Dichotomic reactivity of a stable silylene toward terminal alkynes: Facile C-H bond insertion versus autocatalytic formation of silacycloprop-3-ene

Article abstract of DOI:10.1002/anie.200704939

(Chemical Equation Presented) Jekyll and Hyde: Stable silylene 1 reacts readily with acetylene at room temperature to give 1,1-adduct 2, whereas at -78°C only silacycloprop-3-ene 3 is formed. Once 3 is present in the reaction mixture, it autocatalyzes its own generation even at room temperature. Both the facile silylene C-H bond insertion for terminal alkynes and the autocatalytic formation of silacycloprop-3-enes are unprecedented. R = 2,6-iPr₂C₆H₃.

Full text of DOI:10.1002/anie.200704939

Communications
DOI: 10.1002/anie.200704939
N-Heterocyclic Silylenes
Dichotomic Reactivity of a Stable Silylene toward Terminal Alkynes:
ꢀ
Facile C H Bond Insertion versus Autocatalytic Formation of
Silacycloprop-3-ene**
Shenglai Yao, Christoph van Wüllen, Xiao-Ying Sun, and Matthias Driess*
Dedicated to Professor Philip P. Power
bulky organic groups (A, B). Recently we described the
Silylenes, the silicon analogues of carbenes, are key inter-
mediates in numerous thermal and photochemical reactions.[1]
They are indispensable building blocks for the synthesis of
stable ylide-like silylene 1, in which significant participation
of the 6p-aromatic resonance form 1’ is found in the electronic
ground state (Scheme 1).[3]
ꢀ
ꢀ
ꢀ
organosilanes through X Y insertion reactions (X Y= C H,
Owing to its unusual ylide-like character, silylene 1 shows
a remarkably distinct reactivity toward both electrophiles and
nucleophiles, in comparison with A and B, respectively.
Striking results include the facile formation of silyliumylidene
[1b–h]
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
C C , C O, C Cl, Si H, O H, for example).
Until two
decades ago, silylenes were generally considered to be
extremely unstable species, decomposing readily at low
temperature (> ꢀ1968C). The situation changed profoundly
in 1994, when West et al. described the synthesis of the first
stable N-heterocyclic silylenes A with two-coordinate silicon
which were persistent at room temperature (Scheme 1).[2a,b]
Following the successful isolation of stable silylenes A, only a
few other stable silylenes have been reported as yet,[2c–e]
including the remarkable nonsupported dialkylsilylene B
(Scheme 1).[2f] The latter systems benefit from p-donor
stabilization of the low-valent silicon by the nitrogen atoms,
pseudoaromaticity (A), and/or steric protection through
cations by addition of H+ or other Lewis acid centers at the
[4a]
=
nucleophilic carbon atom of the terminal C Cmoiety in 1,
synthesis of a stable siloxysilylene through heterolytic addi-
tion of water to 1,[4b] and insertion of the divalent silicon atom
[4c]
ꢀ
in 1 into P P bonds of P4.
Another specific feature
represents the addition of simple electrophiles RX (R = H,
Me3Si; X = halogen, OSO2CF3) to 1, which resulted in the
isolation of the corresponding 1,1-adducts (insertion prod-
ucts).[3] Similarly, the 1,4-dipolar nature of 1 could for the first
ꢀ
time facilitate a marked preference for C H insertion of
ꢁ
divalent silicon versus C C-p bond addition with terminal
ꢀ
alkynes. To our knowledge, neither facile silylene C H bond
insertion for terminal alkynes[5] nor formation of isolable
silacycloprop-3-enes, starting from isolable silylenes A have
been reported.[2f] Herein we report the remarkably different
reactivity of 1 toward terminal alkynes. although conversion
ꢀ
at room temperature results solely in the formation of the C
H insertion adducts 2a, 2b, and 3 as thermodynamic products,
the same starting materials surprisingly undergo [2+1] cyclo-
addition after a induction period of a few hours at low
temperature (ꢀ788C) to furnish the corresponding silacyclo-
prop-3-enes 4a and 4b in high yield (Scheme 2).
Scheme 1. Stable silylenes A and B, and mesomeric forms of 1.
R=2,6-iPr2C6H3.
