Mechanism of silver-mediated di-tert-butylsilylene transfer from a silacyclopropane to an alkene

Article abstract of DOI:10.1021/ja0306563

Kinetic studies of the reactions of cyclohexene silacyclopropane 1 and monosubstituted alkenes in the presence of 5 mol % of (Ph₃P) ₂AgOTf suggested a possible mechanism for silver-mediated di-tert-butylsilylene transfer. The kinetic

Full text of DOI:10.1021/ja0306563

Published on Web 07/22/2004
Mechanism of Silver-Mediated Di-tert-butylsilylene Transfer
from a Silacyclopropane to an Alkene
Tom G. Driver and K. A. Woerpel*
Contribution from the Department of Chemistry, UniVersity of California,
IrVine, California 92697-2025
Received December 10, 2003; E-mail: kwoerpel@uci.edu
Abstract: Kinetic studies of the reactions of cyclohexene silacyclopropane 1 and monosubstituted alkenes
in the presence of 5 mol % of (Ph3P)2AgOTf suggested a possible mechanism for silver-mediated di-tert-
butylsilylene transfer. The kinetic order in cyclohexene silacyclopropane 1 was determined to be one. Inverse
kinetic saturation behavior (rate inhibition) was observed in monosubstituted alkene and cyclohexene
concentrations. Saturation kinetic behavior in catalyst concentration was observed. A reactive intermediate,
a silylsilver complex, was observed using low temperature 29Si NMR spectroscopy. Competition experiments
between substituted styrenes and a deficient amount of 1 correlated well with the Hammett equation and
provided a F value of -0.62 ( 0.02 using σp constants. These data support a mechanism involving reversible
silver-promoted di-tert-butylsilylene extrusion from 1 followed by irreversible concerted electrophilic attack
of the silylsilver intermediate on the alkene.
Introduction
intermediates can be generated through photolysis of cyclic
silanes or trisilanes33,34 or the use of strong reductants (such as
The unique reactivity of silacyclopropanes1-6 has prompted
interest in developing efficient methods for their construction.7,8
The syntheses of silacyclopropanes often rely on the transfer
of silylene intermediates (R2Si)9-24 to alkenes. For example,
thermolysis of various cyclic silanes25-28 produces silylene
intermediates that can be trapped by alkenes to form silacyclo-
propanes.29-32 Alternatively, the required silylene or silylenoid
lithium or potassium) with dihalosilanes.26,28,35-38 We recently
reported a mild method for silacyclopropane synthesis by silver-
catalyzed silylene transfer at -27 °C from cyclohexene sila-
(23) For our mechanistic study of di-tert-butylsilylene transfer from cyclohexene
silacyclopropane 1 to allylbenzene, see: Driver, T. G.; Woerpel, K. A. J.
Am. Chem. Soc. 2003, 125, 10659-10663. Conclusions of this study: (a)
Kinetic evidence suggested that the mechanism involved the reversible
extrusion of silylene from 1 followed by irreversible cyclization with an
alkene. (b) The activation parameters for silylene extrusion were determined
to be ∆Gq ) 27.6 kcal‚mol-1, ∆Hq ) 22.1 kcal‚mol-1, and ∆Sq ) -15
eu. (c) A Hammett study revealed that cycloaddition was under entropic
control and occurred through a concerted, electrophilic attack of di-tert-
butylsilylene on the alkene.
(1) Skell, P. S.; Goldstein, E. J. J. Am. Chem. Soc. 1964, 86, 1442-1443.
(2) Seyferth, D.; Annarelli, D. C. J. Am. Chem. Soc. 1975, 97, 7162-7163.
(3) Saso, H.; Ando, W.; Ueno, K. Tetrahedron 1989, 45, 1929-1940.
(4) Kroke, E.; Willms, S.; Weidenbruch, M.; Saak, W.; Pohl, S.; Marsmann,
H. Tetrahedron Lett. 1996, 37, 3675-3678.
(5) Franz, A. K.; Woerpel, K. A. Acc. Chem. Res. 2000, 33, 813-820.
(6) Weidenbruch, M.; Meiners, F.; Saak, W. Can. J. Chem. 2000, 78, 1469-
1473.
(7) Roberts, J. D.; Dev, S. J. Am. Chem. Soc. 1951, 73, 1879-1880.
(8) Lambert, R. L., Jr.; Seyferth, D. J. Am. Chem. Soc. 1972, 94, 9246-9248.
(9) Scha¨fer, A.; Weidenbruch, M.; Pohl, S. J. Organomet. Chem. 1985, 282,
305-313.
(24) For leading references to the coordination of Lewis bases with silylenes,
refer to (a) Steele, K. P.; Weber, W. P. J. Am. Chem. Soc. 1980, 102, 6095-
6097. (b) Weidenbruch, M.; Piel, H.; Lesch, A.; Peters, K.; von Schnering,
H. G. J. Organomet. Chem. 1993, 454, 35-43. (c) Bott, S. G.; Marshall,
P.; Wagenseller, P. E.; Wang, Y.; Conlin, R. T. J. Organomet. Chem. 1995,
499, 11-16. (d) Takeda, N.; Suzuki, H.; Tokitoh, N.; Okazaki, R.; Nagase,
S. J. Am. Chem. Soc. 1997, 119, 1456-1457. (e) Belzner, J.; Ihmels, H.
AdV. Organomet. Chem. 1999, 43, 1-42. (f) Bharatam, P. V.; Moudgil,
R.; Kaur, D. Organometallics 2002, 21, 3683-3690.
(10) Weidenbruch, M. Coord. Chem. ReV. 1994, 130, 275-300.
(11) Weidenbruch, M. Chem. ReV. 1995, 95, 1479-1493.
(12) Gaspar, P. P.; West, R. The Chemistry of Organosilicon Compounds;
Rappoport, Z., Apeloig, Y., Eds.; Wiley: Chichester, 1998; Vol. 2, p 2463-
2468.
(25) Seyferth, D.; Annarelli, D. C. J. Am. Chem. Soc. 1975, 97, 7162-7163.
(26) Pae, D. H.; Xiao, M.; Chiang, M. Y.; Gaspar, P. P. J. Am. Chem. Soc.
1991, 113, 1281-1288.
(27) Belzner, J. J. Organomet. Chem. 1992, 430, C51-C55.
(28) Sakamoto, K.; Tsutsui, S.; Sakurai, H.; Kira, M. Bull. Chem. Soc. Jpn.
1997, 70, 253-260.
(13) Denk, M. K.; Hatano, K.; Lough, A. J. Eur. J. Inorg. Chem. 1998, 1067-
1070.
(14) Kira, M.; Ishida, S.; Iwamoto, T.; Kabuto, C. J. Am. Chem. Soc. 1999,
121, 9722-9723.
(29) Seyferth, D.; Annarelli, D. C. J. Organomet. Chem. 1976, 117, C51-C54.
(30) Scha¨fer, A.; Weidenbruch, M.; Peters, K.; von Schnering, H.-G. Angew.
Chem., Int. Ed. Engl. 1984, 23, 302-303.
(15) Lehmann, J. F.; Urquhart, S. G.; Ennis, L. E.; Hitchcock, A. P.; Hatano,
K.; Gupta, S.; Denk, M. K. Organometallics 1999, 18, 1862-1872.
(16) Haaf, M.; Schmedake, T. A.; West, R. Acc. Chem. Res. 2000, 33, 704-
714.
(31) Boudjouk, P.; Black, E.; Kumarathasan, R. Organometallics 1991, 10,
2095-2096.
