Cycloaddition and nucleophilic substitution reactions of the monomeric titanocene sulfido complex (η5-C5Me5)2(C5H5N)Ti=S

Article abstract of DOI:10.1021/ja980877m

The titanocene sulfido complex Cp*₂(py)Ti=S (1, py = pyridine) reacts reversibly with terminal alkynes to give the thiametallacyclobutenes Cp*₂Ti(SC(R)=CH) (R = H (2), Ph (3), Tol (4), TMS (5)). Complex 1 also reacts with alkyl halides to give products derived from formal 1,2-addition across the titanium-sulfur bond. Treatment of 1 with allyl halides results in the formation of Cp*₂Ti(X)SCH₂CHCH₂ (X = Cl (10), F (16), Br (17), I (18)). Reaction of 1 with alkyl-substituted allyl chlorides showed that displacement of the halide anion occurs with S(N)2' regiochemistry. A kinetic study of the conversion of 1 to 10 is consistent with a mechanism in which pyridine dissociation from the metal center precedes reaction of allyl chloride with the Cp*₂Ti=S fragment. The S(N)2' regiochemistry of the reaction is explained by postulating coordination of the halogen atom to the metal center in the reaction transition state. Complex 1 also reacts with allyl tosylate, but labeling experiments show (in contrast to the allyl chloride reaction) that substitution of the tosylate functionality occurs via a S(N)2 pathway. Heterocycles formed from the reaction of 1 with α,β- unsaturated aldehydes were also isolated and characterized.

Full text of DOI:10.1021/ja980877m

J. Am. Chem. Soc. 1998, 120, 7825-7834
7825
Cycloaddition and Nucleophilic Substitution Reactions of the
Monomeric Titanocene Sulfido Complex (η5-C5Me5)2(C5H5N)TidS
Zachary K. Sweeney, Jennifer L. Polse, Richard A. Andersen,* and Robert G. Bergman*
Contribution from the Department of Chemistry, UniVersity of California, Berkeley, California 94720
ReceiVed March 16, 1998
Abstract: The titanocene sulfido complex Cp*2(py)TidS (1, py ) pyridine) reacts reversibly with terminal
alkynes to give the thiametallacyclobutenes Cp*2Ti(SC(R)dCH) (R ) H (2), Ph (3), Tol (4), TMS (5)).
Complex 1 also reacts with alkyl halides to give products derived from formal 1,2-addition across the titanium-
sulfur bond. Treatment of 1 with allyl halides results in the formation of Cp*2Ti(X)SCH2CHCH2 (X ) Cl
(10), F (16), Br (17), I (18)). Reaction of 1 with alkyl-substituted allyl chlorides showed that displacement of
the halide anion occurs with SN2′ regiochemistry. A kinetic study of the conversion of 1 to 10 is consistent
with a mechanism in which pyridine dissociation from the metal center precedes reaction of allyl chloride
with the Cp*2TidS fragment. The SN2′ regiochemistry of the reaction is explained by postulating coordination
of the halogen atom to the metal center in the reaction transition state. Complex 1 also reacts with allyl
tosylate, but labeling experiments show (in contrast to the allyl chloride reaction) that substitution of the tosylate
functionality occurs via a SN2 pathway. Heterocycles formed from the reaction of 1 with R,â-unsaturated
aldehydes were also isolated and characterized.
Introduction
Scheme 1
In contrast to the inertness of many transition metal multiple
bonds,1 the MdX (X ) O, NR, S, PR) linkage in titanocene
and zirconocene complexes is unusually reactive. A wide range
of transformations has now been established for these com-
plexes, including overall [2 + 2] cycloadditions with olefins,2,3
alkynes,4-7 imines,8 and carbonyls9 and C-H10 and H-H11 bond
cleavage reactions with hydrocarbons and dihydrogen.12
In an attempt to extend this chemistry to titanium-sulfur
multiple bonds, we recently synthesized Cp*2(py)TidS (1, Cp*
) η5-C5Me5) and studied its reactivity toward dihydrogen and
silanes.13 We have now found that the TidS bond in 1 exhibits
cycloaddition reactivity, and in this paper we report the
regioselective reactions of 1 with alkynes and nitriles. In
addition, we have investigated the nucleophilicity of 1 and have
found that this complex undergoes an especially regioselective
series of SN2′ substitution reactions with allyl chlorides.
Results
Reactions of 1 with Terminal Alkynes and Nitriles.
Although no reaction occurs when 1 is treated with dipheny-
lacetylene, the sulfido complex reacts with terminal alkynes at
room temperature over a period of several hours to give the
orange thiametallacyclobutenes 2-5 (Scheme 1) in good yield.
1
In all cases the H NMR spectrum of the product shows only
(1) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; John
Wiley and Sons: New York, 1988.
(2) Walsh, P. J.; Hollander, F. J.; Bergman, R. G. Organometallics 1993,
12, 3705.
(3) Schwartz, D. J.; Smith, M. R.; Andersen, R. A. Organometallics 1996,
15, 1446.
(4) Walsh, P. J.; Baranger, A. M.; Bergman, R. G. J. Am. Chem. Soc.
1992, 114, 1708.
(5) Baranger, A. M.; Walsh, P. J.; Bergman, R. G. J. Am. Chem. Soc.
1993, 115, 2753.
(6) Lee, S.; Bergman, R. G. Tetrahedron 1995, 51, 4255.
(7) Carney, M. J.; Walsh, P. J.; Hollander, F. J.; Bergman, R. G.
Organometallics 1992, 11, 761.
(8) Meyer, K. E.; Walsh, P. J.; Bergman, R. G. J. Am. Chem. Soc. 1995,
117, 974.
(9) Lee, S. Y.; Bergman, R. G. J. Am. Chem. Soc. 1996, 118, 6396.
(10) Lee, S. Y.; Bergman, R. G. J. Am. Chem. Soc. 1995, 117, 5877.
(11) Howard, W. A.; Trnka, T. M.; Waters, M.; Parkin, G. J. Organomet.
Chem. 1997, 528, 95.
one Cp* resonance, suggesting that a single regioisomer has
been formed in the reaction. Attempts to determine the regio-
chemistry of the cycloaddition by hydrolysis of the metallacycle
to the corresponding ketones with dilute aqueous HCl5 led to
the formation of several products (as determined by GC
analysis). However, the large downfield shifts of the alkynyl
protons in the 1H NMR spectra of 2-5 (to between 8.5 and 7.8
ppm) and the appearance of methine carbon resonances between
220 and 190 ppm in the 13C{1H} NMR spectra of these products
are consistent with formation of thiametallacyclobutene com-
pounds in which the terminal methine carbon is bound to
titanium.
To further confirm the regiochemistry of these cycloadditions,
a single-crystal X-ray diffraction study of 5 was performed.
Crystal and data collection parameters are shown in Table 1,
and details of the structure determination are given in the
Experimental Section and as Supporting Information. An
ORTEP diagram of 5 is presented in Figure 1, and selected bond
(12) For leading references to reactive early-metal nonmetallocene imido
complexes see: Bennett, J.; Wolczanski, P. J. Am. Chem. Soc. 1997, 119,
10696.
(13) Sweeney, Z. K.; Polse, J. L.; Andersen, R. A.; Bergman, R. G.;
Kubinec, M. G. J. Am. Chem. Soc. 1997, 119, 4543.
S0002-7863(98)00877-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 07/23/1998

7826 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Sweeney et al.
Scheme 2
4 (as determined by 1H NMR spectroscopy). Although we have
not investigated the mechanism of the alkyne exchange process
in this system, we assume that it involves initial cycloreversion
of the metallacycle to [Cp*2TidS] followed by reaction of this
intermediate with another alkyne.16
Interestingly, metallacycles 2-5 were thermally stable when
heated in the absence of added alkyne in benzene at 105 °C for
24 h. These results contrast with the thermal chemistry of the
analogous oxametallacylobutenes, which rearrange to hydroxy-
acetylide complexes upon thermolysis.14 This difference is
especially pronounced in a comparison of the reactivity of 1
and Cp*2(py)TidO toward (trimethylsilyl)acetylene. Com-
pound 1 reacts with (trimethylsilyl)acetylene to form thermally
stable 5, while treatment of Cp*2(py)TidO with (trimethylsilyl)-
acetylene at room temperature results in the quantitative
formation of the hydroxy-acetylide complex Cp*2Ti(OH)-
CCSiMe3.15
Figure 1. ORTEP diagram of Cp*2Ti(SC(Si(CH3)3)CH) (5).
