Regioselectivity of metallacycle formation via intramolecular CH activation

Article abstract of DOI:10.1016/S0277-5387(00)00325-9

Carbometalation of alkynes by dimethyltitanocene affords vinyl complexes which serve as intermediates for the formation of titanacyclic products through subsequent hydrogen abstraction/methane elimination. Although multiple sites for hydrogen abstraction

Full text of DOI:10.1016/S0277-5387(00)00325-9

www.elsevier.nl/locate/poly
Polyhedron 19 (2000) 879–890
Regioselectivity of metallacycle formation via intramolecular
CH activation
*
Kenneth M. Doxsee , Jerrick J.J. Juliette
Department of Chemistry, University of Oregon, Eugene, OR 97403, USA
Received 6 July 1999; accepted 26 January 2000
Abstract
Carbometalation of alkynes by dimethyltitanocene affords vinyl complexes which serve as intermediates for the formation of titanacyclic
products through subsequent hydrogen abstraction/methane elimination. Although multiple sites for hydrogen abstraction are present in the
intermediate vinyl complexes, the hydrogen abstraction step is in most cases highly regioselective, leading to the formation of either
titanacyclobutenes (via g-H abstraction) or titanacyclopropanes (via b-H abstraction). Rotational isomerization, restricted in the case of
larger vinyl group substituents, provides a simple and consistent explanation for the observed regioselectivities, and this effect is sufficiently
pronounced to lead to preferential g-aliphatic CH activation even in the presence of g-aromatic hydrogens. The titanacyclopropanes may be
alternatively formulated as allene complexes of titanocene. An independently prepared trimethylphosphine-ligated analog of 3-methyl-1,2-
butadiene, representing the first titanocene allene complex to be reported, undergoes reductive coupling reactions with excess 3-methyl-1,2-
butadiene and with alkynes.
q2000 Elsevier Science Ltd All rights reserved.
Keywords: C–H activation; Intramolecular bonds; Titanium; Metallacycles; Allenes; Carbometalation
1. Introduction
A sufficient body of information regarding intramolecular
CH-activation reactions has been amassed to allow some
generalizations to be drawn. Aromatic CH bonds appear to
be more reactive toward such processes than aliphatic CH
bonds, despite the higher bond strength of the aromatic CH
bond. Thus, for example, (o-tolyl)(methyl)zirconocene(1)
undergoes aromatic b-CH bond activation to produce an
aryne complex rather than the isomeric benzometalla-
cyclobutene which would be formed through activation of
a benzylic g-CH bond (Eq. (1)) [13]. Dibenzylhafnocene
Reactions involving the insertion of transition metals into
carbon–hydrogen bonds continue to receive considerable
attention, at least in part given their potentialfortheactivation
of hydrocarbons toward catalytic functionalization [1].
Numerous examples of both intermolecular and intramolec-
ular CH bond activation have been reported, including the
well known a- and b-hydride elimination processes [2,3]
and related cyclometalation reactions of heteroatom-
containing ligands [4–6]. Longer-range intramolecular CH
activation reactions have also been documented. Thus,
orthometalation reactions of arylphosphine [4–7] and aryl-
silane [8] complexes represent g-elimination processes, and
thermolysis of a Ta(V) aryloxide complex is reported to
proceed via sequential ´-elimination processes [9,10]. For
recent examples of d- and ´-activation processes in aniline,
benzylamine and phenethylamine complexes of palladium,
see [11,12]. These longer-range activation reactions, how-
ever, remain uncommon in comparison with the seemingly
ubiquitous a- and b-hydride elimination processes.
(1)
complex 2 undergoes kinetic a-elimination, while the result-
ing ‘tucked in’ benzyl complex then reacts through activation
of an aromatic g-CH bond, forming abenzohafnacyclobutene
(Eq. (2)) [14]. Similarly, Petasis and Bzowej have noted
apparent a-elimination reactions of dibenzyltitanocene and
related derivatives [15,16].
* Corresponding author. Tel.: q1-541-346-4628; fax: q1-541-346-2874;
e-mail: doxsee@oregon.uoregon.edu
0277-5387/00/$ - see front matter q2000 Elsevier Science Ltd All rights reserved.
PII S0277-5387(00)00325-9
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

880
K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
b-activation process and the other a g-activation process
[24]. Herein we present further studies of these competing
allylic CH activation processes as well as an unusual example
wherein allylic g-CH activation is strongly favored over aro-
matic g-CH activation.
2. Results and discussion
2.1. Selective g-activation processes
Diphenylacetylene and bis(trimethylsilyl)acetylene react
cleanly with dimethyltitanocene in dry, deoxygenated ben-
zene at moderate temperatures (65–808C, 14–30 h) to pro-
duce the corresponding titanacyclobutenes (3a, RsPh; 3b,
RsSiMe3) in high yield (Eq. (5)) [22,25]. Although these
(2)
Teuben and co-workers have reported selective a-, b-, and
d-hydrogen elimination reactions of paramagnetic vana-
dium(III) dialkyls [17] (e.g., Eq. (3)) and several a- and
(5)
metallacycles may also be prepared through the addition of
the alkynes to Cp2TiCH2PAl(CH3)2Cl (Tebbe’s reagent)
[26], removal of the aluminum-containing byproducts can
be problematic, reducing isolated yields of pure metallacy-
cles. In contrast, thedimethyltitanocenerouteallowsisolation
of pure metallacycles in near-quantitative yield after a simple
recrystallization while also avoiding the requirement for use
of the highly air- and moisture-sensitive Tebbe’s reagent. (In
marked contrast, dimethyltitanocene may be isolatedthrough
standard bench top techniques, requiring no exclusion of air
or moisture.)
(3)
b-hydrogen elimination reactions of permethyltitanocene
alkyls [18] (Eq. (4)). However, information is scarce
Shorter reaction times in the diphenylacetylene reaction
lead to formation of carbometalation product 4, together with
a small amount of the titanacyclobutene (Eq. (6)). Complex
(4)
(6)
regarding the regiochemical preferences of intramolecular
CH activation processes in which there is a lessdefinedchem-
ical difference between two (or more) potentially reactive
sites, with the selective g-elimination reactions of neopentyl
platinum complexes providing perhaps the best known exam-
ples [19,20]. Interestingly, whereas the neopentyl platinum
complexes undergo selective g-elimination, tetraneopentyl-
titanium has recently been shown to undergo selective a-
elimination [21].
In the course of our studiesofthecarbometalationreactions
of dimethyltitanocene [22], which contrast with an early
account reporting no thermal reaction between dimethyl-
titanocene and alkynes [23], we were afforded the opportu-
nity to examine the intramolecular competition between two
alternative sites for allylic CH activation, one representing a
4, which may be isolated as a pure solid by crystallization at
y208C, serves as a precursor to titanacyclobutene 3a, with
thermolysis at ca. 808C resulting in clean conversion to the
metallacycle.
The allylic methyl group resonance appears at 1.48 ppm
in the H NMR spectrum of 4, while the Ti–CH3 group
appears at y0.42 ppm. Similar upfield shifts are seen for
other methyltitanocene complexes and may be attributed to
an agostic interaction with a vacant titanium wedge plane
orbital [27,28]. In the 13C NMR spectrum of 4, the allylic
methyl resonance appears at 25.7 ppm and the Ti–CH3 group
at 50.3 ppm. High resolution mass spectral and elemental
analytical data provide final confirmation of the proposed
structure.
1
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
881
Both the carbometalation reaction and the subsequent
methane elimination process are noteworthy. Carbometala-
tion reactions of group 4 metallocene alkyl complexes are
often sluggish, with formation of cationic alkyls typically
used to accelerate the reactions [29,30]. Insertion reactions
of dimethyltitanocene havescarcelybeenexamined,although
CO2 and SO2 insertions have been reported [31,32], and NO
insertion reactions of dimethylzirconocene have also been
noted [33]. The carbometalation of diphenylacetylene by
dimethyltitanocene to form 4 is quite rapid. The reasons for
the facility of this reaction are not clear, but may be rooted
in the propensity of methyltitanocene derivatives to undergo
Ti–CH3 bond homolysis [34], with addition of CH3x to the
alkyne, followed by addition of the resulting vinyl radical to
titanium providing a convenient pathway to the observed
product.
this result, given the possibility of slow exchange of bis(4-
chlorophenyl)acetylene into the coordination sphere of the
metal center. Further information regarding these mechanis-
tic possibilities is provided in the discussion of the reactions
of 2-butyne and 3-hexyne with dimethyltitanocene (vide
infra).