(32) Belzner, J.; Ihmels, H.; Kneisel, B. O.; Gould, R. O.; Herbst-Irmer, R.
Organometallics 1995, 14, 305-311.
(17) Belzner, J.; Dehnert, U.; Ihmels, H. Tetrahedron 2001, 57, 511-517.
(18) Gaspar, P. P.; Xiao, M.; Pae, D. H.; Berger, D. J.; Haile, T.; Chen, T.; Lei,
D.; Winchester, W. R.; Jiang, P. J. Organomet. Chem. 2002, 646, 68-79.
(19) Kang, S.-H.; Han, J. S.; Lee, M. E.; Yoo, B. R.; Jung, I. N. Organometallics
2003, 22, 2551-2553.
(20) Apeloig, Y.; Pauncz, R.; Karni, M.; West, R.; Steiner, W.; Chapman, D.
Organometallics 2003, 22, 3250-3256.
(21) Jiang, P.; Trieber, D., II; Gaspar, P. P. Organometallics 2003, 22, 2233-
2239.
(33) Ando, W.; Fujita, M.; Yoshida, H.; Sekiguchi, A. J. Am. Chem. Soc. 1988,
110, 3310-3311.
(34) Zhang, S.; Conlin, R. T. J. Am. Chem. Soc. 1991, 113, 4272-4278.
(35) Boudjouk, P.; Samaraweera, U.; Sooriyakumaran, R.; Chrusciel, J.;
Anderson, K. R. Angew. Chem., Int. Ed. Engl. 1988, 27, 1355-1356.
(36) Tsutsui, S.; Sakamoto, K.; Kira, M. J. Am. Chem. Soc. 1998, 120, 9955-
9956.
(37) Gaspar, P. P.; Beatty, A. M.; Chen, T.; Haile, T.; Lei, D.; Winchester, W.
R.; Braddock-Wilking, J.; Rath, N. P.; Klooster, W. T.; Koetzle, T. F.;
Mason, S. A.; Albinati, A. Organometallics 1999, 18, 3921-3932.
(22) Ohshita, J.; Honda, N.; Nada, K.; Iida, T.; Mihara, T.; Matsuo, Y.; Kunai,
A.; Naka, A.; Ishikawa, M. Organometallics 2003, 22, 2436-2441.
9
10.1021/ja0306563 CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 9993-10002
9993

A R T I C L E S
Driver and Woerpel
cyclopropane 131,39 to various chiral, functionalized alkenes (eq
1).40 This reaction is stereospecific, stereoselective, and tolerant
of various functional groups. Because silver-catalyzed silylene
transfer has proven to be a mild method for the silacyclopro-
panation of alkenes, we believed that a detailed mechanistic
understanding of this reaction would guide future improvements
in silacyclopropane synthesis.
To understand the fundamental reactivity of the intermediates
in the silver-catalyzed formation of silacyclopropanes, we
conducted a quantitative analysis of the behavior of cyclohexene
silacyclopropane 1 in the presence of an alkene and a catalytic
amount of a silver salt. Analysis of the kinetic behavior and
the spectroscopic observation of a silylsilver intermediate led
us to propose a mechanism likely involving reversible silver-
catalyzed extrusion of di-tert-butylsilylenoid from silacyclo-
propane 1 followed by irreversible cyclization with an alkene.
The electrophilicity of the silylsilver intermediate was estab-
lished by the identification of a Hammett F value. These results
and data from competition experiments permitted construction
of a reasonable catalytic cycle to describe the silver-mediated
silylene transfer from 1 to an alkene.
The mechanisms of analogous silver-promoted carbene
transfer reactions remain poorly understood,41-43 although other
metal-mediated carbene transfer reactions44-49 have been studied
extensively. In general, silver-catalyzed reactions50 provide
different products from copper-catalyzed processes, because the
silver-mediated reactions likely proceed by free carbenes, not
metal carbenoids.51-55 By analogy, the silver-promoted silylene
transfer could involve a free silylene intermediate, or it could
proceed via a silver silylenoid species.56,57 Because of these
ambiguities, we believed a quantitative study was needed to
eliminate a mechanism involving free silylene as a reactive
intermediate.
Results and Discussion
Influence of Catalyst on Reaction Rate. Insight into the
mechanism of silver-promoted di-tert-butylsilylene transfer from
cyclohexene silacyclopropane 1 to an alkene began with the
optimization of experimental conditions. To facilitate kinetic
1
analysis using H NMR spectroscopy, a reproducible reaction
that occurred at a moderate rate around 0 °C was sought.
Extensive experimentation was required to identify the optimum
silver salt to fulfill this requirement. Many silver salts promoted
this reaction at low temperatures (e25 °C, eq 2). The duration
of the reaction time, however, was found to be dependent upon
the counterion. The half-life of silylene transfer varied from 30
min at -35 °C employing 5 mol % of AgOTf or AgOCOCF3
to greater than 8 h at 25 °C with 5 mol % of Ag3PO4.
(38) Lee, M. E.; Cho, H. M.; Ryu, M. S.; Kim, C. H.; Ando, W. J. Am. Chem.
Soc. 2001, 123, 7732-7733.
(39) Driver, T. G.; Franz, A. K.; Woerpel, K. A. J. Am. Chem. Soc. 2002, 124,
6524-6525.
(40) CÄ irakovic´, J.; Driver, T. G.; Woerpel. K. J. Am. Chem. Soc. 2002, 124,
9370-9371.
(41) Takebayashi, M.; Ibata, T. Bull. Chem. Soc. Jpn. 1968, 41, 1700-1707.
(42) Duggleby, P. McM.; Holt, G.; Hope, M. A.; Lewis, A. J. Chem. Soc., Perkin
Trans. 1 1972, 3020-3024.
(43) Sudrik, S. G.; Maddanimath, T.; Chaki, N. K.; Chavan, S. P.; Chavan, S.
P.; Sonawane, H. R.; Vijayamohanan, K. Org. Lett. 2003, 5, 2355-2358.
(44) Salomon, R. G.; Kochi, J. K. J. Am. Chem. Soc. 1973, 95, 3300-3310.
(45) Doyle, M. P.; Griffin, J. H.; Bagheri, V.; Dorow, R. L. Organometallics
1984, 3, 53-61.
(46) Maxwell, J. L.; Brown, K. C.; Bartley, D. W.; Kodadek, T. Science 1992,
256, 1544-1547.
(47) D´ıaz-Requejo, M. M.; Belderrain, T. R.; Nicasio, M. C.; Prieto, F.; Pe´rez,
P. J. Organometallics 1999, 18, 2601-2609.
(48) Straub, B. F.; Hofmann, P. Angew. Chem., Int. Ed. 2001, 40, 1288-1290.
(49) Fraile, J. M.; Garc´ıa, J. I.; Mart´ınez-Merino, V.; Mayoral, J. A.; Salvatella,
L. J. Am. Chem. Soc. 2001, 123, 7616-7625.
(50) For recent reports of other silver or gold catalyzed reactions, see (a)
Longmire, J. M.; Wang, B.; Zhang, X. J. Am. Chem. Soc. 2002, 124,
13400-13401. (b) Momiyama, N.; Yamamoto, H. J. Am. Chem. Soc. 2003,
125, 6038-6039. (c) Dias, H. V. R.; Browning, R. G.; Polach, S. A.;
Diyabalanage, H. V. K.; Lovely, C. J. J. Am. Chem. Soc. 2003, 125, 9270-
9271. (d) Wei, C.; Li, C.-J. J. Am. Chem. Soc. 2003, 125, 9584-9585. (e)
Chen, C.; Li, X.; Schreiber, S. L. J. Am. Chem. Soc. 2003, 125, 10174-
10175.