Table 1. Crystal Data for 5, 14, and 21
compd
5
14
21
formula
C25H40SSiTi
9.2973(2)
18.24120(10)
14.7021(3)
95.212(1)
2483.08(6)
5.66
C25H39ClSTi
9.5053(1)
16.9754(3)
15.0822(3)
100.797(1)
2390.53(6)
5.66
C30H40OSTi
9.6350(5)
10.8004(6)
25.828(1)
100.724(1)
2640.8(2)
4.23
a (Å)
b (Å)
c (Å)
â (deg)
V (Å3)
µ(Mo KR) (cm-1
space group
T (°C)
)
P21/c
P21/c
P21/c
-100
-93
-99
Reaction of 1 with an excess of benzonitrile or tolunitrile
results in conversion to products that are tentatively identified
as the four-membered metallacycles 6 and 7 (Scheme 1). The
NMR spectra of these compounds, as well as elemental analysis
and mass spectral information, indicate that further insertion of
a nitrile to form six-membered metallacycles (incorporating 2
equiv of benzonitrile) does not occur. Formation of a six-
membered metallacycle was observed when Cp*2(py)ZrdS was
treated with excess benzonitrile, although a four-membered
metallacycle resulting from reaction of the zirconium-sulfido
complex with 1 equiv of benzonitrile was postulated as an
intermediate in this transformation.7 The bound nitriles in 6
and 7 can be replaced with other nitriles or silanes. Thus,
treatment of a benzene solution of 6 with 2 equiv of m-tolunitrile
gives a mixture of 6 and 7, and reaction of 6 with H2SiMe2
produces Cp*2Ti(H)SSiHMe2 13 and free benzonitrile, as identi-
reflns collcd
unique reflns
obsd reflns
I > 3.00σ(I) var
reflns/param
R
11 074
4534
11 366
4317
10 623
3989
2713
2416
2635
253
253
298
10.72
9.55
8.84
0.049
0.049
0.033
Rw
0.056
0.055
0.040
goodness of fit
p-factor
1.89
1.65
1.47
0.030
0.030
0.030
Table 2. Selected Bond Distances (Å) and Angles (deg) for 5
bond
dist
bond
angle
Ti-S
2.403(1)
2.157(5)
1.840(4)
1.423(6)
2.118
S-Ti-C21
78.0(1)
71.6(1)
123.2
87.1(3)
112.3(2)
124.6(3)
Ti-C21
S-C22
Ti-S-C22
S-C22-C21
Ti-C21-C22
S-C22-Si
C21-C22
Ti-Cp1
Ti-Cp2
2.116
Si-C22-C21
1
fied by H NMR spectroscopy.
Reaction of 1 with Organic Halides. We have found that
1 reacts cleanly with a number of organic halides to give
products resulting from formal 1,2-addition across the titanium-
sulfur bond. Addition of methyl iodide to a benzene solution
of 1 results in a color change from red to brown and formation
of Cp*2Ti(I)SCH3 (8), isolated in 55% yield (Scheme 2). This
lengths and bond angles are shown in Table 2. The trimeth-
ylsilyl group is indeed located as far as possible from the
sterically demanding Cp* ligands. Other structural features are
similar to those of previously characterized metallacyclobutenes.
For example, the Ti-C21 bond length of 2.157(5) Å is close
to the titanium-carbon bond length of 2.144(1) Å in the related
oxametallacycle Cp*2Ti(OC(Ph)dCH),14 and the Ti-S-C bond
angle of 71.6(1)° is similar to the Zr-S-C bond angle in Cp*2-
Zr(SC(Ph)dCPh) (74.2(2)°).7
As was observed with the related oxo complexes and their
cycloadducts, the thiametallacyclobutenes were found to ex-
change alkyne fragments when heated in the presence of a
second alkyne.5,14,15 For example, heating phenylacetylene
adduct 3 to 75 °C in benzene in the presence of 1 equiv of
p-tolylacetylene resulted in the formation of a mixture of 3 and
1
product is identified in the H NMR spectrum by two singlets
corresponding to the Cp* and SCH3 protons. The mass
spectrum of 8 shows the molecular ion and a base peak
corresponding to loss of the SCH3 fragment. Similarly, reaction
of 1 with benzyl chloride results in formation of Cp*2Ti(Cl)-
SCH2C6H5 (9), isolated in 78% yield. The 13C{1H} NMR
spectrum of 9 includes a single methylene signal at 48.7 ppm
assigned to the benzylic methylene carbon, and the mass
spectrum shows a molecular ion and a peak corresponding to
loss of SCH2C6H5. Attempts to investigate the mechanism of
the reaction of 1 and benzyl chloride were complicated by the
(14) Polse, J. L.; Andersen, R. A.; Bergman, R. G. J. Am. Chem. Soc.
1995, 117, 5393.
(16) Exchange of the alkyne fragment of the analagous oxametallacy-
clobutenes was also explained by initial cycloreversion of the metallacycles.
See ref 14.
(15) Polse, J. L.; Andersen, R. A.; Bergman, R. G. Manuscript in
preparation.

Reactions of (η5-C5Me5)2(C5H5N)TidS
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7827
Scheme 3
Scheme 4
Figure 2. ORTEP diagram of Cp*2Ti(Cl)SC(CH3)2CHCH2 (14).
gradual decomposition of the product in the presence of excess
benzyl chloride. However, several experiments in which the
reaction was monitored by UV-visible spectroscopy indicate
that the reaction rate is inhibited by increasing pyridine
concentration (vide infra).
Scheme 5
No reaction is observed when 1 is treated with allyl ethyl
ether or allyl acetate. However, addition of allyl chloride to a
benzene solution of 1 results in an immediate color change from
red to blue and clean formation of 10 (Scheme 3). This
compound was isolated as blue crystals from cold toluene/
pentane and showed peaks in the 1H NMR spectrum consistent
with the presence of an allyl functionality. No intermediates
were observed when the reaction was monitored by NMR
spectroscopy at -30 °C.
Table 3. Selected Bond Distances (Å) and Angles (deg) for 14
bond
dist
bond
angle
Ti-S
2.444(2)
2.346(2)
1.314(8)
2.140
2.146
1.489(7)
Cl-Ti-S
95.69(6)
124.1(2)
134.5
126.3(5)
108.4
Ti-Cl
Ti-S-C3
C1-C2
Ti-Cp1
Ti-Cp2
C2-C3
Cp1-Ti-Cp2
C3-C2-C1
S1-Ti-Cp1
S1-Ti-Cp2
100.9
To determine if 1 reacts with allyl chlorides in an SN2 or
SN2′ fashion17 (Scheme 4), 1 was treated with several substituted
allyl chlorides. All of these reactions were essentially quantita-
tive (g95% as judged by 1H NMR spectroscopy), and in most
cases the products could be isolated in good yield as blue
crystals. To our surprise, analysis of the products of these
reactions indicated that substitution takes place exclusiVely in
an SN2′ fashion, independent of the substituents on the allyl
chloride. The products of reactions with isomeric allyl chlorides
are easily distinguished by their 1H NMR spectra. For example,
To further confirm that substitution occurred in an SN2′
manner, a single-crystal X-ray diffraction study of 14 was
performed. An ORTEP diagram of 14 is shown in Figure 2,
and Table 3 provides intramolecular bond lengths and angles.
As expected, the structure shows that attack of the sulfido ligand
has occurred at the electronically and sterically disfavored
γ-position. The double bond in the product is localized between
the terminal carbons of the allyl fragment (C1-C2 ) 1.314(8)
Å), and the -SC(CH3)2CHdCH2 ligand is oriented at an angle
that minimizes unfavorable steric interaction with the Cp*
ligands. Although the orientation of this ligand makes the Cp*
ligands inequivalent in the solid state, the presence of a single
Cp* resonance in the 1H NMR spectrum indicates that the ligand
rotates rapidly in solution.
A kinetic study of the reaction between 1 and allyl chloride
was performed in order to gain further insight into the
mechanism of these unusual substitution reactions. Production
of 10 was followed by UV-visible spectroscopy. In the
presence of excess pyridine and allyl chloride, good pseudo-
first-order rate behavior was observed. The change in the
pseudo-first-order rate constants kobs with respect to the
concentration of pyridine was studied at constant concentrations
of 1 and allyl chloride.18 A plot of 1/kobs versus [py]/[allyl
chloride] is shown in Figure 3. A linear fit of the data has a
goodness-of-fit factor of 0.993.
1
the H NMR spectrum of 14 shows a single methyl resonance
integrating to 6 protons at 1.68 ppm, while the 1H NMR
spectrum of 13 shows two methyl resonances assigned to the
cis and trans methyl groups at 1.70 and 1.66 ppm. The products
can also be distinguished by their different thermal stability.
Compound 14, in which the sulfur atom is connected to a tertiary
carbon, decomposes to a number of unidentified products over
several hours in solution at 20 °C, while isomer 13 is stable for
several days under the same conditions.
As a final test of the selectivity of these substitution reactions,
1 was treated with 1-chloro-2,4-pentadiene (Scheme 5). The
1H and 13C NMR spectra of the product of this reaction clearly
show that the sulfido ligand reacts exclusively at the γ-position
of the dienyl chloride. In particular, the 13C{1H} NMR spectrum
of 15 shows only 3 resonances for the Cs-symmetric thiolate
ligand. This result provides further evidence of the preference
for γ-substitution in this system, since either SN2 or SN2′′
substitution would result in formation of a n-pentadienethiolate
ligand.
Reaction of 1 with allyl halides appears to be quite general.
Substitution reactions also occurred when 1 was treated with
(17) DeWolfe, R. H.; Young, W. G. Chem. ReV. 1956, 56, 753.
