Synthesis, Cycloaddition, and Cycloreversion Reactions of Mononuclear Titanocene-oxo Complexes

Article abstract of DOI:10.1021/acs.organomet.6b00111

Titanocene-oxo complexes of the type Cpx₂Ti=O(L) (Cp^x = pentamethylcyclopentadienyl; tetramethylcyclopentadienyl; L = pyridine or derivatives) are synthesized from the corresponding titanocene-ethylene complexes via oxidation with pyridine N-oxides or styrene oxide. These oxo complexes react with alkynes, nitriles, and α,β-unsaturated carbonyls to form titanacycles, which undergo exchange reactions with organic substrates or react with 4-dimethylaminopyridine to regenerate the titanocene-oxo. Mechanistic experiments support a dissociative mechanism in which the first step is rate-determining retrocycloaddition followed by trapping of the reactive [Cpx₂Ti=O] species. In the case of the retro-[4+2]-cycloaddition from dioxatitanacyclohexene complexes, a Hammett study gives ρ values of -1.18 and -1.04 for substituents on two different phenyl rings on the metallacycles, suggesting positive charge buildup and a slightly asynchronous cycloreversion in the rate-determining step.

Full text of DOI:10.1021/acs.organomet.6b00111

Article
pubs.acs.org/Organometallics
Synthesis, Cycloaddition, and Cycloreversion Reactions of
Mononuclear Titanocene−oxo Complexes
Trang T. Nguyen, Gregory D. Kortman, and Kami L. Hull*
Department of Chemistry, University of Illinois at Urbana−Champaign, 600 S. Mathews Avenue, Urbana, Illinois 61801, United
States
S
* Supporting Information
ABSTRACT: Titanocene−oxo complexes of the type Cpx2Ti
O(L) (Cpx = pentamethylcyclopentadienyl; tetramethylcyclo-
pentadienyl; L = pyridine or derivatives) are synthesized from
the corresponding titanocene−ethylene complexes via oxidation
with pyridine N-oxides or styrene oxide. These oxo complexes
react with alkynes, nitriles, and α,β-unsaturated carbonyls to
form titanacycles, which undergo exchange reactions with organic substrates or react with 4-dimethylaminopyridine to regenerate
the titanocene−oxo. Mechanistic experiments support a dissociative mechanism in which the first step is rate-determining
retrocycloaddition followed by trapping of the reactive [Cpx2TiO] species. In the case of the retro-[4+2]-cycloaddition from
dioxatitanacyclohexene complexes, a Hammett study gives ρ values of −1.18 and −1.04 for substituents on two different phenyl
rings on the metallacycles, suggesting positive charge buildup and a slightly asynchronous cycloreversion in the rate-determining
step.
INTRODUCTION
■
α,β-Unsaturated aldehydes and ketones are both prominent
functionalities in and synthetic precursors for pharmaceuticals,
natural products, and organic materials.1−3 Olefination
reactions are powerful methodologies for the synthesis of
enones; unfortunately, these reactions often require several
synthetic steps to access the requisite starting materials and
generate stoichiometric waste.4 The goal of the research
presented herein was to develop a new transition-metal-
catalyzed cross-coupling reaction for the synthesis of α,β-
unsaturated carbonyls from readily available building blocks
without the generation of any stoichiometric byproducts in a
redox-neutral, atom-economical process.5 This should be
possible by employing a cycle that uses a Ti/ZrO catalyst
to couple a carbonyl with an alkyne, through (i) [2+2]-
cycloaddition between the MO and an alkyne, (ii) insertion
of carbonyl into the M−C bond, and, finally, (iii) [4+2]-
retrocycloaddition, releasing the α,β-unsaturated carbonyl
product and regenerating the MO catalyst (Figure 1). As
the starting materials are already at the requisite oxidation
states, the reaction will not require either a stoichiometric
oxidant or reductant.
Titanium− and zirconium−imido complexes have shown
great utility in both stoichiometric and catalytic reactions.
However, similar reactivity with related TiO or ZrO
complexes has been relatively underdeveloped. There are many
reports for the synthesis of both Ti− and Zr−oxo complexes,
which are unstable to isolation, as they are prone to undergo
oligomerization to generate μ-oxo products. Therefore, they are
generally trapped by the addition of a Lewis basic ligand, such a
pyridine, or a substrate that can undergo reaction with the M
O, such as an alkyne or carbonyls, to generate stable
Figure 1. Proposed metal−oxo-catalyzed alkyne−aldehyde coupling
reaction.
metallacyclic intermediates. When the Cp* ligand is replaced
with the relatively small Cp ligand, dimeric products of the
oxometallacyclobutene are observed. Alternatively, it has been
shown that sterically encumbering ligands inhibit TiO
oligomerization and crystal structures of the free [Ti]O
have been obtained.
Recently, we demonstrated each of the steps of our proposed
catalytic cycle with a Cp*2ZrO complex.6 Unfortunately,
catalytic turnover was not achieved, as the final step, retro-
[4+2]-cycloaddition was reversible and thermodynamically
disfavored. Further, the aldehyde coupling partner was not
stable to the high reaction temperatures required for the retro-
[4+2] (150 °C). As such, we were interested in investigating
each of the proposed steps with Ti−oxo complexes, as the
Received: February 25, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
shorter Cp*−Ti bond length may promote the desired
retrocycloaddition, as it would be releasing steric strain.
The few known monometallic titanium−oxo complexes have
been synthesized with bulky, stabilizing ligands including
tetraaza[14]annulene7,8 and pentasubstituted cyclopentadien-
yl.9,10 In addition to their propensity to react with water, it has
been reported that these complexes undergo [2+2]-cyclo-
additions with alkynes and allenes.11−13 In contrast, the
isoelectronic titanium−imido complexes have been used in a
number of catalytic reactions.14−17 Key steps in these catalytic
cycles are cycloaddition and cycloreversion involving the Ti
N multiple bond. Cycloreversion has been proposed in
reactions of oxatitanacyclohexenes18 and oxatitanacyclobu-
tenes,12 but these processes have not been studied in detail.
In order to develop the proposed titanocene−oxo-mediated
alkyne−aldehyde coupling reaction, it is important to under-
stand the cycloaddition and cycloreversion reactivity of these
complexes.
Herein, we report our stoichiometric investigation toward the
development of a TiO-catalyzed alkyne−aldehyde coupling
reaction. Concurrently, we also investigated the overall
reactivity of TiO complexes and demonstrate an improved
synthesis of titanocene−oxo complexes, metallacycle formation
reactions of these oxo complexes with unsaturated organic
compounds, and mechanistic studies on the retrocycloaddition
reactions.
Scheme 1. Synthesis of Titanium−oxo Complexes 2a and 2b
via Oxidation with Pyridine N-Oxides
We found that the stoichiometry of reagents was critical.
Excess oxidant gave rise to byproducts, including titanium μ-
oxo clusters,9,22 which were difficult to separate from 2a or 2b.
The use of N-oxide without added pyridine generated a number
of byproducts and very low yield of the terminal oxo. This
suggests that a high concentration of pyridine is required to
increase the rate of the formation of 2a relative to the
oligomerization. Finally, replacing pyridine with 2-picoline and
2,6-lutidine did not afford the desired products, presumably due
to steric hindrance caused by the methyl group(s).
After the initial study by Andersen, titanocene−oxo
complexes bearing bulkier ligands than Cp* have been
reported,18 but terminal oxo complexes with less sterically
hindered ligands, such as the Cp′, are not known. Oxidation of
ethylene complex 3 with nitrous oxide gave product 4a in the
20−30% range, while the use of pyridine N-oxide produced a
reduced yield (<10%) of the desired product. The main
products in both cases were likely Cp′ titanium μ-oxo clusters
resulting from decomposition of the [Cp′2TiO] species. This
result can be attributed to the smaller size of the Cp′ compared
to the Cp* ligand, making it less effective in stabilizing the
reactive monomeric titanium−oxo. We hypothesized that the
use of nonpolar solvents would cause the desired oxo product
to precipitate and thereby prevent oligomerization. However,
both nitrous oxide and pyridine N-oxide have low solubility in
hexane and octane. Styrene oxide, previously used by Chirik in
the synthesis of base-free titanocene−oxo complexes,12 offers a
solution to this problem. Reaction of 3 with styrene oxide in
the presence of pyridine or 3,5-lutidine in hexane generated 4a
and 4b in 60% and 62% isolated yield, respectively (Scheme 2).
RESULTS AND DISCUSSION
■
Synthesis of Titanocene−oxo Complexes. Our initial
investigations of the proposed alkyne−aldehyde coupling
reaction studied the reactivity of Cp*2TiO and Cp′2TiO
complexes (Cp*2 = pentamethylcyclopentadienyl; Cp′2
=
tetramethylcyclopentadienyl). Cp*2TiO(py) (py = pyridine)
is a known complex that has been demonstrated to participate
in the proposed [2+2]-cycloaddition with both alkynes and
allenes; however the reactivity of the resulting oxatitanacyclo-
butenes is unknown. Additionally, the related Cp′2TiO
complex was sought, as the TiO is slightly less hindered and
may be more reactive.
The pyridine-stabilized titanocene−oxo complex Cp*2Ti
O(py) was first prepared via oxidation of the corresponding
titanocene−ethylene complex with nitrous oxide in 59% yield.7
In our previous report on the reactivity of oxatitanacyclobu-
tenes prepared from titanocene−oxo,19 attempts to reproduce
this result were complicated by impurities in the nitrous oxide
gas, which led to low yields (30−40%) in our hands. Given
these results we sought to employ more conventional reagents
to affect the oxidation of the titanocene−ethylene complex.
Pyridine N-oxide is a readily available oxidant that is used in a
number of catalytic oxidation reactions involving metal−oxo
complexes.20 Further, purification and quantification of this
crystalline solid is facile relative to gaseous nitrous oxide.
Previously, it was reported that pyridine N-oxide oxidizes a
titanium−dinitrogen complex, supported by benzamidinate
ligands, to the corresponding titanium−oxo. The product was
isolated in moderate yield (37%) and contains a coordinated N-
oxide ligand, even in the presence of pyridine.21 The oxidation
of Cp*2Ti(CH2CH2) (1) (1.0 equiv) with pyridine N-oxide
(1.0 equiv) in the presence of pyridine (2.0 equiv) produced
oxo complex 2a in 80% isolated yield. Similarly, 3,5-lutidine N-
oxide was an effective oxidizing reagent for this transformation,
generating oxo complex 2b in 84% isolated yield (Scheme 1).
