Reversible cycloaddition of isocyanates to ruthenium silylene complexes

Article abstract of DOI:10.1021/ja972020f

Reactions of ruthenium silylene complexes of the type [Cp*(PMe₃)₂Ru=SiR₂]⁺ with unsaturated compounds were investigated. Nonpolar, unsaturated substrates such as ethylene, acetylene, and 2-butyne do not react with [Cp*(PMe₃)₂Ru=SiR₂]B(C₆F₅)₄ (1a,R = Me; 1b, R = Ph). However, methyl isocyanate inserts into an Si-S bond of the silylene complex [Cp*(PMe₃)₂Ru=Si(STol)₂][BP₄] (5) to give the 1,2-dipolar addition product {Cp*(PMe₃)₂-RuSi(STol)[η²-O(MeN)C(STol)]}[BPh₄] (6a) in 87% yield. The product was characterized by X-ray crystallography as possessing a base-stabilized silylene ligand with thiolate and thiocarbamate substituents. Compound 1a reacts with methyl and phenyl isocyanate to give the 2 + 2 cycloaddition products [Cp*(PMe₃)₂RuSiMe₂NRC=O][B(C₆F₅)₄] (7, R = Me; 8, R = Ph). The analogous triflate complexes [Cp*(PMe₃)₂RuSiR₂NMeC=O][OTf] (9, R = Ph; 10, R = Me) were prepared by reaction of the appropriate ruthenium silyls Cp*(PMe₃)₂RuSiR₂OTf with methyl isocyanate. Heating 9 to 100°C in toluene resulted in dissociation of the isocyanate and regeneration of the triflate Cp*(PMe₃)₂-RuSiPh₂OTf. Competition reactions of various para-substituted phenyl isocyanates with 1b show that the rate of cycloaddition increases with the electron-donating ability of the incoming isocyanate. This is consistent with a stepwise cycloaddition mechanism involving initial coordination of the isocyanate nitrogen atom to the Lewis acidic silylene silicon atom.

Full text of DOI:10.1021/ja972020f

11236
J. Am. Chem. Soc. 1997, 119, 11236-11243
Reversible Cycloaddition of Isocyanates to Ruthenium Silylene
Complexes
Gregory P. Mitchell and T. Don Tilley*
Contribution from the Department of Chemistry, UniVersity of California, Berkeley,
Berkeley, California 94720-1460
ReceiVed June 18, 1997X
Abstract: Reactions of ruthenium silylene complexes of the type [Cp*(PMe3)2RudSiR2]+ with unsaturated compounds
were investigated. Nonpolar, unsaturated substrates such as ethylene, acetylene, and 2-butyne do not react with
[Cp*(PMe3)2RudSiR2]B(C6F5)4 (1a, R ) Me; 1b, R ) Ph). However, methyl isocyanate inserts into an Si-S bond
of the silylene complex [Cp*(PMe3)2RudSi(STol)2][BPh4] (5) to give the 1,2-dipolar addition product {Cp*(PMe3)2-
RuSi(STol)[η2-O(MeN)C(STol)]}[BPh4] (6a) in 87% yield. This product was characterized by X-ray crystallography
as possessing a base-stabilized silylene ligand with thiolate and thiocarbamate substituents. Compound 1a reacts
with methyl and phenyl isocyanate to give the 2 + 2 cycloaddition products [Cp*(PMe3)2RuSiMe2NRCdO][B(C6F5)4]
(7, R ) Me; 8, R ) Ph). The analogous triflate complexes [Cp*(PMe3)2RuSiR2NMeCdO][OTf] (9, R ) Ph; 10,
R ) Me) were prepared by reaction of the appropriate ruthenium silyls Cp*(PMe3)2RuSiR2OTf with methyl isocyanate.
Heating 9 to 100 °C in toluene resulted in dissociation of the isocyanate and regeneration of the triflate Cp*(PMe3)2-
RuSiPh2OTf. Competition reactions of various para-substituted phenyl isocyanates with 1b show that the rate of
cycloaddition increases with the electron-donating ability of the incoming isocyanate. This is consistent with a
stepwise cycloaddition mechanism involving initial coordination of the isocyanate nitrogen atom to the Lewis acidic
silylene silicon atom.
Introduction
plexes has been made in recent years. In 1987, closely related
base adducts containing tetrahedral silicon centers, of the type
LnMSiR2(donor), were described.4 Since then, numerous base-
stabilized silylene complexes have been reported, and in 1990
we were able to isolate the first examples of base-free silylene
complexes (with sp2 silicon), [Cp*(PMe3)2RudSi(SR)2]BPh4
(Cp* ) η5-C5Me5).5 Subsequently, we described the synthesis
and structural characterizations of the transition metal substituted
silylene complex (CO)4OsSi(STol-p)[RuCp*(PMe3)2],6 hetero-
atom-stabilized [trans-(Cy3P)2(H)PtdSi(SEt)2]BPh4,7 and
[Cp*(PMe3)2RudSiR2]B(C6F5)4 (1a, R ) Me; 1b, R ) Ph).8
Silylene complexes have also been obtained by coordination
Transition metal silylene complexes have been the target of
numerous synthetic investigations over many years since these
compounds are expected to mediate useful transformations of
organosilicon species.1,2 In fact, a number of stoichiometric
and catalytic reactions based on transition metal reagents appear
to involve silylene complexes as intermediates.1-3 Considerable
progress toward understanding the properties of silylene com-
X Abstract published in AdVance ACS Abstracts, November 1, 1997.
(1) (a) Tilley, T. D. In The Silicon-Heteroatom Bond; Patai, S.,
Rappoport, Z., Eds.; Wiley: New York, 1991; Chapters 9 and 10, pp 245
and 309. (b) Tilley, T. D. In The Chemistry of Organic Silicon Compounds;
Patai, S., Rappoport, Z., Eds.; Wiley: New York, 1989; Chapter 24, p 1415.
(c) Tilley, T. D. Comments Inorg. Chem. 1990, 10, 37. (d) Zybill, C. Top.
Curr. Chem. 1991, 160, 1. (e) Lickiss, P. D. Chem. Soc. ReVs. 1992, 271.
(f) Corey, J. In AdVances in Silicon Chemistry; Larson, G., Ed.; JAI Press,
Inc.: Greenwich, CT, 1991; Vol. 1, p 327. (g) Pannell, K. H.; Sharma, H.
K. Chem. ReV. 1995, 95, 1351. (h) Schubert, U. Angew. Chem., Int. Ed.
Engl. 1994, 33, 419.
(2) (a) Curtis, M. D.; Epstein, P. S. AdV. Organomet. Chem. 1981, 19,
213. (b) Choe, S. B.; Sanai, H.; Klabunde, J. J. Am. Chem. Soc. 1989, 111,
2875. (c) Okinoshima, H.; Yamamota, K.; Kumada, M. J. Am. Chem. Soc.
1972, 94, 9273. (d) Kobayashi, T.; Hayahi, T.; Yamahita, H.; Tanaka, M.
Chem. Lett. 1988, 1411. (e) Okinoshima, H.; Yamamota, K.; Kumada, M.
J. Organomet. Chem. 1971, 27, C31. (f) Ishikawa, M.; Sakamoto, H.;
Kanetani, F. Organometallics 1989, 8, 2767.
(3) (a) Seyferth, D.; Shannon, M. L.; Vick, S. C.; Lim, T. F. O.
Organometallics 1985, 4, 57. (b) Clarke, M. P.; Davidson, I. M. T. J.