ꢁ
Exposure of yellow solutions of 1 in hexane to HC C H at
room temperature leads to colorless solutions within a few
ꢀ
minutes from which the C H insertion product 2a can be
[*] Dr. S. Yao, Prof. Dr. M. Driess
isolated in the form of colorless crystals in 82% yield.[6]
Additionally, prolonged reaction time with molar excess of
1 or conversion of 2a with 1 in molar ratio of 1:1 affords the
double insertion product 3 which was isolated in 85% yield.
Technische Universität Berlin
Institute of Chemistry: Metalorganic and Inorganic Materials
Sekr. C2, Strasse des 17. Juni 135, 10623 Berlin (Germany)
Fax: (+49)30-314-29732
E-mail: matthias.driess@tu-berlin.de
Homepage: http://www.driess.tu-berlin.de
ꢁ
Furthermore, addition of PhC C H to1 leads exclusively to
the corresponding alkynylsilane 2b, which was isolated in
93% yield. The composition and constitution of the products
was confirmed by NMR spectroscopy (1H, 13C, 29Si), mass
spectrometry, and elemental analysis (C, H, N). The six-
membered C3N2Si rings in 2a and 3 are only slightly puckered,
and the Si Cand C Cbond lengths (Figures 1 and 2) are
similar to those reported for other alkynylsilanes.[7] As
expected, the most important geometrical features of 2b are
practically identical to those of 2a.[6]
Prof. Dr. C. van Wüllen, Dr. X.-Y. Sun
Technische Universität Kaiserslautern
Fachbereich Chemie
Erwin-Schrödinger-Strasse, 67663 Kaiserslautern (Germany)
[**] Financial support from the Deutsche Forschungsgemeinschaft and
the Cluster of Excellence “Unifying Concepts in Catalysis” (EXC 314/
1) are gratefully acknowledged.
ꢀ
ꢁ
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3250
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3250 –3253

Angewandte
Chemie
In contrast, the related stable silylenes A and B do not
ꢀ
lead to C H insertion products with terminal alkynes, and
only B is capable of generating an isolable silacycloprop-3-
ene.[2a,f] Additionally, gas-phase studies of transient silylenes
ꢀ
in the presence of terminal alkynes revealed that C H bond
insertion is not facile and results merely from rearrangement
of silacycloprop-3-enes as primary intermediates at relatively
high temperature (> 5008C).[5] To probe whether silacyclo-
prop-3-enes are possible intermediates for the generation of
ꢁ
ꢁ
2a and 2b, reaction of 1 with HC CH and PhC CH was also
carried out at lower temperature (ꢀ788C). The reaction does
take a different course, which can be monitored by 1H NMR
spectroscopy. It turns out that conversion proceeds slowly,
and is still incomplete even after 5 h, affording the corre-
sponding silacycloprop-3-enes 4a and 4b as sole products
(Scheme 2). To our surprise, completion of the reaction at
room temperature leads exclusively to 4a and to 4b, which
can be isolated in the form of colorless cubes in high yield
(82% (4a), 85% (4b)). Accordingly, conversion of 1 with
Scheme 2. Synthesis of 2a, 2b, 3, and 4a–4c from reaction of 1 with
the corresponding alkynes. R=2,6-iPr2C6H3.
ꢁ
HC CH at room temperature in the presence of 4a (initial
molar ratio 1:4a = 5:1) is complete after about 3 h to give 4a
in quantitative yield (1H, 13CNMR); 4a appears to play an
autocatalytic role. The same is true for 4b, which autocata-
ꢁ
lyzes the transformation of 1 with PhC C H to4b at room
temperature.[6] The formation of 4a and 4b can be seen in the
characteristic deshielding of the vinylic protons at the SiC2
1
ring in the H NMR spectrum[8] (d(1H) = 7.91 (4a), 7.86 ppm
(4b)) and by the typical upfield signal in the 29Si NMR
spectrum at d = ꢀ91.9 (4a) and ꢀ86.4 ppm (4b).[8] Remark-
ably, the synthesis of 4a and 4b can even be catalyzed using
the related silacycloprop-3-ene 4c (Scheme 2). The latter is
ꢁ
easily accessible from 1 and PhC CPh at room temperature in
Figure 1. Molecular structure of 2a. Hydrogen atoms (except for those
at C1, C31, and Si1) omitted for clarity. Selected bond lengths [pm]
and angles [8]: C30–C31 115.5(4), Si1–N1 170.3(2), Si1–N2 172.2(2),
Si1–C30 182.1(3), N1–C2 141.9(3), N2–C4 141.0(3); N1-Si1-N2
104.7(1), N1-Si1-C30 110.3(1), N2-Si1-C30 112.0(1), C31-C30-Si1
175.6(3).