Scheme 2. Synthesis of Titanocene−oxo Complexes 4a and
4b via Oxidation with Styrene Oxide
In both cases, the products contained a small amount of an
unidentified byproduct that persisted after repeated recrystal-
lization, but it appeared to not interfere with other reactions of
4a and 4b. The ratio of product to byproduct was greater than
12:1 based on NMR integration.23
Single crystals suitable for X-ray diffraction were grown by
slow diffusion of pentane into a toluene solution of 4b. An
ORTEP representation of 4b is shown in Figure 2, and selected
bond lengths and angles are given in Table 1. The TiO bond
length in 4b is 1.686 Å, within the TiO bond length range of
known terminal titanium−oxo complexes. Compared to the
Cp*−Ti−Cp* angle of 135° in complex 2a,9 the Cp′−Ti−Cp′
B
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. Scope of Terminal Alkynes in [2+2]-Cycloaddition
a
Reaction with Titanocene−oxo Complexes
b
entry
Cpx
R′ (product)
Ph− (5a)
p-MeO-C6H4− (5b)
3,5-(CF3)2-C6H3− (5c)
p-F-C6H4− (5d)
yield
1
2
3
4
5
6
Cp*
Cp*
Cp*
Cp*
Cp*
Cp′
95%
92%
88%
70%
Figure 2. ORTEP diagram of Cp′2TiO(3,5-lutidine) (4b). Hydro-
gen atoms and a second molecule are omitted for clarity. Ellipsoids are
drawn at the 50% probability level.
c
p-NO2-C6H4− (5e)
3,5-(CF3)2-C6H3− (6a)
0%
94%
a
Reaction conditions: 1.0 equiv of titanocene−oxo, 1.2 equiv of
b
c
alkyne. Isolated yields. Starting materials were consumed, resulting
in an intractable mixture.
Table 1. Selected Bond Lengths and Angles for Oxo
Complex 4b
bond
length (Å)
bond
angle (deg)
Scheme 3. [2+2]-Cycloaddition Reaction of 2a with Internal
Alkynes
Ti1−O1
Ti1−N1
1.6868(18)
2.208(2)
O1−Ti1−N1
CNT1−Ti1−CNT2
91.05(9)
131.03(2)
angle in 4b is smaller (131°), as expected with a less sterically
hindered metal center.
[2+2]-Cycloaddition and -Cycloreversion of Titano-
cene−oxo with Alkynes. With the TiO complexes in
hand, our next step was to synthesize a variety of sterically and
electronically differentiated oxatitanacyclobutenes via [2+2]-
cycloaddition with alkynes, with the eventual goal of
investigating their aptitude to undergo carbonyl insertion.
Previously, Bergman demonstrated that oxo complex 2a
undergoes cycloaddition reactions with terminal alkynes
(RCCH; R = Ph, p-tolyl, t-Bu, and Me) and diphenylacetylene;
terminal alkynes react preferentially over internal alkynes, and
the [2+2]-cycloaddition is reversible. Experimental data
provided in the report is limited;12 as such, for known
compounds full experimental details are provided. We found
that 2a and 4a reacted with aryl acetylenes to give quantitative
NMR yields of the oxatitanacyclobutene products in less than
an hour at room temperature. Both electron-donating (p-
MeO−) and -withdrawing (3,5-di-CF3− and p-F−) substitu-
ents were tolerated. Isolated yields were lower for some of the
Cp* products due to the high solubility of the resulting
metallacycles (Table 2). Product formation was not observed
with p-nitrophenylacetylene (Table 2, entry 6); rather, an
intractable mixture was observed. This result could be
attributed to the strongly electron withdrawing nitro group,
which makes the terminal alkyne sufficiently acidic to undergo
deprotonation by the titanocene−oxo.
This facile [2+2]-cycloaddition reaction of terminal alkynes
was in sharp contrast with the reaction of internal alkynes,
which required higher equivalents of alkynes and more forcing
conditions. At room temperature, reactions of 2a and 4a with 1-
phenyl-1-propyne and diphenylacetylene (4 equiv) proceeded
very slowly, generating the titanacycle products in less than 5%
NMR yield after 7 days. At elevated temperature (60 °C), 2a
reacted with these internal alkynes to generate products 7a and
7b in 50% and 43% isolated yield, respectively (Scheme 3).
Attempts to increase conversion at higher temperature and
longer reaction time were not successful, as decomposition of
2a was observed.
Interestingly, reaction of 2a with the asymmetric alkyne 1-
phenyl-1-propyne produced titanacycle 7a, in which the phenyl
group was α to the titanium and methyl group β. This structure
of 7a was supported by NMR as well as experimental data:
treatment of 7a with pyridine hydrochloride resulted in
formation of benzyl methyl ketone; the other regioisomer of
7a should generate ethyl phenyl ketone instead (Scheme 4).
Scheme 4. Protonation of Oxatitanacyclobutene 7a
This result suggests the dominance of a steric effect, where the
relatively flat sp2 carbon of the phenyl ring is preferred between
the two Cp* rings over the tetrahedral methyl group. Higher
temperatures (>60 °C) did not facilitate the reactions of 4a
with internal alkynes; rather decomposition of starting material
was observed.
Oxatitanacyclobutene complex 6a has been shown to
undergo alkyne exchange with para-tolylacetylene at 75 °C to
generate oxatitanacyclobutene 6 where R = tolyl and phenyl
acetlyene.12 We found that the alkyne exchange reaction
occurred at room temperature without formation of any
byproducts (Table 3). Investigation of the exchange between
oxatitanacyclobutenes and free terminal alkynes can provide
information about the relative stability of the various titana-
cycles. With strongly electron deficient metallacycle 5c (Ar1 =
3,5-(CF3)2-C6H3−), no exchange was observed even at elevated
temperature, suggesting that 5c is significantly more stable than
5a,b due to the electron-withdrawing CF3 groups (Table 3,
C
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
similar reactions of titanocene−sulfido but different from that
of zirconocene−oxo, which reacts with benzonitrile to form a
six-membered metallacycle.25
Table 3. Alkyne Exchange Reaction of
Oxatitanacyclobutenes
a
X-ray quality crystals were grown by slow diffusion of hexane
into a toluene solution of 8b. An ORTEP representation is
shown in Figure 3, with selected bond lengths and angles in
b
entry
Ar1
Ar2
conversion
1
2
3
4
Ph (5a)
Ph (5a)
4-OMe-C6H4
4-F-C6H4
24%
24%
52%
0%
4-OMe-C6H4 (5b)
4-F-C6H4
3,5-(CF3)2-C6H3 (5c)
4-F-C6H4
a
Reaction conditions: 1.0 equiv of oxatitanacyclobutene, 1.0 equiv of
b
alkyne. Determined by NMR integration.
entry 4). This metallacycle is also more thermally stable, as no
reactions occurred upon heating at 80 °C for 3 days, while 5a
was previously reported to rearrange to a hydroxyacetylide
complex at elevated temperatures (45−75 °C).12
The reaction rate is dependent on the substituent on the
metallacycle. When 5a and 5b were treated with para-
fluorophenylacetylene, the more electron rich 5b reacted faster,
achieving 52% conversion after 3 days (Table 3, entry 3), while
5a gave 24% conversion (entry 2). When 5a was treated with
two different phenylacetylene derivatives, both reactions had
the same conversion of 24% after 3 days (Table 3, entries 1 and
2). This is consistent with electron-donating groups on Ar1
accelerating the rate of the exchange by promoting the retro-
[2+2]- and supports a unimolecular [2+2]-cycloreversion rate-
determining step, followed by trapping of the free titanocene−
oxo with alkyne.
[2+2]-Cycloaddition Reaction of Titanocene−oxo
Complexes with Nitriles. With the TiO complexes in
hand, we next chose to investigate other [2+2]-cycloadditions,
so that their reactivity could be directly compared to related
complexes.
Figure 3. ORTEP diagram of Cp*2Ti[OC(3,5-(CF3)2C6H3)N] (8b).
Hydrogen atoms are omitted for clarity. Ellipsoids are drawn at the
50% probability level.
Table 4. Selected Bond Lengths and Angles for 8b
bond
length (Å)
bond
angle (deg)
Ti1−O1
O1−C21
C21−N1
N1−Ti1
2.070(11)
1.278(7)
1.383(8)
1.950(13)
Ti1−O1−C21
O1−C21−N1
C21−N1−Ti1
N1−Ti1−O1
83.7(5)
120.4(3)
85.9(5)
70.0(3)
Table 4. The titanium center is in a pseudotetrahedral
environment with a flat metallacycle. The Ti1−O1 and Ti1−
N1 bond lengths are within the range of known titanium−
oxygen and titanium−nitrogen single bonds.
The titanocene−sulfido complex Cp*2TiS(py) has been
reported to react with benzonitrile in 2 h to form a product
assigned as an azathiatitanacyclobutene in 65% isolated yield.24
On the other hand, treatment of titanocene−oxo complex 2b
with benzonitrile resulted in less than 10% conversion after 24
h. To explore potential effects of aryl substitution on this
reaction, a solution of 2b in C6D6 was treated with different
benzonitrile derivatives, and the reactions were monitored by
1H NMR. Complete conversion to a single product was
observed with 3-bromobenzonitrile and 3,5-bis-
(trifluoromethyl)benzonitrile, generating products 8a and 8b
in 82% and 85% isolated yield, respectively (Scheme 5).
This reaction appeared to be sensitive to both the electronic
property and position of the substituents, as electron-deficient,
meta-substituted benzonitriles formed the [2+2] products in
good yields while ortho- and para-substituted ones gave no
reactions. In the proton NMR spectra of the metallacycles 8a
and 8b the Cp* methyl peak is shifted approximately 0.25 ppm
upfield, compared to the oxo. On the other hand, when 2b was
treated with ortho- and para-bromobenzonitriles, proton NMR
spectra showed only one Cp* peak, which shifted very slightly
(0.01−0.03 ppm) from that of 2b, and broadening of the nitrile
and amine peaks.23 These observations suggest fast pyridine−
nitrile exchange on the NMR time scale and no [2+2]-
cycloaddition. It appears that the stereoelectronic properties of
the substituents have a strong influence on the nitrile group.