Organomet. Chem. 1991, 408, 149. (c) Pannell, K. H.; Cervantes, J.;
Hernandez, C.; Cassias, J.; Vincenti, S. Organometallics 1986, 5, 1056.
(d) Pannell, K. H.; Rozell, J. M, Jr.; Hernandez, C. J. Am. Chem. Soc. 1989,
111, 4482. (e) Pannell, K. H.; Wang, L.-J.; Rozell, J. M. Organometallics
1989, 8, 550. (f) Tobita, H.; Ueno, K.; Ogino, H. Bull Chem. Soc. Jpn.
1988, 61, 2797. (g) Ueno, K.; Tobita, H.; Ogino, H. Chem. Lett. 1990,
369. (h) Haynes, A.; George, M. W.; Haward, M. T.; Poliakoff, M.; Turner,
J. J.; Boag, N. M.; Green, M. J. Am. Chem. Soc. 1991, 113, 2011. (i) Ueno,
K.; Hamasima, N.; Shimoi, M.; Ogino, H. Organometallics 1991, 10, 959.
(j) Pestana, D. C.; Koloski, T. S.; Berry, D. H. Organometallics 1994, 13,
4173. (k) Yamashita, H.; Tanaka, M.; Goto, M. Organometallics 1992, 11,
3227. (l) Tanaka, Y.; Yamashita, H.; Tanaka, M. Organometallics 1995,
14, 530. (m) Tamao, K.; Sun, G.-R.; Kawachi, A. J. Am. Chem. Soc. 1995,
117, 8043.
of the remarkably stable free silylene :SiN(tBu)CHdCHN(tBu)
to nickel.9
Reactivity studies for directly observed LnMdSiR2 complexes
are so far quite limited. The pronounced Lewis acid character
for such compounds is well documented,10,11 and we recently
described redistributions at silicon which occur by a bimolecular
mechanism mediated by silylene complexes.12 By analogy with
transition metal carbene complexes,13 cycloaddition reactions
(4) (a) Straus, D. A.; Tilley, T. D.; Rheingold, A. L.; Geib, S. J. J. Am.
Chem. Soc. 1987, 109, 5872. (b) Zybill, C.; Mu¨ller, G. Angew. Chem., Int.
Ed. Engl. 1987, 26, 669.
(5) Straus, D. A.; Grumbine, S. D.; Tilley, T. D. J. Am. Chem. Soc. 1990,
112, 7801.
(6) Gumbine, S. D.; Tilley, T. D.; Rheingold, A. L. J. Am. Chem. Soc.
1993, 115, 358.
(7) Grumbine, S. D.; Tilley, T. D.; Arnold, F. P.; Rheingold, A. L. J.
Am. Chem Soc. 1993, 115, 7884.
(8) Grumbine, S. K.; Tilley, T. D.; Arnold, F. P.; Rheingold, A. L. J.
Am. Chem. Soc. 1994, 116, 5495.
(9) Denk, M.; Hayashi, R. K.; West, R. J. Chem. Soc., Chem. Commun.
1994, 33.
(10) Nakatsuji, H.; Hada, M.; Kondo, K. Chem. Phys. Lett. 1992, 196,
404.
(11) Grumbine, S. K.; Straus, D. A.; Tilley, T. D.; Rheingold, A. L.
Polyhedron 1995, 14, 127.
(12) Grumbine, S. K.; Tilley, T. D. J. Am.Chem. Soc. 1994, 116, 6951.
S0002-7863(97)02020-9 CCC: $14.00 © 1997 American Chemical Society

Cycloaddition of Isocyanates to [Cp*(PMe3)2RuSiR2]+
J. Am. Chem. Soc., Vol. 119, No. 46, 1997 11237
of silylene complexes with unsaturated substrates might be
expected to provide important silicon-element bond-forming
processes. Here we report our observations on the interaction
of unsaturated compounds with silylene complexes of the type
[Cp*(PMe3)2RudSiR2]+ (R ) Me, Ph, S-p-Tol), and the first
directly observed cycloaddition reactions for silylene complexes.
Complexes 1a and 1b also did not react over 4 days at -78
°C with acetylene or 2-butyne in dichloromethane, as nearly
the same mixture of decomposition products was eventually
observed by 1H and 31P NMR spectroscopy. Also, no reaction
was observed between 1a and CO2 at 0 °C in dichloromethane
after 6 h. In contrast, azobenzene reacts with 1a over a 12-h
period to produce a dark orange oil, which could not be purified
by crystallization. However, an NMR analysis of the crude
reaction mixture revealed that it was composed of a major
product, formed in 50% yield. In the 31P NMR spectrum, this
product gave rise to two doublets at 3.11 and -3.54 ppm (JPP
) 46.8 Hz), and inequivalent 1H resonances for the SiMe2 group,
at 0.44 and 0.43 ppm, were also observed. Although it was
not possible to isolate and fully characterize this product, its
spectroscopic properties suggested that it might be a metalla-
cycle, formed via a 2 + 2 cycloaddition with the RudSi double
bond.
Results and Discussion
Transition metal silylene complexes are in general highly
reactive, due primarily to their very electron-deficient silicon
centers.8,11 These complexes may, however, be stabilized by
π-donation from a heteroatom substituent, or via coordination
of a Lewis base to silicon.5-9,11,12 The non-heteroatom-
stabilized silylene complexes that are known do not form stable
solutions at room temperature, making them somewhat difficult
to study.8 Fortunately, the triflate complexes Cp*(PMe3)2-
RuSiR2OTf may be used as convenient, stable precursors for
generating silylene complexes in situ. In this investigation,
dichloromethane solutions of the silylene complexes 1a and 1b
were generated quantitatively by reaction of the corresponding
triflate complexes with (Et2O)LiB(C6F5)4.8 The precipitated
LiOTf was then removed by filtration, giving a light yellow
(1a) or orange (1b) solution of the desired silylene complex.
Compounds 1a (t1/2 ) 7 h at 298 K) and 1b (t1/2 ) 3 h at 298
K) were typically generated at 0 °C immediately before use.
On the other hand, the heteroatom-stabilized complex [Cp*-
(PMe3)2RuSi(S-p-Tol)2]BPh4 (2) is stable enough to be stored
for long periods at room temperature.5
Reactions with Polar Unsaturated Substrates. Given the
lack of reactivity of 1a and 1b toward nonpolar substrates and
the polar nature of the RudSi double bond, we turned our
attention to reactions of silylene complexes with polar multiple
bonds. Methyl isocyanate reacted cleanly with the heteroatom-
stabilized silylene complex [Cp*(PMe3)2RudSi(STol)2][BPh4]
(2) to form one product (6a), isolated in 85% yield from
dichloromethane. In a similar manner, the analogous triflate
salt 6b was prepared from the corresponding triflate Cp*-
1
(PMe3)2RuSi(STol)2OTf.5,11 The H NMR spectra of 6a and
6b indicate the presence of inequivalent thiolate and phosphine
groups, and the 31P NMR spectra contain a set of doublets
corresponding to inequivalent phosphorus atoms. Although the
spectroscopic data did not allow an unambiguous assignment
of structures for 6a and 6b, they appeared to be consistent with
2 + 2 cycloaddition products. Subsequently, a single crystal
structure determination for 6b (vide infra) revealed the structure
depicted in eq 2. These reactions may therefore be described
Attempted Reactions with Nonpolar Unsaturated Sub-
strates. Initial reactivity studies with 1a and 1b focused on
nonpolar, unsaturated hydrocarbons. Under 1 atm of ethylene
at -78 °C, a solution of 1a in dichloromethane-d2 converted to
a complex mixture of products over 4 days (by NMR spectros-
copy). During the reaction, the SiMe2 1H NMR resonances of
1a were replaced by multiple peaks in the region from 0.0 to
0.2 ppm, which presumably represent species derived from the
silylene fragment. The 31P NMR spectrum contained three
+
primary resonances, assigned to the cations Cp*(PMe3)2RuD2
(3),14 Cp*(PMe3)2RuCl(D)+ (4),14,15 and Cp*(PMe3)3Ru+ (5),14
and other unidentified signals (eq 1). However, no hydride
as 1,2-dipolar additions of an Si-S bond to the isocyanate. The
products may be regarded as base-stabilized silylene complexes,
with a chelating thiocarbamate group bound to silicon. A similar
process has been observed for addition of an isocyanate to a
Fischer-type carbene complex,16a and related 1,2-dipolar addi-
tions of M-OR derivatives to isocyanates are also known.16b,c
This reaction probably occurs in a stepwise manner, via initial
coordination of the isocyanate nitrogen atom to silicon, followed
by migration of the thiolate group to the carbonyl carbon atom.