(51) Kirmse, W. Carbene Chemistry; Academic Press: New York, 1971; Vol.
1, pp 257-260 and 475-493.
(52) Jones, M., Jr.; Moss, R. A. ReactiVe Intermediates in Organic Chemistry;
Wiley: New York, 1973; Vol. 1, p 107-114.
(53) Agosta, W. C.; Wolff, S. J. Org. Chem. 1975, 40, 1027-1030.
(54) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic Methods for
Organic Synthesis with Diazo Compounds; Wiley: New York, 1998.
(55) Julian, R. R.; May, J. A.; Stoltz, B. M.; Beauchamp, J. L. J. Am. Chem.
Soc. 2003, 125, 4478-4486.
(56) For recent examples of metal silylenoids refer to: (a) Shimada, S.; Rao,
M. L. N.; Hayashi, T.; Tanaka, M. Angew. Chem., Int. Ed. 2001, 40, 213-
216. (b) Zhang, Y.; Cervantes-Lee, F.; Pannell, K. H. Organometallics 2003,
22, 2517-2524. (c) Antolini, F.; Gehrhus, B.; Hitchcock, P. B.; Lappert,
M. F. Angew. Chem., Int. Ed. 2002, 41, 2568-2571. (d) Klei, S. R.; Tilley,
T. D.; Bergman, R. G. Organometallics 2002, 21, 4648-4661. (e) Theil,
M.; Jutzi, P.; Neumann, B.; Stammler, A.; Stammler, H.-G. J. Organomet.
Chem. 2002, 662, 34-42. (f) Amoroso, D.; Haaf, M.; Yap, G. P. A.; West,
R.; Fogg, D. E. Organometallics 2002, 21, 534-540. (g) Sato, T.; Okazaki,
M.; Tobita, H.; Ogino, H. J. Organomet. Chem. 2003, 669, 189-199. (h)
Okazaki, M.; Tobita, H.; Ogino, H. J. Chem. Soc., Dalton Trans. 2003, 4,
493-506.
(57) For additional metal-catalyzed reactions of silacycles see: (a) Ohshita, J.;
Ishikawa, M. J. Organomet. Chem. 1991, 407, 157-165. (b) Naka, A.;
Lee, K. K.; Yoshizawa, K.; Yamabe, T.; Ishikawa, M. J. Organomet. Chem.
1999, 587, 1-8. (c) Kang, S.-Y.; Yamabe, T.; Naka, A.; Ishikawa, M.;
Yoshizawa, K. Organometallics 2002, 21, 150-160. (d) Kang, S. O.; Lee,
J.; Ko, J. Coord. Chem. ReV. 2002, 231, 47-65.
During the screening of different silver salts, several chal-
lenges arose. Monitoring the progress of reactions using more-
reactive silver salts such as AgOTf or AgOCOCF3 was difficult.
Agitation of the cold (-78 °C) reaction mixture before
placement in a cold (-40 °C) NMR probe induced reaction up
to 20%. The accumulation of Ag(0) as a mirror or precipitate
as the reaction progressed represented an additional obstacle.
The combination of these challenges led to irreproducible data.
The addition of ancillary ligands was postulated to address
the aforementioned challenges by limiting the decomposition
of the catalyst through the stabilization of reactive intermediates.
The use of phosphine ligands allowed the rate of silylene transfer
1
to be studied conveniently using H NMR spectroscopy. The
substituents on phosphine were varied to observe their impact
on reaction rate (eq 3). Alkyl phosphines were found to inhibit
the reaction. While no reaction was observed employing t-Bu3P,
silylene transfer took place using Cy3P only at temperatures in
excess of 60 °C. At this temperature, silylene transfer occurs
thermally.23 Aryl substituents on phosphine, conversely, facili-
tated silylene transfer at 25 °C without significant catalyst
decomposition.
9
9994 J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004

Silver-Mediated Di-tert-butylsilylene Transfer
A R T I C L E S
presence of styrene and a catalytic amount of (Ph3P)2AgOTf
were examined to determine if cyclohexene silacyclopropane 1
is the most efficient source of di-tert-butylsilylene and if silylene
transfer were reversible (eq 4, Table 1). Of the bicyclic
silacyclopropanes examined, cyclohexene silacyclopropane 1
exhibited the highest reactivity, with a half-life of 1 h at 10 °C
(entry 1). Decreasing or increasing the size of the cycloalkene
portion of the silacyclopropane diminished the rate of reaction.
Cyclopentene silacyclopropane 4 reacted slowly at 25 °C (t1/2
≈ 25 h, entry 2), and cyclooctene silacyclopropane 5 required
temperatures in excess of 60 °C to afford any styrene silacy-
clopropane 3 (entry 3). The different rates for silylene extrusion
from silacyclopropanes 1, 4, and 5 were consistent with the
thermal stabilities of 1, 4, and 5. The cause for the different
relative stabilities of cycloalkene silacyclopropanes, however,
has not been determined.
The electronic nature of the phosphine appears to control the
rate of the reaction.58 Steric effects can be discounted, because
the fastest reaction occurs using (C6F5)3P as the ancillary ligand,
which is similar in size to Cy3P.59 Due to its increased solubility
and the moderate rate of reaction at 10 °C, (Ph3P)2AgOTf was
chosen as the catalyst to study the transfer of di-tert-butylsilylene
from cyclohexene silacyclopropane 1 to an alkene.
Structure of [(Ph3P)2AgOTf]n. The mechanistic study of
silver-mediated silylene transfer commenced with structural
determination of (Ph3P)2AgOTf. In the solid state, the catalyst
exists as the dinuclear complex [(Ph3P)2AgOTf]2, where the
triflates serve as bridges between the two distorted tetrahedral
silver atoms.60 Silver phosphine complexes often exist as higher-
order oligomers in the solid state. With counterions such as
halides or perchlorate, infinite polymeric chains,61 or cubane
structures62,63 are typically observed.
Table 1. Relative Reactivity of Silacyclopropanes
In solution, the silver phosphine complex used for catalysis
exists as a dynamic species.60,64 IR spectroscopy of our catalyst
in solution closely resembled its solid state (KBr pellet)
spectrum: prominent SdO stretches at 1293 and 1096 cm-1
indicated the ionic nature of the triflate counterion.65 Analysis
of (Ph3P)2AgOTf using variable temperature 31P NMR spec-
troscopy further illuminated the catalyst solution structure. At
-40 °C, the 31P {1H} NMR spectrum displayed two doublets
at δ 12.4 ppm, corresponding to 31P-107Ag and 31P-109Ag
coupling (J ) 487.4 Hz and J ) 560.0 Hz, respectively).66 At
temperatures above -20 °C,67 the doublets at δ 12.4 ppm
coalesced into broad peaks at δ 13.5 and 11.0 ppm, indicating
the rapid exchange of triphenylphosphine between 107Ag and
109Ag atoms.68 These studies reveal that at 10 °C, the temper-
ature at which catalysis occurs, a molecule of free triphenyl-
phosphine is rapidly exchanging with the silver phosphine
complex.