(18) Details of the kinetic study are provided as Supporting Information.

7828 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Sweeney et al.
Figure 4. Possible products of the reaction of 1 with 2-methyl-2-
butenal.
Figure 3. Plot of 1/kobs vs [py]/[allyl chloride] for the reaction of 1
with allyl chloride.
Scheme 6
Table 4. 1H NMR Chemical Shift Data for
Cp*2Ti(X)SCHa CHbdC(Hc)Hd
2
Figure 5. ORTEP diagram of Cp*2Ti(SC(Ph)HC(Me)C(H)O) (21).
cmpd
X
δ(H)a
δ(H)b
δ(H)c
δ(H)d
Scheme 8
16
10
17
18
19
F
3.53
3.76
3.82
3.90
4.72
6.16
6.13
6.13
6.15
6.45
4.97
4.95
4.95
4.95
5.10
5.20
5.19
5.18
5.18
5.35
Cl
Br
I
OTs
Scheme 7
Reactions of 1 with r,â-Unsaturated Aldehydes. To
determine whether the γ-attack observed in the reaction of 1
with allyl halides could be generalized to other organic reactants,
the sulfido complex was treated with R,â-unsaturated aldehydes.
These reactions led to the six-membered cyclic products 20 and
allyl fluoride, allyl bromide, or allyl iodide (Scheme 6). The
product of each of these reactions was isolated and characterized
(see Experimental Section for details). The reaction of 1 with
allyl fluoride is significantly slower than the reaction of 1 with
other allyl halides. Competition reactions between 1 and
equimolar amounts of allyl bromide and allyl chloride indicate
that the reaction of 1 with allyl bromide occurs approximately
five times more rapidly than the reaction of 1 with allyl chloride.
Reactivity of 1 toward Allyl p-Toluenesulfonate. To
determine if the regioselectivity observed in the reactions of 1
with allyl chlorides was dependent on the halide functionality,
1 was treated with allyl tosylate to form compound 19 (Scheme
7). In sharp contrast to the reactions of 1 with allyl chlorides,
which are complete in minutes, the tosylate reaction requires
several days at 20 °C to go to completion. The similarities
1
21 shown in Scheme 8. Important features of the H NMR
spectra of these products are two resonances corresponding to
diastereotopic Cp* protons and a large upfield shift of the
formerly aldehydic proton. To distinguish between products
arising from overall [2 + 2] and [4 + 2] cycloadditions
1
(illustrated as A and B in Figure 4), a H-13C HMQC NMR
study19 of 20 was performed. The results indicate that the
hydrogen adjacent to a methyl group, which appears as a quartet
at 4.15 ppm in the 1H NMR spectrum, is connected to a carbon
that appears at 43.2 ppm in the 13C{1H} NMR spectrum. This
strongly suggests that this carbon is sp3 hybridized and that the
structure of the product is the six-membered metallacycle (type
B in Figure 4). To further confirm this conclusion, the structure
of 21 was determined by X-ray crystallography. Crystal and
data collection parameters are shown in Table 1, and details of
the structure determination are given in the Experimental
Section. An ORTEP diagram of 21 is shown in Figure 5, and
Table 5 provides intramolecular bond lengths and bond angles.
The structure shows that the titanium-sulfur and titanium-
oxygen bond lengths are in the range expected for Ti-S and
Ti-O single bonds. The six-membered ring is strongly
puckered, and the phenyl group resides in an equatorial position.
1
between the H and 13C{1H} NMR spectra of 19 and 10 and
16-18 (Table 4) suggest that 19 has the same connectivity as
its halogenated analogues. The regiochemistry of this substitu-
tion was probed by treating 1 with allyl tosylate-1-d, and the
product of this reaction (19-d) was analyzed by NMR spec-
1
2
troscopy. The H, H, and 13C{1H} spectra clearly show that
reaction takes place predominately in an SN2 fashion (Scheme
7). In particular, the 2H spectrum of the product in C6H6 shows
only a single signal at 4.7 ppm, and the 13C{1H} spectrum,
which was otherwise identical to the fully protiated analogue,
shows a 1:1:1 triplet instead of a singlet at 44 ppm.
(19) Bax, A.; Subramanian, S. J. Magn. Reson. 1986, 67, 565.

Reactions of (η5-C5Me5)2(C5H5N)TidS
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7829
Similar conversion of the thiametallacyclobutenes to hydro-
sulfido acetylide compounds would produce a relatively weak
SH bond. Because of this, we suspect that rearrangement is
not observed because the thiametallacylclobutenes are thermo-
dynamically more stable than their hydrosulfido-acetylide
isomers.
The detailed mechanism of cycloaddition reactions between
Cp2MdX complexes (M ) Ti, Zr; X ) CR2, NR, O, S) and
alkynes and alkenes is now well understood. Previous workers
in our group have shown that the reaction of a zirconocene imido
pyridine complex with diphenylacetylene is preceded by pyridine
dissociation (eq 2).6
Figure 6. Molecular orbital diagram showing the frontier orbitals of
Cp*2TidS.
Table 5. Selected Bond Distances (Å) and Angles (deg) for 21
bond
dist
bond
S-Ti-O
angle
Ti-S
Ti-O
C22-C21
C22-C23
S-C23
O-C21
Ti-Cp1
Ti-Cp2
2.4231(9)
1.890(2)
1.319(4)
1.532(4)
1.857(3)
1.350(3)
2.151
87.70(6)
111.6(2)
119.6(2)
125.4(3)
137.3(2)
102.7(1)
103.7
S-C23-C22
C23-C22-C21
C22-C21-O
C21-O-Ti
C23-S-Ti
As mentioned above, dissociation of the labile ligand from
the metal center releases a metal-based orbital adjacent to the
reactive metal-heteroatom bond. Norton et al. have recently
suggested that apparent [2 + 2] organometallic cycloadditions
always involve initial coordination of one partner to the metal,23
and we believe that ligand dissociation is required in order for
the above-mentioned cyclization reactions to occur because they
are mediated by this available empty orbital. This postulate is
consistent with the calculations of Upton and Rappe´, who
studied the formation of titanacyclobutanes from olefins and
titanium alkylidene complexes.24 An empty metal-based orbital
adjacent to the reactive ligand is also thought to participate in
the hydrogenation25 and dealkylation26 reactions of d0 metal-
carbon bonds.
The apparent nucleophilicity of early metal complexes
containing metal-ligand multiple bonds is generally thought
to be a consequence of the high energy of the metal frontier
orbitals.1 The resulting localization of electron density on the
ligand enables some of these complexes to form metallacycles
with heterocumulenes or simple adducts with strong Lewis acids.
For example, Geoffroy et al. found that (tmtaa)TidS reacts with
perfluorinated acetone,27 and Green et al.28 and Floriani et al.29
have recently characterized adducts formed from the reaction
of titanium oxo complexes with boranes.30-38 However, as
mentioned above, theoretical calculations have shown that the
TidS bond in Cp2Ti(S) is probably best described as covalent,
and we were surprised by the facility with which 1 underwent
nucleophilic substitution reactions with alkyl halides. Since we
were interested in the possibility that the metal might interact
O-Ti-Cp1
S-Ti-Cp1
2.136
107.3
Discussion
The synthesis and structure of Cp*2Ti(S)py (1) have been
reported in preliminary form.13 The pyridine ligand in this
complex is labile in solution and is rapidly displaced by stronger-
binding ligands such as p-(dimethylamino)pyridine. Dissocia-
tion of pyridine from the 18-electron complex20 results in
formation of a coordinatively unsaturated 16-electron molecule
that contains a vacant metal-based orbital perpendicular to the
metal-sulfur bond.21 Calculations on the cyclopentadiene
analogue of this compound using ab initio methods predict that
the Ti-S bond polarity expected from the electronegativity
difference between the sulfur ligand and the metallocene
fragment is offset somewhat by π donation from the ligand to
the metal center and that the bond is best described as covalent.22
A simple molecular orbital diagram showing the frontier orbital
configuration of Cp*2TidS is illustrated in Figure 6.
We have found that 1 readily forms metallacycles when
treated with alkynes and nitriles. The regiochemistry of the
alkyne cyclization reaction appears to be controlled by steric
factors. Complex 1, like the closely related oxo compound
Cp*2Ti(O)py, reacts with terminal alkynes to place the alkyne
substituent in the â position, away from the sterically demanding
Cp*2Ti center. The ready reversibility of these cyclization
reactions has also been observed in the analogous oxo system.
However, the oxatitanacyclobutenes rearrange to the corre-
sponding hydroxy-acetylide complexes upon thermolysis
(eq 1).14
(23) Bender, B. R.; Ramage, D. L.; Norton, J. R.; Wiser, D. C.; Rappe´,
A. K. J. Am. Chem. Soc. 1997, 119, 5628.
(24) Upton, T. H.; Rappe´, A. K. J. Am. Chem. Soc. 1985, 107, 1206.
(25) Gell, K. I.; Posin, B.; Schwartz, J.; Williams, G. M. J. Am. Chem.
Soc. 1982, 104, 1846.