Higher reaction temperatures in the thermolysis of 4 do
lead to the formation of the isomeric titanacyclobutene (5)
resulting from the less favorable aromatic CH activationstep,
although this metallacycle is formed in only low yield (up to
ca. 25% at 1108C), with the allylic activation product still the
major product of the reaction (Eq. (8)).
Vinyl complex 4 was previously reported as the minor
byproduct in the photochemical reaction between dimethyl-
titanocene and diphenylacetylene, the majorproductofwhich
was a titanacyclopentadiene formed through the reductive
coupling of 2 equiv. of the alkyne by transiently generated
titanocene (Eq. (7)) [35–37]. Formation of4inthisreaction
(8)
Given the anticipated steric preference of the vinyl group
to lie in the wedge plane, two rotamers of 4 might be antici-
pated, one (4a) placing the g-methyl group in close prox-
imity to the departing Ti–CH3 group and the other (4b)
placing the aromatic g-CH in this position (Eq. (9)). With
(7)
perhaps provides circumstantial support for the radical path-
way for its formation, given the likelihood of generation of
methyl radicals under photolytic conditions, although direct
thermal insertion via s-bond metathesis [38] is not ruled out.
Given the generally preferred reactivity of aromatic CH
bonds relative to aliphatic CH bonds, the formation of the
titanacyclobutene from 4 is intriguing, especially since pref-
erential activation of the g-aliphatic position occurs in the
presence of a g-aromatic CH bond. This may point toward a
radical mechanism for the CH activation process, with initial
Ti–CH3 bond homolysis followed by preferential abstraction
of an allylic hydrogen by the resulting methyl radical and
subsequent collapse to the metallacyclobutene.
Formation of the titanacyclobutene could conceivably be
initiated by the thermal extrusion of diphenylacetylene from
4, regenerating dimethyltitanocene. Thermal decomposition
of dimethyltitanocene can serve to generate Cp2TisCH2
[39,40], although this decomposition reaction is appreciably
more complex than that of the permethyltitanocene analog
[41], and this methylidene complex, based on the known
chemistry of Tebbe’s reagent and of titanacyclobutanes
[26,42], would be expected to react with diphenylacetylene
to form the titanacyclobutene. In order to address this possi-
bility, vinyl complex 4 was thermolyzed at 908C in the pres-
ence of an excess of bis(4-chlorophenyl)acetylene.Complex
4 is cleanly converted to diphenyltitanacyclobutene 3a, pro-
viding support for the g-CH activation process, although the
alternate mechanism cannot be unequivocally ruled out by
(9)
either concerted methane elimination or with Ti–CH3 bond
homolysis followed by hydrogen abstraction, rotamer 4a
would be expected to lead to formation of the titanacyclo-
butene (3a) and rotamer 4b to the benzotitanacyclobutene
(5). Restricted rotation of vinyl groups in permethyltitano-
cene chloride derivatives has been previously noted [43].
Although steric effects can be less pronounced in simple
titanocenes than in the permethyl analogs, they are nonethe-
less often profound. For example, steric effects play an
extreme role in the nitrile and aldehyde insertion reactions of
titanacylobutenes [44,45]. It seems likely that the initial car-
bometalation step in the reaction of dimethyltitanocene with
diphenylacetylene affords a single rotamer (4a), presenting
a g-H poised for abstraction and leading to the diphenyl-
titanacyclobutene. Thermolysis at higher temperatures facil-
itates rotational equilibration of 4a with 4b, leading to the
formation of observable quantitiesofbenzotitanacyclobutene
5. Consistent with this hypothesis is the recent report of g-
elimination from (Z)-CpNi(CPh_CPh–CH3), which deliv-
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

882
K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
ers the g-hydrogen close to the nickel center, but not from
the (E) isomer, which maintains a longer distance between
the nickel center and the g-hydrogens [46].
The reaction between dimethyltitanocene and bis(tri-
methylsilyl)acetylene (Eq. (5), RsSiMe3) affords no spec-
trally observable intermediates, even when carried out at
comparatively low temperature (358C), but presumably pro-
ceeds via an initial carbometalation step, affording vinyl
complex 6 (Eq. (10)). An allylic CH activation process
mass spectral and elemental analytical data are also suppor-
tive of this structure. Particularly revealing are two 13C res-
onances in the downfield region of 180–195 ppm, highly
characteristic of vinylic carbons attached to titanium [51].
In order to provide additional information about the for-
mation of 9, the reaction of 2-butyne with Cp2Ti(CD3)2 was
examined. The 1H and 13C NMR spectra of 9-d3 are virtually
identical to those of 9, with the exception of the loss of the
highest field 1H methyl resonance (0.8 ppm) and lowest field
13C methyl resonance (24.8 ppm). (Loss of 13C signal inten-
sity for CD3 groups due to spin–spin splitting is well known
[52–54].) These chemical shifts unambiguously define the
labeled methyl group as the isopropylidene methyl cis to the
titanocene fragment [55]. An agostic interaction between
this methyl group and the empty titanium wedge plane orbital
(10)
1
seems likely, with the opposing shifts of the H and 13C
analogous to that seen for the diphenyl analog would then
close up the titanacyclobutene ring. Again, a high degree of
regiocontrol is apparent in this CH activation process. The
isomeric metallacycle(7)whichwouldresultfromactivation
of a silyl methyl group (Eq. (10)) is not observed, even
though such reactivity of a silyl methyl group is well prece-
dented for the later transition metals [47–49] and has been
reported for a bis(trimethylsilyl)amide complex of titano-
cene as well [50]. As in the case of the diphenyl derivative,
this is consistent with initial formation of a single reactive
rotamer of the intermediate vinyl complex which displays
significant steric inhibition of bond rotation. In contrast to
the diphenyl case, in which formation of the anticipated iso-
meric titanacycle (5) is observed at higher reaction temper-
atures, rotation of the vinyl group in the bis(trimethylsilyl)
derivative (6) is sufficiently hindered to preclude formation
of detectable quantities of the silatitanacyclobutane (7) even
at the highest temperatures examined (ca. 1208C, in toluene-
d8 in a sealed NMR tube).
resonances consistent with other reports of such agosticinter-
actions [27,28].
In addition to allowing for the assignment of the outlying
NMR resonances to this cis-methyl group, the deuterium
labeling study also provides compelling information regard-
ing the mechanism of formation of 9. Thermal elimination of
methane from dimethyltitanocene, generating Cp2Ti_CH2,
may be virtually unequivocally ruled out as a mechanistic
step, since neither a CD2 group nor a CD2H group, which
might potentially be formed from a CD2-containing inter-
mediate, is found in 9-d3.
Given the mechanistic significance of the formation of 9,
its structure was further verified through protonolysis and
analysis of the resulting diene 10 (Eq. (12)). (E)-2,5-di-
(12)
2.2. Competitive b- and g-activation processes
methylhepta-2,5-diene is formed with no loss of stereochem-
istry, being obtained as the sole volatile component of the
protonolysis reaction (as demonstrated by gas chromato-
graphic and NMR spectral analysis). Although two methyl
resonances are degenerate in the 1H NMR spectrum, all nine
carbons in 10 are unambiguously resolved in the 13C NMR
spectrum. As in the 13C NMR spectrum of its titanacyclic
precursor, one of the methyl resonances is rather downfield
(25.9 ppm) from the others. In this case, simple correlation
tables point to the assignment of this resonance to the (E)-
methyl of the isopropylidene group, shifted to lower field
than the (Z)-methyl group due to the g effect [56,57]. Pro-
tonolysis of the deuterated titanacycle, 9-d3, affords diene
10-d3, displaying 1H and 13C NMR spectral properties
completely consistent with the location of the CD3 group
anticipated from the proposed structure of its titanacyclic
precursor.
2-Butyne is partially polymerized in the presence of dime-
thyltitanocene at 658C, and use of only 1 equiv. of this alkyne
thus leads to rather intractable mixtures of organic and organo-
metallic products as well as unreacted dimethyltitanocene.
Use of two or more equivalents of 2-butyne, however, leads
to much cleaner reaction chemistry. Unexpectedly, the dom-
inant product ()90%) of this reaction is not the titanacyclo-
butene (8), but rather a new metallacyclic complex (9)
incorporating 2 equiv. of 2-butyne (Eq. (11)).