Monocyclic silacyclopropane 2 was determined to be stable
to the reaction conditions. Exposure of silacyclopropane 2 to
(Ph3P)2AgOTf and styrene afforded neither silacycle 3 nor any
decomposition products (entry 4). After 3 days at 60 °C,
silacyclopropane 2 remained in over 90% of its starting
concentration. The stability of benzyl silacyclopropane 2 towards
any silver species produced from di-tert-butylsilylene transfer
was also established. After silylene transfer from cyclohexene
silacyclopropane 1 to allylbenzene (to form 2 in 86% yield),
styrene was added (eq 5). Neither silylene transfer to styrene
nor decomposition of 2 was observed. Because silacyclo-
propanes 2 and 5 are inert to silver-salt catalysis below 25 °C,
silylene transfer to acyclic and likely most cyclic alkenes must
be irreversible.
Relative Reactivity of Silacyclopropanes with (Ph3P)2-
AgOTf. The reactivities of various silacyclopropanes in the
(58) For studies detailing the role of the electronic nature of phosphine ligands
in metal phosphine complexes, refer to: (a) Rahman, M. M.; Liu, H. Y.;
Prock, A.; Giering, W. P. Organometallics 1987, 6, 650-658. (b) Rahman,
M. M.; Liu, H.-Y.; Eriks, K.; Prock, A.; Giering, W. P. Organometallics
1989, 8, 1-7. (c) Orpen, A. G.; Connelly, N. G. Organometallics 1990, 9,
1206-1210. (d) Pacchioni, G.; Bagus, P. S. Inorg. Chem. 1992, 31, 4391-
4398.
(59) Tolman, C. A. Chem. ReV. 1977, 77, 313-348.
(60) Bardaj´ı, M.; Crespo, O.; Laguna, A.; Fischer, A. K. Inorg. Chim. Acta
2000, 304, 7-16.
(61) Montes, J. A.; Rodr´ıguez, S.; Ferna´ndez, D.; Garc´ıa-Seijo, M. I.; Gould,
R. O.; Garc´ıa-Ferna´ndez, M. E. J. Chem. Soc., Dalton Trans. 2002, 1110-
1118.
Examination of the Initial Rates of Di-tert-butylsilylene
Transfer. A kinetic analysis of (Ph3P)2AgOTf-mediated di-tert-
butylsilylene transfer from cyclohexene silacyclopropane 1 to
an alkene was performed using the optimal experimental
(62) Teo, B.-K.; Calabrese, J. C. Inorg. Chem. 1976, 15, 2474-2486.
(63) Teo, B.-K.; Calabrese, J. C. Inorg. Chem. 1976, 15, 2467-2474.
(64) Muetterties, E. L.; Alegranti, C. W. J. Am. Chem. Soc. 1970, 92, 4114-
4115.
(65) Ion paired metal triflate salts exhibit stretches at 1270 and 1043 cm-1
;
whereas, a monodentate bound triflate shifts the 1270 cm-1 stretch to 1380
cm-1. Lawrance, G. A. Chem. ReV. 1986, 86, 17-33.
(68) The dissociation constants for silver triphenylphosphine complexes have
been elucidated in DMSO, pyridine, and water and show a significant
solvent effect resulting from the difference in Ag(I) solvation. In methylene
chloride, the expected dissociation constant (K2) is expected to be
significantly smaller than the 8.91 × 105 s-1 reported in water. Refer to:
(a) Ahrland, S.; Hulte´n, F.; Persson, I. Acta Chem. Scand. 1986, A40, 595-
600. (b) Ahrland, S.; Hulte´n, F. Inorg. Chem. 1987, 26, 1796-1798. (c)
Di Bernado, P.; Dolcetti, G.; Portanova, R.; Tolazzi, M.; Tomat, G.;
Zanonato, P. Inorg. Chem. 1990, 29, 2859-2862.
(66) Elemental silver exists as a 48:52 mixture of two NMR active (I ) 1/2)
isotopes, 107Ag and 109Ag, with a gyromagnetic ratio of 1.15. Becker, E.
D. High-Resolution NMR: Theory and Chemical Applications; Aca-
demic: San Diego, 1980; pp 281-291.
(67) At low temperatures (below -40 °C), the disproportionation of (Ph3P)2-
AgOTf into (Ph2P)3AgOTf and Ph3PAgOTf was observed in the resolution
of small broad peaks upfield at 12.3 and 10.6 ppm. See Muetterties and
Alegranti64 for further details.
9
J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004 9995

A R T I C L E S
Driver and Woerpel
conditions. To ensure the solubility of the silver salt over a range
of temperatures, CD2Cl2 was selected as the solvent. Because
1
of its high boiling point, low polarity, and distinct H NMR
spectrum, phenyltrimethylsilane was chosen as the internal
standard. The selection of styrene as the di-tert-butylsilylene
acceptor facilitated analysis using 1H NMR spectroscopy.
Several signals, including those of the t-Bu groups, were distinct
for both the product and starting material. The kinetic order
experiments were performed by the dropwise addition of a
CD2Cl2 solution containing the silver catalyst and alkene to a
cold (-78 °C) CD2Cl2 solution of cyclohexene silacyclopropane
1. Because of this procedure, the catalyst is assumed to exist at
the outset of the reaction as the 18-electron alkene complex,
(Ph3P)2Ag(styrene)2OTf.69 After agitation, the NMR tube con-
taining the reaction mixture was placed in a cool (-20 °C) NMR
probe. The reaction progress was monitored at 10 °C.70
Insight into the mechanism of silver-mediated silylene transfer
was obtained through an examination of the initial rates under
pseudo first-order reaction conditions. The kinetic order was
determined to be one in cyclohexene silacyclopropane 1
concentration (Figure 1). Inverse saturation behavior was
observed for both styrene and cyclohexene concentrations
(Figure 2). Inhibition of the reaction by cyclohexene was more
than four times more severe than styrene. Addition of 1 equiv
of triphenylphosphine (relative to catalyst) caused immediate
precipitation of a white solid, presumed to be [(Ph3P)3AgOTf]n,
and the reaction was completely suppressed.
Figure 1. Determination of kinetic order in [cyclohexene silacyclopropane
1].
Figure 2. Determination of kinetic order in [styrene] and [cyclohexene].
Scheme 1. Mechanistic Origin of Rate Suppression by Alkene
and Phosphine Concentration
Inhibition of the reaction rate through the addition of
triphenylphosphine and the rate differences between aryl and
alkyl phosphines (vide supra) suggest that the ancillary ligand
must initially dissociate (Scheme 1). The addition of triphen-
ylphosphine may only have generated insoluble silver oligomeric
complexes. Alternatively, increasing the phosphine concentration
could suppress reaction by decreasing the amount of mono-
phosphine complex 6. The rate acceleration observed with the
perfluorinated aryl phosphine silver complex could reflect the
lability of this electron poor ligand.58 In contrast, the electron-
rich ancillary ligand, tricyclohexylphosphine, could have slowed
the rate of silylene transfer because of its potent coordination.