(26) Caulton, K. G.; Marsella, J. A. J. Am. Chem. Soc. 1982, 104, 2361.
(27) Housmekerides, C. E.; Ramage, D. L.; Kretz, C. M.; Shontz, J. T.;
Geoffroy, G. L. Inorg. Chem. 1992, 31, 4453. tmtaa ) dianion of 7,
16-dihydro-6,8,15,17-tetramethyldibenzo[b,i][1,4,8,11]tetraazacyclotetra-
decane.
(28) Galsworthy, J. R.; Green, M. L. H.; Muller, M.; Prout, K. J. Chem.
Soc., Dalton Trans. 1997, 1309.
(20) Parkin and Bercaw have proposed that 18 electron complexes of
this type be referred to as “class b” complexes. See: Parkin, G.; Bercaw,
J. E. J. Am. Chem. Soc. 1989, 111, 391.
(21) Lauher, J. W.; Hoffman, R. J. Am. Chem. Soc. 1976, 98, 1729.
(22) Fischer, J. M.; Piers, W. E.; Ziegler, T.; MacGillivray, L. R.;
Zaworotko, M. J. Chem. Eur. J. 1996, 2, 1221.
(29) Franceschi, F.; Gallo, E.; Solari, E.; Floriani, C.; Chiesi-Villa, A.;
Rizzoli, C.; Re, N.; Sgamellotti, A. Chem. Eur. J. 1996, 2, 1466.

7830 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Sweeney et al.
Scheme 9
k3k4[RCl][1]
rate )
) kobs[1]
(5)
(6)
k-3[py] + k4[RCl]
k-3
[py]
1
1
)
+
(
)(
)
kobs k3
k3k4
[RCl]
inhibit the reaction rate. Over a range of py and RCl
concentrations, a plot of 1/kobs versus the [py]/[RCl] ratio should
be linear (eq 6).
Qualitatively, we observe that added pyridine does slow the
rate of substitution. A plot of 1/kobs versus [py]/[allyl chloride]
is shown in Figure 3. The plot is linear, consistent with pyridine
dissociation occurring prior to reaction of the Cp*2TidS
intermediate with allyl chloride (path B). The rate constant for
pyridine dissociation from 1 at 20 °C can be determined from
the value of the y-intercept of this plot, which is equal to 1/k3.
with the leaving group in the transition state of these substitution
reactions, we have examined the origins of the apparent
nucleophilicity of 1 more closely.
It is likely that pyridine coordination to the metal center
indirectly increases the electron density on the sulfur ligand by
increasing the electron density at the metal center. Since this
might affect the nucleophilic reactivity of the sulfido ligand,
we attempted to determine whether reactions in which 1
functions as a nucleophile occur directly from the pyridine-
bound complex or whether prior dissociation of pyridine is
required. Two general mechanisms through which 1 might react
as a nucleophile toward allyl chloride are shown in Scheme 9.
Path A involves direct nucleophilic attack of the sulfur atom of
the pyridine-bound complex followed by cation-anion recom-
bination and liberation of pyridine. In path B, initial rapid and
reversible pyridine dissociation generates an unsaturated com-
plex that reacts with allyl chloride in a subsequent step.39
The observed rate constant for the associative reaction (path
A) should be zero order in pyridine concentration and directly
proportional to the concentration of allyl chloride (eq 3). When
This value is approximately 2.2 × 10-2 s-1
.
The fact that added pyridine retards the reaction rate of 1
with allyl chloride and benzyl chloride suggests that these
transformations occur by reaction of the organic halide with
the 16-electron Cp*2Ti(S) fragment. This is significant because
it implies that a reaction pathway in which C-Cl bond breaking
is facilitated by interaction of the halide with the metal center
is preferred over one in which 1 reacts directly with the
electrophile. These conclusions are supported by the regiose-
lectivity of the reaction of 1 with allyl chloride (vide infra).
As mentioned in the Results section, 1 reacts regioselectively
with substituted allyl chlorides to form products resulting from
SN2′ substitution of the halide anion. Exclusive SN2′ substitution
of allyl halides is very rare, since this mode of substitution
requires that the nucleophile attack a relatively electron-rich
carbon center instead of the more electropositive R-carbon.40
Many reactions once thought to occur via SN2′ processes have
been shown to actually occur by reaction of the nucleophile
with an allyl halide ion pair,41 and it is difficult to find examples
of nucleophiles that react exclusively via a SN2′ pathway when
SN2′ substitution is sterically disfavored.42 Various organome-
tallic compounds have been shown to give some SN2′ substitu-
tion product when treated with allyl halides,43 although to our
knowledge RCu‚BF3 is a rare example of an organometallic
reagent that reacts exclusively in an SN2′ manner.44,45
rate ) k1[RCl][1] ) kobs[1]
kobs [RCl]
(3)
(4)
kobs is measured under pseudo-first-order conditions in the
presence of different concentrations of allyl chloride, a plot of
the allyl chloride concentration versus the observed reaction rate
constant (kobs) should be linear (eq 4). The observed rate
constant for a reaction following Path B need not be directly
proportional to the allyl chloride concentration. The rate law
for a reaction following this pathway is shown in eq 5. This
equation indicates that if conditions can be found such that k4-
[RCl] is not always much larger than k-3[py], added py should
The appearance of SN2′ products in reactions of organome-
tallic reagents with allyl halides is often explained by postulating
an interaction between the metal center and the leaving group
in a six-membered transition state.43,46-49 It seems likely that
substitution by the titanocene sulfido complex proceeds via a
similar mechanism in which the halide leaving group interacts
with the Cp*2Ti(S) LUMO. A diagram of this postulated
transition state is shown in Figure 7A. This mechanism is
(30) For other reactions illustrating the basic or nucleophilic properties
of complexes containing metal-oxygen or metal-sulfur multiple bonds
see refs 11 and 31-38. For a review of the nucleophilicity of metal-
heteroatom bonds see: Caulton, K. G. New J. Chem. 1994, 18, 25.
(31) Pasini, A.; Colombo, A.; Marturano, G. Polyhedron 1995, 15, 481.
(32) Brunner, H.; Kubicki, M. M.; Leblanc, J. C.; Moise, C.; Volpato,
F.; Wachter, J. J. Chem. Soc., Chem. Commun. 1993, 851.
(33) Jernakoff, P.; Geoffroy, G. L.; Rheingold, A. L.; Geib, S. J. J. Chem.
Soc., Chem. Commun. 1987, 1610.
(34) Fischer, J.; Kress, J.; Osborn, J. A.; Ricard, L.; Westolek, M.
Polyhedron 1987, 6, 1839.
(35) Bridgeman, A. J.; Davis, L.; Dixon, S. J.; Green, J. C.; Wright, I.
N. J. Chem. Soc., Dalton Trans. 1995, 1023.
(40) Bordwell, F. G. Acc. Chem. Res. 1970, 3, 281.
(41) Bordwell, F. G.; Clemens, A. H.; Cheng, J.-P. J. Am. Chem. Soc.
1987, 109, 1773.
(42) Cyclobutenyl chlorides have been shown to undergo SN2′ reactions
with NaOMe. See: Kirmse, W.; Scheidt, F.; Vater, H. J. J. Am. Chem.
Soc. 1978, 100, 3945.
(36) Pilato, R. S.; Housmekerides, C. E.; Jernakoff, P.; Rubin, D.;
Geoffroy, G. L. Organometallics 1990, 9, 2333.
(37) Pilato, R. S.; Rubin, D.; Geoffroy, G. L.; Rheingold, A. L. Inorg.
Chem. 1990, 29, 1986.
(43) Magid, R. M. Tetrahedron 1980, 36, 1901.
(44) Yamamoto, Y. Angew. Chem., Int. Ed. Engl. 1986, 25, 947.
(45) Yamamoto, Y.; Yamamoto, S.; Yatagai, H.; Maruyama, K. J. Am.
Chem. Soc. 1980, 102, 2318.
(38) Pilato, R. S.; Geoffroy, G. L.; Rheingold, A. L. J. Chem. Soc., Chem.
Commun. 1989, 1287.
(46) Magid, R. M.; Nieh, E. C.; Gandour, R. D. J. Org. Chem. 1971,
36, 2099.
(47) Lenox, R. S.; Katzenellenbogen, J. A. J. Am. Chem. Soc. 1973, 95,
957.
(48) Chaabouni, R.; Laurent, A.; Marquet, B. J. Am. Chem. Soc. 1976,
98, 757.
(49) Fraser-Reid, B.; Tam, S. Y. K.; Radatus, B. Can. J. Chem. 1975,
53, 2005.