Electron-withdrawing groups at the meta position make the
group more electrophilic and reactive in [2+2]-cycloaddition,
while electron-donating substituents increase the electron
density on the nitrogen atom of the nitrile group, making it a
good ligand for the titanium complex. Interestingly, Bergman
reported the cycloaddition reaction of the titanocene−sulfido
complex with meta-tolunitrile, while reactions with the ortho
and para isomers were not discussed.24
1
Integration of H NMR spectra indicated a monoinsertion
product with one benzonitrile molecule incorporated. Using
two or more equivalents of nitriles did not lead to formation of
other products. This monoinsertion mode is consistent with
Scheme 5. [2+2]-Cycloaddition Reaction of 2b with
Benzonitriles
The azaoxatitanacyles 8a and 8b underwent nitrile exchange
reactions at room temperature (Scheme 6). This reaction was
approached from both directions, and an approximately 6:1
D
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
b
Scheme 6. Nitrile Exchange Reaction of 8a and 8b
Scheme 8. [2+2]-Cycloaddition Reaction of 2b with
Benzophenone Imine
step in developing our proposed catalytic cycle was to develop
conditions that promote carbonyl insertion into the Ti−C
bond. These studies would allow direct comparison to the
related oxazirconoacyclobutenes, which are known to insert
into carbonyls to generate dioxazirconacyclohexenes.
a
b
Determined by NMR integration. Reaction conditions: 1.0 equiv of
oxaazatitanacyclobutene, 1.0 equiv of nitrile.
We recently reported facile reductive elimination from 5a
and 5b upon treatment with phenyl ketone derivatives to
generate an oxatitanacyclopropane intermediate (Scheme 9).19
This intermediate has been successfully intercepted in a
number of transformations, including carbonyl insertion and
C−H, C−C, and C−X (X = F, Cl) activation.
ratio of 8b to 8a was observed in both cases after 3 h, as
determined by integration of H NMR spectra. In addition,
titanacycle 8a reacted with para-methoxyphenylacetylene to
generate 5b and free nitrile in almost quantitative yield after 2.5
h (Scheme 7). In comparison, the alkyne exchange reaction of
1
b
Scheme 7. Reaction of 9a with Terminal Alkyne
Scheme 9. Reaction of 6a with (a) Benzophenone and (b)
Trifluoroacetophenone
a
b
Determined by 1H NMR integration. Reaction conditions: 1.0 equiv
of oxaazatitanacyclobutene, 1.0 equiv of alkyne.
1
8b was significantly slower, reaching 50% conversion (by H
NMR) after 24 h. The equilibrium favors the oxatitanacyclo-
butene, as no reaction was observed when 5b was treated with
benzonitriles.
[2+2]-Cycloaddition Reaction of Titanocene−oxo
Complexes with Imines. Alkynes, nitriles, and carbonyl
compounds have been shown to undergo cycloaddition
reactions with a number of terminal chalcogenido complexes
of zirconium and titanium. One class of unsaturated organic
substrates yet to be studied in this reaction is imine. To explore
the reactivity of titanocene−oxo toward this functional group,
2b was treated with N-phenyl benzophenone imine, N-phenyl
para-methoxybenzaldehyde imine, and benzophenone imine in
benzene-d6. While the first two compounds did not show any
change by 1H NMR spectra, reaction with benzophenone imine
To further study the impact of substituents on the
metallacycle on this transformation, disubstituted metallacycle
8a was treated with 2,2,2-trifluoroacetophenone and 4-
methoxybenzaldehyde. In contrast to the facile reaction of 5b
with trifluoroacetophenone, which resulted in C−Cp* bond-
forming reductive elimination and subsequent C−F bond
activation, no reaction was observed between 7a with this 2,2,2-
trifluoroacetophenone (Scheme 10a). One explanation is that
the second substituent on the metallacycle hinders the
approach of the ketone, preventing the Lewis base promoted
ring slip/reductive elimination.
1
immediately resulted in an orange-red solution, and H NMR
showed formation of a new product with a Cp* peak
significantly more upfield (1.761 ppm) than that of 2b (1.912
ppm). This product is assigned to be the cyclic ketal 9 (Scheme
8) based on NMR evidence that the two phenyl rings are
equivalent and no imine signal is found in the 13C NMR
spectrum. This observation that 2b reacts with benzophenone
imine and not with N-phenyl-para-methoxybenzaldehyde imine
can be attributed to the increase in steric bulk on the imine
nitrogen, hindering its approach to the Cp*TiO.
Reaction of 7a with 4-methoxybenzaldehyde resulted in a
mixture of products from which the free alkyne could be
observed by 1H NMR by comparison with an authentic sample
Reactions of Oxatitanacyclobutenes with Carbonyl
Compounds. With oxatitanacyclobutenes in hand, the next
E
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
of reaction conditions (−78 to 25 °C; polar and nonpolar
solvents, etc.). The lack of carbonyl insertion into the Ti−C
bond of oxatitanacyclobutenes under a wide variety of
conditions is a critical setback in our overall goal of developing
the alkyne−aldehyde coupling reaction.
Scheme 10. Reactions of 7a with Carbonyl Compounds
Reaction of Titanocene−oxo with α,β-Unsaturated
Carbonyls. Although we were unable to synthesize the desired
dioxatitanacyclohexene via carbonyl insertion, we were
interested in investigating the reactivity of these complexes
relative to the analogous dioxazirconacyclohexenes.6 On the
basis of our recent investigations into ZrO complexes, we
proposed that these titanacycles could be prepared by [4+2]-
cycloaddition reaction of titanocene−oxo complexes with α,β-
unsaturated carbonyls, as similar reactivity has been reported
with titanocene−sulfido complex Cp*2TiS(py).24 Complexes
2a and 4a reacted with chalcone 10a and its derivatives to
generate dioxatitanacyclohexenes 11 and 12, respectively, in
quantitative NMR yield and good to excellent isolated yields
(Table 5). Both electron-donating and -withdrawing groups on
the enones were well tolerated.
a
No reaction was observed at room temperature.
(Scheme 10b). Since 7a did not undergo reductive elimination,
a different pathway must be invoked to explain its reaction with
aldehyde. As previously proposed in the alkyne exchange
reaction of complexes 5, oxatitanacyclobutene 7a could
undergo a retro-[2+2]-cycloaddition, followed by trapping of
the free titanocene−oxo by an aldehyde (Scheme 11). This
Scheme 11. Proposed Mechanism for the Reaction of 7a
with Aldehydes
Table 5. Scope of Enones in [4+2]-Cycloaddition Reaction
a
with Titanocene−oxo
b
entry
Cpx, Ar1, Ar2
product
yield
1
Cp*, Ar1 = Ar2 = Ph (a)
11a
11b
11c
11d
11e
11f
65%
82%
78%
74%
81%
88%
81%
85%
60%
58%
61%
75%
80%
2
Cp*, p-CF3-C6H4 ; p-MeO-C6H4 (b)
Cp*, p-CF3-C6H4 ; p-Me-C6H4 (c)
Cp*, p-CF3-C6H4 ; Ph (d)
Cp*, p-CF3-C6H4 ; p-Cl-C6H4 (e)
Cp*, Ar1 = Ar2 = p-CF3-C6H4 (f)
Cp*, p-NMe2-C6H4; p-CF3-C6H4 (g)
Cp*, p-MeO-C6H4; p-CF3-C6H4 (h)
Cp*, Ph; p-CF3-C6H4 (i)
Cp*, p-Br-C6H4; p-CF3-C6H4 (j)
Cp′, Ar1 = Ar2 = Ph (a)
Cp′, p-NMe2-C6H4; p-CF3-C6H4 (g)
Cp′, p-MeO-C6H4; p-CF3-C6H4 (h)
hypothesis was supported by observation that reaction of
DMAP-trapped oxo complex 2c with 4-methoxybenzaldehyde
generated a mixture whose NMR spectrum superimposed with
the spectrum of the reaction shown in Scheme 10.23 In the
reactions of 5 with aldehydes, the reductive elimination
pathway was faster and might account for most of the products.
In our recent study on the reactivity of oxatitanacyclobu-
tenes, we suggested that the crowded metal center was crucial
to the reductive elimination step. To test this hypothesis,
complex 7a was treated with 4,4′-bismethoxybenzophenone
(Scheme 12). No reductive elimination was observed; rather
3
4
5
6
7
11g
11h
11i
8
9
10
11
12
13
11j
12a
12g
12h
a
Reaction conditions: oxo complex 0.26 mmol, 1.0 equiv, enone 0.26
Scheme 12. Treatment of Oxatitanacyclobutene 6a with
Benzophenone Resulted in No Reaction23
mmol, 1.0 equiv, toluene 4 mL, reactions were stirred at rt for 3 h.
Isolated yield.
b
Single crystals suitable for X-ray diffraction crystallography
were grown by cooling a concentrated hexane solution of 11a.
Selected bond lengths and angles are included in Table 6. As
illustrated in the ORTEP representation (Figure 4), the metal is
Table 6. Selected Bond Lengths and Angles for 11a
1
only slight shifts were seen in the H and 13C NMR spectra,
suggesting coordination but no reductive elimination or
carbonyl insertion. One possible explanation is that the smaller
size of the Cp′ ligand allows for ketone coordination without
invoking ring slip, which is required for reductive elimination.26
Further, unlike analogous azatitanacyclobutene16,17 and aza-
and oxazirconacyclobutenes,6,27,28 no carbonyl insertion prod-
ucts were observed when 5a−e and 6a were combined with
various aldehydes (RCOH; R = Ph, Bu, H) under a wide array
bond
length (Å)
bond
angle (deg)
Ti−O1
Ti−O2
O1−C11
O2−C13
C11−C12
C12−C13
1.871(7)
1.964(6)
1.410(6)
1.500(13)
1.341(9)
1.511(8)
O1−Ti−O2
89.82(9)
127.0(6)
118.7(7)
105.8(4)
123.7(2)
125.3(5)
Ti−O1−C11
Ti−O2−C13
O2−C13−C12
C13−C12−C11
C12−C11−O1
F
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
state when it is located on the phenol ring. This substituent
effect was further studied with Hammett experiments in the
later part of our study.