resonances were observed in the 1H NMR spectrum, indicating
that deuteride ligands were derived from the solvent. Since
these same decomposition products are observed when 1a is
allowed to decompose in dichloromethane solution, we conclude
that 1a does not react with ethylene.
(13) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.; Principles
and Applications of Organotransition Metal Chemistry; University Sci-
ence: Mill Valley, CA, 1987; Chapter 16.
(14) Grumbine, S. K. Ph.D. Thesis, University of California, San Diego,
1993.
(15) Tilley, T. D.; Grubbs, R. H.; Bercaw, J. E. Organometallics 1984,
3, 274.
(16) (a) Fischer, H.; Schlageter, A.; Bidell, W.; Fru¨h, A. Organometallics
1991, 10, 389. (b) Gibson, V. C.; Redshaw, C.; Clegg, W.; Elsegood, M.
R. J. J. Chem. Soc., Chem. Commun. 1994, 2635. (c) Braunstein, P.; Nobel,
D. Chem. ReV. 1989, 89, 1927.

11238 J. Am. Chem. Soc., Vol. 119, No. 46, 1997
Mitchell and Tilley
is consistent with carbamate character for the OC(NMe)S(Tol)
group,19 and strong contributions from resonance structures A
and B. This picture is also supported by the IR spectrum of
6a, which displays an absorption at 1625 cm-1 attributed to
the carbamate group.19 The Ru-Si distance of 2.249(1) Å is
quite short, inviting description of the compound as a base-
stabilized silylene complex. For comparison, the acetonitrile-
4a
stabilized complex [Cp*(PMe3)2RudSi(STol)2(NCMe)]BPh4
has a Ru-Si distance of 2.284(1) Å, and the base-free
dimethylsilylene complex 1a possesses a Ru-Si bond length
of 2.238(2) Å.8
The reactivity mode described above for 2 is not available
to silylene complexes 1a and 1b, which possess relatively strong
Si-C bonds. Compound 1a reacts rapidly and quantitatively
with methyl isocyanate (MeNdCdO) in dichloromethane to
form a new product (7) that possesses inequivalent phosphine
ligands (by 31P NMR spectroscopy) and silicon methyl groups
Figure 1. ORTEP drawing of the cation in 6b.
1
(by H NMR spectroscopy). The combustion analysis for 7
Table 1. Selected Bond Distances (Å) and Angles (deg) for
6b‚0.5C7H8
confirmed formation of a silylene-isocyanate adduct, and the
infrared spectrum of this complex contains an absorption at 1643
cm-1, which indicated the presence of a carbon-heteroatom
double bond. To obtain more structural information for 7, a
2D 1H/29Si correlation experiment was used to measure the Si-
(a) Bond Distances
Ru-P1
Ru-P2
Ru-Si
Si-N
2.299(1)
2.291(1)
2.249(1)
1.866(3)
Si-O
1.838(3)
2.158(2)
1.308(5)
1.305(5)
C1-S2
N-C2
S1-C4
S2-C3
1.734(4)
1.458(5)
1.790(4)
1.781(4)
Si-S1
C1-N
C1-O
2
N-C-H (3JHSi ) 2.9 Hz) and Si-C-H (average JHSi ) 6.0
3
(b) Bond Angles
93.71(4) Si-O-C1
Hz) coupling constants. The similar coupling constants JHSi
P1-Ru-P2
P1-Ru-Si
P2-Ru-Si
Ru-Si-N
Ru-Si-O
Ru-Si-S1
S1-Si-O
S1-Si-N
O-Si-N
90.6(2)
89.2(3)
2
20
) 3.3 Hz and JHSi ) 6.6 Hz reported for Me2Si(NMe2)2
90.88(4)
88.74(4)
128.0(1)
123.7(1)
118.59(6)
103.1(1)
102.3(1)
70.4(1)
Si-N-C1
Si-N-C2
C1- N-C2
S2- C1-O
S2-C1-N
O-C1-N
Si-S1-C4
C1-S2-C3
suggested the structure given in eq 3, which was later confirmed
145.5(3)
125.2(4)
125.6(3)
124.6(3)
109.7(3)
105.3(1)
100.7(2)
An attempt to observe intermediates in the formation of 6a by
monitoring the reaction at -60 °C by 31P NMR spectroscopy
was not successful.
by X-ray crystallography (vide infra). Reaction of 1a with
phenyl isocyanate (PhNdCdO) in dichloromethane at 0 °C
rapidly produced a similar product (8; eq 3). Inequivalent
phosphine and SiMe groups (by NMR spectroscopy) indicate
that 8 has a structure analogous to that for 7.
Interestingly, isocyanates also react with the complexes Cp*-
(PMe3)2RuSiR2OTf (R) Me, Ph) to displace triflate and produce
the corresponding cycloaddition products. Reactions with
methyl isocyanate occur instantly at room temperature to afford
the metallacycles 9 (R ) Ph) and 10 (R ) Me). We attribute
this reactivity to the equilibrium shown in eq 4, involving triflate
The structure of 6b is shown in Figure 1, and relevant bond
distances and angles are listed in Table 1. The geometry about
ruthenium may be described as a fairly regular three-legged
piano stool, with L-Ru-L angles (L ) P, Si) close to 90°.
The silicon is in a pseudotetrahedral environment, with Si-O
and Si-N bond lengths of 1.838(3) and 1.866(3) Å, respectively.
These distances are relatively long compared to typical single
bond distances of 1.63 (Si-O) and 1.74 Å (Si-N).17 On the
other hand, bonds to the three-coordinate carbon atom C(1) are
significantly shortened with respect to normal single bond
distances. The C(1)-O, C(1)-N, and C(1)-S(2) distances of
1.304(5), 1.308(5), and 1.734(4) Å are considerably shorter than
typical single bond lengths of 1.41, 1.47, and 1.81 Å, respec-
tively.18 The sum of the angles about C(1) (359.9°) reflects
planarity and sp2 hybridization. Internal angles in the Si-O-
C(1)-N ring vary greatly, ranging from an O-Si-N angle of
70.4(1)° to an O-C(1)-N(1) angle of 109.7(3)°, with the Si-
O-C(1) (90.6(2)°) and Si-N-C(1) (89.2(3)°) angles being
intermediate.