Inhibition of the reaction rate by alkene implicates a mech-
anism in which dissociation of a molecule of alkene must occur
before reaction of the catalyst with cyclohexene silacyclopropane
1 (Scheme 1). Given silver’s affinity for alkenes,71-75 increasing
the concentration of either styrene or cyclohexene would have
a deleterious effect on reaction rate by decreasing the amount
of free, reactive catalyst (7). The proposal that an alkene must
dissociate from silver is also consistent with the relative
magnitude of inhibition by addition of different alkenes. The
observation that silylene transfer is inhibited four times more
by cyclohexene than by styrene correlates well with the known
four-fold lower dissociation constants for disubstituted alkenes
versus monosubstituted alkenes.74 The suppressed rate could
originate, additionally, from faster trapping of an intermediate
by cyclohexene than by styrene, as has been observed for
t-Bu2Si.23
Determination of kinetic order concluded with the observation
of saturation behavior in (Ph3P)2AgOTf concentration (Figure
3). Saturation behavior occurs when the concentration of
catalyst, (Ph3P)2Ag(styrene)2OTf, appears in both the numerator
and the denominator of the rate expression. To satisfy this
(69) Typically when olefin ligands are present, the Ag+ ion is tetrahedrally
coordinated. See: (a) Baenziger, N. C.; Haight, H. L.; Alexander, R.; Doyle,
J. R. Inorg. Chem. 1966, 5, 1399-1400. (b) Gray, D.; Wies, R. A.; Closson,
W. D. Tetrahedron Lett. 1968, 9, 5639-5641. (c) Rodesiler, P. F.; Hall
Griffith, E. A.; Amma, B. L. J. Am. Chem. Soc. 1972, 94, 761-766. (d)
Mak, T. C. W.; Ho, W. C.; Huang, N. Z. J. Organomet. Chem. 1983, 251,
413-421.
(70) Refer to the experimental and Supporting Information sections for further
details.
(71) Keller, R. N. Chem. ReV. 1941, 28, 229-267.
(72) Fueno, T.; Okuyama, T.; Deguchi, T.; Furukawa, J. J. Am. Chem. Soc.
1965, 87, 170-174.
(73) Parker, R. G.; Roberts, J. D. J. Am. Chem. Soc. 1970, 92, 743.
(74) Wilcox, C. F., Jr.; Gaal, W. J. Am. Chem. Soc. 1971, 93, 2453-2459.
(75) Salomon, R. G.; Kochi, J. K. J. Am. Chem. Soc. 1973, 95, 1889-1897.
9
9996 J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004

Silver-Mediated Di-tert-butylsilylene Transfer
A R T I C L E S
Scheme 3. Oligomerization as Explanation for Saturation
Behavior in Silver Concentration
Figure 3. Determination of kinetic order in [(Ph3P)2AgOTf].
Scheme 2. Possible Mechanism To Satisfy Saturation Behavior of
Catalyst
formation of and not the ring-opening of silacyclopropanes.
Second, similar activation parameters to thermal silylene
transfer23 would be expected to describe this mechanism, but
they are different (vide infra).
Other mechanisms involving the preactivation of cyclohexene
silacyclopropane 1 before reaction with the silver catalyst also
appear to be inconsistent with our observations (eq 7). While
the formation of the siliconate of 1 should increase the lability
of di-tert-butylsilylene, coordination by triflate or alkene would
lead to kinetic orders in catalyst or alkene that are greater than
one. Unique activation parameters would be observed if the
siliconate is formed with solvent, but these parameters are
consistent in two different solvents (vide infra). In an additional
mechanism, the cleavage of a C-Si bond before reaction with
catalyst could occur (eq 8). The silver-catalyzed silylene transfer,
however, is stereospecific,40 and a Hammett F value inconsistent
with the stepwise extrusion of silylenoid was observed (vide
infra). The activation of cyclohexene silacyclopropane 1 does
not appear to occur before reaction with catalyst.
requirement, a reversible step must precede participation of a
molecule of catalyst in the mechanism. If unreactive silver
catalyst oligomers accumulated at higher concentrations, this
nonlinear behavior could result from a kinetic order less than
one (but greater than zero). All experiments, however, were
performed significantly below concentrations where silver
phosphine complexes have been reported to be monomeric.60,64,76
Mechanisms Consistent with the Kinetic Behavior. Several
mechanisms satisfy the rate expression constraints imposed by
the observed saturation behavior (Schemes 2 and 3).77 In the
first mechanism (Scheme 2, eq 6), cyclohexene silacyclopropane
1 extrudes di-tert-butylsilylene, which is rapidly trapped with
catalyst. Silylsilver complex 9 then reacts with a molecule of
styrene to afford the product styrene silacyclopropane. This
mechanism has several liabilities. First, it violates the principle
of microscopic reversibility, since silver only participates in the
A mechanism that involves the sequestering of the silver
catalyst as a silver oligomer appears to be most consistent with
the observed data (Scheme 3). In this mechanism, increasing
the concentration of silver serves only to favor the equilibrium
toward an unreactive oligomeric reservoir. Upon dissociation,
ligand exchange of triphenylphosphine with a molecule of
styrene occurs to form 6. To liberate a coordination site and
thus allow reaction with cyclohexene silacyclopropane 1, styrene
dissociates from the monophosphine catalyst. These two el-
ementary steps explain the inhibition by alkene and phosphine.
Silver then mediates the reversible extrusion of di-tert-butyl-
(76) Bachman, R. E.; Andretta, D. F. Inorg. Chem. 1998, 37, 5657-5663.
(77) Another conceivable reaction mechanism in which free silylene reacts with
a molecule of (Ph3P)2Ag(styrene)2OTf to produce product would exhibit a
rate law inconsistent with the kinetic data (zero order in styrene concentra-
tion would be predicted). For more details and analyses of other less-likely
mechanisms refer to the Supporting Information.
9
J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004 9997

A R T I C L E S
Driver and Woerpel
Scheme 4. Behavior for Silylsilver Complex 8
even at -50 °C, triphenylphosphine is rapidly exchanging from
the silylsilver intermediate. Analysis of the reaction mixture
using IR spectroscopy revealed prominent new peaks at 1204
and 972 cm-1. The new spectrum appears consistent with that
expected for a monodentate binding of triflate to a silicon
atom.65,84 Attempts to induce crystallization of the silylsilver
complex were unsuccessful.
Figure 4. 29Si {1H} NMR spectroscopic evidence for silylsilver complex
8.
The behavior of the reactive silver intermediate was analyzed
qualitatively (Scheme 4). Warming silylsilver complex 8 to
above -20 °C resulted in the disappearance of the distinctive
29Si NMR doublets and the accumulation of oligomeric silicon
products.85 The silylsilver complex appeared to be catalytically
active. The addition of allylbenzene (4 equiv) to the intermediate
resulted in complete consumption of cyclohexene silacyclopro-
pane 1 and formation of product silacyclopropane 2. Further-
more, analysis of the 29Si {1H} NMR spectrum revealed the
disappearance of the peaks associated with silylsilver complex
8 and the appearance of a new set of two doublets situated at
δ 90 ppm. The observed ratio of 1.155:1 between the coupling
constants (J ) 372.5 and 322.4 Hz) is consistent with the
gyromagnetic ratio expected for 109Ag-29Si and 107Ag-29Si
coupling.66 A different ligand environment around silver (cy-
clohexene replaced with allylbenzene) could account for the
variation of chemical shift and coupling constants between 8
and 10. Since exposure of silacyclopropane 2 to (Ph3P)2AgOTf
did not yield 8, 10, or allylbenzene, formation of silylsilver
complex 8 from cyclohexene silacyclopropane 1 is unique.
Unfortunately, monitoring the rate of formation of silver
intermediate 8 using spectroscopic techniques was not success-
ful.
silylene from 1 to form silylsilver complex 8, which then reacts
with an additional molecule of alkene to form the product,
styrene silacyclopropane 3. The resultant rate expression (eq
9) indicates that saturation behavior in cyclohexene silacyclo-
propane 1 should be observed. Because only a small range of
concentrations was examined for 1, it is possible that saturation
never was reached.
Identification and Analysis of a Silylsilver Intermediate.