The formation of dioxatitanacyclohexene product can be
considered the result of the oxo species [Cp*2TiO] being
trapped by an enone. We hypothesized that other reagents that
react more favorably with titanium−oxo should also promote
the generation of α,β-unsaturated ketones from dioxatitanacy-
clohexenes. One class of such reagents is strong Lewis bases
such as 4-dimethylaminopyridine (DMAP), which can stabilize
the monomeric titanocene−oxo species. Reactions of DMAP
with dioxatitanacyclohexenes 11 and 12 generated the
corresponding enones and the DMAP-trapped titanium−oxo
complexes 2c and 4c, respectively (Table 7). These oxo
complexes were independently synthesized via ligand exchange
reaction of the corresponding pyridine-trapped oxo complexes
2a and 4a with DMAP (Scheme 14).
Figure 4. ORTEP diagram of Cp*2Ti[OC(Ph)CHCH(Ph)O] (11a).
Hydrogen atoms are omitted for clarity. Ellipsoids are drawn at the
50% probability level.
in a pseudotetrahedral environment. This complex contains a
boat-like titanacycle with the C13 phenyl ring at the equatorial
position. The Ti−O bonds are within the range reported for
Ti−O single bonds with the Ti−alkoxide (Ti−O2) bond length
longer than the Ti−enolate (Ti−O1) bond length (1.964(6) Å
compared to 1.871(7) Å).
Table 7. Scope of Dioxatitanacyclohexenes in the Retro-
[4+2]-Cycloaddition and Subsequent Trapping with DMAP
a
Retro-[4+2]-Cycloaddition of Dioxatitanacyclohexene
Complexes 11 and 12. In the preparation of complexes 11,
we observed that with electron-donating groups on both of the
phenyl rings the product was not stable and decomposed to the
enone and metal−oxo cluster. This cluster was the oligomeriza-
tion product of free [Cp*2TiO], which was potentially
generated from retro-[4+2]-cycloaddition. This reaction has
been proposed in the reactions of titanacycles but has not been
studied in detail. To probe the mechanism, dioxatitanacyclo-
hexenes 11 and 12 were treated with α,β-unsaturated ketones,
and enone exchange reactions were observed. The rate of this
reaction was strongly dependent on the electronic nature of the
substituents on the phenyl rings; slower reactions were
observed with more electron deficient metallacycles. Between
the two phenyl rings on the metallacyles, electron-donating
substituents on the enol phenyl ring had a more profound
impact in promoting this enone exchange reaction (Scheme
13). Starting with an equimolar solution of 11b and 10f, 50%
conversion to 11f to 10b was reached in 4.2 h, while reaction
between 11h and 10f had a t1/2 of only 1.5 h. These
observations suggest that in the retro-[2+2]-cycloaddition step
an electron-donating group can better stabilize the transition
b
b
entry
Cpx, R1, R2
yield 10
yield 2c (4c)
1
2
3
4
5
Cp*, H, H (11a)
Cp*, MeO; CF3 (11h)
Cp′, H, H (12a)
Cp′, MeO; CF3 (12h)
Cp′, NMe2; CF3 (12g)
40%
80%
38%
c
37%
d
0%
0%
e
c
34%
35%
f
81%
69%
a
Reaction conditions: 1 equiv of metallacycle (0.02 M in C6D6), 20
b
equiv of DMAP. NMR yield determined by comparison of the
c
product to an internal standard in 1H NMR spectra after 19 h. Lower
yield of oxo compared to enone was attributed to low solubility of oxo
product, as it precipitated out of solution. No reactions at 80 °C and
longer reaction time. Reaction was run at 60 °C. Yield determined
d
e
f
after 5 h.
Scheme 13. Enone Exchange Reaction of 11b and 11h
Similar to the enone exchange reaction, the rate of reactions
between dioxatitanacyclohexenes 11 and 12 with DMAP is
strongly influenced by electronic effects: slower reactions were
Scheme 14. Synthesis of DMAP-Trapped Titanocene−oxo
Complexes 2c and 4c via Ligand Exchange
G
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
observed with more electron deficient metallacycles. Sub-
stitutions on the phenyl rings and ligands on the metal were
both important. Electron-donating substituents led to signifi-
cant increases in yield (Table 7, entries 1 and 2, entries 4 and
5). Reactions of metallacycles bearing the Cp′ ligand were
much slower, as evident from reactions of 11h and 12h:
reaction of 11h with DMAP afforded an 80% yield of 10h in 19
h at room temperature (Table 7, entry 2), while 12h only gives
34% of product 10h at 60 °C in the same reaction time (Table
7, entry 4). This facile cycloreversion and its dependence on
electronic effects are different from reactions of dioxazircona-
cyclohexenes, which require heating over 100 °C.19 A similar
retro-[4+2]-cycloaddition from diazatitanacyclohexenes has
been reported to be part of a titanium−imido-catalyzed
imine−alkyne coupling reaction; in this case also, high
temperature is required to promote the cycloreversion.3 In
comparison to these systems, the facile cycloreversion of
oxatitanacycles bearing a Cp* ligand is quite remarkable. One
possible explanation is the release of strain. Observation that
dioxatitanacyclohexenes bearing the less bulky Cp′ ligand
undergo significantly slower reactions supports this hypothesis.
Next, mechanistic experiments were undertaken to gain
insight into the cycloreversion reaction. All kinetic inves-
tigations were preformed using initial rate kinetics, as clean
kinetic profiles were not observed to three half-lives, as
decomposition was observed after extended reaction times.27
The reaction of 11b with DMAP is first order in 11b (Figure 5)
Figure 6. Inversed observed rate constants increase with the enone/
DMAP ratio. The linear relationship is consistent with the rate law in
eq 1.
k1k2[metallacycle][DMAP]
rate =
k−1[enone] + k2[DMAP]
(1)
Hammett plots were constructed for substituents on each of
the two phenyl rings while keeping the other ring as p-
CF3C6H4. Negative ρ values of −1.18 and −1.09 were obtained
when σ+ parameters were used (Figures 7 and 8), similar to that
Figure 5. Natural log plot of the retro-[4+2]-cycloaddition reaction
showing the reaction is first order in titanacycle 11b.
Figure 7. Hammett plot for substituents on the enolic phenol ring
with σ+ (top) and σp (bottom).
and zero order in DMAP, supporting a dissociative mechanism
in which the retro-[4+2]-cycloaddition is the rate-determining
step (Scheme 15). Product inhibition was observed when the
enone product was added to the reaction (Figure 6), suggesting
a reversible cycloreversion step. These empirical data are
consistent with the rate law (eq 1) derived from the proposed
mechanism.
previously obtained in the retro-[4+2]-cycloaddition reactions
of azaoxazirconacyclohexenes.17 A plot of σp rather than σ+
does not give a linear correlation for substituents on the allylic
phenol ring but produces ρ values of −2.07 with substituents
on the enolic phenol ring.23 The negative values indicate that
positive charge is developed in the transition state, and the
small absolute values imply slight deviation from a synchronous
retro-[4+2]-cycloaddition. The more negative ρ value asso-
ciated with the enolic phenyl ring is consistent with
experimental observation that placing electron-donating sub-
stituents on this ring increases the reaction rate more than
when it is placed on the other ring.
Scheme 15. Proposed Mechanism of Retro-[4+2]-
Cycloaddition Reaction of Dioxatitanacyclohexenes
CONCLUSION
■
The synthesis and reactivity of Ti−oxo complexes have been
explored. We have demonstrated that pyridine N-oxide and
styrene oxide are excellent alternatives to nitrous oxide as
oxidants in the synthesis of monomeric titanocene−oxo
H
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
for the synthesis of enones. Over the course of these
experiments, we demonstrated that carbonyl insertion into
titanacyclobutenes does not occur; rather an array of other
reactions, including reductive elimination and retro-[2+2]-
cycloaddition, take place preferentially. This is especially
unfortunate, as the [4+2]-retrocycloaddition from the dioxati-
tanacyclohexene is relatively facile, especially in comparison to
the analogous zirconium complex.
In general, titanocene complexes are more reactive in both
cycloaddition and retro-cycloaddition reactions and less
reactive in insertion reactions compared to zirconocene
complexes. This difference in reactivity can be attributed to
the more crowded metal center in titanocene complexes. These
facile cycloaddition and cycloreversion reactions offer useful
insights into the development of titanium−oxo-mediated
reactions.
EXPERIMENTAL SECTION
■
General Experimental Procedures. All reactions were carried
out in oven-dried (at 140 °C, for at least 4 h) glassware under an
atmosphere of nitrogen unless otherwise indicated. Air- or moisture-
sensitive materials were synthesized and stored in a nitrogen-filled
glovebox. Column chromatography was performed with silica gel from
Grace Davison Discovery Sciences (35−75 μm), packed as a slurry,
and run under positive pressure. Analytical thin-layer chromatography
(TLC) was performed on precoated glass silica gel plates with F-254
indicator purchased from EMD Chemicals Inc. Visualization was done
by short-wave (254 nm) ultraviolet light. Distillations were performed
using a 3 cm short-path column either under reduced pressure or
under positive pressure of nitrogen.
Figure 8. Hammett plot for substituents on the allylic phenyl ring.
complexes with high yield and simple operation. Titanocene−
oxo complexes bearing less bulky ligands than Cp* have been
successfully synthesized through judicious oxidant and solvent
selection. These methods should facilitate future research on
these reactive oxo complexes. Further, our studies demonstrate
that titantium−oxo complexes undergo reversible cycloaddition
reactions with alkynes, nitriles, and α,β-unsaturated ketones to
form isolable metallacycle products. Finally, these complexes
undergo cycloreversion at room temperature in the presence of
reagents that can react with and trap the [Cpx2TiO] species.
The reactivity of these titanocene complexes differs
significantly from those of analogous zirconocene complexes.
Titanocene−oxo complexes react preferably with terminal
alkynes and nitriles to form four-membered metallacycles.