Considerable delocalization of electron density in the silicon-
based ligand is indicated both by the observed metrical
parameters and by the number of resonance structures that can
be drawn for 6b (Scheme 1). In particular π-donation from
oxygen, nitrogen, and sulfur, as evidenced by the short bond
distances, results in partial double bond character in the bonds
to C(1). The 13C NMR chemical shift of 177.72 ppm for C(1)
(19) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C., Spectroscopic
Identification of Organic Compounds, 4th ed.; Wiley: New York, 1981;
Chapter 5.
(20) Van Der Kelen, G. P.; Van Den Berghe, E. V. J. Organomet. Chem.
1976, 122, 329.
(17) Sheldrick, W. S. In The Chemistry of Organic Silicon Compounds;
Patai, S., Rappoport, Z., Eds.; Wiley: New York, 1989; Chapter 3, p 227.
(18) March, J. AdVanced Organic Chemistry, 3rd ed.; Wiley-Inter-
science: New York, 1985; p 19.

Cycloaddition of Isocyanates to [Cp*(PMe3)2RuSiR2]+
J. Am. Chem. Soc., Vol. 119, No. 46, 1997 11239
Scheme 1
Table 2. Selected Bond Distances (Å) and Angles (deg) for
9‚2CH2Cl2
The molecular structure of 9 was unequivocally established
by X-ray crystallography (Figure 2). Selected bond distances
and angles are listed in Table 2. The coordination geometry
for ruthenium may be described as a four-legged piano stool.
The Ru-P(1) and Ru-P(2) distances are 2.374(1) and 2.362-
(1) Å, respectively, while the P(1)-Ru-P(2) angle is 85.59-
(5)° and the C(1)-Ru-Si angle is only 62.1(1)°. The metal-
lacycle ring is nearly planar, with the largest deviation from
the least-squares plane being 0.068(5) Å (for C(1)), but it is
also highly asymmetric, with bond lengths of 1.351(6) (C(1)-
N), 1.752(4) (Si-N), 2.171(5) (Ru-C(1)), and 2.483(1) Å (Ru-
Si). The C(1)-N distance of 1.351(6) Å is greater than the
NdC distance of 1.168(5) Å in free methyl isocyanate,22 but
still significantly less than the N-C(2) single bond distance of
1.460(6) Å.
(a) Bond Distances
Ru-P1
Ru-P2
Ru-Si
2.374(1)
2.362(1)
2.483(1)
Ru-C1
2.171(5)
1.752(4)
1.894(5)
1.901(5)
O-C1
N-C1
N-C2
1.216(5)
1.351(6)
1.460(6)
Si-N
Si-C3
Si-C4
(b) Bond Angles
85.59(5) Ru-Si-N
P1-Ru-P2
P1-Ru-Si
P1-Ru-C1
P2-Ru-Si
P2-Ru-C1
Si-Ru-C1
Ru-C1-O
Ru-C1-N
O-C1-N
85.6(1)
120.5(2)
129.6(2)
109.1(2)
104.1(2)
102.9(2)
101.4(3)
136.9(3)
121.0(4)
81.22(5)
129.3(1)
110.86(5)
77.1(1)
Ru-Si-C3
Ru-Si-C4
N- Si-C3
N- Si-C4
C3-Si-C4
Si-N-C1
Si-N-C2
C1-N-C2
62.1(1)
126.4(4)
110.3(3)
123.3(4)
Compounds 7-10 are relatively stable and show no signs of
decomposition at room temperature after 3 days in dichlo-
romethane-d2. However, we were intrigued by the possibility
that these metallacycles might undergo ring opening to give
a silaimine (e.g., Me2SidNMe) and the carbonyl complex
Cp*(PMe3)2RuCO+. It seemed that a potentially strong driving
force for this reaction might be extrusion of the strongly
π-accepting carbonyl ligand from the metallacycle, which would
greatly stabilize the electron-rich Cp*(PMe3)2Ru+ fragment.
Of course, the silaimine would be a high-energy species
which would dimerize to more stable structures containing
tetrahedral silicon centers.23 To investigate this possibility, we
examined the pyrolytic decomposition of complex 9. Heating
9 in toluene-d8 at 100 °C for 1 day produced predominately
one product, which was identified by 1H and 31P NMR
spectroscopy as the triflate Cp*(PMe3)2RuSiPh2OTf. Similar
results were obtained with compounds 7 and 8, although the
starting silylene complexes were not recovered cleanly (presum-
ably because of their thermal instability). Additionally, a
solid sample of 9 heated to 150 °C under vacuum produced
methyl isocyanate, trapped at -196 °C and identified by GC/
MS, as the only volatile thermal decomposition product. The
cycloaddition reactions are therefore reversible, as indicated in
eq 5.
Figure 2. ORTEP drawing of the cation in 9.
dissociation to produce a transient ruthenium silylene complex,
which is then rapidly trapped by isocyanate.11 It is difficult at
this time to entirely rule out an SN2 substitution mechanism.
Attempts to determine the effect of added [nBu4N][OTf] on the
rate for the reaction of Cp*(PMe3)2RuSiPh2OTf with PhNCO
were unsuccessful due to the extremely rapid reaction at low
temperature.
This chemistry suggests that these metallacyles might be
characterized by weak Ru-C and Si-N bonds. However,
analysis of the bond distances in 9 does not support this
view. For comparison, the Ru-C(1) distance of 2.171(5) Å
in 9 is slightly longer than the corresponding Ru-C bond
(2.084(7) Å) in the complex CpRu(CO)2CONH2.24 In addition,
the CdO and C-N bond lengths of 1.216(5) and 1.351(6)
Å in 9 are close to the corresponding values of 1.244(8) and
The cycloaddition products 7-10 exhibit fluxional behavior
(by variable-temperature 1H NMR spectroscopy). At room
temperature, one of the PMe3 groups of 7 appears as a sharp
doublet at 1.54 ppm (2JHP ) 9.3 Hz), while the other phosphine
ligand gives rise to a broad singlet at 1.65 ppm. Heating a
sample of 7 in toluene-d8 to 50 °C resulted in sharpening of
the latter resonance to a well-defined doublet, and cooling to
-30 °C resolved this peak into three distinct doublets for each
methyl group of the PMe3 ligand. We attribute this behavior
to hindered rotation about one of the Ru-P bonds. The
coalescence temperatures are insensitive to the nature of the
group bound to nitrogen (Me or Ph). However as expected,
substituents at silicon play a strong role in determining the
coalescence temperature. The difference in coalescence tem-
peratures for 7 and 9, -20 and 40 °C, reflects a greater activation
barrier for rotation in 9 (by ca. 2 kcal mol-1).21
(21) This estimate is based on a two-site exchange formula: Sandstro¨m,
J.; Dynamic NMR Spectroscopy, Academic Press: London, 1982; Chapter
7.
(22) Anderson, D. W. W.; Rankin, D. W. H.; Robertson, A. J. Mol. Struct.
1972, 14, 385.
(23) (a) Parker, D. R.; Sommer, L. H. J. Organomet. Chem. 1976, 110,
C1. (b) Zigler, S.; Johnson, L. M.; West, R. J. Organomet. Chem. 1988,
341, 187. (c) Wiberg, N.; Schurz, K. J. Organomet. Chem. 1988, 341, 145.
(d) Tamao, K.; Nakagawa, Y.; Ito, Y. J. Am. Chem. Soc. 1992, 114, 218.
(24) Wagner, H.; Jungbauer, A.; Thiele, G.; Behrens, H. Z. Naturforsch.,
Teil B 1979, 34, 1487.

