N. Ghavtadze, F. S. Melkonyan, A. V.
6819.
Gulevich, C. Huang, V. Gevorgyan, Nat. Chem. 2014, 6, 122.
[24] E. M. Simmons, J. F. Hartwig, Angew. Chem. Int. Ed. 2012, 51,
3066; Angew. Chem. 2012, 124, 3120.
[22] For a review see: J. Y. Corey in Adv. Organomet. Chem., Vol. 59
(Eds.: F. H. Anthony, J. F. Mark), Academic Press, 2011, pp. 1 –
328.
À
[25] For recent examples for catalytic activation of non-strained Si C
À
[23] For selected examples of silylation of C(sp3) H bonds, see: a) F.
bond, see: a) Y. Liang, W. Geng, J. Wei, Z. Xi, Angew. Chem. Int.
Ed. 2012, 51, 1934; Angew. Chem. 2012, 124, 1970; b) Y. Liang, S.
Zhang, Z. Xi, J. Am. Chem. Soc. 2011, 133, 9204; c) M. Tobisu,
M. Onoe, Y. Kita, N. Chatani, J. Am. Chem. Soc. 2009, 131, 7506;
d) M. Onoe, K. Baba, Y. Kim, Y. Kita, M. Tobisu, N. Chatani, J.
Am. Chem. Soc. 2012, 134, 19477. Also see ref. [10a].
Kakiuchi, K. Tsuchiya, M. Matsumoto, E. Mizushima, N.
Chatani, J. Am. Chem. Soc. 2004, 126, 12792; b) H. Ihara, A.
Ueda, M. Suginome, Chem. Lett. 2011, 40, 916; c) T. Mita, K.
Michigami, Y. Sato, Org. Lett. 2012, 14, 3462; d) E. M. Simmons,
J. F. Hartwig, Nature 2012, 483, 70; e) T. G. Frihed, M. Heuck-
endorff, C. M. Pedersen, M. Bols, Angew. Chem. Int. Ed. 2012,
51, 12285; Angew. Chem. 2012, 124, 12451; f) T. Mita, K.
Michigami, Y. Sato, Chem. Asian J. 2013, 8, 2970; g) Y.
Kuninobu, T. Nakahara, H. Takeshima, K. Takai, Org. Lett.
2013, 15, 426; h) B. Li, M. Driess, J. F. Hartwig, J. Am. Chem.
Received: March 8, 2016
Published online: April 13, 2016
Angew. Chem. Int. Ed. 2016, 55, 6319 –6323
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6323