The resulting titanacyclobutenes undergo retro-[2+2]-cyclo-
addition reaction at room temperature, as evident by alkyne and
nitrile exchange reactions. Zirconocene−oxo complexes, on the
other hand, react preferentially with internal alkynes and form
double-addition products with benzonitrile, to afford the six-
membered metallacycle.25 The [2+2]-cycloaddition products
oxazirconacyclobutenes undergo facile aldehyde insertion to
form dioxazirconacyclohexenes, while oxatitanacyclobutenes
react with aldehydes in a nonconstructive fashion due to a
combination of favorable retro-[2+2]-cycloaddition and reduc-
tive elimination of the Cp* ligand. The facile reductive
elimination of a Cp* ligand and a Ti−C bond from the
titanacycle is a unique and fascinating reactivity of oxatitana-
cyclobutenes.8 In comparison, the Cp* ligand in oxazircona-
cyclobutenes reacts only at 160 °C and through C−H
activation at one of the methyl groups. Finally, dioxametalla-
cyclohexene complexes of these two metals offer direct
comparison in retro-[4+2]-cycloaddition. Titanocene metalla-
cycles 11 and 12 undergo reaction at room temperature with
high yields and recapture of the oxo complex, while their
zirconium counterparts require temperatures above 120 °C to
affect this transformation. In addition, stabilizing the reactive
zirconocene oxo with dative ligands is difficult, and chalcone
has been used instead as a trapping reagent, forming another
dioxazirconacyclohexene.6
Instrumentation. 1H, 13C, and 19F NMR spectra were recorded on
a Varian Unity 500 MHz (125 MHz for 13C, 470 MHz for 19F)
spectrometer. Spectra were collected in CDCl3 or C6D6 and were
referenced using residual protic solvent (1H NMR: 7.26, 13C NMR:
1
77.16 ppm for CDCl3, H NMR: 7.15, 13C NMR: 128.06 ppm for
C6D6). 19F NMR were referenced internally using C6F6 (19F NMR:
−163.04 ppm). Chemical shifts were reported in parts per million
(ppm), and multiplicities are reported as follows: s (singlet), d
(doublet), t (triplet), q (quartet), p (pentet), and m (multiplet).
Coupling constants (J) are reported in hertz, and integrations are
provided. High-resolution mass spectrometry was performed at the
School of Chemical Sciences Mass Spectrometry Laboratory located at
the University of Illinois at Urbana−Champaign. X-ray crystallography
was done at the George L. Clark X-ray Facility and 3M Materials
Laboratory at the University of Illinois at Urbana−Champaign.
Microanalysis was performed at the School of Chemical Sciences
Microanalysis Laboratory located at the University of Illinois at
Urbana−Champaign. Bulk purity of samples is represented in
1
Supporting Information by H, 13C, and 19F NMR spectra.
Materials. Solvents used for extraction, column chromatography,
and recrystallizations of air-stable materials were reagent grade and
used as received. Solvents for reactions, extractions, and recrystalliza-
tions of air- and water-sensitive materials were dried on a Pure Process
Technology Glass S3 Contour Solvent Purification system equipped
with activated stainless steel columns following manufacture’s
recommendations for solvent preparation and dispensing. Solvents
were then further dried by storing over 4 Å molecular sieves that had
been activated by heating to 200 °C under dynamic vacuum for at least
24 h. Pyridine, 3,5-lutidine, phenylacetylene, hexafluorobenzene, and
styrene oxide were distilled under an atmosphere of nitrogen,
transferred to a nitrogen-filled glovebox, and stored over activated
molecular sieves for at least 24 h prior to use. C6D6 was degassed by
freeze−pump−thaw cycles, stored in a nitrogen-filled glovebox, and
dried over activated molecular sieves for at least 24 h prior to use.
(TiCl3)3AlCl3, diphenylacetylene (Alfa Aesar), C2H4 (Specialty Gases
of America), n-BuLi, 1-phenyl-1-propyne, 3-phenyl-1-propyne, chal-
cone, benzonitrile and derivatives (Aldrich), DMAP (ACROS
Organic), 4′-methoxyphenylacetylene, and 4′-fluorophenylacetylene
These studies were undertaken with the eventual goal of
developing a Ti-catalyzed aldehyde−alkyne coupling reaction
I
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(Oakwook Chemical) were used as received. Pentamethylcyclopenta-
diene was obtained from Boulder Scientific and distilled prior to use.
Lithium cyclopentadienides were prepared by dissolving the
corresponding cyclopentadiene in hexane followed by slow addition
of 1.11 equiv of 1.6 M n-BuLi in hexane (Aldrich) at 0 °C. The
mixture was allowed to warm to room temperature overnight with
stirring. The product was collected by filtration, washed with hexane,
and dried under vacuum. The following compounds were synthesized
by literature procedures: TiCl3(THF)3,29 3,5-lutidine N-oxide,30 3,5-
bis(trifluoromethyl)phenylacetylene,31 chalcone derivatives 9b to 9j.32
Bis(pentamethylcyclopentadienyl)titanium(IV) Dichloride.
This procedure was adopted from a previously reported synthesis of
titanocene dichloride with TiCl3(THF)3.33 For an earlier synthesis of
Cp*2TiCl2 reported by Bercaw see ref 34. In a glovebox, TiCl3(THF)3
(5.03 g, 13.58 mmol, 1.0 equiv), Cp*Li (4.82 g, 33.94 mmol, 2.5
equiv), and THF (60 mL) were combined in a 250 mL round-
bottomed, thick-wall vessel. The vessel was sealed with a screw cap,
and the reaction was heated at reflux overnight in a fume hood. The
reaction was allowed to cool to room temperature, then further cooled
in an ice bath. Concentrated HCl (55 mL) was added with stirring,
and the reaction mixture went from purple-blue to reddish-brown. The
mixture was extracted with chloroform (3 × 60 mL), dried over
anhydrous magnesium sulfate, filtered, and evaporated to dryness. The
dark brown solid residue was dissolved in chloroform (30−40 mL) in
a 500 mL beaker; then methanol (300 mL) was carefully layered on
top of this solution. After a day or two at room temperature, large
needles of Cp*2TiCl2 formed and were collected by filtration, washed
with methanol, and dried under vacuum. Dark reddish-purple, long
needles, 3.59 g, 68% yield. 1H NMR (500 MHz, chloroform-d): δ 2.00
(s, 30H). 13C NMR (126 MHz, chloroform-d): δ 128.64, 13.06.
Bis(tetramethylcyclopentadienyl)titanium(IV) Dichloride,
Cp′2TiCl2. This complex was prepared following the procedure
described above for Cp*2TiCl2 using TiCl3(THF)3 and Cp′Li. Dark
Bis(pentamethylcyclopentadienyl)titanium(IV)−oxo,
Cp*2TiO(3,5-lutidine) (2b). 2b was prepared following the
procedure described above for 2a, using 1a (1.0 equiv), 3,5-lutidine
(2.3 equiv), and 3,5-lutidine N-oxide (1.0 equiv). Light orange solid,
1
84% yield. H NMR (500 MHz, benzene-d6): δ 8.70 (s, 1H), 7.29 (s,
1H), 6.50 (s, 1H), 1.91 (s, 30H), 1.81 (s, 3H), 1.64 (s, 3H). 13C NMR
(126 MHz, benzene-d6): δ 153.54, 147.90, 138.00, 134.40, 131.49,
117.06, 17.89, 17.61, 12.10. Anal. Calcd for C27H39NOTi: C, 73.46; H,
8.90; N, 3.17. Found: C, 73.05; H, 8.78; N, 3.46.
Bis(pentamethylcyclopentadienyl)titanium(IV)−oxo,
Cp*2TiO(DMAP) (2c). Titanocene−oxo complex 2a (0.1 g, 0.242
mmol, 1.0 equiv) was dissolved in toluene (4 mL) in a 20 mL
scintillation vial containing a magnetic stirbar. DMAP (29 mg, 0.242
mmol, 1.0 equiv) was added with stirring. The reaction was stirred at
room temperature for 2 h, and the product was collected by filtration
as a light yellow solid, 99 mg, 90% yield. 1H NMR (500 MHz,
3
3
benzene-d6): δ 8.70 (d, JHH = 6.6 Hz, 1H), 7.34 (d, JHH = 6.5 Hz,
1H), 5.82 (m, 2H), 2.03 (s, 30H), 2.02 (s, 6H). Low solubility did not
allow for a 13C NMR. Anal. Calcd for C27H40N2OTi: C, 71.04; H,
8.83; N, 6.14. Found: C,70.96; H, 8.69; N, 6.22.
Bis(tetramethylcyclopentadienyl)titanium−Ethylene,
Cp′2Ti(C2H4) (3). 3 was prepared following the procedure described
1
for 1 starting with Cp′2TiCl2. Green crystalline solid, 69% yield. H
NMR (500 MHz, benzene-d6): δ 4.10 (s, 2H), 2.44 (s, 12H), 2.03 (s,
4H), 1.07 (s, 12H). 13C NMR (126 MHz, benzene-d6): δ 123.91,
119.05, 111.91, 102.69, 13.48, 12.65. Anal. Calcd for C20H30Ti: C,
75.46;H, 9.50. Found: C, 75.21; H, 9.50.
Bis(tetramethylcyclopentadienyl)titanium(IV)−oxo,
Cp′2TiO(py) (4a). Titanocene−ethylene complex 3 (0.92 g, 2.89
mmol, 1.0 equiv) was dissolved in n-hexane (28 mL) in a 100 mL
round-bottom flask containing a magnetic stirbar, followed by addition
of pyridine (0.56 mL, 6.94 mmol, 2.4 equiv), and the mixture was
stirred for 30 min. A solution of styrene oxide (0.33 mL, 2.89 mmol,
1.0 equiv) in hexane (2 mL) was added dropwise. The reaction
mixture gradually turned dark yellow with formation of solid particles.
The reaction mixture was stirred overnight, and a yellow powder
product was collected by filtration and washed with cold pentane.
1
red flakes, 65% yield. H NMR (500 MHz, chloroform-d): δ 6.00 (s,
2H), 2.05 (s, 12H), 1.90 (s, 12H). 13C NMR (126 MHz, chloroform-
d): δ 135.89, 125.22, 113.19, 14.98, 13.20.