11240 J. Am. Chem. Soc., Vol. 119, No. 46, 1997
Mitchell and Tilley
Scheme 2
1.336(10) Å observed for CpRu(CO)2CONH2. Also, the
Si-N distance of 1.752(4) Å in 9 is not unusual, as seen by
comparison to the average Si-N distance of 1.73 Å in [Me2-
SiNMe]4.25
A possible mechanism for these cycloaddition reactions is
shown in Scheme 2. Given the strong Lewis acidic nature of
the silylene complexes under consideration,11 we favor initial
coordination of the isocyanate to silicon via the nitrogen lone
pair. This produces an intermediate that may be considered a
base-stabilized silylene complex, or a Ru(II) silyl complex with
the positive charge localized principally on nitrogen. In the
final step of this mechanism, the electron rich Ru(II) center adds
to the carbonyl carbon atom to form the metallacycle. This
reaction could therefore be described as a stepwise cycloaddition
involving polar intermediates.26 So far, we have not been able
to detect an intermediate (by 31P NMR spectroscopy) in these
extremely rapid cycloadditions.
Figure 3. Hammett plot. Open points represent rates relative to phenyl
isocyanate. Solid points represent rates relative to p-chlorophenyl
isocyanate, adjusted to phenyl isocyanate. Averaged results are as
follows: p-MeOPhNCO/PhNCO (1.86:1), p-ClPhNCO/PhNCO (0.30:
1), p-CF3PhNCO/PhNCO (0.11:1), p-MeOPhNCO/p-ClPhCNO (7.33:
1), p-CF3PhNCO/p-ClPhNCO (0.32:1). Slope ) -1.7, R2 ) 0.98.
For cycloadditions involving rate-determining coordination
to silicon, the rates should scale with the nucleophilicity of the
incoming isocyanate. To investigate this possibility, a series
of competition reactions between pairs of para-substituted
phenyl isocyanates and the diphenylsilylene complex 1b were
examined. Rates for p-methoxyphenyl isocyanate, p-chlorophe-
nyl isocyanate, and p-(trifluoromethyl)phenyl isocyanate were
determined relative to the rate for phenyl isocyanate. Addition-
ally, rates for p-methoxyphenyl isocyanate and p-(trifluorom-
ethyl)phenyl isocyanate were determined relative to that for
p-chlorophenyl isocyanate. For each reaction, 10 equiv of each
isocyanate were combined and then added to a solution of the
silylene complex in 0.5 mL of dichloromethane-d2. The relative
phenyl isocyanate was added to Cp*(PMe3)2RuSiPh2OTf, no
reaction occurred after 30 min in dichloromethane-d2, or after
heating to 80 °C for 1 day in toluene-d8. However, when the
triflate was precipitated from these reaction solutions by addi-
tion of (Et2O)LiB(C6F5)4, the solution immediately turned
from very light yellow to the bright orange color of 1b, then
back to pale yellow within a span of a few seconds. NMR
spectroscopy confirmed clean conversion to the corresponding
cycloaddition products. As expected, [Cp*(PMe3)2RuSiPh2N-
(p-ClC6H4)CdO]B(C6F5)4 reacts with [nBu4N][OTf] in toluene-
d8 at 80 °C to regenerate the silyl triflate complex Cp*(PMe3)2-
RuSiPh2OTf and the isocyanate. Presumably, the electron
withdrawing effect of the p-chloro and p-CF3 substituents
reduces the nucleophilicity of the isocyanate nitrogen such that
these isocyanates do not compete effectively with triflate for
binding to the silicon center. As a consequence, all competition
studies were performed with the free silylene complex 1b to
prevent interference from the triflate counterion.
The averaged results from the competition reactions are
compiled in the Hammett plot of Figure 3. The linearity of
this plot indicates that the same mechanism is involved in each
reaction, and the negative F value of -1.7 describes the
accelerating effect of electron-donating groups in the para
position. These data are therefore consistent with the mecha-
nism of Scheme 2 involving initial nucleophilic attack of the
isocyanate at the silylene silicon atom.
rates of reaction were then determined by integration of the 31
P
NMR spectra, with each competition being examined twice. The
product ratios were measured within 0.5 h of initiation of
reaction, and are assumed to represent kinetically determined
product distributions. The validity of this assumption was
supported by the competition reaction involving p-methoxy-
phenyl and phenyl isocyanate. This competition resulted
in a product ratio of 1.86:1 in favor of the p-methoxyphenyl
isocyanate adduct. After standing at room temperature for
1 day the ratio of products did not change, but heating the
sample to 60 °C for 8 h resulted in an altered ratio of 1.5:1,
still in favor of the p-methoxyphenyl isocyanate adduct.
Continued heating resulted in no further change in the product
ratio, indicating that a thermodynamic distribution had been
reached.
Unlike methyl isocyanate, isocyanates with electron-with-
drawing substituents do not react with the silyl triflate com-
plexes. When p-chlorophenyl isocyanate or p-(trifluoromethyl)-
Conclusions
Given the very limited availability of transition metal silylene
complexes, little is known concerning their reactivity. In this
contribution, we describe the first direct observation of cy-
cloaddition reactions involving silylene complexes. These
reactions occur in a stepwise manner via a mechanism that
(25) Edwards, B. P. E.; Harrison, W.; Nowell, I. W.; Post, M. L.; Shearer,
H. M. M.; Trotter, J. Acta Crystallogr., Sect. B 1976, 32, 648.
(26) (a) Lim, D.; Jorgensen, W. L. J. Phys. Chem. 1996, 100, 17490.
(b) Huisgen R. Pure Appl. Chem. 1980, 52, 2283. (c) Huisgen R. Acc. Chem.
Res. 1977, 10, 117. (d) Huisgen R. Acc. Chem. Res. 1977, 10, 199.
(27) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1964, 2, 245.

Cycloaddition of Isocyanates to [Cp*(PMe3)2RuSiR2]+
J. Am. Chem. Soc., Vol. 119, No. 46, 1997 11241
STol), 131.05 (s, ring STol), 134.95 (s, ring STol), 135.73 (s, ring STol),
involves nucleophilic attack onto the electrophilic silicon center.
The ruthenium silylene complexes discussed here are unreactive
toward nonpolar multiple bonds which do not bind initially to
the silylene ligand. This lack of reactivity is probably also
associated with the inert nature of the metal’s coordination
sphere, which does not allow the binding of alkene or alkyne
substrates. Future investigations on cycloaddition reactions of
silylene complexes will focus on 16-electron platinum centers,7
which should more readily bind (and activate) unsaturated
substrates.
138.65 (s, ring STol), 142.82 (s, ring STol), 177.72 (s, CSTol). 31P-
2
2
{1H} NMR (162.0 MHz) δ 2.52 (d, JPP ) 35 Hz), 4.24 (d, JPP ) 35
Hz). IR: 1754 m, 1625 s, 1260 m, 1144 s, 926 w, 793 m, 750 m, 726
m, 654 w.
[Cp*(PMe3)2RuSiMe2NMeCdO][B(C6F5)4] (7). A 50-mL round-
bottom Schlenk flask was charged with Cp*(PMe3)2RuSiMe2OTf (0.150
g, 0.252 mmol) and (Et2O)LiB(C6F5)4 (0.192 g, 0.252 mmol). Dichlo-
romethane (5 mL) was added after the flask was cooled to 0 °C, and
the mixture was stirred until all reactants had dissolved. A cloudy
yellow solution formed within 30 s, indicating the formation of 1a.