(11)
The new complex is easily purified through recrystalliza-
tion from hexanes at y308C. 1H and 13C NMR spectra pro-
vide unambiguous proof of the structure of complex 9, and
Dimethyltitanocene displays analogous reactivity toward
3-hexyne, but the titanacyclobutene is formed in appreciably
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
883
higher yield in this reaction than in that of 2-butyne. Again,
2.3. Metallacyclopropane (allene complex) intermediates
methane is formed as a byproduct and two metallacyclic
complexes, one incorporating a second equivalent of 3-hex-
yne, are obtained (Eq. (13)).
The titanacyclopropane intermediate may be represented
in an alternative resonance form as an allene complex of
titanocene[59,60]. Significantly, anintermediateisobserved
by NMR spectroscopy in both the 2-butyne and 3-hexyne
reactions which displays a cyclopentadienyl 1H resonance at
ca. 4.7 ppm, a chemical shift similar to that displayed by
isolable titanocene(II) derivatives, including Cp2Ti(PMe3)2
(4.51 ppm) [61] and Cp2Ti(CO)2 (4.58 ppm) [62], and a
cyclopentadienyl 13C resonance at 98 ppm (1JCHs168 Hz),
in range with that reported for Cp2Ti(PMe3)2 (91.2 ppm).
Observation of these NMR resonances is consistent with
intermediate formation of a titanocene(II) complex. While
this observation does not unequivocally point to the allene
complex as the observed intermediate, given the possibility
of generation of other titanocene(II) species under the reac-
tion conditions, the close resemblance of these spectral fea-
tures to those of an independently prepared allene complex
(vide infra) is noteworthy.
(13)
The most characteristic 1H NMR spectral features of
titanacyclopentene 12 are a methine quartet at 2.83 ppm cou-
3
pled to a methyl doublet (1.16 ppm, JHHs6 Hz), a high-
field methyl singlet (0.65 ppm), and two cyclopentadienyl
resonances, clearly indicating the lack of a plane of symmetry
in the complex. The 13C NMR spectrum of 12 displayssimilar
features to that of 2-butyne-derived complex 9, most notably
two downfield singlets (192.6, 193.0 ppm) for the sp2-car-
bons attached to the titanium center, with the other vinyl
carbons appearing near 132 ppm; assignment of resonances
was facilitated by use of the DEPT (distortionless enhance-
ment by polarization transfer) technique [58]. The methine
carbon appears as a doublet (1JCHs125 Hz) at 41.6 ppm.
Formation of both titanacyclobutenes and titanacyclopen-
tenes in the reaction of dimethyltitanocene with alkynes is
conveniently rationalized by initial alkyne carbometalation,
followed by competitive allylic g-CH activation, affording
the titanacyclobutene, and b-CH activation, affording a
titanacyclopropane (Scheme 1). Rotamer isomerization,
affording access to elimination from the b-allylic position, is
more facile in these cases than in the more hindered phenyl
and trimethylsilyl derivatives. The product of this b-activa-
tion becomes dominant in the case of the 2-butyne reaction,
while the increased steric bulk of the 3-hexyne derived inter-
mediate allows for appreciable formation of the g-activation
product. Whereas the titanacyclobutenes are comparatively
unreactive toward alkynes, the titanacyclopropanes undergo
facile alkyne insertion, with steric issues directing the incom-
ing alkyne to the observed insertion site rather than to the
more crowded site, which would form unobserved isomer 13
(Scheme 1).
While this work was in progress, Jones and co-workers
reported related chemistry in the zirconocene series [63].
Thermolysis of bis(methylcyclohexenyl)zirconocene in the
presence of trimethylphosphine affords a 1,2-cyclohexadiene
complex of zirconocene through a b-activation process (Eq.
(14)). No evidence was reported for the formation of the
(14)
isomeric zirconacyclobutene which would arise from com-
petitive g-activation. The allene complex reacts with un-
saturated substrates (e.g. diphenylacetylene) to afford
zirconacyclic complexes analogous to 9 and 12.
Given the likelihood of the intermediacy of allene com-
plexes in the reactions of dimethyltitanocene with 2-butyne
and 3-hexyne, the possibility of additional isomeric products
is raised by the ability of allene complexes to undergo coor-
dination isomerization [64](for discussions of related issues
for ketene complexes see Ref. [65]), as suggested in Scheme
2. In each case, however, the isomeric allene complex would
Scheme 1.
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

884
K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
Scheme 2.
be expected to be sterically destabilized due to additional
interactions between the cyclopentadienyl rings and the
allene substituents.
Attempted isolation of 14 through removal of solvent at
higher temperature (258C instead of y58C) results in its
partial conversion to the metallacyclopentane 15. Treatment
of Cp2Ti(PMe3)2 with 2 equiv. of 3-methyl-1,2-butadiene
permits the preparation of 15 in pure form. An agostic inter-
action between the methyl groups cis to the titanium center
appears likely given the appearance of one methyl resonance
The possibility that titanacyclopentene 9 arises from a fol-
low-up reaction of titanacyclobutene 8 was ruled out by
examining the reactivity of independently prepared dime-
thyltitanacyclobutene [26,66,67] toward 2-butyne. No spec-
tral evidence was obtained for the formation of 9, and the
only noticeable reaction after 24 h at 708C was the partial
decomposition of the titanacyclobutene. The appearance of
1
in the H NMR spectrum of 15 at appreciably higher field
(0.75 ppm) than the other (1.48 ppm). Reductive coupling
of allenes by zirconocene(II) has been noted previously
[69,70], and a simple titanium(II) complex, (iPrO)2Ti(1-
propene), has very recently been reported to effectanalogous
allene coupling [71]. Allenes have also recently been incor-
porated successfully as reactants in the Pauson–Khand reac-
tion [72]; although mechanistically unrelated, also note-
worthy is the reductive coupling of allenes and alkynes medi-
ated by hydrozirconation [73].
1
broad peaks in the allyl region of the H NMR spectrum is
suggestive of the polymerization of 2-butyne by the titana-
cyclobutene, supportive of earlier proposals regarding the
possible intermediacy of metallacyclobutenes in certain cat-
alytic alkyne polymerization mechanisms [68].
In order to establish more firmly the titanacyclopropane
(allene complex) intermediate, the reaction between
Cp2Ti(PMe3)2 and 3-methyl-1,2-butadiene was examined.
Two metallocene products are obtained, the allene complex
14 and titanacyclopentane 15, incorporating 2 equiv. of the
allene (Eq. (15)).
Importantly, allene complex 14 reacts readily with 2-
butyne (1 h at 658C) to form titanacyclopentene9(RsCH3)
(Eq. (16)), providing circumstantial evidence for the inter-
(16)
(15)
mediacy of an allene complex in the reaction between 2-
butyne and dimethyltitanocene. Diphenylacetylene also
couples efficiently with allene complex 14 to form
titanacyclopentene 16 (RsPh) (Eq. (16)), suggesting
potential synthetic applications in the form of crosscouplings
of unsaturated substrates [71].
When the reaction iscarriedoutinthepresenceofanexcess
of trimethylphosphine, complex 14, apparently representing
the first example of a titanocene allene complex (the firstand,
prior to our report, only preparation of a group 4 allene com-
plex was that by Jones and co-workers [63]), is obtained
cleanly and may be isolated in analytically pure form by
simple evaporation of solvent and excess phosphine at ca.
y58C. In contrast to vinyl complex 4 and metallacycles 9
and 12, for which NMR analysis suggests an agostic inter-
action between a proximal methyl group and the titanocene
2.4. Reaction kinetics
Kinetic analysis of the 2-butyne and 3-hexyne reactions is
complicated by non-linearity at longer reaction times, at least
in part due to the thermal decomposition of the titanacyclo-
butenes, but nonetheless qualitatively reveals some signifi-
cant features of the reactions. First order consumption of
dimethyltitanocene is observed at all temperatures examined
1
center, the methyl H (2.02 and 2.41 ppm) and 13C (24.2
and 29.7 ppm) resonances of complex 14 appear at rather
typical chemical shifts. Clearly, the trimethylphosphine
ligand in 14 overrides the weaker agostic interaction present
in 4, 9, and 12.
