Disilane-catalyzed cyclotrimerization of acetylenes

Full text of DOI:10.1021/ja980356z

6834
J. Am. Chem. Soc. 1998, 120, 6834-6835
Scheme 1
Disilane-Catalyzed Cyclotrimerization of Acetylenes
Jinchao Yang and John G. Verkade*
Department of Chemistry, Gilman Hall
Iowa State UniVersity, Ames, Iowa 50011
ReceiVed February 2, 1998
Eighty-two years after the discovery by Bertholet in 1866 that
acetylene thermally trimerizes to benzene in low yield at
temperatures in excess of 400 °C,1 Reppe reported that this
reaction is catalyzed near room temperature in solutions of nickel
complexes.2 Since then, alkyne trimerization has become one of
the most intensely studied synthetically useful transformations,
and many transition-metal systems including salts, oxides, orga-
nometallic derivatives, and zerovalent metals have been found
to catalyze this reaction via a mechanism involving alkyne
coordination to the metal.3 In 1980 it was reported that the
nonmetallic compound diethylamine catalyzes the cyclotrimer-
ization of aryl ethynyl ketones to 1,3,5-triaroyl benzenes.4
Evidence for an ionic mechanism was put forth in 1994 for this
reaction in which Michael addition of diethylamine to the CH
carbon of the triple bond to form an eneamineone was followed
by addition of two aryl ethynyl ketone molecules with subsequent
regeneration of diethylamine by elimination from the trimer.5
Because this reaction depends on the formation of an enamineone
resonance-stabilized by an aroyl carbonyl, this cyclotrimerization
appears to be restricted to aroylethynes. We report here the
second example of alkyne cyclotrimerizations catalyzed by a
nonmetal compound and the first example of a catalyst that
operates by a free-radical mechanism. The alkynes in the
reactions reported herein include disubstituted as well as mono-
substituted ethynes.
Cl6 in a sealed tube at this temperature, only an insoluble black
solid was formed. However, by lowering the temperature,
hexaphenylbenzene was obtained in reasonable yield (eq 1).7 That
In 1971 it was reported that the gas-phase reaction of acetylene
with hexachlorodisilane at 450 °C gave 1 in 30% yield, presum-
ably via SiCl2 diradicals produced by the disproportionation of
Si2Cl6.6 When we allowed diphenylacetylene to react with Si2-
Si2Cl6 acted as a procatalyst that undergoes no net change in our
reaction was shown by the single 29Si NMR peak at δ 12.52 ppm
for Si2Cl6 in the reaction mixture, which was confirmed by
recording this spectrum again after addition of authentic Si2Cl6.
Further confirmation was secured by quantitatively converting
the Si2Cl6 in the reaction mixture to Si2(NMe2)6 with excess
HNMe2 followed by comparison of the 1H NMR, 13C NMR, and
EI mass spectra with those of an authentic sample. Recycling
the same sample of Si2Cl6 five times in separate reactions with
diphenyl acetylene revealed no loss in C6Ph6 yield.
While the formation of 1 from acetylene and Si2Cl6 at 450 °C
was attributed to SiCl2 diradical formation,6 it is reasonable to
suppose that, under our milder conditions, SiCl3 radicals induce
the trimerization of PhCtCPh by an addition-elimination
pathway depicted in Scheme 1. The presence of radicals in eq 1
was substantiated by the lack of detectable product when
hydroquinone or 9,10-dihydroanthracene was added to the reaction
(1) Bertholet, M. C. R. Held. Seances Acad. Sci. 1866, 905.
(2) Reppe, W.; Schlichting, O.; Klager, K.; Toepel, T. Liebigs Ann. Chem.
1948, 560, 1.
(3) See, for example: (a) Takahashi, T.; Xi, Z.; Yamazaki, A.; Liu, Y.;
Nakajima, K.; Kotora, M. J. Am. Chem. Soc. 1998, 120, 1672. (b) Yokota,
T.; Sakurai, Y.; Sakaguchi, S.; Ishii, Y. Tetrahedron Lett. 1997, 38, 3923. (c)
Marx, H. W.; Moulines, F.; Wagner, T.; Astruc, D. Angew. Chem., Int. Ed.
Engl. 1996, 35, 1701. (d) Bose, R.; Matzger, A. J.; Mohler, D. L.; Vollhardt,
K. P. C. Angew. Chem., Int. Ed. Engl. 1995, 34, 1478. (e) Bianchini, C.;
Caulton, K. G.; Chardon, C.; Doublet, M. L.; Eisenstein, O.; Jackson, S. A.;
Johnson, T. J.; Meli, A.; Peruzzini, M.; Streib, W. E.; Vacca, A.; Vizza, F.
Organometallics 1994, 13, 2010. (f) Vollhardt, K. P. C. Pure Appl. Chem.