quantitative yield. The composition of the novel silacyclo-
prop-3-enes 4a–4c was confirmed by H, 13C, and 29Si NMR
1
spectroscopy, EI-MS, and correct combustion analyses (C, H,
N). Their molecular structures have been established by
single-crystal X-ray diffraction.[6] Although a few silacyclo-
prop-3-enes have already been characterized by X-ray
diffraction,[9,10] 4a and 4b represent the first structurally
characterized derivatives with hydrogen attached to the
olefinic carbon atoms in the three-membered SiC2 ring. The
C3N2Si-ring in 4a (and in 4b) is practically planar and
perpendicular to the SiC2 ring (Figure 3).
The SiC2 ring in 4a–4c is strikingly stable and survives
prolonged heating of solutions at 1108Cfor several hours.
This can be explained by pronounced s* aromaticity of the
SiC2-ring, in accordance with theoretical calculations.[11] This
implies some masked silacyclopropenylium ylide-like charac-
ter of 4a–4c, as depicted in the resonance structures in
Scheme 3. However, the high stability clearly precludes that
4a and 4b play any role as intermediates for the formation of
ꢀ
the respective C H insertion products 2a and 2b. Accord-
ingly, attempts to generate 2a and 2b by thermolysis of 4a and
4b were unsuccessful. Although the pathway to 4a and 4b is
dominant at low temperatures, it plays no role at higher
temperatures, which implies a lower (i.e., more negative)
activation entropy. In other words, the conversion of 1 into the
Figure 2. Molecular structure of 3. Hydrogen atoms (except for those
at C1, C30, Si1, and Si2) omitted for clarity. Selected bond lengths
[pm] and angles [8]: Si1–N1 171.6(2), Si1–N2 171.7(2), Si2–N4
171.3(2), Si2–N3 171.3(2), Si1–C58 183.0(3), Si2–C59 183.8(3), C58–
C59 121.1(3); N1-Si1-N2 105.7(1), N4-Si2-N3 105.2(1), C59-C58-Si1
178.5(3), C58-C59-Si2 176.7(3).
ꢀ
C H insertion products 2a and 2b proceeds by a completely
different mechanism than previously reported for transient
Angew. Chem. Int. Ed. 2008, 47, 3250 –3253
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3251

Communications
Figure 3. Molecular structure of 4a. Hydrogen atoms
(except for those at C1, C30, and C31) omitted for clarity.
Selected bond lengths [pm] and angles [8]: C30–C31
131.1(5), Si1–N1 170.7(2), Si1–N2 170.9(2), Si1–C30
173.7(4), Si1–C31 177.1(4), N1–C2 140.9(3), N2–C4
140.7(3); N1-Si1-N2 105.1(1), C30-Si1-C31 43.9(2), C31-
C30-Si1 69.4(3), C30-C31-Si1 66.7(3).
ꢁ
Figure 4. Energy profile for the reaction of 1a with HC CH, including DFT-
calculated structures of the model compounds 2A, 4A, and 5A as well as their
corresponding transition states TS-2A, TS-4A, and TS-5A, respectively.[6]
silylenes. This fact is supported by theoretical calcu-
lations using the model system 1a, in which the 2,6-
iPr2C6H3 groups at nitrogen have been replaced by phenyl
play a crucial role (Scheme 3). This is in accordance with the
prediction of relatively large proton affinities of the terminal
H2Cgroup in the model compounds 4A, 1a, and in the
groups.[6] We initially considered concerted [2+1] cycloaddi-
ꢀ
tion versus C H insertion of the divalent silicon in 1a towards
ꢁ
HC CH (Figure 4): whereas the calculations revealed a
relatively small barrier of 45.1 kJmolꢀ1 for the [2+1] cyclo-
addition to give the corresponding silacycloprop-3-ene 4 A,
ꢀ
ꢁ
direct insertion into the C H bond in HC CH to form 2 A
encounters a much higher barrier of 172.8 kJmolꢀ1. As
expected, 2A is thermodynamically favored by 22.5 kJmolꢀ1
over 4A.[6] Consistent with the experimental results, there is
also no facile interconversion 4A!2A under thermal
conditions.[6] Thus it seems reasonable to assume that the
ꢀ
mechanism for C H insertion must be at least a two-step
process. We propose the ylide-like structure 5A (Figure 4) as
a key intermediate, which is virtually as stable as 4A (DE <
0.1 kJmolꢀ1).[6] Compund 5A is most likely formed after
ꢁ
deprotonation of HC CH by the basic terminal CH2 group of
ꢁ
1a, followed by a transfer of the C CH anion to the silicon
atom. In the gas phase, charge separation is disfavored, thus
our calculations find a transition state 100.7 kJmolꢀ1 above
the starting compounds, and the proton is transferred
relatively late on the reaction coordinate.[6] In the condensed
phase, deprotonation will occur much earlier, and formation
of 5A be even more facile. Contrary to the direct formation of
Scheme 3. Proposed autocatalytic cycle of 4a–c for the conversion of 1
with terminal alkynes. R’=H, Ph.