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
885
(ranging from 41 to 1208C). Interestingly, the rate of disap-
pearance of dimethyltitanocene is significantly greater in the
2-butyne reaction than in the 3-hexyne reaction (a five-fold
difference at 638C), providing further circumstantial evi-
dence against rate-determining formation of Cp2Ti_CH2 as
a mechanistic step.
nation for the observed regioselectivity in the subsequent
hydrogen abstraction/methane elimination step. Intermedi-
ates bearing vinyl groups able to undergo rotation favor
formation of titanacyclopropanes through a b-activation
process, while those unable to undergo rotation undergo a g-
activation process, leading to titanacyclobutenes. This selec-
tivity is observed both for intermediate vinyl complexes
presenting chemically very similar b- and g-hydrogens (the
2-butyne and 3-hexyne reactions) and for those in which the
competing sites for hydrogen abstraction are inequivalent
(the diphenylacetylene and bis(trimethylsilyl)acetylene
reactions) and appears to override the frequently observed
preference for aromatic over aliphatic CH abstraction. Alter-
native formation of titanacyclobutenes through thermal
decomposition of dimethyltitanocene, affording Cp2Tis
CH2, appears inconsistent with labeling and kinetic studies
at moderate reaction temperatures, although it may appear as
a competitive side reaction at very high temperatures.
In contrast with the reaction of 2-butyne with dimethyl-
titanocene, which affords low yields of titanacyclobutene 8
together with titanacyclopentene 9 (Eq. (11)), the reaction
with Cp2Ti(CD3)2 affords only the titanacyclopentene at all
temperatures examined. This suggests a substantial kinetic
isotope effect, slowing the g-CH(D) activation process, con-
sistent with rate-determining CH(D) activation. A firm num-
ber for the isotope effect cannot be obtained due to the
complete disappearance of the titanacyclobutene product in
the Cp2Ti(CD3)2 reaction. However, given the assumption
that formation of )2% of the titanacyclobutene product
would be detectable by 1H NMR and ignoring any secondary
isotope effects, a lower limit of kg-H/kg-Ds3.2 may be cal-
culated. Interestingly, Janssens et al. have recently noted a
similar dramatic change in reaction pathway for neopentyl
groups on Pt(111) surfaces, with exclusive a-elimination
seen for the protio derivative and exclusive g-elimination for
the a,a-d2 derivative [74].
The titanacyclopropanes may be alternatively formulated
as allene complexes of titanocene. An independently pre-
pared trimethylphosphine-ligated analog of 3-methyl-1,2-
butadiene, representing the first titanocene allene complex to
be reported, undergoes reductive coupling reactions with
excess 3-methyl-1,2-butadiene and with alkynes.
Formation of the titanacyclopentene product in the 2-
butyne reaction was studied over a 508C temperature range.
An Eyring plot provides values for DH‡s12.9 kcal moly1
and DS‡sy28 eu, with the relatively large and negative
entropy of activation consistent with a fairly ordered transi-
tion state. This could point toward a concerted methane elim-
ination, requiring the proper alignment of Ti–CH3 and b-H
groups, but may also arise from rapid Ti–CH3 bond homo-
lysis, followed by rate-determining b-hydride abstraction.
The 3-hexyne reaction affords a unique opportunity to
examine the competition between b- and g-CH activation,
since products arising from both processes are obtained at all
reaction temperaturesstudied. AnEyringanalysisoftheprod-
uct ratio as a function of temperature allows determinationof
DDH‡b/gs9.6 kcal moly1 and DDS‡b/gsy8 eu. The posi-
tive DDH‡b/g is consistent withtheformationofamorehighly
strained titanacyclopropane via b-hydride activation versus
a lessstrainedtitanacyclobuteneviag-hydrideactivation.The
small but negative DDS‡b/g isin accord withamoreorganized
transition state to bring the ethyl group bearing the b-hydro-
gen into proper alignment for elimination than is required to
align the sterically less demanding methyl group bearing the
g-hydrogen.
4. Experimental
4.1. General
After purification and drying, all solids and solvents were
stored under nitrogen in a Vacuum Atmospheres glove box
equipped with a Dri-Train MO 40-1 inert gas purifier unit,
RGF-1 regeneration flow unit, Pedatrol pressure control unit,
DK-3E Dri-Kool refrigeration unit and Dri-Cold freezer.
Unless otherwise noted, all operations were carried out under
an atmosphere of argon and nitrogen, purified by passagefirst
through activated BASF catalyst, then through activated
molecular sieves.
4.2. Materials
Titanocene dichloride was purchased from Boulder Sci-
entific and used without purification. Diphenylacetylene,
bis(trimethylsilyl)acetylene, 3-hexyne, MeLi (1.4 M in die-
thyl ether), and CD3Li (1.5 M in diethyl ether) were used as
received from Aldrich. 2-Butyne was obtained from Farchan
and used as received. Benzene-d6 was purchased from MSD
and distilled under a pre-purified nitrogen atmosphere from
Na/K alloy and benzophenone. Toluene, benzene, THF and
diethyl ether were distilled from Na/K alloy and benzophe-
none under a pre-purified nitrogen atmosphere. Alkanes
(pentane and hexane) were washed with concentrated
H2SO4, then with aqueous NaHCO3, dried over calcium
hydride, and distilled from Na/K alloy and benzophenone
under a pre-purified nitrogen atmosphere. Trimethylphos-
3. Summary and conclusions
Alkynes undergo carbometalation by dimethyltitanocene,
affording vinyl complexes which serve as intermediates for
the formation of titanacyclic productsthroughtheelimination
of methane. Rotational isomerization, restricted in the case
of larger substituents, provides a simple and consistentexpla-
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

886
K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
phine (Aldrich) was stored in the glove box freezer and used
without purification. Bis(trimethylphosphine)titanocene
was prepared as described [61].
workers in the photochemical reaction between dimethylti-
tanocene and diphenylacetylene [35–37].) 1H NMR
(C6D6): d y0.42 (s, 3H, Ti–CH3), 1.45 (s, 3H, CH3), 5.84
(s, 10H, Cp), 6.72 (d, 2H, 3JHHs7.5 Hz, ArH), 6.82–6.87
(m, 4H, ArH), 6.91–7.01 (m, 4H, ArH). 13C{1H} NMR
(C6D6): d 25.7 (CH3), 50.3 (Ti–CH3), 112.5 (Cp), 123.7
(Ar), 123.9 (Ar), 124.9 (Ar), 127.3 (Ar), 127.5 (Ar),
128.3 (Ar), 135.5 (Ti–C_C), 146.4 (Ar), 147.2 (Ar),
189.6 (Ti–C_C). 13C NMR (C6D6): d 25.7 (q, 1JCHs125
4.3. Instrumentation
Nuclear magnetic resonance (NMR)spectraofallsamples
were obtained on a GE QE-300 (300.152 MHz, 1H; 75.481
MHz, 13C). The following notation is used in reporting NMR
data: (1) chemical shifts are reported in parts per million
(ppm); (2) multiplicity of signal where sssinglet, ds
doublet, tstriplet, msmultiplet, brsbroad, dtsdoublet of
triplets, br ssbroad singlet, etc; (3) coupling constants are
reported in Hertz (Hz). Chemical shifts are referenced to
residual solvent signals in benzene-d6 (d 7.150, 1H; d 128.00,
13C), chloroform-d (d 7.26, 1H; d 77.0, 13C) and toluene-d8
(d 2.09, 1H; d 21.4, 13C). Gas chromatographicanalyseswere
performed on a Shimadzu GC-9A with He as the carrier gas
with a flow rate of 50 ml miny1, using an FID detector and
an injection port temperature of 2258C. Mass spectroscopy
was performed at the University of Oregon Mass Spectrom-
etry Facility.
1
Hz, CH3), 50.3 (q, JCHs126 Hz, Ti–CH3), 112.5 (d,
1JCHs174 Hz, Cp), 123.7 (d, 1JCHs160 Hz, Ar), 123.9 (m,
Ar), 124.9 (d, 1JCHs163 Hz, Ar), 127.3 (d, 1JCHs161 Hz,
Ar), 127.5 (d, 1JCHs158 Hz, Ar), 128.3 (d, 1JCHs161 Hz,
Ar), 135.5 (s, Ti–C_C), 146.4 (s, Ar), 147.2 (s, Ar), 189.6
(s, Ti–C_C). HRMS. Calc. for C26H26Ti: 386.1514. Found:
386.1508. Anal. Calc. for C26H26Ti: C, 80.82; H, 6.78. Found:
C, 80.67; H, 6.86%.