Silylsilver complex 8 was identified as a reactive intermediate
using low-temperature NMR and IR spectroscopy. For these
experiments, the reaction between (Ph3P)2AgOTf (1 equiv) and
cyclohexene silacyclopropane 1 (2 equiv) was analyzed at -40
°C (Scheme 4). Free cyclohexene was observed in the 1H NMR
spectrum,78 and analysis of the 29Si {1H} NMR spectrum
revealed the appearance of two new doublets situated at δ 97
ppm (Figure 4). The observed 1.155:1 ratio between the coupling
constants (J ) 259.9 and 224.9 Hz) implicates 107Ag-29Si and
109Ag-29Si coupling, since their relative gyromagnetic ratio is
1.15.66 The downfield location of this silicon species is
consistent with a Lewis base-stabilized metal silylenoid.79,80
Recently, the silicon atom of the silylgold complex, Ph3PAu-
SiCp*2Cl,56e was reported to appear at δ 77.6 ppm.81 In contrast
to the silylgold species, no 31P-29Si coupling was observed in
either the 29Si or 31P {1H} NMR spectrum of our silver
intermediate.82,83 The lack of phosphorus coupling indicates that
(82) At -60 °C, the peaks present in 31P {1H} NMR spectrum sharpened, and
the area of the doublet at 6.23 ppm was noticeably smaller. Examination
of the NMR tube, however, revealed the formation of extensive precipita-
tion. Further analysis at this (or lower) temperatures was hampered by the
insolubility of the intermediates.
(83) In addition to a large singlet at δ -1.95 ppm, the 31P {1H} NMR spectrum
also exhibited smaller peaks at δ 6.23 ppm (d, J ) 248 Hz) and 4.52 ppm.
The identity of the doublet at δ 6.23 ppm could be (Ph3P)4AgOTf (data
consistent with the report of Laguna et al.60). Since there were no other
identifiable peaks in the 29Si NMR spectrum, the smaller broad singlet at
δ 4.52 could either be another product that does not contain silicon or
evidence of the formation of an oligomeric silicon phospine compound.
(84) A 0.4 M solution of t-BuMe2SiOTf in CH2Cl2 exhibited an IR spectrum
with peaks at 1390, 1247, 1212, 1154, and 970 cm-1. Refer to the
Supporting Information for an overlay of the silylsilver intermediate and
t-BuMe2SiOTf IR spectra.
(85) While the isolation of any distinct products was not possible, no
spectroscopic evidence for the formation of any Si-H products was
observed. For references on various oligomeric silicon products resulting
from the decomposition of silylenes refer to: (a) Masamune, S.; Murakami,
S.; Tobita, H. Organometallics 1983, 2, 1464-1466. (b) Watanabe, H.;
Muraoka, T.; Kageyama, M.; Yochizumi, K.; Nagai, Y. Organometallics
1984, 3, 141-147. (c) Kyushin, S.; Sakurai, H.; Matsumoto, H. J.
Organomet. Chem. 1995, 499, 235-240.
(78) The absence of a diagnostic δ ppm shift in the 1H NMR spectrum of
cyclohexene in the presence of silylsilver complex 8 (relative to cyclohexene
without silver) did not provide evidence for the coordination of cyclohexene
to silver.
(79) Lewis base stabilized silylenoid complexes have been reported to appear
at 29Si δ 88.6 ppm for (Et3P)2PtSi(St-Bu)2(OEt2) by Mitchell, G. P.; Tilley,
T. D. J. Am. Chem. Soc. 1998, 120, 7635-7636 or 29Si δ 56.0 ppm for
Fp-SiMe2Ot-Bu by Sharma, H. K.; Pannell, K. H. Organometallics 2001,
20, 7-9.
(80) Neutral metal silylenoids appear further downfield: 29Si δ 367 ppm for
(Cy3P)2PtdSiMes2 by Feldman, J. D.; Mitchell, G. P.; Nolte, J.-O.; Tilley,
T. D. J. Am. Chem. Soc. 1998, 120, 11184-11185 and 29Si δ 241 ppm for
[PhB(CH2PPh2)3Ir(H)2(dSiMes2)]+ by Peters, J. C.; Feldman, J. D.; Tilley,
T. D. J. Am. Chem. Soc. 1999, 121, 9871-9872.
(81) Exposure of 1 equiv of Ph3PAuCl to 2 equiv of cyclohexene silacyclo-
propane 1 led to the appearance of a broad peak between 80 and 82 ppm
in the 29Si NMR spectrum attributable to the formation of Ph3PAu-Si(t-
Bu)2(Cl).
9
9998 J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004

Silver-Mediated Di-tert-butylsilylene Transfer
A R T I C L E S
sign and magnitude of the F-values, however, are all similar,
indicating that the values are not describing styrene dissociating
from a metal-alkene complex. If such were the case, work by
Roberts73 and Kochi75 would predict a change in the F-value
sign when copper was substituted for silver as the silylene
transfer catalyst.88 In these reports, a silver-alkene complex was
described as having less electron density on the alkene (as
compared to the free alkene),89,90 a copper-alkene complex has
increased electron density;75 and a gold-alkene complex a
mixture (electron density on one carbon increased, the other
reduced).91,92
Figure 5. Linear correlation of krel with the Hammett equation using σp
constants.
Competition Experiments between para-Substituted Sty-
renes. A series of competition experiments between various
para-substituted styrenes (p-OMe, p-Me, p-F, p-Cl, and p-CF3)
and a deficient amount of 1 in the presence of 5 mol % of
(Ph3P)2AgOTf provided insight into the electronic nature of the
transition state for metal-mediated cycloaddition. Electron-rich
styrenes were found to be more reactive. The rate difference
for formation of substituted styrene silacyclopropanes correlated
linearly with the Hammett equation using σp86 constants to result
in a F value of -0.62 ( 0.02 at 25 °C (Figure 5). The negative
value reveals the nucleophile to be the alkene source and
establishes the electrophilicity of the silylsilver intermediate.
The magnitude of the F value suggests a concerted reaction in
which a small partial positive charge is generated on the benzylic
carbon in the transition state.87
The magnitude and sign of the Hammett F value were
validated through examination of temperatures within 30 °C that
did not reveal an isokinetic point. The F value at 8 °C and -8
°C were found to be -0.71 and -0.79, respectively. The linear
relationship observed between temperature and F value allowed
estimation of the isokinetic temperature to be 129 °C. The
isokinetic relationship observed in silver-mediated silylene
transfer is opposite to what was observed for the metal-free,
thermal reaction.23 Silver-mediated silylene transfer is therefore
under enthalpic control, whereas thermal silylene transfer is
entropically driven. Enthalpic control of cyclization requires a
relatively large ∆Hq (vide infra) and implicates formation of a
stable reactive intermediate, which is consistent with the
proposal that the observed silylsilver complex (vide supra) is
an intermediate prior to cyclization.
The reduction of electron density on the alkene in a silver-
alkene complex is also inconsistent with it acting as the
nucleophile in the mechanism. While the presence of an
electron-donating ancillary ligand impacts the electron density
of metal-olefin complex,93-96 we observed little variation in the
1H NMR spectra of the vinylic protons of styrene when it was
exposed to various concentrations of (Ph3P)2AgOTf.97 While
1
the lack of diagnostic H NMR spectroscopic shifts prevented
a definitive conclusion on the nucleophilicity of the phosphine
silver-alkene complex, they did establish that the alkene portion
of the (Ph3P)2Ag(styrene)2OTf complex does not contain
significantly more electron density (i.e., is not more nucleophilic)
than uncomplexed styrene.