1993, 65, 153. (g) Baidossi, W.; Goren, N.; Blum, J. J. Mol. Catal. 1993, 85,
153. (h) Smith, D. P.; Strickler, J. R.; Gray, S. D.; Bruck, M. A.; Holmes, R.
S.; Wigley, D. E. Organometallics, 1992, 11, 1275. (i) Bianchini, C.; Caulton,
K.; Chardon, C.; Eisenstein, O.; Folting, K.; Johnson, T. J.; Meli, A.; Peruzzini,
M.; Rauscher, D. J.; Streib, W. E.; Vizza, F. J. Am. Chem. Soc. 1991, 113,
5127. (j) DuToit, C. J.; Du Plessis, J. A. K.; Lackmann, G. J. Mol. Catal.
1989, 53, 67. (k) Borrini, A.; Dweisi, P.; Ingrosso, G.; Luckermi, A.; Serra,
G. J. Mol. Catal. 1985, 30, 181. (l) Tysoe, W. T.; Nyberg, G. L.; Lambert, R.
M. J. Chem. Soc., Chem. Commun. 1983, 11, 623.
(7) In a thick-walled quartz tube was placed diphenylacetylene (0.50 g,
2.8 mmol) followed by 0.50 g (1.8 mmol) of hexachlorodisilane which was
added under nitrogen by a syringe. The tube was cooled in liquid nitrogen
and was flame sealed under vacuum. The tube was heated to 170-180 °C in
an oil bath for 2 days, during which time the color of the solution changed to
brown after 1 h and then to black after 1 day. Single crystals were observed
to form at the bottom of the tube. After allowing the tube to cool to room
temperature, it was opened and the solid product was washed with 3 × 0.5
mL of CHCl3 and dried under vacuum to give 0.3 g (60% yield) of crystalline
product which was identified as hexaphenylbenzene: mp 455-457 °C. 1H
NMR (300 MHz, C6D6): δ 6.69-7.15 (m, C6H5, 30H). 13C NMR (75 MHz,
C5D6): δ 140.60, 140.28, 131.41, 126.55, 125.16. MS (EI, 70 eV): m/z (ion,
rel intensity) 534.2 (M+, 22.53), 457.2 (M+ - C6H5, 0.80). Anal. Calcd for
C42H30: C, 94.38; H, 5.62. Found: C, 93.25; H, 5.73. The NMR data are
identical to those reported: The Aldrich library of 13C and 1H FT NMR Spectra.
1st ed.; Aldrich Chemical Co.: Milwaukee, WI, 1993.
(4) Balasubramanian, K.; Selvaraj, P.; Venkataramani, S. Synthesis 1980,
29.
(5) (a) Matsuda, K.; Inoue, K.; Koga, N.; Nakamura, N.; Iwamura, H. Mol.
Cryst. Liq. Cryst. 1994, 253, 33. (b) Matsuda, K.; Nakamura, N.; Iwamura,
H. Chem. Lett. 1994, 1765.
(6) Chernyshev, E. A.; Komalenkova, N. G.; Bashkuova, S. A. Zh. Obshch.
Khim. 1971, 41, 1175.
S0002-7863(98)00356-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/23/1998

Communications to the Editor
J. Am. Chem. Soc., Vol. 120, No. 27, 1998 6835
mixture as radical traps. Acid-catalyzed cyclo-trimerization of
diphenylacetylene by HCl formed by hydrolysis of Si2Cl6 in the
presence of adventitious water was ruled out by failure of the
reaction with HCl (in Et2O) or with SiCl4 that had been partially
hydrolyzed by exposure to moist air. That silicon tetrachloride
by itself is ineffective as a catalyst in our reactions is reasonable
in view of its exceedingly high first dissociation energy (ca. 111
kcal/mol8). Because the dissociation energy of SiCl3 (that is
generated from Si2Cl6) to SiCl2 is only ca. 66 kcal/mol,8 SiCl2 as
a reactive intermediate cannot be ruled out a priori. However,
the products obtained under our experimental conditions, when
compared with the more drastic conditions resulting in the
formation of 1,6 favor the relatively simple pathway involving
SiCl3 radicals depicted in Scheme 1. Although attempts to trap
one or more of the intermediates A-D in Scheme 1 with hydrogen
donors such as 9,10-dihydroanthracene or 9,10-dihydrophenan-
threne failed, we were able to detect HSiCl3 and the corresponding
aromatized hydrogen donor dehydrogenated product by 1H NMR
spectroscopy. This result further substantiates the formation of
SiCl3 radicals via Si-Si bond cleavage. Trichlorosilane and
anthracene were also detected when a mixture consisting only of
Si2Cl6 and 9,10-dihydroanthracene was heated to 190 °C for 80
h. As noted earlier, the presence of 9,10-dihydroanthracene in
Scheme 1 inhibited hexaphenylbenzene formation (reaction 1).