4.6. Thermolysis of 4
Thermolysis of solutions of vinyl complex 4 in benzene-
d6 or toluene-d8 at 708C for 3 days resulted in the formation
of methane (1H NMR (C6D6): d 0.15) and clean conversion
to deep-red solutions of the titanacyclobutene (3a). Removal
of solvent in vacuo afforded the titanacyclobutene in quan-
titative yield, with spectral properties identical to those pre-
4.4. Preparation of dimethyltitanocene [39,40]
A Schlenk flask was charged with titanocene dichloride
(5.0 g, 20.1 mmol) and diethyl ether. To the red suspension,
cooled to y408C, was added dropwise 32.0 ml of a 1.4 M
solution of MeLi in diethyl ether (44.8 mmol, 2.23 equiv.).
The mixture was stirred for 80 min while slowly warming to
ca. 08C, affording an orange solution. After cooling to
y408C, 40 ml of water was cautiously added. The organic
phase was washed with water (2=25 ml) and brine (1=25
ml) and dried over MgSO4. Removal of the solvent on a
rotary evaporator afforded the crude product as an orange
solid (3.9 g, 95%). Recrystallization from ca. 15 ml of pen-
tane at y308C afforded crystalline product (3.5 g, 85%). 1H
1
viously reported [26,66,67]. H NMR (C6D6): d 3.45 (s,
2H, CH2), 5.67(s, 10H, Cp), 6.91–7.14(m, 8H, ArH),7.31–
7.33 (d, 2H, ArH). 13C{1H} NMR (C6D6): d 73.4 (CH2),
100.6 (Ti–C_C), 112.4 (Cp), 124.5 (Ar), 125.9 (Ar),
126.2 (Ar), 129.5 (Ar), 129.6 (Ar), 138.7 (Ar), 147.6
(Ar), 210.9 (Ti–C_C).
Prolonged thermolysis of vinyl complex 4 at higher tem-
peratures (28 h at 80–858C) leads to the formation of a new
product in up to a ca. 1:4 ratio to the titanacyclobutene.
Although this complex has not been isolated in pure form, its
1H NMR spectrum is consistent with the benzotitanacyclo-
NMR (C6D6): d 0.05 (s, 6H, CH3), 5.68 (s, 10H, Cp). 13
NMR: d 45.9 (q, JCHs124 Hz, CH3), 113.2 (dm,
1JCHs173 Hz, Cp). Dimethyltitanocene-d6 was prepared
analogously, using MeLi-d3 in place of MeLi.
C
butene 5. H NMR (C6D6): d 0.70 (s, 3H, CH3), 5.57 (s,
1
1
10H, Cp), 6.8–7.2 (m, ArH). Control experiments indicate
that complex 5 is not formed via thermolysis of the
titanacyclobutene.
4.5. Reaction of dimethyltitanocene with diphenylacetylene:
4.7. Direct preparation of titanacyclobutene 3a from
formation of vinyl intermediate 4
dimethyltitanocene
To aSchlenkflask containingasolutionofCp2TiMe2 (1.00
g, 4.80 mmol) in toluene (25 ml) was added diphenylace-
tylene (1.71 g, 9.59 mmol, 2 equiv.). The flask was wrapped
in aluminum foil to exclude light and heated at 728C for 5 h.
The solution underwent a color change from orange to deep-
red. The solvent was then removed in vacuo, affording a
brownish solid. This solid was mixed with 10 ml of hexanes
and cooled at y208C for 2 days, resulting in the precipitation
of vinyl complex 4, which was isolated as a reddish-brown
solid by filtration (400 mg, 22%). (This product appears
identical to the minor product obtained by Rausch and co-
Titanacyclobutene 3a may also be prepared without iso-
lation of the intermediate vinyl complex. Thus, Cp2TiMe2
(0.998 mg, 4.79 mmol) reacted with 0.902 g of diphenyl-
acetylene (5.06 mmol, 1.06 equiv.) in 10 ml of toluene to
afford, after heating for 16 h at 65–688C and removal of
solvent in vacuo, a dark-purple solid (1.815 g, 98%) com-
posed of titanacyclobutene 3a (ca. 80%) and vinyl complex
4 (ca. 20%). This material was redissolved in 10 ml of tol-
uene and heated for 12–14 h at 80–858C. Removal of solvent
in vacuo quantitativelyaffordedthecrudemetallacycle.Crys-
tallization of the crude product twice from hexanes afforded
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
887
the pure titanacyclobutene as a nicely crystalline, purplesolid
(60%).
equiv.) and heating at 708C for 24 h, then stirring at room
temperature for an additional 12 h. 1H NMR (C6D6): d 1.17
(s, 3H, CH3), 1.36 (s, 3H, CH3), 1.49 (s, 3H, CH3), 2.54
(br s, 2H, CH2), 5.94 (s, 10H, Cp). 13C{1H} NMR (C6D6):
d 18.4 (CH3), 20.3 (CH3), 20.9 (CH3), 42.0 (CH2), 113.8
(Cp), 119.7 (Ti–C_C), 125.0 (Ti–C_C), 185.5 (Ti–
C_C), 188.0 (Ti–C_C). Anal. Calc. for C19H24Ti: C, 75.99;
H, 8.06. Found: C, 75.88; H, 8.17%.
4.8. Preparation of bis(trimethylsilyl)titanacyclobutene (3b)
[26,66,67,75,76]
To a solution of 1.05 g of Cp2TiMe2 (5.0 mmol) in 30 ml
of toluene in a 100 ml Schlenk flask was added 0.83 g of
bis(trimethylsilyl)acetylene (1 equiv.). The flask was
wrapped in aluminum foil and heated in an oil bath at 808C
4.11. Protonolysis of titanacyclopentene 9
1
for 14 h, at which point H NMR analysis of an aliquot
showed clean conversion to the titanacyclobutene. Solvent
was removed in vacuo, yielding the product in a satisfactory
state of purity as an oil in quantitative yield. Crystallization
from hexanes at y308C afforded a dark-red solid (1.46 g,
80%). 1H NMR (C6D6): d 0.21 (s, 9H, Si(CH3)3), 0.32 (s,
9H, Si(CH3)3), 4.67 (s, 2H, CH2), 5.22 (s, 10H, Cp).
13C{1H} NMR (C6D6): d 1.49 (CH3), 2.80 (CH3), 107.6
(Cp), 108.8 (CH2).
A side-arm round bottom flask was charged with titana-
cycle 9 (500 mg, 2.77 mmol) and 0.75 ml of C6D6. Excess
dry HCl gas was gently introduced to the solution over the
course of 5 min, then the volatile portion of the reaction
mixture was isolated by vacuum transfer. Gas chromato-
graphic analysis on a 30 m=0.25 mm Carbowax column,
50–2108C at 108C miny1, revealed only a single component
(retention time 5.2 min) in addition to the solvent. (E)-2,5-
1
dimethylhepta-2,5-diene: H NMR (C6D6): d 1.54 (s, 3H,
4.9. Preparation of titanacyclopentene 9
CH3), 1.56 (s, 6H, 2=CH3), 1.66 (s, 3H, CH3), 2.71 (d,
2H, Js7.2 Hz, CH2), 5.23–5.33 (m, 2H, 2=sCH).
13C{1H} NMR (C6D6): d 13.5 (CH3), 15.9 (CH3), 17.6
(CH3), 25.8 (CH3), 38.6 (CH2), 118.5 (_CH), 123.3
(_CH), 132.2 (_C(CH3)2), 135.3 (sC(CH3)CH2). 13C
NMR (C6D6): d 13.5 (q of d, 1JCHs126 Hz, 3JCHs4.5 Hz,
CH3), 15.9 (q, 1JCHs125 Hz, 3JCH;4 Hz, CH3), 17.6 (m,
1JCHs125 Hz, 3JCH;4 Hz, CH3), 25.8 (m, 1JCHs125 Hz,
A side-arm round bottom flask was charged with 50 ml of
dry toluene and 1.38 g of dimethyltitanocene (6.62 mmol).