Bis(pentamethylcyclopentadienyl)titanium−Ethylene,
Cp*2Ti(C2H4) (1). This complex was prepared using a modified
procedure from a previously reported method.35 In a 250 mL Schlenk
flask containing a magnetic stirbar and nitrogen inlet were combined
Cp*2TiCl2 (1.97g, 5.06 mmol, 1.0 equiv) and toluene (60 mL). To the
reaction mixture was added a 1.5 M solution of n-BuLi in hexane (7.4
mL, 2.2 equiv) over 30 min at −78 °C, and the solution was stirred for
1 h. The headspace was quickly evacuated and put under static
vacuum; then an ethylene balloon was attached. After removal of the
balloon, the cooling bath was removed and the reaction was allowed to
warm to room temperature. The reaction was accompanied by the
generation of gas and a color change from red to brownish-green. After
48 h, all volatiles were removed under vacuum and the resulting
brownish-green residue was taken up in hexane and filtered. The
filtrate was evaporated to dryness, and the resulting green solid was
washed with cold pentane. Green crystalline solid, 1.4 g, 80% yield. 1H
NMR (500 MHz, benzene-d6): δ 2.04 (s, 4H), 1.68 (s, 30H). 13C
NMR (126 MHz, benzene-d6): δ 119.75, 104.96, 11.84.
1
Yellow solid, 0.67 g, 60% yield. H NMR (500 MHz, benzene-d6): δ
8.96 (d, 3JHH = 5.9 Hz, 1H), 7.53 (d, 3JHH = 5.8 Hz, 1H), 6.72 (t, 3JHH
= 7.7 Hz, 1H), 6.41 (m, 2H), 5.45 (s, 2H), 2.31 (s, 6H), 2.20 (s, 6H),
1.64 (s, 6H), 1.41 (s, 6H). 13C NMR (126 MHz, benzene-d6): δ
156.93, 149.95, 136.80, 124.86, 122.92, 121.64, 120.06, 116.91, 112.05,
110.02, 14.86, 13.19, 12.11, 11.18. Anal. Calcd for C23H31NOTi: C,
71.68; H, 8.11; N, 3.63. Found: C, 71.43; H, 7.99; N, 3.34.
Bis(tetramethylcyclopentadienyl)titanium(IV)−oxo,
Cp′2TiO(3,5-lutidine) (4b). 4b was prepared following the
procedure described for 4a, starting with 3, 3,5-lutidine, and styrene
1
oxide. Yellow solid, 62% yield. H NMR (500 MHz, benzene-d6): δ
8.77 (s, 1H), 7.37 (s, 1H), 6.48 (s, 1H), 5.52 (s, 2H), 2.36 (s, 6H),
2.24 (s, 6H), 1.76 (s, 3H), 1.72 (s, 6H), 1.60 (s, 3H), 1.44 (s, 6H). 13C
NMR (126 MHz, benzene-d6): δ 154.04, 147.67, 138.37, 134.62,
131.99, 121.40, 120.09, 116.65, 112.03, 109.96, 17.84, 17.64, 14.92,
13.27, 12.14, 11.20. Anal. Calcd for C25H35NOTi: C, 72.63; H, 8.53;
N, 3.39. Found: C, 72.68; H, 8.52; N, 3.61.
Bis(tetramethylcyclopentadienyl)titanium(IV)−oxo,
Cp′2TiO(DMAP) (4c). 4c was prepared following the procedure
described for 2c, starting with 4a and DMAP. Light yellow solid, 79%
Bis(pentamethylcyclopentadienyl)titanium(IV)−oxo,
Cp*2TiO(py) (2a). In a glovebox, titanocene−ethylene complex 1
(0.5 g, 1.45 mmol, 1.0 equiv) was dissolved in THF (10 mL) in a 100
mL round-bottom flask containing a magnetic stirbar. To this solution
was added pyridine (0.23 mL, 2.89 mmol, 2.0 equiv), and the mixture
was stirred for 30 min before a solution of pyridine N-oxide (1.45
mmol, 1.0 equiv) in THF (2.0 mL) was added dropwise. The reaction
mixture was stirred at room temperature for 5 h; then all volatiles were
removed under vacuum, resulting in a yellow solid. This solid was
scrapped onto a fritted funnel and washed with hexane, giving the
1
3
yield. H NMR (500 MHz, benzene-d6): δ 8.72 (d, JHH = 6.5 Hz,
1H), 7.37 (d, 3JHH = 6.4 Hz, 1H), 5.79 (m, 2H), 5.58 (s, 2H), 2.41 (s,
6H), 2.33 (s, 6H), 2.02 (s, 6H), 1.86 (s, 6H), 1.67 (s, 6H). Low
solubility did not allow for 13C NMR. Anal. Calcd for C25H36N2OTi:
C, 70.09; H, 8.47; N, 6.54. Found: C, 70.18; H, 8.24; N, 6.63.
Bis(pentamethylcyclopentadienyl)phenyloxatitana-
cyclobutene, Cp*2Ti[OC(Ph)CH] (5a). To a 20 mL scintillation vial
was added titanocene−oxo 2a (330 mg, 0.80 mmol, 1.0 equiv),
toluene (5 mL), and a magnetic stirbar. Phenylacetylene (105 μL, 0.96
mmol, 1.2 equiv) was added dropwise with stirring; an immediate
color change from orange to green was observed. The reaction mixture
was stirred at room temperature for 2 h before all volatiles were
removed under vacuum. The dark green residue was taken up in
1
product as a yellow solid, 0.48 g, 80% yield. H NMR (500 MHz,
benzene-d6): δ 8.91 (s, br, 1H), 7.50 (s, br, 1H), 6.75 (t, 3JHH = 7.5 Hz,
3
3
1H), 6.46 (dd, JHH = 7.4 Hz, JHH = 6.6 Hz, 2H), 1.86 (s, 30H). 13C
NMR (126 MHz, benzene-d6): δ 156.48, 150.01, 136.46, 124.66,
122.57, 117.13, 12.06.
J
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
hexane and filtered to remove any solid. The filtrate was concentrated
and cooled to −30 °C overnight, affording dark green crystals. The
product was collected by filtration, washed with cold hexane, and dried
d6): δ 169.05, 143.70, 128.49, 128.04, 127.22, 122.38, 121.95, 18.73,
11.97. HRMS (ESI): calcd for C29H38OTi 450.2402 [M]+, found
450.2405. Anal. Calcd for C29H38OTi·(1/6)H2O: C, 76.81; H, 8.52.
Found: C, 76.81; H, 8.45.
1
under vacuum overnight, 331 mg, 95% yield. H NMR (500 MHz,
3
5
Cp*2Ti[OC(C6H5)C(C6H5)] (7b). 7b was prepared following the
procedure described for 7a, starting with 2b and diphenylacetylene.
Dark brown solid, 43% yield. 1H NMR (500 MHz, benzene-d6): δ 8.02
benzene-d6): δ 8.01 (dd, JHH = 8.2 Hz, JHH = 1.5 Hz, 2H), 7.67 (s,
1H), 7.30 (dd, 3JHH = 8.0 Hz, 3JHH = 7.6 Hz, 2H), 7.12 (tt, 3JHH = 7.3
Hz, 5JHH = 1.3 Hz, 1H), 1.71 (s, 30H). 13C NMR (126 MHz, benzene-
d6): δ 167.24, 135.68, 128.30, 126.53, 125.23, 121.45, 11.63. One
signal is underneath solvent peaks. NMR data were in agreement with
a previous report.12
3
4
(dd, JHH = 8.3 Hz, JHH = 1.4 Hz 2H), 7.17 (m, 2H), 7.14 (m, 2H),
7.07 (m, 3H), 6.91 (tt, J = 7.3, 1.3 Hz, 1H), 1.73 (s, 30H). 13C NMR
(126 MHz, benzene-d6): δ 170.62, 143.44, 136.36, 131.96, 128.98,
128.73, 128.17, 127.06, 126.41, 122.86, 122.74, 11.92. NMR data were
in agreement with a previous report.12
Bis(pentamethylcyclopentadienyl)(4-methoxyphenyl)-
oxatitanacyclobutene, Cp*2Ti[OC(4-OMe-C6H4)CH] (5b). 5b was
prepared following the procedure described for 6a, starting with 2a
and 4-methoxyphenylacetylene. Dark green solid, 92% yield. 1H NMR
Cp*2Ti[OC(3-Br-C6H4)N] (8a). To a 20 mL scintillation vial was
added titanocene−oxo 2b (168 mg, 0.38 mmol, 1.0 equiv), toluene (3
mL), and a magnetic stirbar. 3-Bromobenzonitrile (173 mg, 0.95
mmol, 2.5 equiv) was added with stirring; an immediate color change
from orange to red was observed. The reaction was stirred at room
temperature overnight, then kept at −30 °C for 3 h. The product was
collected by filtration, washed with cold hexane, and dried under
vacuum. Red powder, 161 mg, 82% yield. 1H NMR (500 MHz,
benzene-d6): δ 8.79 (s, 1H), 8.34 (d, 3JHH = 8.0 Hz, 1H), 7.32 (d, 3JHH
3
(500 MHz, benzene-d6): δ 7.94 (d, JHH = 9.0 Hz, 2H), 7.60 (s, 1H),
3
6.93 (d, JHH = 8.8 Hz, 2H), 3.34 (s, 3H), 1.75 (s, 30H). 13C NMR
(126 MHz, benzene-d6) δ 166.19, 159.32, 126.40, 121.37, 113.98,
54.82, 11.65, two signals are underneath solvent peaks. HRMS (ESI):
calcd for C29H38O2Ti 467.2430 [MH]+, found 467.2433. Anal. Calcd
for C29H38O2Ti·1/9(H2O): C, 74.35; 8.22. Found: C, 74.09; H, 7.82.
Bis(pentamethylcyclopentadienyl)[3,5-bis(trifluoromethyl)-
phenyl]oxatitanacyclobutene, Cp*2Ti[OC(3,5-(CF3)2-C6H3)CH]
(5c). 5c was prepared following the procedure described for 5a,
starting with 2a and 3,5-bis(trifluoromethyl)phenylacetylene. Dark
3
= 8.0 Hz, 1H), 6.92 (t, JHH = 7.8 Hz, 1H), 1.68 (s, 30H). 13C NMR
(126 MHz, benzene-d6): δ 143.53, 137.85, 132.00, 130.24, 129.97,
125.76, 123.61, 122.94, 11.38. Anal. Calcd for C27H34BrNOTi·(1/
3)H2O: C, 62.08; H, 6.69; N, 2.68. Found: C, 62.04; H, 6.51; N, 2.89.