(39) Theoretically, either pathway could occur via single electron transfer
(SET) steps; however, the regioselectivity observed in reactions of 1 with
substituted allyl chlorides indicates that in these reactions an SET mechanism
is not operative. For an interesting study of electron-transfer reactions of a
titanocene carbene complex see: Buchwald, S. L.; Ansyln, E. V.; Grubbs,
R. H. J. Am. Chem. Soc. 1985, 107, 1766.

Reactions of (η5-C5Me5)2(C5H5N)TidS
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7831
Experimental Section
General Methods. Unless otherwise noted, all reactions and
manipulations were carried out in dry glassware under a nitrogen or
argon atmosphere using standard Schlenk techniques or at 20 °C in a
Vacuum atmospheres 553-2 drybox equipped with a MO-40-2 inert
gas purifier. The amount of O2 in the drybox atmosphere was
monitored with a Teledyne model no. 316 trace oxygen analyzer. Some
reactions were carried out in thick-walled glass vessels fused to vacuum
stopcocks. These vessels are referred to in the procedures as bombs.
The other instrumentation and general procedures used have been
described previously.53
Unless otherwise specified, all reagents were purchased from
commercial suppliers and used without further purification. Pyridine
was distilled sequentially from CaH2 and sodium and was stored over
activated 4 Å molecular sieves. Sulfur was dissolved in dry benzene
containing activated 4 Å molecular sieves, stirred overnight, and then
recrystallized from the filtered benzene solution. Pentane and hexane
(UV grade) were distilled from sodium benzophenone ketyl/tetraglyme
under nitrogen. Deuterated solvents for NMR experiments were dried
in the same way as their protiated analogues but were vacuum
transferred from the drying agent. Acetylene was bubbled through
concentrated H2SO4 and passed through two -78 °C cold traps. Methyl
iodide, phenylacetylene, tolylacetylene, nitriles, R,â-unsaturated alde-
hydes, allyl fluoride, and all allyl chlorides were dried over activated
4 Å molecular sieves before use. Cp*2Ti(C2H4) was prepared by the
literature method, except that Cp*2TiCl was used instead of Cp*2TiCl2.54
3-Chloro-3-methylbut-1-ene,55 allyl tosylate, and allyl tosylate-1-d56
were prepared according to literature procedures.
Figure 7. Postulated transition states for the reaction of 1 with allyl
chlorides (A) and R,â-unsaturated aldehydes (B).
consistent with the observed pyridine inhibition of the reaction
rate, since pyridine dissociation must precede interaction of the
halide with the metal center. It should be noted that Parkin et
al. have recently suggested that the deprotonation of acetone
t
and BuI by Cp*2Zr(O)py may occur via similar transition
states.11
This hypothesis also explains the different substitution
regioselectivity (SN2 rather than SN2′) observed when 1 reacts
with allyl tosylate. Apparently, the tosylate moiety is unable
to interact directly with the metal center in a transition state
leading to SN2′ substitution, and so the lowest energy reaction
pathway involves direct attack on the substituted carbon. The
observation that the reaction of 1 with allyl tosylate is
considerably slower than the reaction of 1 with allyl chloride,
even though tosylates are generally more easily displaced than
chlorides,50,51 provides further evidence that these reactions do
not proceed via the same mechanism.52
The HMQC experiment was recorded on a Brucker AMX spec-
trometer resonating at 300.13 MHz for 1H and 75.42 MHz for 13C that
was equipped with an inverse probe. One-dimensional NMR spectra
were recorded on the Brucker AMX spectrometer described above or
It is also possible that the reactions of 1 with R,â-unsaturated
aldehydes occur via six-membered transition states (Figure 7B).
In this case, the carbonyl functionality would interact with the
metal center through one of the oxygen lone pairs. However,
attack of “soft” nucleophiles at the â position of R,â-unsaturated
aldehydes is much more common than SN2′ substitution of allyl
halides, so the postulate of a concerted reaction proceeding via
a six-membered transition state may not be necessary to explain
the regioselectivity observed in these transformations.
1
on a Brucker AMX spectrometer resonating at 400 MHz for H and
100 MHz for 13C that was equipped with a QNP probe. 1H and 13C-
{1H} NMR spectra were referenced to resonances corresponding to
residual protons or carbon atoms of the solvent. 19F NMR spectra were
externally referenced to CCl3F (δ ) 0.0 ppm).
Cp*2Ti(dS)py (1). A solution of Cp*2Ti(CH2CH2) (486 mg, 1.40
mmol) and pyridine (0.56 mL, 5 equiv) in toluene (20 mL) was cooled
to -40 °C. To this stirred solution, a suspension of freshly recrystal-
lized sulfur (33 mg, 1.0 equiv) in cold toluene was added dropwise.
The green solution immediately turned dark. After 40 h, the solvent
was reduced in vacuo (to approximately 10 mL) and pentane was
layered onto the toluene solution. This mixture was cooled to -40 °C
to yield 1 as red crystals (459 mg, 76%). 1H NMR (C6D6): δ 8.00 (d,
2H), 6.64 (m, 1H), 6.25 (m, 2H), 1.90 (s, 30H) ppm. 13C NMR: δ
135.9 (CH), 128.3 (CH), 122.4 (CH), 120.3 (C-CH3), 13.3 (C-CH3)
ppm. IR (Nujol): 2730, 1598, 1209, 1014, 802, 755, 709 cm-1. MS
(EI): m/z 430 (M+). Anal. Calcd for C25H35NSTi: C, 69.90; H, 8.22;
N, 3.26. Found: C, 69.88; H, 8.25; N, 3.23.
Conclusion
We have shown that the (decamethylcyclopentadienyl)-
titanium sulfido complex 1 undergoes regiospecific cycloaddi-
tion reactions with nitriles and terminal alkynes. These
transformations are probably mediated by the vacant metal
orbital adjacent to the sulfido ligand in the 16-electron Cp*2-
TidS species. We have also found that 1 undergoes very
regioselective nucleophilic substitution reactions with alkyl
halides, alkyl tosylates, and R,â-unsaturated aldehydes.
A
Cp*2Ti(SC(H)CH) (2). A bomb was charged with 1 (95 mg, 0.22
mmol) dissolved in benzene (10 mL). Acetylene was bubbled through
the solution for 5 min, and the resulting mixture was allowed to stand
overnight. The solution was filtered several times through fiberglass
filter paper, and the volatile materials were removed in vacuo.
Crystallization of the residue from a toluene/hexane solution at -40
°C gave 52 mg (62%) of 2 as brown crystals. 1H NMR (C6D6): δ
7.42 (d, J ) 9.9 Hz, 1H), 6.95 (d, J ) 9.9 Hz, 1H), 1.69 (s, 30H) ppm.
13C NMR: δ 194.9 (CH), 120.0 (C-CH3), 96.5 (CH), 11.8 (C-CH3)
ppm. IR (Nujol): 2723, 1490, 1446, 1187, 1020, 806, 673 cm-1. MS
(EI): m/z 376 (M+). Anal. Calcd for C22H32STi: C, 70.20; H, 8.57.
Found: C, 70.10; H, 8.67.
kinetic study has established that the reaction of 1 with allyl
chlorides involves initial reversible dissociation of pyridine. This
is most likely followed by coordination of the halogen atom at
the metal center and attack of the sulfur atom at the γ-carbon,
leading to product via a six-membered transition state. In this
way the vacant metal-based orbital also plays an important role
in substitution reactions of these metal-heteroatom complexes.
(50) March, J. AdVanced Organic Chemistry, 4th ed.; John Wiley &
Sons: New York, 1992.
(51) Streitwieser, A. SolVolytic Displacement Reactions; McGraw-Hill:
New York, 1962.
Cp*2Ti(SC(Ph)CH) (3). A Schlenk flask equipped with a Teflon
stir bar was charged with 1 (51 mg, 0.12 mmol) dissolved in benzene
(52) The relative rates of the reactions of 1 with benzyl chloride and
allyl tosylate suggest that benzyl chloride substitution also does not occur
via a simple SN2 mechanism. The rate of this reaction is inhibited by
pyridine, indicating that halide dissociation might be assisted by coordination
of the halide to the metal center in the transition state of the substitution.
It is possible that formation of 9 occurs via Claisen-type attack of sulfur at
the ortho position of the aromatic ring, followed by 1,3-migration to give
the observed product. At this time we are unable to distinguish between
this mechanism or other processes such as those proceeding via electron
transfer.
(53) Meyer, K. E.; Walsh, P. J.; Bergman, R. G. J. Am. Chem. Soc. 1995,
117, 3749.
(54) Cohen, S. A.; Auburn, P. R.; Bercaw, J. E. J. Am. Chem. Soc. 1983,
105, 1136.
(55) Sneen, R. A.; Key, P. S. J. Am. Chem. Soc. 1972, 94, 6983.
(56) Baldwin, J. E.; Bradley, M.; Turner, N. J.; Adlington, R. M.; Pitt,
A. R.; Sheridan, H. Tetrahedron 1991, 47, 8203.

7832 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Sweeney et al.
(5 mL). Phenylacetylene (80 µL, 0.72 mmol) was dissolved in benzene
(2 mL), and the solutions were combined. The reaction mixture was
stirred overnight, and the volatile materials were removed to yield 52.1
mg of analytically pure 3 (95%). 1H NMR (C6D6): δ 8.06 (m, 2H),
7.80 (s, 1H), 7.27 (m, 2H), 7.11 (m, 1H), 1.69 (s, 30H) ppm. 13C
NMR: δ 191.4 (CH), 137.5 (C), 128.2 (CH), 126.6 (CH), 126.4 (C-
CH3), 122.0 (CH), 115.3 (C), 12.1 (C-CH3) ppm. IR (Nujol): 3100,
2717, 1968, 1874, 1789, 1737, 1164, 1020, 904, 730 cm-1. MS (EI):
m/z 452 (M+). Anal. Calcd for C28H36STi: C, 74.32; H, 8.02.