To this solution was added 2-butyne (1.04 ml, 717 mg, 2
equiv.). The flask was wrapped in foil and heated in an oil
1
bath at 708C for 12 h. H NMR analysis of the reaction
mixture at this point revealed conversion to a mixture of
titanacyclopentene 9 and titanacyclobutene 8, which displays
identical spectral properties to those of an authentic sample
[26,66,67,75,76]. The solvent was removed in vacuo and the
residue was extracted with 25 ml of hexanes. The resulting
solution was cooled to y308C, thenfiltered. Thepureproduct
was isolated in ca. 35% yield by evaporation of the filtrand
1
3JCH;3.7 Hz, CH3), 38.6 (t, JCHs124 Hz, CH2), 118.5
(m, 1JCHs152 Hz, 3JCH;6Hz, _CH), 123.3(m, 1JCHs151
Hz, 3JCH;5 Hz, _CH), 132.2 (heptet, 3JCH;6 Hz,
_C(CH3)2), 135.3 (s, _C(CH3)CH2). MS (70 eV):
m/zs97 (M–C2H3, 7%), 79 (10%), 69 (12%), 67 (15%),
65 (10%), 63(10%), 56(C4H8, 31%), 55(23%), 54(C4H6,
43%), 53 (18%), 51 (29%), 50 (19%), 44 (18%), 43
(65%), 42 (22%), 41 (C3H5, 100%), 40 (11%).
1
in vacuo. H NMR (C6D6): d 0.82 (s, 3H, CH3), 1.17 (s,
3H, CH3), 1.39 (s, 3H, CH3), 1.52 (s, 3H, CH3), 2.57 (br
s, 2H, CH2), 5.94 (s, 10H, Cp). 13C{1H} NMR (C6D6): d
18.4 (CH3), 20.4 (CH3), 20.9 (CH3), 24.8 (CH3), 42.0
(CH2), 113.8 (Cp), 119.6 (Ti–C_C), 125.1 (Ti–C_C),
185.6 (Ti–C_C), 188.1 (Ti–C_C). 13C NMR (C6D6): d
4.12. Protonolysis of titanacyclopentene 9-d3
1
1
18.4 (q, JCHs124 Hz, CH3), 20.4 (q of t, JCHs124 Hz,
3JCHs4 Hz, CH3), 20.9 (q, 1JCHs124 Hz, CH3), 24.8 (q of
q, 1JCHs123 Hz, 3JCHs6 Hz, CH3), 42.0 (t of q, 1JCHs121
Treatment of 9-d3 with excess dry HCl, as above, afforded
the corresponding diene, (E,E)-1,1,1-d3-2,5-dimethylhepta-
1
3
2,5-diene, as the sole volatile product. H NMR (C6D6): d
Hz, JCHs4 Hz, CH2), 113.8 (m, Cp), 119.6 (hextet,
3JCHs6.5 Hz, Ti–C_C), 125.1 (heptet, 3JCHs5.8 Hz, Ti–
C_C), 185.6 (s, Ti–C_C), 188.1 (s, Ti–C_C). MS:
m/zs300 (Mq, 24%), 232 (M–C5H8, 14%), 218 (28%),
192 (17%), 178 (Cp2Ti, 100%), 152 (26%), 128 (18%),
113 (CpTi, 76%). HRMS. Calc. for C19H24Ti: 300.1357.
Found: 300.1370. Anal. Calc. for C19H24Ti: C, 75.99; H, 8.06.
Found: C, 75.83; H, 8.12%.
1.54 (s, 3H, CH3), 1.56(s, 6H, 2=CH3), 2.70(d, 2H, Js7.5
Hz, CH2), 5.25 (t, 1H, Js7.5 Hz, _CH–CH2), 5.28–5.34
(m, 1H, _CH). 13C{1H} NMR (C6D6): d 13.5 (CH3), 15.9
1
(CH3), 17.6 (CH3), 24.7 (m, JCD;18 Hz, CD3), 38.6
(CH2), 118.5 (_CH), 123.3 (_CH), 132.1 (_C(CH3)2),
135.3 (_C(CH3)CH2). MS (70 eV): m/zs127 (Mq, 5%),
109 (M–CD3, 6%), 100 (M–C2H3, 7%), 95 (11%), 79
(11%), 73 (25%), 69 (14%), 67 (22%), 65 (12%), 63
(12%), 60 (10%), 59 (27%), 58 (16%), 57 (30%), 56
(C4H8, 57%), 55 (46%), 54 (C4H6, 56%), 53 (46%), 51
(46%), 50 (46%), 47 (43%), 46 (46%), 45 (46%), 44
(46%), 43 (94%), 42 (59%), 41 (C3H5, 100%), 40 (33%).
4.10. Preparation of titanacyclopentene 9-d3
This complex was prepared analogously, using 400 mg of
Cp2Ti(CD3)2 (1.88 mmol) and 0.292 ml of 2-butyne (2
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

888
K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
4.13. Reaction of dimethyltitanocene with 3-hexyne
phine and 30 ml of 3-methyl-1,2-butadiene (1 equiv.). The
mixture was stirred at room temperature for 1 h, then frozen
in a y58C bath. Volatiles were removed from the frozen
sample in vacuo, yielding 14 in analytically pure form as a
greenish-brown solid (0.079 g, 81%), m.p. 1208C. 1H NMR
(C6D6): d 0.29 (br s, 2H, CH2), 0.83 (br s, 9H, PMe3), 2.02
(s, 3H, CH3), 2.41 (s, 3H, CH3), 4.82 (s, 10H, Cp). 13C{1H}
NMR (C6D6): d 7.2 (PMe3), 24.2 (CH3), 29.7 (CH3), 99.6
(Cp), 113.7 (Ti–CH2), 118.5 (_C(CH3)2), 172.0 (Ti–
C_C(CH3)2). 13C NMR (C6D6): d 7.2 (m, Japps150 Hz,
An NMR tube was charged with 0.5 ml of C6D6, 42 mg of
dimethyltitanocene (0.2 mmol), and 23 ml of 3-hexyne (1
equiv.). The tube was wrapped in foil and heated in an oil
bath at 608C for 12 h, with periodic 1H NMR monitoring of
the progress of the reaction. Upon complete consumption of
the dimethyltitanocene, a 58:42 mixture (based on integra-
tion of the cyclopentadienyl resonances) of titanacyclo-
pentene 12 and titanacyclobutene 11 was obtained. The
titanacyclobutene displays spectral properties identical to
1
1
PMe3), 24.2 (q, JCHs123 Hz, CH3), 29.7 (q, JCHs127
1
1
those reported previously [51]. Titanacyclobutene 11: H
Hz, CH3), 99.6 (d, JCHs174 Hz, Cp), 113.7 (m, Japps3
NMR (C6D6): d 0.92 (t, 3H, 3JHHs7.5 Hz, CH3), 1.04 (t,
3H, 3JHHs7.5 Hz, CH3), 1.95 (q, 2H, 3JHHs7.5 Hz, CH2),
Hz, Ti–CH2), 118.5 (s, _C(CH3)2), 172.0 (s, Ti–
C_C(CH3)2). Anal. Calc. for C18H27PTi: C, 67.08; H, 8.44;
P, 9.61. Found: C, 67.12; H, 8.33; P, 9.50%.
3
2.53 (q, 2H, JHHs7.5 Hz, CH2), 3.30 (s, 2H, Ti–CH2),
5.53 (s, 10H, Cp). 13C{1H} NMR (C6D6): d 12.4 (CH3),
15.8 (CH3), 22.4 (CH2), 29.3 (CH2), 77.6 (Ti–CH2), 92.5
(Ti–C_C), 110.0 (Cp), 221.0 (Ti–C_C). 13C NMR
(C6D6): d 12.4 (q, 1JCHs126 Hz, CH3), 15.8 (q, 1JCHs125
Hz, CH3), 22.4 (t, 1JCHs127 Hz, CH2), 29.3 (t, 1JCHs124
Hz, CH2), 77.6 (t, Ti–CH2), 92.5 (s, Ti–C_C), 110.0 (dm,
1JCHs172 Hz, Cp), 221.0 (s, Ti–C_C). Titanacyclopentene
4.16. Reaction of (trimethylphosphine)(3-methyl-1,2-
butadiene)titanocene with 2-butyne
To a solution of 20 mg of allene complex 14 (0.062 mmol)
in 0.5 ml of C6D6 in an NMR tube was added 9.5 ml of 2-
butyne (2 equiv.). The tube was sealed, then heated at 658C
for 1.5 h, at which time spectral analysis revealed complete
conversion of 14 to titanacyclopentene 9 (ca. 97%) and a
trace of the product of reductive coupling of two equivalents
of the allene (15, vide infra).