Cp*2Ti[OC(3,5(CF3)2-C6H3)N] (8b). 8b was prepared following the
procedure described for 9a, starting with 2b and one equivalent of 3,5-
bis(trifluoromethyl)benzonitrile. Red powder, 85% yield. 1H NMR
(500 MHz, benzene-d6) δ 8.99 (s, 2H), 7.83 (d, 1H), 1.62 (s, 30H).
1
green solid, 88% yield. H NMR (500 MHz, benzene-d6): δ 8.44 (s,
2H), 7.69 (s, 1H), 7.58 (s, 1H), 1.57 (s, 30H). 13C NMR (126 MHz,
2
benzene-d6): δ 168.64, 138.48, 131.72 (q, JCF = 32.6 Hz), 126.97,
124.57 (q, 1JCF = 273.4 Hz), 124.33 (q, 3JCF = 3.9 Hz), 122.17, 119.36
3
(hept, JCF = 3.9 Hz), 11.47. 19F NMR (470 MHz, benzene-d6): δ
−62.92. HRMS (ESI): calcd for C30H34F6OTi 573.2071 [MH]+, found
573.2072. Anal. Calcd for C30H34F6OTi: C, 62.94; H, 5.99. Found: C,
62.79; H, 6.00.
13C NMR (126 MHz, benzene-d6): δ 142.53, 138.18, 131.81 (q, 2JCF
=
3
1
33.1 Hz), 126.89 (q, JCF = 4.3 Hz), 124.33 (q, JCF = 273.42 Hz),
124.08, 122.25 (hept, JCF = 3.9 Hz), 11.29. 19F NMR (470 MHz,
3
Bis(pentamethylcyclopentadienyl)(4-fluorophenyl)-
oxatitanacyclobutene, Cp*2Ti[OC(4-F-C6H4)CH] (5d). 5d was
prepared following the procedure described for 5a, starting with 2a
benzene-d6): δ −62.86. Anal. Calcd for C29H33F6NOTi·(3/11)H2O:
C, 60.23; H, 5.85; N, 2.42. Found: C, 60.25; H, 6.07; N, 2.56.
Cp*2Ti[OC(3,5(CF3)2-C6H3)N] (9). To a 20 mL scintillation vial
were added benzophenone imine (51 mg, 0.28 mmol, 1.0 equiv),
titanocene−oxo 2b (150 mg, 0.34 mmol, 1.2 equiv), toluene (1.4 mL),
and a magnetic stirbar. The resulting reddish-orange solution was
stirred for 24 h at room temperature before all volatiles were removed
under vacuum. The orange-red residue was taken up in hexane and
filtered to remove excess 2b. The filtrate was evaporated to yield 9 as a
very dense red oil, 117 mg, 81% yield. A small amount of
benzophenone imine proved to be difficult to remove from this oil.
1H NMR (500 MHz, benzene-d6): δ 7.65 (ddt, J = 13.8, 6.0, 1.4 Hz,
4H), 7.20 (td, J = 7.6, 6.1 Hz, 4H), 7.12 (m, 2H), 3.54 (s, 1H), 1.76 (s,
30H). 13C NMR (126 MHz, benzene-d6): δ 129.54, 128.64, 127.99,
127.79, 119.54, 11.88.
1
and 4-fluorophenylacetylene. Dark green solid, 70% yield. H NMR
(500 MHz, benzene-d6): δ 7.80 (dd, 3JHH = 8.8 Hz, 4JHF = 5.6 Hz, 2H),
7.50 (s, 1H), 6.95 (t, J = 8.8 Hz, 2H), 1.70 (s, 30H). 13C NMR (126
MHz, benzene-d6): δ 166.20 (d, 4JCF = 2.2 Hz), 163.54, 161.60, 131.91
(d, 3JCF = 2.8 Hz), 126.57 (d, 2JCF = 7.6 Hz), 121.56, 115.03 (d, 1JCF
=
21.3 Hz), 11.59. 19F NMR (470 MHz, benzene-d6): δ −117.66. HRMS
(ESI): calcd for C28H35FOTi 455.2230 [MH]+, found 455.2225. Anal.
Calcd for C28H35FOTi·2/9(H2O): C: 73.36; H, 7.79. Found: C, 73.37;
H, 7.66.
Bis(tetramethylcyclopentadienyl)[3,5-bis(trifluoromethyl)-
phenyl]oxatitanacyclobutene, Cp′2Ti[OC(3,5-(CF3)2-C6H3)CH]
(6a). 6a was prepared following the procedure described for 5a,
starting with 4a and 3,5-bis(trifluoromethyl)phenylacetylene. Dark
Cp*2Ti[OC(C6H5)CHCH(C6H5)O] (11a). To a 20 mL scintillation
vial were added chalcone 10a (54 mg, 0.26 mmol, 1.0 equiv),
titanocene−oxo 2b (115 mg, 0.26 mmol, 1.0 equiv), toluene (4 mL),
and a magnetic stirbar. The reaction mixture was stirred at room
temperature for 3 h before all volatiles were removed under vacuum.
Hexane (10 mL) was added to the vial, and the mixture was stirred
thoroughly and kept at −30 °C overnight. The product containing
trapped hexane (as shown in NMR spectra and crystal structure) was
collected by filtration, washed with cold hexane, and dried under
1
green solid, 94% yield. H NMR (500 MHz, benzene-d6): δ 8.43 (s,
2H), 7.70 (s, 1H), 7.59 (s, 1H), 5.37 (s, 2H), 1.72 (s, 6H), 1.51 (s,
6H), 1.50 (s, 6H), 1.40 (s, 6H). 13C NMR (126 MHz, benzene-d6): δ
2
167.62, 138.11, 131.73 (q, JCF = 32.6 Hz), 129.25, 127.98, 127.70,
124.58 (q, 3JCF = 3.8 Hz), 124.51 (q, 1JCF = 273.29 Hz), 120.60, 119.75
3
(hept, JCF = 3.9 Hz), 119.34, 111.02, 13.63, 13.49, 11.96, 10.81. 19F
NMR (470 MHz, benzene-d6): δ −62.94. HRMS (ESI): calcd for
C28H30F6OTi 545.1759 [MH]+, found 545.1750. Anal. Calcd for
C28H30F6OTi: C, 61.78; H, 5.55. Found: C, 61.36; H, 5.48.
1
vacuum. Dark brown solid, 99 mg, 65% yield. H NMR (500 MHz,
benzene-d6): δ 7.86 (dd, 3JHH = 8.4 Hz, 4JHH = 1.3 Hz, 2H), 7.74 (dd,
Cp*2Ti[OC(CH3)C(C6H5)] (7a). In a glovebox, titanium−oxo
complex 2b (213 mg, 0.484 mmol, 1.0 equiv) and 1-phenyl-1-propyne
(243 μL, 1.94 mmol, 4.0 equiv) were dissolved in toluene (5 mL) in a
20 mL scintillation vial containing a magnetic stirbar. The vial was
sealed with a Teflon-lined screw cap and heated at 60 °C for 3 days.
All volatiles were then removed under vacuum, resulting in a dark
brown, oily residue. This residue was dissolved in pentane (2.0 mL)
and kept at −30 °C for 3 days, affording light brown crystals. The
product was collected by filtration, washed with cold pentane, and
4
3
3
3JHH = 8.3 Hz, JHH = 1.3 Hz, 2H), 7.38 (t, JHH = 8.1 Hz, JHH = 7.4
3
3
3
Hz, 2H), 7.30 (t, JHH = 8.0 Hz, JHH = 7.5 Hz, 2H), 7.21 (tt, JHH
=
4
3
4
7.4 Hz, JHH = 1.5 Hz, 1H), 7.17 (tt, JHH = 7.4 Hz, JHH = 1.5 Hz,
1H), 6.64 (s, br, 1H), 5.55 (d, 3JHH = 2.0 Hz, 1H), 1.89 (s, 15H), 1.79
(s, 15H). 13C NMR (126 MHz, benzene-d6): δ 160.81, 148.97, 142.92,
128.32, 128.16, 127.09, 127.04, 126.53, 125.70, 125.28, 124.53, 100.02,
83.17, 12.40, 11.92. Anal. Calcd for C35H42O2Ti·1/2(C6H14): C,
77.93; H, 8.43. Found: C, 78.03; H, 8.44.
1
dried under vacuum. Light brown solid, 109 mg, 50% yield. H NMR
Cp*2Ti[OC(4-CF3-C6H4)CHCH(4-OMe-C6H4)O] (11b). 11b was
prepared following the procedure described for 11a, starting with 2b
and enone 10b. Brown solid, 82% yield. 1H NMR (500 MHz,
(500 MHz, benzene-d6): δ 7.31 (dd, 3JHH = 8.3, 7.3 Hz, 2H), 6.99 (tt,
3JHH = 7.3 Hz, 4JHH = 1.3 Hz, 1H), 6.97 (dd, 3JHH = 7.4 Hz, 4JHH = 1.4
Hz, 2H), 2.28 (s, 3H), 1.73 (s, 30H). 13C NMR (126 MHz, benzene-
3
3
benzene-d6): δ 7.75 (d, JHH = 8.0 Hz, 2H), 7.62 (d, JHH = 8.6 Hz,
K
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
2H), 7.49 (d, 3JHH = 8.0 Hz, 2H), 7.00 (d, 3JHH = 8.6 Hz, 2H), 6.59 (s,
br, 1H), 5.53 (d, 3JHH = 2.0 Hz, 1H), 3.41 (s, 3H), 1.88 (s, 15H), 1.76
(s, 15H). 13C NMR (126 MHz, benzene-d6): δ 159.02, 158.78, 146.17,
125.57 (q, 1JCF = 283.0 Hz), 125.41, 125.22 (q, 3JCF = 3.8 Hz), 124.61,
112.43, 95.89, 82.33, 40.38, 12.43, 11.92. 19F NMR (470 MHz,
benzene-d6): δ −62.07. HRMS (ESI): calcd for C38H46F3NO2Ti
654.3038 [MH]+, found 654.3057. Anal. Calcd for C38H46F3NO2Ti·1/
3(H2O): C, 69.19; H, 7.13; N, 2.12. Found: C, 68.9; H, 7.06; N, 2.27.