The solution was then filtered into a precooled (0 °C) flask and MeNCO
(21 µL, 0.356 mmol) was added via syringe. The reaction mixture
was stirred for 5 min, after which the volatile material was removed
under reduced pressure. The crude product was crystallized from 1:1
dichloromethane/diethyl ether (10 mL) at -40 °C. Yield 85% (0.298
g). Anal. Calcd for C44H42F20P2BNORuSi: C, 44.68; H, 3.58; N, 1.18.
Found: C, 44.39; H, 3.57; N, 1.33. Mp 150-155 °C dec. 1H NMR:
Experimental Section
All manipulations were performed under an inert atmosphere with
standard Schlenk techniques. Dry, oxygen-free solvents were employed
throughout. Diethyl ether, pentane, and benzene-d6 were distilled from
sodium/benzophenone and stored under nitrogen. Dichloromethane was
distilled from CaH2 and degassed with two freeze-pump-thaw cycles
prior to use. The compounds Cp*(PMe3)2RuSiMe2OTf,11 Cp*(PMe3)2-
RuSiPh2OTf,4a (Et2O)LiB(C6F5)4,27 and [Cp*(PMe3)2RudSi(STol)2]-
[BPh4]5 were prepared according to known procedures. All isocyanates
were purchased from Aldrich. Methyl isocyanate, p-methoxyphenyl
isocyanate, and p-(trifluoromethyl)phenyl isocyanate were used as
received. Phenyl isocyanate was distilled prior to use, and p-
chlorophenyl isocyanate was recrystallized prior to use. Elemental
analyses were performed by Desert Microanalytical Laboratories or in
the Microanalytical Laboratory in the College of Chemistry at the
University of California, Berkeley. All IR samples were prepared as
2
δ 0.55 (s, 3 H, SiMe2), 0.59 (s, 3 H, SiMe2), 1.54 (d, JHP ) 9.3 Hz,
9 H, PMe3), 1.65 (br s, 9 H, PMe3), 1.77 (s, 15 H, Cp*), 2.74 (s, 3 H,
NMe). 13C{1H} NMR (100 MHz) δ 5.26 (s, SiMe2), 10.06 (s, SiMe2),
1
11.21 (s, Cp*), 19.70 (d, JCP ) 40 Hz, PMe3), 23.40 (br s, PMe3),
29.78 (s, NMe), 104.35 (s, ring Cp*), 113.10 (m, C6F5), 137.7 (dm,
1JCF ) 258 Hz, C6F5), 145.15 (dm, 1JCF ) 243 Hz, C6F5), 148.44 (dm,
2
2
1JCF ) 249 Hz, C6F5), 163.9 (dd, JCPcis ) 5.2 Hz, JCPtrans ) 49 Hz).
31P{1H} NMR (121.5 MHz) δ -4.75 (d, JPP ) 55 Hz), 4.94 (d, JPP
2
2
) 55 Hz). 29Si{1H} NMR (59.6 MHz) δ 14.16 (dd, JSiPcis ) 3.0 Hz,
2
2JSiPtrans ) 33 Hz). IR: 1643 m, 1616 s, 1512 s, 1294 s, 1261 m, 1216
w, 1087 s, 978 s, 831 m, 773 w, 758 m, 683 w, 663 m, 575 w.
Nujol mulls on CsI plates, and all absorptions are reported in cm-1
.
All NMR spectra were recorded at room temperature in dichlo-
romethane-d2, unless otherwise noted. All 1H NMR spectra were
obtained at 300 MHz, unless otherwise stated. The compounds Cp*-
(PMe3)2RuD2+ (3), Cp*(PMe3)2RuCl(D)+ (4), and Cp*(PMe3)3Ru+ (5)
were identified by their 31P{1H} NMR chemical shifts of δ 10.10, 8.26,
and 1.98, respectively.14
[Cp*(PMe3)2RuSiMe2NPhCdO][B(C6F5)4] (8). The procedure
employed was identical with that used to synthesize 7, with the
following quantities: Cp*(PMe3)2RuSiMe2OTf (0.140 g, 0.235 mmol),
(Et2O)LiB(C6F5)4 (0.179 g, 0.235 mmol) and PhNCO (45 µL, 0.435
mmol). Yield 70% (0.205 g). Anal. Calcd for C49H44F20P2-
BNORuSi: C, 47.28; H, 3.56; N, 1.13. Found: C, 45.28; H, 3.82; N,
1.05. Mp 147-150 °C. 1H NMR: δ 0.65 (s, 3 H, SiMe2), 0.82 (s, 3
H, SiMe2), 1.63 (d, 2JHP ) 9.6 Hz, 9 H, PMe3), 1.72 (br s, 9 H, PMe3),
1.83 (s, 15 H, Cp*), 7.25-7.33 (m, 5 H, NC6H5). 13C{1H} NMR (100
MHz) δ 6.60 (s, SiMe2), 10.50 (s, SiMe2), 11.20 (s, Cp*), 19.60 (d,
1JCP ) 40 Hz, PMe3), 22.50 (br s, PMe3), 104.8 (s, ring Cp*), 123.2 (s,
{Cp*(PMe3)2RuSi(STol)[η2-O(MeN)C(STol)]}[BPh4] (6a). To
a solution of [Cp*(PMe3)2RudSi(STol)2][BPh4] (0.200 g, 0.201 mmol)
in 15 mL of dichloromethane was added MeNCO (16 µL, 0.27 mmol)
via syringe. After the solution was stirred for 5 min, the volatile
materials were removed under reduced pressure and the resulting
yellow residue was crystallized from 15 mL of 1:1 dichloromethane/
diethyl ether at -40 °C. Yield 87% (0.181 g). Anal. Calcd for
C56H66S2P2BNORuSi: C, 64.98; H, 6.43. Found: C, 64.72; H, 6.13.
Mp 162-165 °C. 1H NMR (400 MHz) δ 1.31 (d, 2JHP ) 8.4 Hz, 9 H,
2
NC6H5), 125.7 (s, NC6H5), 129.4 (s, NC6H5), 164.7 (dd, JCPcis ) 5.1
2
Hz, JCPtrans ) 50 Hz). The ipso carbon was not observed. 31P{1H}
NMR (121.5 MHz) δ -5.29 (d, 2JPP ) 56 Hz), 4.03 (d, 2JPP ) 56 Hz).
IR: 1643 m, 1622 s, 1269 s, 1086 m, 978 s, 739 m, 683 w, 660 m.