1
12: H NMR (C6D6): d 0.65 (s, 3H, CH3), 1.16 (d, 3H,
3JHHs6 Hz, CH3–CH), 2.83 (q, 1H, 3JHHs6 Hz, CH3–CH),
5.96 (s, 5H, Cp), 6.05 (s, 5H, Cp). (The remaining aliphatic
resonances appear as a set of complex multiplets, partially
obscured by the resonances of the titanacyclobutene,between
0.8–2.6 ppm.) 13C{1H} NMR (C6D6): d 12.8 (CH3), 13.2
(CH3), 14.8 (CH3), 21.1 (CH3), 22.3 (CH3), 22.9 (CH2),
27.1 (CH2), 27.6 (CH2), 41.6 (CH–CH3), 113.4 (Cp),
114.7 (Cp), 131.3 (Ti–C_C), 132.7 (Ti–C_C), 192.6 (Ti–
C_C), 193.0 (Ti–C_C). 13C NMR (C6D6): d 12.8 (q,
1JCH;127 Hz, CH3), 13.2 (q, 1JCH;123 Hz, CH3), 14.8 (q,
1JCH;120 Hz, CH3), 21.1 (q, 1JCH;123 Hz, CH3), 22.3 (q,
1JCH;126 Hz, CH3), 22.9 (CH2), 27.1 (CH2), 27.6 (CH2),
4.17. Reaction of (trimethylphosphine)(3-methyl-1,2-
butadiene)titanocene with diphenylacetylene
Treatment of allene complex 14 with 1 equiv. of diphen-
ylacetylene as described above afforded a ca. 9:1 mixture of
titanacyclopentene 16 (RsPh) and reductive couplingprod-
1
uct 15 (vide infra). H NMR (C6D6): d 0.8 (s, 3H, CH3),
1.52 (s, 3H, CH3), 3.16 (s, 2H, CH2), 6.02 (s, 10H, Cp),
6.8–7.4 (m, 10H, Ph).
1
1
41.6 (d, JCH;122 Hz, CH–CH3), 113.4 (dm, JCHs173
Hz, Cp), 114.7 (dm, 1JCHs173 Hz, Cp), 131.3 (Ti–C_C),
132.7 (Ti–C_C), 192.6 (Ti–C_C), 193.0 (Ti–C_C).
(Resonances for which no multiplicity information is pro-
vided were of insufficient intensity to permit such a
determination.)
4.18. Reductive coupling of 3-methyl-1,2-butadiene by
Cp2Ti(PMe3)2
To a solution of 250 mg of Cp2Ti(PMe3)2 (0.758 mmol)
in 8 ml of diethyl ether was added 100 ml of trimethylphos-
phine and 148 ml of 3-methyl-1,2-butadiene (2 equiv.). The
mixture was kept at room temperature, with periodic moni-
4.14. Protonolysis of titanacyclopentene 12
Treatment of 12 with excess dry HCl, as above, afforded
the corresponding diene, (Z,E)-6-ethyl-3,5-dimethyl-3,6-
1
toring of aliquots by H NMR. After the reaction appeared
complete(ca. 24h), removalofallvolatilesinvacuoafforded
an orange solid. This crude product was extracted into pen-
tane and the resulting solution was filtered to remove insol-
uble residues and evaporated, affording the titanacyclo-
pentane 15 as a bright-orange solid (186 mg, 78%). 1H NMR
(C6D6): d 0.75 (s, 6H, CH3), 1.48 (s, 6H, 2=CH3), 2.23
(s, 4H, CH2), 6.05 (s, 10H, Cp). 13C{1H} NMR (C6D6): d
21.4 (CH3), 25.9 (CH3), 31.3 (CH2), 114.2 (Cp), 124.5
(Ti–C_C), 186.7 (Ti–C_C). MS (70 eV): m/zs314
(Mq), 312, 310, 296, 272, 243, 192, 178 (Cp2Ti), 113
(CpTi). HRMS. Calc. for C20H26Ti: 314.1514. Found:
1
nonadiene. H NMR (C6D6): d 0.87–0.99 (m, 9H, CH3–
CH2), 1.18–1.29 (m, 3H, CH3–CH), 1.39 (m, 1H, CH–
CH3), 1.65 (d, 2H, Js0.6 Hz, CH3–C_), 1.90–1.97 (m,
6H, CH2CH3), 4.81 (d, 1H, Js5.1 Hz, _CH–CH(CH3)),
5.15 (t, 1H, Js7.2 Hz, _CH–CH2CH3).
4.15. Preparation of (trimethylphosphine)(3-methyl-1,2-
butadiene)titanocene (14)
To a stirred solution of 100 mg of Cp2Ti(PMe3)2 (0.303
mmol) in 2 ml of C6D6 was added 100 ml of trimethylphos-
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
889
[18] G.A. Luinstra, J.H. Teuben, Organometallics 11 (1992) 1793.
[19] P. Foley, R. DiCosimo, G.M. Whitesides, J. Am. Chem. Soc. 102
(1980) 6713.
314.1529. Anal. Calc. for C20H26Ti: C, 76.42; H, 8.34. Found:
C, 76.34; H, 8.41%.
[20] S.S. Moore, R. DiCosimo, A.F. Sowinski, G.M. Whitesides, J. Am.
Chem. Soc. 103 (1981) 948.
4.19. Kinetic studies
[21] J. Cheon, L.H. Dubois, G.S. Girolami, J. Am. Chem. Soc. 119 (1997)
6814.
[22] K.M. Doxsee, J.J.J. Juliette, J.K.M. Mouser, K. Zientara, Organo-
metallics 12 (1993) 4682.
[23] H. Masai, K. Sonogashira, N. Hagihara, Bull. Chem. Soc. Jpn. 41
(1968) 750.
[24] K.M. Doxsee, J.J.J. Juliette, G. Nieckarz, K. Zientara, J. Am. Chem.
Soc. 116 (1994) 2147.
[25] N.A. Petasis, D.-K. Fu, Organometallics 12 (1993) 3776.
[26] F.N. Tebbe, R.L. Harlow, J. Am. Chem. Soc. 102 (1980) 6149.
[27] M. Brookhart, M.L.H. Green, J. Organomet. Chem. 250 (1983) 395.
[28] M.L.H. Green, L.-L. Wong, J. Chem. Soc., Chem. Commun. (1984)
1442 and Refs. therein.
Reaction mixtures comprised 0.288 mM Cp2TiMe2 in
C6D6, 2–10 equiv. of the alkyne, and 10 mol% of ferrocene
as an internal standard were sealed under dry nitrogen in
5 mm NMR tubes. Reactions at )708C were carried out
directly in the NMR probe ("0.58C), while reactions at
lower temperatures were carried out in thermostatted oil
baths, with the tubes wrapped in aluminum foil to preclude
possible photochemical reactions. Reaction progress was
monitored using the cyclopentadienyl resonancesoftheprod-
ucts, which were all clearly resolved.
[29] J.J.W. Eshuis, Y.Y. Tan, A. Meetsma, J.H. Teuben, Organometallics
11 (1992) 362.
[30] A. Bertuleit, C. Fritze, G. Erker, R. Frohlich, Organometallics 16
(1997) 2891 and Refs. therein.
Acknowledgements
[31] R.F. Johnston, J.C. Cooper, Organometallics 6 (1987) 2448.
[32] P.C. Wailes, H. Weigold, A.P. Bell, J. Organomet. Chem. 33 (1971)
181.
[33] P.C. Wailes, H. Weigold, A.P. Bell, J. Organomet. Chem. 34 (1972)
155.
[34] P.W.N.M. VanLeeuwen, H. Van der Heijden, C.G.C.F. Roobeek,
J.H.G. Frijns, J. Organomet. Chem. 209 (1981) 169.
[35] W.H. Boon, M.D. Rausch, J. Chem. Soc., Chem. Commun. (1977)
397.
This work was supported in part by the National Institutes
of Health and in part by the National Science Foundation.
J.J.J.J. was the recipient of a fellowship from the US Depart-
ment of Education Graduate Assistance in Areas of National
Need Program. Experimental assistance by Kenneth Zientara
and Cynthia Ferguson is gratefully acknowledged.