ꢁ
2A from 1a and HC CH, our DFT calculations neither find
excessive barriers nor instable intermediates along this path-
way. Subsequently, 1,4-adduct 5A could easily tautomerize to
2A. This is also consistent with previous experimental results,
in which the comparatively basic terminal CH2 group at the
zwitterionic silylene 1 plays an important role toward electro-
philes.[3] How can the astonishing autocatalytic role of the
silacycloprop-3-enes 4a–4c on the conversion of 1 with
terminal alkynes be explained? Although the mechanism is
still unknown, the resonance ylide structures 4a’–4c’ could
corresponding silicon dibromide [1aBr2] in the order 4A
(1111.6) > 1a (1099) > [1aBr2] (1088.1 kJmolꢀ1).[3,6] Thus, the
latter compounds are relatively strong neutral bases.
In turn, this suggests that the alkyne is preferably captured
in terminal fashion by 4a to give the intermediate complex 6,
but then no 1,4-adduct can be formed. Thus we believe that
4a–4c can serve as auxiliary bases which are capable of
mediating the [2+1] cycloaddition of 1 via the postulated
complexes 6, 7, and 8 (Scheme 3). Accordingly, conversion of
3252
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3250 –3253

Angewandte
Chemie
ꢁ
Chem. 2004, 689, 4165 – 4183; c) M. Weidenbruch, Angew.
1 with R’C C H (R’ = H, Ph) at room temperature in the
Chem. 2006, 118, 4347 – 4348; Angew. Chem. Int. Ed. 2006, 45,
4241 – 4242; d) B. Gehrhus, M. F. Lappert, J. Organomet. Chem.
2001, 617–618, 209 – 223; e) C.-W. So, H. W. Roesky, P. M.
Gurubasavaraj, R. B. Oswald, M. T. Gamer, P. G. Jones, S.
Blaurock, J. Am. Chem. Soc. 2007, 129, 12049 – 12054; f) M. Kira,
S. Ishida, T. Iwamoto, C. Kabuto, J. Am. Chem. Soc. 1999, 121,
9722 – 9723.
presence of similarly strong neutral bases such as the
dibromide [1Br2] leads also exclusively to 4a and to 4b. In
contrast, organic nitrogen-containing bases with significantly
lower proton affinities, such as NEt3 and pyridine, show no
catalytic activity. However, further experiments are necessary
to elucidate the mechanism of the autocatalytic process.
In summary, we report the unprecedented reactivity of the
ylide-like silylene 1 toward terminal alkynes. The first facile
[3] M. Driess, S. Yao, M. Brym, C. van Wꢀllen, D. Lentz, J. Am.
Chem. Soc. 2006, 128, 9628 – 9629.
[4] a) M. Driess, S. Yao, M. Brym, C. van Wꢀllen, Angew. Chem.
2006, 118, 6882 – 6885; Angew. Chem. Int. Ed. 2006, 45, 6730 –
6733; b) S. Yao, M. Brym, C. van Wꢀllen, M. Driess, Angew.
Chem. 2007, 119, 4237 – 4240; Angew. Chem. Int. Ed. 2007, 46,
4159 – 4162; c) Y. Xiong, S. Yao, M. Brym, M. Driess, Angew.
Chem. 2007, 119, 4595 – 4597; Angew. Chem. Int. Ed. 2007, 46,
4511 – 4513.