2
PMe3), 1.40 (d, JHP ) 8.8 Hz, 9 H, PMe3), 1.89 (s, 15 H, Cp*), 2.41
[Cp*(PMe3)2RuSiPh2NMeCdO][OTf] (9). MeNCO (24 mL, 0.400
mmol) was added via syringe to a solution of Cp*(PMe3)2RuSiPh2OTf
(0.220 g, 0.306 mmol) in dichloromethane (5 mL). After the reaction
mixture was stirred for approximately 5 min, the volatiles were removed
in Vacuo to leave a light green residue that was then crystallized from
1:1 dichloromethane/diethyl ether (20 mL) at -40 °C. Yield 82%
(0.195 g). Anal. Calcd for C31H46O4F3P2NSSiRu: C, 47.93; H, 5.97;
(s, 3 H, STol), 2.43 (s, 3 H, STol), 2.80 (s, 3 H, NMe), 6.88 (t, 3JHH
)
3
14.4 Hz, 4 H, BPh4), 7.03 (t, JHH ) 14.8 Hz, 4 H, BPh4), 7.14-7.28
(m, 16 H, BPh4), 7.30-7.34 (m, 8 H, STol). 13C{1H} NMR (100 MHz)
δ 11.97 (s, Cp*), 21.38 (s, STol), 21.58 (s, STol), 23.20 (d, 1JCP ) 24
1
Hz, PMe3), 23.50 (d, JCP ) 23 Hz, PMe3), 32.48 (s, NMe), 96.20 (s,
ring Cp*), 117.72 (s, BPh4), 125.75 (s, BPh4), 127.21 (s, BPh4), 128.67
(s, ring STol), 129.48 (s, ring STol), 130.52 (s, ring STol), 131.05 (s,
ring STol), 134.95 (s, ring STol), 135.73 (s, ring STol), 138.46 (s, BPh4),
138.65 (s, ring STol), 142.82 (s, ring STol), 177.72 (s,CSTol). 31P-
N, 1.80. Found: C, 46.78; H, 6.19; N, 1.70. Mp 120-125 °C. 1H
2
NMR: δ 0.81 (br s, 3 H, PMe3), 1.57 (s, 15 H, Cp*), 1.64 (d, JHP
)
9.6 Hz, 12 H, PMe3), 1.97 (br s, 3 H, PMe3), 3.07 (s, 3 H, NMe),
7.38-7.41, 7.49-7.52, 7.70-7.73 (m, 10 H, C6H5). 13C{1H} NMR
(100 MHz) δ 10.74 (s, Cp*), 20.31 (d, 1JCP ) 31 Hz, PMe3), 32.38 (s,
NMe), 104.3 (s, ring Cp*), 129.0, 129.3, 130.2, 130.6, 134.4, 135.7,
2
2
{1H} NMR (162.0 MHz) δ 2.52 (d, JPP ) 35 Hz), 4.24 (d, JPP ) 35
Hz). IR: 1754 m, 1625 s, 1312 m, 1075 s, 992 w, 793 m, 750 m, 726
m, 654 w.
{Cp*(PMe3)2RuSi(STol)[η2-O(MeN)C(STol)]}[OTf] (6b). To a
solution of Cp*(PMe3)2RuSi(STol)2OTf (0.300 g, 0.369 mmol) in 10
mL of dichloromethane was added MeNCO (40 µL, 0.668 mmol) via
syringe. After the solution was stirred for 15 min, the volatile materials
were removed under reduced pressure and the resulting green-yellow
residue was crystallized from 20 mL of toluene at -80 °C. Yield 90%
(0.289 g). Anal. Calcd for C33H50RuO4F3S3P2NSi: C, 45.61; H, 5.80.
Found: C, 45.42; H, 5.94. mp 158-160 °C. 1H NMR (400 MHz) δ
1.31 (d, 2JHP ) 8.4 Hz, 9 H, PMe3), 1.40 (d, 2JHP ) 8.8 Hz, 9 H, PMe3),
1.89 (s, 15 H, Cp*), 2.41 (s, 3 H, STol), 2.43 (s, 3 H, STol), 2.80 (s,
3 H, NMe), 7.30-7.34 (m, 8 H, STol). 13C{1H} NMR (100 MHz) δ
2
2
136.7, 139.1 (all s, C6H5), 162.1 (dd, JCPcis ) 5.2 Hz, JCPtrans ) 52
Hz). 31P{1H} NMR (161.98 MHz) δ -2.04 (d, JPP ) 57 Hz), 4.80
2
(d, 2JPP ) 57 Hz). IR: 1614 s, 1269 s, 1221 m, 1142 m, 1030 m, 943
w, 825 m, 752 m, 714 w, 685 w, 634 m, 571 m, 507 w.
[Cp*(PMe3)2RuSiMe2NMeCdO][OTf] (10). The procedure em-
ployed was identical with that used for 9, with Cp*(PMe3)2RuSiMe2-
OTf (0.300 g, 0.504 mmol) and MeNCO (40 µL, 0.667 mmol). Yield
90% (0.296 g). Anal. Calcd for C21H42O4F3P2NSRuSi: C, 38.64; H,
6.49; N, 2.15. Found: C, 38.69; H, 6.81; N, 2.34. Mp 138-140 °C.
1H and 31P{1H} NMR spectra are identical with those for 7. The 13C-
{1H} spectrum is identical with that for 7, except for the absence of
B(C6F5)4- resonances. IR: 1614 m, 1260 s, 1144 m, 1030 m, 964 m,
944 w, 825 m, 636 s, 517 m.
1
11.97 (s, Cp*), 21.38 (s, STol), 21.58 (s, STol), 23.20 (d, JCP ) 24
1
Hz, PMe3), 23.50 (d, JCP ) 23 Hz, PMe3), 32.48 (s, NMe), 96.20 (s,
ring Cp*), 128.67 (s, ring STol), 129.48 (s, ring STol), 130.52 (s, ring

11242 J. Am. Chem. Soc., Vol. 119, No. 46, 1997
Mitchell and Tilley
Table 3. Crystal and Data Collection Parameters for 6b‚0.5C7H8 and 9‚2CH2Cl2
6b‚0.5C7H8
9‚2CH2Cl2
(a) Crystal Parameters
formula
RuO4F3S3P2NSiC36.5H54
915.11
RuSP2SiF3O4NC33Cl4H50
946.73
colorless, columnar
0.15 × 0.30 × 0.30
orthorhombic
Pbca (No. 61)
14.5771(1)
formula weight
crystal color, habit
crystal size, mm
crystal system
space group
a, Å
colorless block
0.20 × 0.20 × 0.30
monoclinic
P21/c (No. 14)
12.1313(3)
18.7362(3)
19.2248(4)
100.599(1)
4295.1(1)
4
b, Å
c, Å
18.4240(1)
31.1253(3)
â, deg
V, Å3
8359.27(9)
8
1.504
8.35
Z
D(calc), g cm-3
µ(Mo KR), cm-1
temp, K
1.415
6.59
151
158
(b) Data Collection
Siemens SMART
Mo KR (λ ) 0.71069 Å)
ω (0.30° per frame)
10.0
18051
6614 (Rint ) 0.032)
4992 (I > 3.00σ(I))
diffractometer
radiation
scan type
scan rate, s per frame
total no of reflcns collected
no of unique reflcns
no. of observations
Siemens SMART
Mo KR (λ ) 0.71069 Å)
ω (0.30° per frame)
30.0
33799
6646 (Rint ) 0.032)
4843 (I > 3.00σ(I))
(c) Refinement
10.76
3.6
reflection/parameter ratio
R(F), %
10.64
4.5
R(wF), %
4.7
5.9
goodness of fit indicator
max peak in final diff map, e-/Å3
min peak in final diff map, e-/Å3
1.80
0.92
-0.73
2.84
0.90
-0.81
Competition Experiments. Adducts were generated in solution via
reaction of silylene 1b, generated from Cp*(PMe3)2RuSiR2OTf (0.015
g, 0.021 mmol) and (Et2O)LiB(C6F5)4 (0.016 g, 0.021 mmol), with the
appropriate isocyanate. Addition of p-ClC6H4NCO (0.005 g, 0.030
mmol) to a solution of 1b in 0.5 mL of dichloromethane-d2 yielded
two experiments were averaged. An identical procedure was employed
for the other experiments, but with the following quantities: p-
methoxyphenyl isocyanate (0.057 g, 0.38 mmol)/phenyl isocyanate
(0.045 g, 0.37 mmol), p-(trifluoromethyl)phenyl isocyanate (0.072g,
0.38 mmol)/phenyl isocyanate (0.048 g, 0.40 mmol), p-methoxyphenyl
isocyanate (0.057 g, 0.38 mmol)/p-chlorophenyl isocyanate (0.061 g,
0.39 mmol), p-(trifluoromethyl)phenyl isocyanate (0.074 g, 0.39 mmol)/
p-chlorophenyl isocyanate (0.059 g, 0.38 mmol).