However, when 1-decyne was employed in the presence of this
hydrogen donor or 9,10-dihydrophenanthrene, quantitative tri-
merization was detected as well as the formation of HSiCl3 and
aromatization of the hydrogen donor. This result is consistent
with the idea that alkyl ethynes are more competitive for SiCl3
radicals (even at low concentrations) than diphenyl acetylene.
Equation 1 has been extended to other alkyne substrates, among
which are those shown in eqs 2-5. Apparently terminal alkynes
naphthalene) along with an isomeric mixture of trimers under the
conditions of eq 4, whereas 3-hexyne and 1-decyne were thermally
stable in the absence of Si2Cl6.9
We found that Si2(OMe)6 also catalyzes alkyne cyclotrimer-
ization, although conversions are substantially lower than with
Si2Cl6 under the same conditions (e.g., 40% and 30% conversion
in eqs 3 and 4, respectively). On the other hand, no reactions
were detected using Si2Me6 as a potential procatalyst, even when
present in high molar ratios and in reactions lasting up to 150 h.
We suggest that the decreasing catalytic activity in the order Si2-
Cl6 > Si2(OMe)6 > Si2Me6 is attributable to an accompanying
decrease in the electron-withdrawing power of the silicon sub-
stituents. Such an electronic effect can be expected to strengthen
the Si-Si bond, thus diminishing thermal production of silyl
radicals. In all three of the above disilanes, the substituent-
silicon bonds8,10 are considerably stronger than the Si-Si link-
age.10 Interestingly, the reaction of Si2(NMe2)6 with alkynes
follows a stoichiometric course leading to 1,1,4,4-tetrakis-
(dimethylamino)-1,4-disilacyclohexadienes.9
In Scheme 1, cleavage of an Si-substituent bond in intermedi-
ate A by an SiCl3 radical to give SiCl4 and SiCl2 is apparently
less favored than sequential attack of A by two additional PhCt
CPh molecules. When intermediate C is formed, the drive to a
stable aromatic product causes an SiCl3 radical to be extruded
from this intermediate. Further substantiation for Si-Si bond
cleavage in the action of Si2Cl6 and Si2(OMe)6 in these reactions
was sought by heating a mixture of these compounds in hopes of
forming the crossover product Cl3SiSi(OMe)3. Only a compli-
cated mixture of Si2Cl6-x(OMe)x products could be detected by
1H NMR spectroscopy, however.
Others have reported that Me3SiCl and Pd/C (10% palladium
on carbon) is a catalyst for cyclo-trimerizing alkynes to the
corresponding benzene derivatives.11 It was found in this system
that both trimethylchlorosilane and palladium were necessary for
the reaction to occur. Although the authors suggest that the
seemingly heterogeneous process might in fact be a homogeneous
one (possibly through the formation of a soluble catalytic
palladium species), they were not able to detect or isolate an
intermediate. To determine if silyl radicals produced under these
conditions could be catalyzing the cyclo-trimerization, we repeated
the reaction with Si2Cl6 and 1-octyne at 140 °C using a mixture
of Si2Cl6 and 10% palladium on carbon, and also an Si2Cl6/PdCl2
mixture. Surprisingly, the conversions to trimers were only 7.00%
for Pd/C/Si2Cl6 system and 11.0% for PdCl2/Si2Cl6 system; much
lower than Si2Cl6 by itself under the same conditions (70%
conversion). At the higher temperature these palladium catalysts
apparently act as radical traps, perhaps by an oxidative addition
processsa possibility we are currently examining.
Experiments are underway to evaluate the scope and limitations
of the alkynes and group 14 procatalysts congeneric with disilanes
that can be employed in these novel reactions.
undergo cyclo-trimerization in high conversion more easily than
nonterminal examples for steric reasons. At 20 mol % Si2Cl6,
eq 2 gave a 30% conversion after 2 days, and for reasons that
are not clear, this conversion was not appreciably increased after
6 days. Equation 3 gives an 80% conversion to products even at
5 mol % Si2Cl6 after 30 h. To rule out catalysis by the walls of
the steel pressure vessel in eq 5, the reaction was run in the
absence of Si2Cl6, whereupon only starting material was recov-
ered. Interestingly, in the absence of Si2Cl6, phenyl acetylene
was found to give mainly dimerized products (R and â-phenyl
Acknowledgment. The authors are grateful to the NSF and the Iowa
State Center for Advanced Technology Development for grant support.
Helpful discussions with T. J. Barton and (the late) G. A. Russell are
gratefully acknowledged.
JA980356Z
(9) Yang, J.; Verkade, J. To be published.
(10) Moeller, T. Inorganic Chemistry; John Wiley & Sons: New York,
1982.
(8) Walsh, R. Acc. Chem. Res. 1981, 14, 246.
(11) Jhingan, A. K.; Maier, W. F. J. Org. Chem. 1987, 52, 1161.