Found: C, 74.35; H 8.02.
was crystallized from toluene at -40 °C to yield 87 mg of 9 (78%) as
blue crystals.1H NMR (C6D6): δ 7.48 (m, 2H), 7.01 (m, 3H), 4.32 (s,
2H), 1.88 (s, 30H) ppm. 13C NMR: δ 144.8 (CH), 129.2 (C), 128.7
(CH), 126.0 (CH), 125.4 (C-CH3), 48.7 (CH2), 12.8 (C-CH3) ppm.
IR (Nujol): 3058, 2707, 1598, 1492, 1218, 1182, 1066, 1020, 707 cm-1
.
MS (EI) m/z 476 (M+) Anal. Calcd for C27H37ClSTi: C, 67.99; H,
7.82. Found: C, 67.89; 7.64.
Cp*2Ti(Cl)SCH2CHCH2 (10). A bomb containing a frozen solution
of 1 (164 mg, 0.38 mmol) in benzene was evacuated, and 1.5 equiv
(0.57 mmol) of 3-chloropropene was condensed onto the frozen solution
at -196 °C. Upon thawing, the solution color changed immediately
from red to green. The solution was allowed to stand for 2 h, and the
volatile materials were removed. The resulting green solid was
recrystallized from ether/pentane at -40 °C to yield 10 as purple crystals
(99 mg, 61%). 1H NMR (C6D6): δ 6.13 (m, 1H), 5.19 (m, 1H), 4.95
(m, 1H), 3.76 (m, 2H), 1.88 (s, 30H) ppm. 13C NMR: δ 139.0 (CH),
125.3 (C-CH3, 113.3 (CH), 47.0 (CH2), 12.7 (C-CH3) ppm. IR
(Nujol): 3054, 2719, 1631, 1209, 1072, 1018, 917, 744 cm-1. MS
(EI): m/z 426 (M+). Anal. Calcd for C23H35ClSTi: C, 64.71; H, 8.26.
Found: C, 64.74; H, 8.28.
Cp*2Ti(Cl)SCH2CHCHCH3 (11). A bomb containing a frozen
solution of 1 (151 mg, 0.35 mmol) in benzene was evacuated, and 1.1
equiv (0.39 mmol) of 3-chloro-1-butene was condensed in at -196
°C. Upon thawing, the solution color changed immediately from red
to green. The solution was allowed to stand for several hours. The
volatile materials were removed to give a green solid that was
crystallized from toluene/pentane at -40 °C to give 11 as a blue,
crystalline solid (120 mg, 77%). 1H NMR (C6D6): δ 5.77 (m, 1H),
5.56 (m, 1H), 3.74 (m, 2H), 1.90 (s, 30H), 1.59 (d, J ) 6.4 Hz, 3H)
ppm. 13C NMR: δ 132.1 (CH), 125.2 (C-CH3), 124.2 (CH), 46.6
(CH), 17.7 (CH3), 12.7 (C-CH3) ppm. IR (Nujol): 3067, 2719, 1195,
1020, 960, 809, 694 cm-1. MS (EI): m/z 440 (M+). Anal. Calcd for
C24H37ClSTi: C, 65.37; H, 8.46. Found: C, 65.40; H, 8.63.
Cp*2Ti(Cl)SCH(CH3)CHCH2 (12). A bomb containing a frozen
solution of 1 (151 mg, 0.27 mmol) in benzene was evacuated, and 1
equiv of crotyl chloride (0.27 mmol) was condensed into the bomb at
-196 °C. Upon thawing, the solution color changed immediately from
red to green. The reaction mixture was allowed to stand for 2 h. The
volatile materials were removed to yield a green solid that was
crystallized from ether at -40 °C to yield 121 mg (77%) of 12 as blue
crystals. 1H NMR (C6D6): δ 6.45 (m, 1H), 5.31 (m, 1H), 5.08 (m,
1H), 4.32 (m, 2H), 1.90 (s, 15H), 1.88 (s, 15H), 1.65 (d, J ) 6.8 Hz,
3H) ppm. 13C NMR: δ 144.4 (CH), 125.3 (C-CH3), 110.1 (CH2),
48.1 (CH), 21.3 (CH3), 12.9 (C-CH3), 12.9 (C-CH3) ppm. IR
(cyclohexane): 3087, 2732, 1635, 1498, 1378, 1029, 809, 727, 690
cm-1. Anal. Calcd for C24H37ClSTi: C, 65.37; H, 8.46. Found: C,
65.60; H, 8.59.
Cp*2Ti(Cl)SCH2CHC(CH3)2 (13). To a vial containing a stirred
solution of 1 (186 mg, 0.43 mmol) in benzene was added 3-chloro-3-
methylbut-1-ene (55 µL, 0.52 mmol) via syringe. The solution color
immediately changed from red to green. The solution was stirred for
30 min, and then the volatile materials were removed to give a green-
blue solid that was recrystallized from ether at -40 °C to give 13 as
blue crystals (131 mg, 70%). 1H NMR (C6D6): δ 5.66 (m, 1H), 3.80
(d, J ) 7.9 Hz, 2H), 1.90 (s, 30H), 1.70 (s, 3H), 1.66 (s, 3H). 13C
NMR: δ 131.3 (C), 126.4 (CH), 125.4 (C-CH3), 43.1 (CH2), 25.8
(CH3), 18.2 (CH3), 13.0 (C-CH3) ppm. IR (Nujol): 3021, 2719, 1666,
1199, 1103, 1018, 848 cm-1. Anal. Calcd for C25H39ClSTi: C, 66.00;
H, 8.64. Found: C, 65.83; H, 8.64.
Cp*2Ti(Cl)SC(CH3)2CHCH2 (14). Neat 1-chloro-3-methyl-2-
butene (105 µL, 0.93 mmol) was added to a solution of 1 (100 mg,
0.23 mmol) in C6H6 (5 mL). The reaction mixture was agitated and
then allowed to stand for 15 min. The volatile materials were removed,
and the remaining solid was crystallized from toluene/pentane at -40
°C to yield 14 as large blue crystals (57 mg, 54%). 1H NMR (C6D6):
δ 6.63 (m, 1H), 5.10 (m, 1H), 4.98 (m, 1H), 1.89 (s, 30H), 1.68 (s,
6H) ppm. 13C NMR: δ 151.4 (CH), 127.1 (C-CH3), 107.5 (CH2),
55.5 (C), 33.1 (CH3), 13.4 (C-CH3) ppm. IR (Nujol): 2709, 1625,
1376, 1105, 1014, 904, 794, 694, 592 cm-1. MS (EI): m/z 454 (M+).
Anal. Calcd for C25H39ClSTi: C, 66.00; H, 8.64. Found: C, 65.91;
H, 8.64.
Cp*2Ti(SC(Tol)CH) (4). A scintillation vial was charged with 1
(122 mg, 0.28 mmol) in benzene (5 mL). p-Tolylacetylene (165 mg,
1.4 mmol) was added via syringe, and the solution was stirred at RT
(room temperature) for 2 d. The volatile materials were removed, and
the remaining solid was crystallized from a toluene/pentane mixture at
-40 °C to yield 4 as dark orange crystals (98 mg, 74%). 1H NMR
(C6D6): δ 7.98 (d, 2H, J ) 8.0 Hz), 7.79 (s, 1H), 7.10 (d, 2H, J ) 8.0
Hz), 1.80 (s, 3H), 1.70 (s, 30 H) ppm. 13C NMR: δ 190.3 (CH), 135.4
(C), 134.5 (C), 129.0 (CH), 126.3 (CH), 121.7 (C-CH3), 114.8 (CH),
20.8 (CH3), 11.9 (C-CH3) ppm. IR (Nujol): 3020, 2713, 1565, 1506,
1118, 1020, 817, 757, 636 cm-1. MS (EI): m/z 466 (M+). Anal. Calcd
for C29H38STi: C, 74.66; H, 8.21. Found: C, 74.31; H, 8.13.
Cp*2Ti(SC(SiMe3)CH) (5). A scintillation vial was charged with
a solution of 1 (110 mg, 0.26 mmol) in benzene (5 mL). (Trimeth-
ylsilyl)acetylene (150 mg, 1.5 mmol) was added via syringe. The
reaction mixture was allowed to stir for 3 d. The volatile materials
were removed, and crystallization of the residue from pentane at -40
1
°C gave 5 as large orange crystals (98 mg, 85%) H NMR (C6D6): δ
8.20 (s, 1H), 1.66 (s, 30H), 0.47 (s, 9H) ppm. 13C NMR: δ 210.4
(CH), 118.9 (C-CH3), 97.6 (C), 12.4 (C-CH3), 1.2 (CH3) ppm. IR
(cyclohexane): 3008, 1498, 1459, 1373, 1240, 831, 742 cm-1. MS
(EI): m/z 448 (M+). Anal. Calcd for C25H40SSiTi: C, 66.93; H, 8.99.