[5] a) R. T. Conlin, P. P. Gaspar, J. Am. Chem. Soc. 1976, 98, 3715 –
3716; b) M. Tho Nguyen, D. Sengupta, L. G. Vanquickenborne,
Chem. Phys. Lett. 1995, 240, 513 – 520; c) R. Becerra, H. M. Frey,
B. P. Mason, R. Walsh, J. Chem. Soc. Chem. Commun. 1993,
1050 – 1052; d) C. H. Haas, M. A. Ring, Inorg. Chem. 1975, 14,
2253 – 2256.
ꢀ
silylene C H bond insertion for terminal alkynes was thereby
discovered, and, at the same time, a unique autocatalytic
generation of the stable silacycloprop-3-enes 4a and 4b. The
easily accessible compounds 4a–4c are promising building
blocks for cyclopropene syntheses by catalytic s-bond meta-
thesis.[12] Respective investigations to elucidate the autocata-
lytic formation of 4a and 4b are currently underway.
Received: October 24, 2007
Revised: November 22, 2007
Published online: March 13, 2008
[6] Experimental and computational details as well as X-ray
diffraction analyses and geometrical parameters of 2a 2b, 3,
and 4a–c are provided in the Supporting Information.
[7] W. S. Sheldrick in The Chemistry of Organosilicon Compounds,
Part 1 (Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1989,
p. 246, and references therein.
[8] J. Belzner, H. Ihmels, Tetrahedron Lett. 1993, 34, 6541 – 6544.
[9] D. Ostendorf, W. Saak, M. Weidenbruch, H. Marsmann,
Organometallics 2002, 21, 636 – 640.
[10] a) K. Hirotsu, T. Higuchi, M. Ishikawa, H. Sugisawa, M.
Kumada, J. Chem. Soc. Chem. Commun. 1982, 726 – 727; b) F.
Hojo, S. Sekigawa, N. Nakayama, T. Shimizu, W. Ando, Organo-
metallics 1993, 12, 803 – 810; c) S. Tsutsui, K. Sakamoto, C.
Kabuto, M. Kira, Organometallics 1998, 17, 3819 – 3821; d) D.
Ostendorf, L. Kirmaier, W. Saak, H. Marsmann, M. Weiden-
bruch, Eur. J. Inorg. Chem. 1999, 2301 – 2307; see also refer-
ence [9].
Keywords: cycloaddition · heterocycles · silicon · silylation ·
small ring systems
.
[1] a) Review on silylenes: P. P. Gaspar, R. West, in The Chemistry
of Organosilicon Compounds, Vol. 2 (Eds.: Z. Rappoport, Y.
Apeloig), Wiley, New York, 1998, chap. 43; see also references
[2b–d]; b) L. E. Bourque, P. A. Cleary, K. A. Woerpel, J. Am.
Chem. Soc. 2007, 129, 12602 – 12603, and references therein;
c) T. Kajiwara, N. Takeda, T. Saamori, N. Tokitoh, Organo-
metallics 2004, 23, 4723 – 4734; d) P. Jiang, D. Trieber II, P. P.
Gaspar, Organometallics 2003, 22, 2233 – 2239; e) H. Ottosson,
P. G. Steel, Chem. Eur. J. 2006, 12, 1576 – 1585; f) M. Kira, S.
Ishida, T. Iwamoto, C. Kabuto, J. Am. Chem. Soc. 2002, 124,
3830 – 3831; g) D. F. Moser, T. Bosse, J. Olson, J. L. Moser, I. A.
Guzei, R. West, J. Am. Chem. Soc. 2002, 124, 4186 – 4187; h) M.
Weidenbruch, H. Piel, K. Peters, H. G. von Schnering, Organo-
metallics 1994, 13, 3990 – 3994.
[11] A. Gꢁllner, H. Heydt, T. Clark, J. Org. Chem. 1996, 61, 5840 –
5846.
[12] I. Marek, S. Simaan, A. Masarwa, Angew. Chem. 2007, 119,
7508 – 7520; Angew. Chem. Int. Ed. 2007, 46, 7364 – 7376.
[2] a) M. Denk, R. Lennon, R. Hayashi, R. West, A. V. Belyakov,
H. P. Verne, A. Haaland, M. Wagner, N. Metzler, J. Am. Chem.
Soc. 1994, 116, 2691 – 2692; b) N. J. Hill, R. West, J. Organomet.
Angew. Chem. Int. Ed. 2008, 47, 3250 –3253
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3253