[36] W.H. Boon, M.D. Rausch, H. Alt, J. Organomet. Chem. 141 (1977)
299.
[37] H. Alt, M.D. Rausch, J. Am. Chem. Soc. 96 (1974) 5936.
[38] N.M. Kablaoui, S.L. Buchwald, J. Am. Chem. Soc. 118 (1996) 3182.
[39] T.S. Piper, G. Wilkinson, J. Inorg. Nucl. Chem. 3 (1956) 104.
[40] K. Clauss, H. Bestian, Liebigs Ann. Chem. 654 (1962) 8.
[41] C. McDade, J.C. Green, J.E. Bercaw, Organometallics 1 (1982)1629.
[42] J.B. Lee, G.J. Gajda, W.P. Schaefer, T.R. Howard, T. Ikariya, D.A.
Straus, R.H. Grubbs, J. Am. Chem. Soc. 103 (1981) 7358.
[43] G.A. Luinstra, J. Vogelzang, J.H. Teuben, Organometallics11(1992)
2273.
[44] K.M. Doxsee, J.K.M. Mouser, Organometallics 9 (1990) 3012.
[45] K.M. Doxsee, J.K.M. Mouser, Tetrahedron Lett. 32 (1991) 1687.
[46] A. Pietrzykowski, S. Pasynkiewicz, P. Buchalski, Coll. Czech. Chem.
Commun. 62 (1997) 318.
References
[1] J.A. Davies, P.L. Watson, A. Greenberg, J.F. Liebman, Selective
Hydrocarbon Activation, Principles and Progress, VCH, New York,
1990.
[2] T.J. Marks, R.D. Ernst, in: G. Wilkinson, F.G.A. Stone, E.W. Abel
(Eds.), Comprehensive Organometallic Chemistry, Pergamon,
Oxford, 1982, Ch. 21.
[3] M. Bothill, P.D. Gavens, J.W. Kelland, J. McMeeking, in: G. Wilkin-
son, F.G.A. Stone, E.W. Abel (Eds.), ComprehensiveOrganometallic
Chemistry, Pergamon, Oxford, 1982, Ch. 22.3.
[4] M.I. Bruce, Angew. Chem., Int. Ed. Engl. 16 (1977) 73.
[5] E.C. Constable, Polyhedron 3 (1984) 1037.
[47] B.C. Ankianiec, V. Christou, D.T. Hardy, Sk.K. Thomson, G.B.
Young, J. Am. Chem. Soc. 116 (1994) 9963.
[6] C. Hinderling, D.A. Plattner, P. Chen, Angew. Chem., Int. Ed. Engl.
36 (1997) 243.
[48] T. Behling, G.S. Girolami, G. Wilkinson, R.G. Somerville, M.B.
Hursthouse, J. Chem. Soc., Dalton Trans. (1984) 877.
[49] R.A. Andersen, R.A. Jones, G. Wilkinson, J. Chem. Soc., Dalton
Trans. (1978) 446.
[50] C.R. Bennett, D.C. Bradley, J. Chem. Soc., Chem. Commun. (1974)
29.
[51] J.K.M. Mouser, Ph.D. Dissertation, University of SouthernCalifornia,
Los Angeles, CA, 1991.
[52] H.J. Reich, M. Jautelat, M.T. Messe, J.J. Weigert, J.D. Roberts, J. Am.
Chem. Soc. 91 (1969) 7445.
[53] N.S. Bhacca, D.D. Giannini, W.S. Jankowski, M.E. Wolff, J. Am.
Chem. Soc. 95 (1973) 8421.
[54] H. Eggert, C. Djerassi, J. Org. Chem. 38 (1973) 3788.
[55] E. Breitmaier, W. Voelter, Carbon-13 NMR Spectroscopy, VCH,New
York, 3rd edn., 1987.
[56] G.C. Levy, R.L. Lichter, G.L. Nelson, Carbon-13 Nuclear Magnetic
Resonance Spectroscopy, Wiley Interscience, New York, 2nd edn.,
1980.
[7] M.A. Bennett, T. Dirnberger, D.C.R. Hockless, E. Wenger, A.C. Wil-
lis, J. Chem. Soc., Dalton Trans. (1998) 271.
[8] M. Aizenberg, D. Milstein, Organometallics 15 (1996) 3317.
[9] I.P. Rothwell, Acc. Chem. Res. 21 (1988) 153.
[10] J.S. Vilardo, M.A. Lockwood, L.G. Hanson, J.R. Clark, B.C. Parkin,
P.E. Fanwick, I.P. Rothwell, J. Chem. Soc., Dalton Trans. (1997)
3353.
[11] J. Albert, J. Granell, A. Luque, J. Minguez, R. Moragas, M.
Fontbardia, X. Solans, J. Organomet. Chem. 522 (1996) 87.
[12] J. Vicento, I. Saurallamas, M.G. Palin, P.G. Jones, M.C.R. Dearellano,
Organometallics 16 (1997) 826.
[13] S.L. Buchwald, R.B. Nielson, Chem. Rev. 88 (1988) 1047.
[14] J.W. Bulls, W.P. Schaefer, M. Serfas, J.E. Bercaw, Organometallics
6 (1987) 1219.
[15] N.A. Petasis, E.I. Bzowej, J. Org. Chem. 57 (1992) 1327.
[16] N.A. Petasis, E.I. Bzowej, Tetrahedron Lett. 34 (1993) 943.
[17] B. Hessen, J.-K.F. Buijink, A. Meetsma, J.H. Teuben, G. Helgesson,
M. Ha˚kansson, S. Jagner, A.L. Spek, Organometallics 12 (1993)
2268.
[57] J.B. Stothers, Q. Rev. Chem. Soc. 19 (1965) 144.
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401

890
K.M. Doxsee, J.J.J. Juliette / Polyhedron 19 (2000) 879–890
[58] D.M. Doddrell, D.T. Pegg, M.R. Bedall, J. Magn. Reson. 48 (1982)
323.
[67] F.N. Tebbe, G.W. Parshall, G.S. Reddy, J. Am. Chem. Soc. 100
(1978) 3611.
[59] M.J.S. Dewar, Bull. Soc. Chim. Fr. 18 (1951) C71.
[60] J. Chatt, L.A. Duncanson, J. Chem. Soc. (1953) 2939.
[61] L.B. Kool, M.D. Rausch, H.G. Alt, M. Herberhold, U. Thewalt, B.
Wolf, Angew. Chem., Int. Ed. Engl. 24 (1985) 394.
[62] D.J. Sikora, D.W. Macomber, M.D. Rausch, Adv. Organomet. Chem.
25 (1986) 317.
[68] C.-C. Han, T. Katz, J. Organometallics 4 (1985) 2186.
[69] D.M. Duggan, Inorg. Chem. 20 (1981) 1164.
[70] J.R. Schmidt, D.M. Duggan, Inorg. Chem. 20 (1981) 318.
[71] D. Hideura, H. Urabe, F. Sato, Chem. Commun. (1998) 271.
[72] M. Ahmar, O. Chabanis, J. Gauthier, B. Cazes, Tetrahedron Lett. 38
(1997) 5277.
[63] J. Yin, K.A. Abboud, W.M. Jones, J. Am. Chem. Soc. 115 (1993)
3810.
[73] K. Suzuki, T. Imai, S. Yamanoi, M. Chino, T. Matsumoto, Angew.
Chem., Int. Ed. Engl. 36 (1997) 2469.
[64] D.J. Pasto, N. Huang, C.W. Eigenbrot, J. Am. Chem. Soc. 107 (1985)
3160.
[74] T.V.W. Janssens, G.L. Jin, F. Zaera, J. Am. Chem. Soc. 119 (1997)
1169.
[65] H. Hoberg, J. Korff, J. Organomet. Chem. 152 (1978) 255 and Refs.
therein.
[75] J.D. Meinhart, B.D. Santarsiero, R.H. Grubbs, J. Am. Chem. Soc. 108
(1986) 3319.
[66] F.N. Tebbe, G.W. Parshall, D.W. Ovenall, J. Am. Chem. Soc. 101
(1979) 5074.
[76] T.L. Howard, J.B. Lee, R.H. Grubbs, J. Am. Chem. Soc. 102 (1980)
6876.
Thursday May 04 02:02 PM
StyleTag -- Journal: POLY (Polyhedron) Article: 3401