Found: C, 66.88; H, 9.07.
Cp*2Ti(SC(Ph)N) (6). A scintillation vial was charged with 1 (192
mg, 0.45 mmol) in benzene (5 mL). Benzonitrile (46 µL, 0.45 mmol)
was added via syringe. After 2 h, the volatile materials were removed
to give an orange solid that was recrystallized from toluene/pentane at
-40 °C to yield 6 (132 mg, 65%) as orange crystals. 1H NMR
(C6D6): δ 8.47 (m, 2H), 7.27 (t, 2H), 7.14 (m, 1 H), 1.72 (s, 30H)
ppm. 13C NMR: δ 143.9 (C), 137.7 (C), 129.0 (CH), 128.3 (CH),
125.4 (CH), 123.1 (C-CH3), 11.8 (C-CH3) ppm. IR (Nujol): 3020,
2730, 2036, 1882, 1818, 1641, 1454, 1375, 1155, 1066, 881, 759, 692,
609 cm-1. MS (EI): m/z 454 (M + H+). Anal. Calcd for C27H35-
NSTi: C, 71.51; H, 7.78; N, 3.09. Found: C, 71.38; H, 7.84; N, 3.15.
Cp*2Ti(SC(m-Tol)N) (7). A scintillation vial was charged with 1
(103 mg, 0.24 mmol) in benzene (5 mL). m-Tolunitrile (29 uL, 0.24
mmol)) was added via syringe. After 6 h, the volatile materials were
removed to give an orange solid that was crystallized from toluene/
pentane at -40 °C to yield 73 mg of 7 (56%) as an orange powder. 1H
NMR (C6D6): δ 8.52 (m, 1H), 7.34 (t, 1H), 7.15 (m, 1H), 6.97 (d,
1H), 2.16 (s, 3H), 1.73 (s, 30H). 13C NMR: δ 143.9 (C), 137.8 (C),
137.4 (CH), 129.8 (CH), 128.8 (CH), 124.8 (CH), 123.0 (C-CH3),
21.2 (CH3), 11.8 (C-CH3) ppm. IR (Nujol): 3045, 2719, 1795, 1619,
1020, 929, 784, 692, 559 cm-1. MS (EI): m/z 468 (M + 1). Anal.
Calcd for C28H37NSTi: C, 71.93; H, 7.98; N, 3.00. Found: C, 71.80;
H, 7.86; N, 2.76.
Cp*2Ti(I)SMe (8). A bomb was charged with 1 (135 mg, 0.31
mmol) in benzene (5 mL). Methyl iodide (0.35 mmol) was vacuum
transferred onto the frozen solution at -196 °C. The solution turned
green after standing for 5 min at RT. After 12 h, the volatile materials
were removed to yield a green solid. This material was crystallized
from toluene/pentane at -40 °C to give 86 mg of 8 (55%) as a green
solid. 1H NMR (C6D6): 2.89 (s, 3H), 1.94 (s, 30H) ppm. 13C NMR:
δ 125.2 (C-CH3) 28.3 (CH3), 14.2 (C-CH3) ppm. IR (Nujol): 2721,
1544, 1308, 1018, 582 cm-1. MS (EI) m/z 492 (M+). Anal. Calcd
for C21H33ISTi: C, 51.23; H, 6.75. Found: C, 51.23; H, 6.76.
Cp*2Ti(Cl)SCH2Ph (9). Benzyl chloride (29 µL, 0.25 mmol) was
added via syringe to a stirred solution of 1 (100 mg, 0.23 mmol) in
benzene (5 mL). After 5 min the color of the solution had changed
from red to purple. The reaction mixture was allowed to stir for 30
min, and the volatile materials were removed. The remaining solid

Reactions of (η5-C5Me5)2(C5H5N)TidS
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7833
Cp*2Ti(Cl)SCH(CHCH2)CHCH2 (15). 5-Chloro-1,3-pentadiene
(30 µL, 0.27 mmol) was added to a solution of 1 (98 mg, 0.23 mmol)
in benzene (7 mL). The reaction mixture was stirred for 4 h. The
volatile materials were removed, and the remaining solid was crystal-
lized from Et2O/pentane at -40 °C to yield 15 as a dark green solid
(68 mg, 65%). 1H NMR (C6D6): δ 6.22 (m, 2H), 5.24 (m, 2H), 5.08
(m, 2H), 4.64 (m, 1H), 1.88 (s, 30H) ppm. 13C NMR: δ 140.5 (CH),
125.8 (C-CH3), 112.8 (CH2), 55.5 (CH), 13.15 (C-CH3) ppm. IR
(C6D6): 2983, 2718, 1631, 1525, 1390, 911, 773. Anal. Calcd for
C25H37ClSTi: C, 66.30; H, 8.23. Found: C, 66.12; H, 8.32.
Cp*2Ti(F)CH2CHCH2 (16). 3-Fluoropropene (0.63 mmol) was
condensed into a bomb containing a frozen benzene solution of 1 (90
mg, 0.21 mmol). The solution was thawed and was allowed to stand
for 5 h. The volatile materials were removed to give a red solid that
was crystallized from ether/pentane at -40 °C to give 16 as dark red
crystals (49 mg, 54%). 1H NMR (C6D6): δ 6.16 (m, 1H), 5.22 (m,
1H), 4.98 (m, 1H), 3.52 (m, 2H), 1.84 (s, 30H) ppm. 13C NMR: δ
139.8 (CH), 124.1 (C-CH3), 112.6 (CH2), 41.9 (d, CH, JCF ) 7 Hz),
11.7 (C-CH3) ppm. 19F NMR: δ 20.4 ppm. IR (Nujol): 3072, 2719,
1629, 1207, 1019, 907, 734 cm-1. Anal. Calcd for C23H35FSTi: C,
67.30; H, 8.59. Found: C, 67.10; H, 8.58.
1214, 1159, 1020, 846, 653 cm-1. MS (EI): m/z 434 (M+). Anal.
Calcd for C25H38OSTi: C, 69.11; H, 8.81. Found: C, 69.14; H, 8.79.
Cp*2TiSCH(Ph)C(CH3)CHO (21). A small scintillation vial was
charged with a solution of 1 (148 mg, 0.36 mmol) in 5 mL of benzene.
R-Methyl-trans-cinnamaldehyde (72 µL, 0.52 mmol) was added via
syringe, and the solution was agitated with a pipet. After 30 min, the
volatile materials were removed, and the resulting green oil was
crystallized from pentane at -40 °C to give 70 mg of 21 (41%). 1H
NMR (C6D6): δ 7.63 (m, 2H), 7.29 (s, 1H), 7.24 (t, 2H), 7.09 (t, 1H),
5.18 (s, 1H), 1.90 (s, 15H), 1.85 (s, 15H), 1.43 (s, 3H) ppm. 13C
NMR: δ 158.3 (C(O)(H)), 146.6 (C), 129.5 (CH), 127.8 (CH), 126.7
(CH). 124.3 (C-CH3), 123.6 (C-CH3), 106.0 (C), 53.3 (CH), 17.5
(CH3), 12.3 (C-CH3), 11.9 (C-CH3) ppm. IR (cyclohexane): 3064,
3029, 2950, 2732, 1614, 1380, 1211, 1174, 709, 611 cm-1. MS (EI):
m/z 496 (M+). Anal. Calcd for C30H40OSTi: C, 72.56; H, 8.12.
Found: C, 72.68; H, 7.94.
X-ray Crystallographic Study of Cp*2Ti(SC(SiMe3)CHCH2) (5).
Crystals of 5 suitable for X-ray diffraction were obtained by slow
cooling of a pentane solution of 5 to -40 °C. A fragment of one of
these reddish lumps with the approximate dimensions of 0.15 × 0.20
× 0.27 mm was mounted on a glass fiber using Paratone N hydrocarbon
oil. All measurements were made on a Siemens SMART diffractometer
with a CCD area detector using graphite-monochromated Mo KR
radiation. All calculations were performed using the TEXSAN
crystallographic software package.
Cell constants and an orientation matrix obtained from a least-squares
refinement using the measured positions of 4907 reflections in the range
3.00 < 2θ < 45.00° corresponded to a primitive monoclinic cell with
dimensions given in the Supporting Information. The final cell
parameters and specific data collection parameters for this data set are
given in Table 1 or as Supporting Information. The systematic absences
of h01, 1 * 2n, and 0k0, k * 2n, uniquely determined the space group
to be P21/c (No. 14). Data were collected at a temperature of -100 (
1 °C.