Our observations are not consistent with a mechanism
involving free silylene. The reduction in reaction temperature
from the addition of catalyst implicates silver involvement in
silylene extrusion. The identification of the silylsilver reactive
intermediate suggests that silylene is shuttled from cyclohexene
(88) The shielding or deshielding of vinylic protons (∆δ ppm) between an alkene
and a metal‚alkene complex was used to generalize the amount of π electron
density present on the alkene moiety.
(89) Quinn, H. W.; McIntyre, J. S.; Peterson, D. J. Can. J. Chem. 1965, 43,
2896-2910.
In addition to (Ph3P)2AgOTf, other salts, including (CuOTf)2‚
PhH, AgOTf, and Ph3PAuCl, were examined as catalysts (eq
10). In each case, a unique negative Hammett F value was
obtained. The different values indicate that neither counterion
nor phosphine is responsible for cyclization. The values are also
different in magnitude from the F value (-0.415) calculated
from the thermal, metal-free isokinetic studies,23 eliminating
cyclization involving free silylene as a possible mechanism. The
(90) Quinn, H. W.; VanGilder, R. L. Can. J. Chem. 1969, 47, 4691-4694.
(91) Hu¨ttel, R.; Reinheimer, H. Chem. Ber. 1966, 99, 2778-2781.
(92) Hu¨ttel, R.; Reinheimer, H.; Nowak, K. Chem. Ber. 1968, 101, 3761-3776.
(93) Solodar, J.; Petrovich, J. P. Inorg. Chem. 1971, 10, 395-397.
(94) For the Dewar-Chatt model of metal-olefin bonding, see: (a) Dewar, M.
J. S. Bull. Soc. Chim. Fr. 1951, 18, C71-C79. (b) Chatt, J.; Duncanson,
L. A. J. Chem. Soc. 1953, 2939-2947.
(95) Bennett, M. A.; Kneen, W. R.; Nyholm, R. S. Inorg. Chem. 1968, 7, 552-
556.
(96) Electron-rich 2,2′-bipyridine ancillary ligands on copper(I) alkene complexes
augmented the electron density present on the bound alkene (relative to
free alkene) through increased π back-donation in the copper(I)-ethylene
bonding. See: Munakata, M.; Kitagawa, S.; Kosome, S.; Asahara, A. Inorg.
Chem. 1986, 25, 2622-2627.
(86) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165-195.
(87) Stepwise reactions that involve charged intermediates, such as the hydration
and bromination of styrene, are typified by large negative F values. For
example: hydration F ) -4 (by Schubert, W. M.; Lamm, B.; Keefe, J. R.
J. Am. Chem. Soc. 1964, 86, 4727-4729) and bromination F ) -4.3 (by
Dubois, J. E.; Schwarcz, A. Tetrahedron Lett. 1964, 5, 2167-2173).
(97) In contrast to ∆δ of 0.90, 0.62, and 0.33 ppm reported in Quinn and
VanGilder90 for AgOTf‚tert-butylethylene, we observed ∆δ of 0.019, 0.015,
and 0.015 ppm for (Ph3P)2Ag(styrene)nOTf.
9
J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004 9999

A R T I C L E S
Driver and Woerpel
Scheme 5. Catalytic Cycle To Describe Silver-Catalyzed Transfer
of Di-tert-butylsilylene from 1 to Styrene
be the concerted, electrophilic cyclization of the silylsilver
complex 10 with a molecule of alkene. Because a silylsilver
complex can be observed and the cyclization is enthalpically
driven (vide supra), a considerable ∆Gq is implicated for
cyclization and extrusion. Since the progress of the reaction can
be followed using NMR spectroscopy without evidence of a
paramagnetic species, transition structure 12, formed via trans-
metalation, is postulated. Rate-determining cyclization then
occurs rapidly to form product silacyclopropane 3 and regenerate
the silver-alkene catalyst.
The difference in rate between cycloaddition onto cyclohex-
ene and styrene (k3 and k-2) was estimated by comparison of
the initial rates of silylsilver complex trapping by cyclopentene
and cyclooctene versus styrene (eq 11). Competition experiments
were performed using 10 equiv of alkene (below saturation
behavior). Both cyclopentene and cyclooctene were equally
reactive (within experimental error) and slightly more efficient
than styrene in consuming cyclohexene silacyclopropane 1.
Since the difference in mono- versus disubstituted alkene
dissociation from Ag+ has been reported to be four,74 the krel
value of 1.1 may reflect the difference between cyclization of
silylsilver complex 10 onto a cyclic alkene and styrene.
to styrene without dissociation from silver. Since silver is
involved in extrusion it must also be involved in cyclization.
The Hammett F value describing the silver-promoted transfer
of silylene is different than the F value estimated for 25 °C
from the thermal transfer of silylene. Additionally, the identi-
fication that the silver-catalyzed cyclization of silylene is under
enthalpic control is different from the entropic control of thermal
silylene transfer. Enthalpic control is not consistent with free
silylene as an intermediate prior to cyclization.
Determination of Activation Parameters. The activation
parameters of silver-mediated di-tert-butylsilylene transfer from
cyclohexene silacyclopropane 1 to styrene were determined in
both toluene-d8 and CD2Cl2. Unfortunately, neither the extrusion
nor cyclization mechanistic step could be isolated for analysis
through the manipulation of reagent concentrations. For the
activation parameters to represent the entire mechanism, and
not alkene dissociation from the silver-alkene complex, an
alkene concentration below saturation behavior (5 equiv) was
chosen for investigation. A correlation between the observed
rate and temperature was obtained between 10 °C and 30 °C
(Figures 6 and 7).99 In toluene-d8, Arrhenius and Eyring
activation parameters were determined to be: Ea ) 30 ( 1
Derivation of a Catalytic Cycle. Analysis of the experi-
mental data in its entirety allows construction of a catalytic cycle
that describes the silver-mediated di-tert-butylsilylene transfer
from cyclohexene silacyclopropane 1 to styrene (Scheme 5).
The solid [(Ph3P)2AgOTf]2 first dissociates into the monomeric
species followed by coordination of styrene. Presumably, a third
molecule of styrene replaces a triphenylphosphine ligand to
generate the active catalyst 6. At any point in the saturation of
silver’s coordination sphere with styrene, oligomerization is
believed to be a competing pathway. A molecule of styrene,
subsequently, is exchanged for silacyclopropane 1. Because of
silver’s affinity for alkenes and from the observation of inverse
saturation behavior in alkene concentration, k-1 is greater than
the rate of silylene extrusion (k2). Upon coordination of 1, silver-
mediated â-silyl elimination98 (i.e., migratory insertion/dein-
sertion) occurs to yield cyclohexene and the silylsilver complex
10 that was observed spectroscopically. Based on low-temper-
ature NMR spectroscopic experiments, the first three mecha-
nistic steps (phosphine dissociation, styrene dissociation, and
silver-mediated silylene extrusion from 1) are significantly more
rapid than electrophilic attack of the silylsilver complex. From
kinetic order experiments, the rate-determining step appears to
kcal‚mol-1, A ) 1.25 × 1019 s-1 and ∆Hq ) 30 ( 1 kcal‚mol-1
,
and ∆Sq ) 27 ( 7 eu.100,101 Similar activation parameters were
observed in CD2Cl2: (Arrhenius) Ea ) 31 ( 1 kcal‚mol-1
A ) 1.26 × 1020 s-1 and (Eyring) ∆Hq ) 30 ( 1 kcal‚mol-1
,
,
and ∆Sq ) 31 ( 7 eu. The similar magnitude of the activation
parameters obtained in CD2Cl2 and toluene-d8 suggest the
(99) Since a limited temperature range (40 °C) was used to determine the
activation parameters, caution should be used in drawing firm conclusions
from them.