the adduct Cp*(PMe3)2RuSiPh2N(p-ClC6H4)CdO][B(C6F5)4] exclu-
1
sively (by H and 31P NMR spectroscopy): 1H NMR (400 MHz) δ
2
0.71 (br s, 3 H, PMe3), 1.59 (s, 15 H, Cp*), 1.64 (d, JHP ) 9.6 Hz,
X-ray Crystal Structure Determinations. Crystals were mounted
on a glass fiber with Paratone N hydrocarbon oil. All X-ray data were
collected on a Siemens SMART diffractometer with a CCD area
detector. Preliminary orientation matrices and unit cell parameters were
determined by collecting 60 10-s frames. Frame data were integrated
by using SAINT and corrected for Lorentz and polarization effects.
12 H, PMe3), 1.94 (br s, 3 H, PMe3), 7.00-7.07 (m, 5 H, SiC6H5),
7.25-7.31, (m, 5 H, SiC6H5), 7.42 (m, 2 H, p-ClPh), 7.61 (m, 2 H,
2
p-ClPh). 31P{1H} NMR (162.0 MHz) δ -3.82 (d, JPP ) 59 Hz),
2
1.54 (d, JPP ) 59 Hz). Other adducts were generated analogously.
Cp*(PMe3)2RuSiPh2N(p-MeOC6H4)CdO][B(C6F5)4]: 1H NMR (400
MHz) δ 0.70 (br s, 3 H, PMe3), 1.62 (s, 15 H, Cp*), 1.71 (d,
2JHP ) 9.6 Hz, 12 H, PMe3), 1.92 (br s, 3 H, PMe3), 7.12-7.16
(m, 5 H, SiC6H5), 7.22-7.28 (m, 5 H, SiC6H5), 7.40 (m, 2 H,
p-MeOPh), 7.58 (m, 2 H, p-MeOPh). 31P{1H} NMR (162.0 MHz)
(a) 6b‚0.5C7H8: A colorless blocklike crystal with approximate
dimensions of 0.20 × 0.20 × 0.30 mm was mounted and placed under
a cold stream of nitrogen on the diffractometer. A hemisphere of data
was collected at a temperature of -121 ( 1 °C with ω scans of 0.30°
and a collection time of 10 s per frame. An absorption correction was
applied by using XPREP (µR ) 0.05, Tmax ) 0.90, Tmin ) 0.82). The
18051 reflections which were integrated were averaged in point group
2/m to yield 6614 unique reflections (Rint ) 0.032). No correction for
decay was necessary. The structure was solved by using direct methods
(SIR92) and refined by full matrix least-squares methods with use of
teXsan software. The non-hydrogen atoms were refined anisotropically,
except for the disordered toluene methyl carbon, which was modeled
at half occupancy and refined isotropically. Hydrogen atoms were
included at calculated positions but not refined. The number of variable
parameters was 464, giving a data/parameter ratio of 10.76. The
maximum and minimum peaks on the final difference Fourier map
corresponded to 0.92 and -0.73 e-/Å3: R ) 0.036, Rw ) 0.047, GOF
) 1.80.
2
2
δ -3.54 (d, JPP ) 59 Hz), 1.68 (d, JPP ) 59 Hz). Cp*-
(PMe3)2RuSiPh2N(p-CF3C6H4)CdO][B(C6F5)4]: 1H NMR (400 MHz)
δ 0.75 (br s, 3 H, PMe3), 1.63 (s, 15 H, Cp*), 1.68 (d, 2JHP ) 9.6 Hz,
12 H, PMe3), 1.98 (br s, 3 H, PMe3), 6.98-7.02 (m, 5 H, SiC6H5),
7.15-7.21 (m, 5 H, SiC6H5), 7.45 (m, 2 H, p-CF3Ph), 7.58 (m, 2 H,
2
p-CF3Ph). 31P{1H} NMR (162.0 MHz) δ -3.84 (d, JPP ) 58 Hz),
2
1.77 (d, JPP ) 58 Hz). Cp*(PMe3)2RuSiPh2N(Ph)CdO][B(C6F5)4]:
1H NMR (400 MHz) δ 0.82 (br s, 3 H, PMe3), 1.65 (s, 15 H, Cp*),
2
1.67 (d, JHP ) 9.6 Hz, 12 H, PMe3), 2.01 (br s, 3 H, PMe3), 7.05-
7.12 (m, 5 H, SiC6H5), 7.19-7.28 (m, 5 H, SiC6H5), 7.49 (m, 2 H,
NPh), 7.63 (m, 2 H, NPh). 31P{1H} NMR (162.0 MHz) δ -3.66 (d,
2
2JPP ) 59 Hz), 1.58 (d, JPP ) 59 Hz).
In a glovebox, a well-mixed solid mixture of Cp*(PMe3)2RuSiPh2-
OTf (0.027 g, 0.037 mmol) and (Et2O)LiB(C6F5)4 (0.028 g, 0.037 mmol)
was dissolved in 0.5 mL of dichloromethane-d2 to generate the silylene
complex 1b. The orange solution was then added to a mixture of
p-chlorophenyl isocyanate (0.060 g, 0.39 mmol) and phenyl isocyanate
(0.047 g, 0.39 mmol), resulting in immediate formation of a light-yellow
solution. The ratio of cycloaddition products for each sample was then
determined by integration of the 31P NMR spectra, and the results of
(b) 9‚2CH2Cl2: A colorless columnar crystal with approximate
dimensions of 0.15 × 0.30 × 0.30 mm was mounted and placed under
a cold stream of nitrogen on the diffractometer. A hemisphere of data
was collected at a temperature of -115 ( 1 °C with ω scans of 0.30°
and a collection time of 30 s per frame. Analysis of the redundant
data with the program XPREP indicated that there was no significant

Cycloaddition of Isocyanates to [Cp*(PMe3)2RuSiR2]+
J. Am. Chem. Soc., Vol. 119, No. 46, 1997 11243
absorption by the crystal and no correction was applied. The 33799
reflections which were integrated were averaged in point group mmm
to yield 6646 unique reflections (Rint ) 0.032). No correction for decay
was necessary. The structure was solved using direct methods
(SAPI91), expanded with Fourier techniques (DIRDIF92), and refined
by full matrix least-squares methods with use of teXsan software. The
non-hydrogen atoms were refined anisotropically. Hydrogen atoms
were included at calculated positions but not refined. The number of
variable parameters was 455, giving a data/parameter ratio of 10.64.
The maximum and minimum peaks on the final difference Fourier map
corresponded to 0.90 and -0.81 e-/Å3: R ) 0.045, Rw ) 0.059, GOF
) 2.84.
Acknowledgment is made to the National Science Founda-
tion for their generous support of this work. We also thank
.
Johnson-Matthey Aesar, Inc. for a loan of RuCl3 xH2O.
Supporting Information Available: Tables of crystal, data
collection, and refinement parameters, bond distances and
angles, and anisotropic displacement parameters (27 pages). See
any current masthead page for ordering and Internet access
instructions.
JA972020F