Data were integrated using SAINT with box parameters of 2.0 ×
2.0 × 0.6 deg to a maximum 2θ values of 51.7°. The data were
corrected for Lorentz and polarization effects. No decay correction
was applied. An empirical absorption correction based on the measure-
ment of redundant and equivalent reflections and an ellipsoidal model
for the absorption surface was applied using XPREP (Tmax ) 0.90, Tmin
) 0.83). The 11 074 integrated reflections were averaged to yield 4534
unique reflections (Rint ) 0.042). The structure was solved by direct
methods and expanded using Fourier techniques. All non-hydrogen
atoms were refined anisotropically. Hydrogen atoms were included at
calculated or in located positions but not refined. The final cycle of
full matrix least-squares refinement was based on 2713 observed
reflections (I > 3.00σ(I)) and 253 variable parameters and converged
(largest parameter shift was 0.01 times its esd) with unweighted and
weighted agreement factors of R ) 0.049 and Rw ) 0.056. The standard
deviation of an observation of unit weight was 1.89. The weighting
scheme was based on counting statistics and included a factor (p )
0.030) to downweight the intense reflections. The maximum and
minimum peaks on the final difference Fourier map corresponded to
0.77 and -0.55 e/Å3, respectively.
Cp*2Ti(Br)SCH2CHCH2 (17). 3-Bromopropene (26 µL, 0.29
mmol) was added via syringe to a vial containing a stirred solution of
1 (115 mg, 0.26 mmol) in benzene (5 mL). The solution color changed
from red to green, and the solution was allowed to stir for 10 min. The
volatile materials were removed to give a blue-green solid that was
crystallized from ether/pentane at -40 °C to give 17 as dark blue
crystals (85 mg, 67%). 1H NMR (C6D6): δ 6.13 (m, 1H), 5.18 (m,
1H), 4.95 (m, 1H), 3.82 (m, 2H), 1.91 (s, 30H) ppm. 13C NMR: δ
138.7 (CH), 125.4 (C-CH3), 113.4 (CH2), 48.2 (CH2), 13.1 (C-CH3)
ppm. IR (Nujol): 3091, 2717, 1633, 1198, 1018, 908, 808, 732 cm-1
.
Anal. Calcd for C23H35BrSTi: C, 58.61; H, 7.48. Found: C, 58.61;
H, 7.70.
Cp*2Ti(I)SCH2CHCH2 (18). 3-Iodopropene (18 µL 0.20 mmol)
was added via syringe to a vial containing a stirred solution of 1 (80
mg, 0.19 mmol) in benzene (5 mL). The solution color changed from
red to green, and the solution was allowed to stir for 10 min. The
volatile materials were removed to give a blue-green solid that was
crystallized from ether/pentane at -40 °C to give 18 as dark blue
crystals (71 mg, 73%). 1H NMR (C6D6): δ 6.15 (m, 1H), 5.18 (m,
1H), 4.95 (m, 1H), 3.90 (m, 2H), 1.97 (s, 30H) ppm. 13C NMR: δ
138.7 (CH), 125.4 (C-CH3), 113.9 (CH2), 49.6 (CH2), 14.2 (C-CH3)
ppm. IR (Nujol): 3079, 2704, 1631, 1199, 1020, 908, 742, 594 cm-1
.
Anal. Calcd for C23H35ISTi: C, 53.29; H, 6.81. Found: C, 53.07; H,
6.81.
Cp*2Ti(OSO2Tol)SCH2CHCH2 (19). Allyl tosylate (57 µL, 0.30
mol) was added to a vial containing a stirred solution of 1 (119 mg,
0.28 mmol) in benzene (10 mL). The solution was stirred for 3 d at
20 °C. The volatile materials were removed to give a brown oily solid
that was washed with 5 mL of Et2O and pentane to give 19 as a brown
solid (87 mg, 56%). 1H NMR (C6D6): δ 7.97 (d, 2H, J ) 8.1 Hz),
6.89 (d, 2H, J ) 8.0 Hz), 6.45 (m, 1H), 5.35 (m, 1H), 5.10 (m, 1H),
4.72 (m, 2H), 1.96 (s, 3H), 1.90 (s, 30H) ppm. 13C NMR: δ 143.9
(C), 140.3 (C), 140.2 (CH), 128.8 (CH), 127.4 (C-CH3), 127.3 (CH),
114.5 (CH2), 44.1 (CH2), 21.1 (CH3), 13.0 (C-CH3) ppm. IR (C6H6):
2900, 2829, 1783, 1670, 1396, 1164, 1058, 874, 858 cm-1. Anal. Calcd
for C30H42S2O3Ti: C, 64.04; H, 7.52. Found: C, 64.28; H, 7.85. The
X-ray Crystal Structure of Cp*2Ti(Cl)SC(CH3)2CHCH2 (14).
Green crystals of 14 suitable for X-ray diffraction were grown by slow
cooling of a toluene/pentane solution of 14 to -40 °C. A block-shaped
crystal having approximate dimensions of 0.10 × 0.11 × 0.17 mm
was mounted as described above. Data collection and calculations were
also performed as described above. The specific data collection
parameters are given in Table 1 or as Supporting Information. The
space group was determined to be P21/c (No. 14). Cell constants and
an orientation matrix were obtained from 4509 reflections. Data were
integrated as described above using box parameters of 1.6 × 1.6 ×
0.6 deg to a maximum 2θ value of 52.1°. The 11 366 integrated and
2
2H NMR spectra of the H-substituted analogue (which was prepared
using 56 mg of 1 and the procedures described above) in C6H6 showed
only a singlet at 4.7 ppm. The 13C{1H} spectrum of this compound
was identical to that of the undeuterated analogue except the singlet at
44 ppm was replaced by a 1:1:1 triplet.
Cp*2TiSCH(CH3)C(CH3)CHO (20). trans-2-Butenal (50 mg, 0.52
mmol) was added via syringe to a stirred solution of 1 (150 mg, 0.35
mmol) in benzene. After 2 h the volatile materials were removed and
the remaining brown solid was crystallized from a toluene/pentane
corrected reflections were averaged to yield 4378 unique data (Rint
)
1
solution at -40 °C to give 128 mg of 20 (84%) as a purple solid. H
0.057). The final cycle of full-matrix least-squares refinement was
based on 2416 observed reflections and converged (largest parameter
shift was 0.45 times its esd) with unweighted and weighted agreement
factors of R ) 0.049 and Rw ) 0.055. The standard deviation of an
observation of unit weight was 1.65 using the weighting scheme
NMR (C6D6): δ 7.25 (s, 1H), 4.15 (q, J ) 6.3 Hz, 1H), 1.89 (s, 15H),
1.85 (s, 15H), 1.70 (d, J ) 6.3 Hz, 3 H), 1.67 (s, 3H) ppm. 13C NMR:
δ 157.5 (CH), 122.0 (C-CH3), 43.2 (CH), 27.3 (CH3), 16.1 (CH3),
12.8 (C-CH3), 12.5 (C-CH3) ppm. IR (Nujol): 2723, 1687, 1612,

7834 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Sweeney et al.
described above. The maximum and minimum peaks on the final
difference Fourier map corresponded to 0.36 and -0.36 e/Å3. All non-
hydrogen atoms were refined anisotropically. Hydrogen atoms were
included at calculated or in located positions but not refined.
0.36 and -0.20 e/Å3. All non-hydrogen atoms were refined anisotro-
pically. Hydrogen atoms were included at calculated or in located
positions but not refined.
X-ray Crystal Structure of Cp*2Ti(SCH(Ph)C(Me)CHO) (21).
Black crystals of 21 suitable for X-ray diffraction were grown by slow
cooling of a toluene/pentane solution of 21 to -40 °C. A fragment of
a block-shaped crystal having approximate dimensions of 0.16 × 0.18
× 0.21 mm was mounted as described above. Data collection and
calculations were performed as described for the X-ray analysis of 5.
The space group was determined to be P21/c (No. 14). The specific
data collection parameters are given in Table 1 or as Supporting
Information. Cell constants and an orientation matrix were obtained
from 6064 reflections. Data were integrated as described above using
box parameters of 1.8 × 1.8 × 0.6 deg to a maximum 2θ value of
46.5 °. The 10 623 integrated and corrected reflections were averaged
to yield 3989 unique data (Rint ) 0.038). The final cycle of full-matrix
least-squares refinement was based on 2635 observed reflections and
converged (largest parameter shift was 0.02 times its esd) with
Acknowledgment. We thank the National Institutes of
Health for generous financial support of this work (Grant No.
GM-25459 to R.G.B.), Dr. F. J. Hollander for solving the crystal
structures of 5 and 21, Dr. Ryan Powers for solving the crystal
structure of 14 (both at CHEXRAY, the U. C. Berkeley X-ray
diffraction facility), and a reviewer for a careful reading of the
manuscript.
Supporting Information Available: Tables giving details
of the data collection, data reduction, and structure solution and
refinement, positional and thermal parameters, bond lengths,
bond angles, and torsion angles for 5, 14, and 21 and a table
and figure presenting data from the kinetic study of the reaction
between 1 and allyl chloride (22 pages, print/PDF). See any
current masthead page for ordering information and Web access
instructions.
unweighted and weighted agreement factors of R ) 0.033 and Rw
)
0.040. The standard deviation of an observation of unit weight was
1.47 using the weighting scheme described above. The maximum and
minimum peaks on the final difference Fourier map corresponded to
JA980877M