(100) The uncertainties in the activation parameters were calculated from the
error propagation formulas derived by Girolami and Binsch. See: (a)
Morse, P. M.; Spencer, M. D.; Wilson, S. R.; Girolami, G. S. Organo-
metallics 1994, 13, 1646-1655. (b) Steigel, A.; Sauer, J.; Kleier, D. A.;
Binsch, G. J. Am. Chem. Soc. 1972, 94, 2770-2779.
(98) For reports of â-silyl elimination see: (a) Randolph, C. L.; Wrighton, M.
S. J. Am. Chem. Soc. 1986, 108, 3366-3374.(b) Wakatsuki, Y.; Yamazaki,
H.; Nakano, M.; Yamamoto, Y. J. Chem. Soc., Chem. Commun. 1991, 703-
704. (c) Marciniec, B.; Pietraszuk, C. J. Chem. Soc., Chem. Commun. 1995,
2003-2004. (d) Marciniec, B.; Kownacki, I.; Kubicki, M. Organometallics
2002, 21, 3263-3270.
(101) The systematic uncertainties δksys were estimated based on subjective
judgments of the sensitivities of the NMR integration, ca. 5% (Morse et
al.100a). The total uncertainties δk were calculated from δk ) [δk(ran)
+
2
2
δk(sys)
]
1/2. See: Li, L.; Hung, M.; Xue, Z. J. Am. Chem. Soc. 1995, 117,
12746-12750.
9
10000 J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004

Silver-Mediated Di-tert-butylsilylene Transfer
A R T I C L E S
tration and observed saturation behavior in catalyst concentra-
tion. The observed small, negative Hammett F value suggested
that rate-determining cycloaddition occurs through a concerted,
electrophilic attack of a silylsilver species on the alkene. Similar
activation parameters obtained in toluene and methylene chloride
suggested the generality of the mechanism across different
solvents. We believe that the mechanistic insight gained in this
study will guide future rational modifications to improve both
the construction and synthetic utility of silacyclopropanes.
Experimental Section
General Comments Concerning the Procedure. Unless otherwise
noted, all reagents were commercially obtained and, where appropriate,
purified prior to use. Toluene-d8 was distilled from CaH2 and degassed
prior to use. Deuterated methylene chloride was dried over activated
molecular sieves and degassed before use. All reagents and cyclohexene
di-tert-butylsilacyclopropane 1 were stored in an Innovative Technolo-
gies nitrogen atmosphere drybox.
Figure 6. Linear Arrhenius correlation between rate and temperature.
The apparatuses (volumetric flasks and NMR tubes) were washed
with a solution of 2:1 HCl (12 N): HNO3 (16 N), rinsed with water
and acetone, washed with ammonium hydroxide, and rinsed with water
and acetone. The NMR tubes were dried at 130 °C. Each experiment
was set up in the nitrogen atmosphere drybox and was carried out in
triplicate (three at a time). The reagents were added, dropwise, to a
cold solution of cyclohexene silacyclopropane 1. After agitation, the
NMR tube containing the reaction mixture was placed in a cold (below
the temperature for kinetic analysis) NMR spectrometer. The reaction
mixture was warmed to the desired temperature in the NMR spectrom-
eter. The reaction progress over the first half-life (or, when appropriate,
five half-lives) was monitored periodically using 1H NMR spectroscopy.
The concentrations of the reactants and products were obtained through
comparison of the area of the standard (PhSiMe3 peak at 0.19 ppm,
area ) 1.0000) and the area of the t-Bu peaks of the silacyclopropanes
1 and 3 (1, 1.00 and 1.19 ppm; 3, 1.06 and 0.83 ppm). The data obtained
were fit to the best straight line using a least-squares program. For
tabular and graphical representation of the data, refer to the Supporting
Information.
Figure 7. Linear Eyring correlation between rate and temperature.
mechanism of silver-mediated silylene transfer does not involve
participation of a solvent.
The large, positive entropy of activation parameter could
reflect several aspects of the mechanism. A significant portion
of its value is the dissociation of the silver oligomer into the
reactive monomer 6. Additionally, the dissociation of a molecule
of styrene from the silver‚alkene complex 7 (step 1) and the
loss of cyclohexene from the transition state 12 (step 2) may
also contribute to the entropy of activation parameter. The
magnitude of the entropy of activation term may also indicate
that in the rate-determining step (step 3), a fragmentation occurs
(such as loss of an alkene from the [Ag] complex).
The general procedure is illustrated for the reaction of cyclohexene
silacyclopropane 1 with excess styrene in the presence of 5 mol % of
(Ph3P)2AgOTf. To a cool (-78 °C) solution of cyclohexene silacyclo-
propane 1 (0.065 mL of a 0.4455 M solution in CD2Cl2, 0.02895 mmol)
in 0.200 mL of a 0.023246 M solution of PhSiMe3 in CD2Cl2 (contained
in a thin wall NMR tube) was added a solution containing styrene (0.065
mL, 2.07 mmol) and (Ph3P)2AgOTf (0.050 mL of a 0.02896 M solution
in CD2Cl2) in 0.250 mL of a 0.023246 M solution of PhSiMe3 in
CD2Cl2. After thermal equilibration to -78 °C, the resulting solution
was shaken 5 times. The reaction mixture was re-cooled to -78 °C,
and then it was placed in a cool (-20 °C) NMR spectrometer. An
Conclusion
1
initial H NMR spectrum was obtained, and the temperature of the
A quantitative analysis of the behavior of cyclohexene
silacyclopropane 1 upon exposure to an alkene and a catalytic
amount of (Ph3P)2AgOTf improved our understanding of the
fundamental reactivity of silacyclopropanes. Qualitative obser-
vations established the unique reactivity of 1 and the irrevers-
ibility of silver-mediated silylene transfer from 1 to cyclic
andacyclic alkenes. A mechanism involving free silylene
cyclization was eliminated through kinetic studies and spectro-
scopic observation of a catalytically active silylsilver intermedi-
ate. Kinetic order experiments implicated a mechanism in which
silver mediates silylene extrusion from silacyclopropane 1. The
data were summarized into a plausible catalytic cycle that
accommodated the inhibition of reaction rate in alkene concen-
experiment was regulated to 10 °C. Periodically the reaction progress
was measured relative to the internal standard of PhSiMe3 using H
NMR spectroscopy.
1
Acknowledgment. This research was supported by the
National Institute of General Medical Sciences of the National
Institutes of Health (GM54909). K. A. W. thanks the Camille
and Henry Dreyfus Foundation, Merck Research Laboratories,
Johnson & Johnson, and the Sloan Foundation for awards to
support research. T.G.D. thanks Novartis Pharmaceuticals
through the ACS Division of Organic Chemistry and Eli Lilly
and Company for predoctoral fellowships. We thank Dr. Phil
Dennison for assistance with NMR spectrometry and Professors
9
J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004 10001

A R T I C L E S
Driver and Woerpel
A. F. Heyduk (UCI) and R. G. Bergman (UC Berkeley) for
insightful discussions.
thermodynamic parameters, and spectroscopic data for the
products (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
Supporting Information Available: Experimental procedures,
tabular, graphical, mathematical derivations of kinetic and
JA0306563
9
10002 J. AM. CHEM. SOC. VOL. 126, NO. 32, 2004