Cp*2Ti[OC(4-OMe-C6H4)CHCH(4-CF3-C6H4)O] (11h). 11h was
prepared following the procedure described for 11a, starting with 2b
and enone 10h. Brown solid, 85% yield. 1H NMR (500 MHz,
2
1
140.69, 128.65 (q, JCF = 31.9 Hz), 128.38, 125.70, 125.48 (q, JCF
=
272.16 Hz), 125.36, 125.10 (q, 3JCF = 3.8 Hz), 124.67, 113.83, 102.18,
82.95, 54.87, 12.36, 11.90. 19F NMR (470 MHz, benzene-d6): δ
−62.19. HRMS (ESI): calcd for C37H43F3O3Ti 641.2722 [MH]+,
found 641.2721. Anal. Calcd for C37H43F3O3Ti·(2/5)H2O: C, 68.6; H,
6.82. Found: C, 68.5; H, 6.68.
3
3
benzene-d6): δ 7.79 (d, JHH = 8.5 Hz, 2H), 7.62 (d, JHH = 8.5 Hz,
2H), 7.57 (d, 3JHH = 8.5 Hz, 2H), 6.92 (d, 3JHH = 8.5 Hz, 2H), 6.53 (s,
br, 1H), 5.29 (d, 3JHH = 2.0 Hz, 1H), 3.37 (s, 3H), 1.88 (s, 15H), 1.79
(s, 15H). 13C NMR (126 MHz, benzene-d6): δ 161.33, 159.56, 153.26,
135.44, 128.61 (q, 2JCF = 31.7 Hz), 127.07, 126.81, 125.52, 125.48 (q,
Cp*2Ti[OC(4-CF3-C6H4)CHCH(4-Me-C6H4)O] (11c). 11c was
prepared following the procedure described for 11a, starting with 2b
and enone 10c. Brown solid, 78% yield. 1H NMR (500 MHz, benzene-
3
3
d6): δ 7.72 (d, JHH = 8.1 Hz, 2H), 7.64 (d, JHH = 7.7 Hz, 2H), 7.48
(d, 3JHH = 8.1 Hz, 2H), 7.22 (d, 3JHH = 7.7 Hz, 2H), 6.61 (d, 3JHH = 2.0
Hz, 1H), 5.54 (d, 3JHH = 2.1 Hz, 1H), 2.24 (s, 3H), 1.87 (s, 15H), 1.76
(s, 15H). 13C NMR (126 MHz, benzene-d6): δ 158.93, 146.19, 145.64,
135.94, 129.08, 128.64 (q, 2JCF = 31.9 Hz), 127.17, 125.70, 125.49 (q,
3
1JCF = 272.2 Hz), 125.24 (q, JCF = 3.7 Hz), 124.74, 113.65, 97.39,
82.18, 54.83, 12.40, 11.88. 19F NMR (470 MHz, benzene-d6): δ
−62.10. HRMS (ESI): calcd for C37H43F3O3Ti 641.2722 [MH]+,
found 641.2729. Anal. Calcd for C37H43F3O3Ti·1/3(H2O): C, 68.73;
H, 6.81. Found: C, 68.64; H, 6.81.
3
1JCF = 272.3 Hz), 125.41, 125.11 (q, JCF = 3.8 Hz), 124.71, 102.28,
83.10, 21.21, 12.36, 11.89. 19F NMR (470 MHz, benzene-d6): δ
−62.19. HRMS (ESI): calcd for C37H43F3O2Ti 625.2773 [MH]+,
found 625.2772. Anal. Calcd for C37H43F3O2Ti·3/5(H2O): C, 69.94;
H, 7.01. Found: C, 69.59; H, 6.77.
Cp*2Ti[OC(C6H5)CHCH(4-CF3-C6H4)O] (11i). 11i was prepared
following the procedure described for 11a, starting with 2b and enone
10i. Brown solid, 60% yield. 1H NMR (500 MHz, benzene-d6): δ 7.84
3
4
3
(dd, JHH = 8.2 Hz, JHH = 1.4 Hz, 2H), 7.60 (d, JHH = 8.2 Hz, 2H),
7.56 (d, 3JHH = 8.2 Hz, 2H), 7.31 (dd, 3JHH = 8.0, 7.5 Hz, 2H), 7.18 (tt,
3JHH = 7.5 Hz, 4JHH = 1.4 Hz, 1H), 6.50 (s, br, 1H), 5.37 (d, 3JHH = 2.1
Hz, 1H), 1.86 (s, 15H), 1.76 (s, 15H). 13C NMR (126 MHz, benzene-
Cp*2Ti[OC(4-CF3-C6H4)CHCH(C6H5)O] (11d). 11d was prepared
following the procedure described for 11a, starting with 2b and enone
10d. Brown solid, 74% yield. 1H NMR (500 MHz, benzene-d6): δ 7.71
(d, 3JHH = 7.8 Hz, 2H) 7.69 (d, 3JHH = 8.0 Hz, 2H) 7.48 (d, 3JHH = 8.1
Hz, 2H), 7.38 (t, J = 7.7 Hz, 2H), 7.21 (t, 3JHH = 7.4 Hz, 1H), 6.60 (d,
3JHH = 2.0 Hz, 1H), 5.51 (d, 3JHH = 2.1 Hz, 1H), 1.86 (s, 15H), 1.75 (s,
15H). 13C NMR (126 MHz, benzene-d6): δ 159.12, 148.52, 146.17,
2
d6): δ 161.54, 153.10, 142.74, 128.53 (q, JCF = 32.1 Hz), 127.30,
127.00, 125.65, 125.60, 125.47 (q, 1JCF = 272.5 Hz), 125.25 (q, 3JCF
=
3.7 Hz), 124.81, 99.26, 82.02, 12.37, 11.85, one signal is underneath
solvent peaks. 19F NMR (470 MHz, benzene-d6): δ −62.11. HRMS
(ESI): calcd for C36H41F3O2Ti 611.2616 [MH]+, found 611.2616.
Anal. Calcd for C36H41F3O2Ti ·1/4(H2O): C, 70.3; H, 6.8. Found: C,
70.31; H, 6.71.
2
128.66 (q, JCF = 32.0 Hz), 128.39, 127.02, 126.74, 125.69, 125.48,
125.47 (q, 1JCF = 272.4 Hz), 125.10 (q, 3JCF = 3.8 Hz), 124.77, 102.23,
83.07, 12.35, 11.87. 19F NMR (470 MHz, benzene-d6): δ −62.20.
HRMS (ESI): calcd for C36H41F3O2Ti 611.2616 [MH]+, found
611.2619. Anal. Calcd for C36H41F3O2Ti·2/5(H2O): C, 69.99; H, 6.82.
Found: C, 70.02; H, 6.62.
Cp*2Ti[OC(4−Br-C6H4)CHCH(4-CF3-C6H4)O] (11j). 11j was
prepared following the procedure described for 11a, starting with 2b
and enone 10j. Brown solid, 58% yield. 1H NMR (500 MHz, benzene-
d6): δ 7.56 (s, 4H), 7.51 (d, 3JHH = 8.4 Hz, 2H), 7.41 (d, 3JHH = 8.4 Hz,
2H), 6.44 (s, br, 1H), 5.24 (d, 3JHH = 2.1 Hz, 1H), 1.82 (s, 15H), 1.73
(s, 15H). 13C NMR (126 MHz, benzene-d6): δ 160.17, 152.81, 141.53,
131.30, 128.66 (q, 2JCF = 31.7 Hz), 127.22, 126.99, 125.70, 125.42 (q,
Cp*2Ti[OC(4-CF3-C6H4)CHCH(4-Cl-C6H4)O] (11e). 11e was
prepared following the procedure described for 11a, starting with 2b
and enone 10e. Brown solid, 81% yield. 1H NMR (500 MHz, benzene-
d6): δ 7.69 (d, J = 8.1 Hz, 3H), 7.49 (d, J = 8.2 Hz, 2H), 7.45 (d, 3JHH
3
1JCF = 272.3 Hz), 125.28 (q, JCF = 3.7 Hz), 124.94, 121.10, 99.63,
3
3
= 8.4 Hz, 2H), 7.34 (d, JHH = 8.5 Hz, 2H), 6.44 (d, JHH = 2.2 Hz,
1H), 5.35 (d, JHH = 2.1 Hz, 1H), 1.83 (s, 15H), 1.71 (s, 15H). 13C
3
81.98, 12.35, 11.82. 19F NMR (470 MHz, benzene-d6): δ −62.15.
HRMS (ESI): calcd for C36H40BrF3O2Ti 691.1701[MH]+, found
691.1706. Anal. Calcd for C36H40BrF3O2Ti: C, 62.71; H, 5.85. Found:
C, 62.86; H, 5.98.
NMR (126 MHz, benzene-d6): δ 159.32, 147.11, 145.99, 132.38,
128.84 (q, 2JCF = 31.8 Hz), 128.50, 128.46, 125.68, 125.63, 125.42 (q,
3
1JCF = 272.4 Hz), 125.14 (q, JCF = 3.8 Hz), 124.91, 101.39, 82.08,
12.32, 11.84. 19F NMR (470 MHz, benzene-d6): δ −62.23. HRMS
(ESI): calcd for C36H40ClF3O2Ti 645.2227 [MH]+, found 645.2232.
Anal. Calcd for C36H40ClF3O2Ti·2/5(H2O): C, 66.29; H, 6.31. Found:
65.96; H, 6.16.
Cp′2Ti[OC(C6H5)CHCH(C6H5)O] (12a). 12a was prepared follow-
ing the procedure described for 11a, starting with 4a and chalcone
1
10a. Reddish-brown solid, 61% yield. H NMR (500 MHz, benzene-
d6): δ 7.80 (dd, 3JHH = 8.3 Hz, 3JHH = 1.4 Hz, 2H), 7.61 (dd, 3JHH = 8.4
4
3
Hz, JHH = 1.5 Hz, 2H), 7.30 (dd, JHH = 8.4, 6.9 Hz, 2H), 7.24 (dd,
Cp*2Ti[OC(4-CF3-C6H4)CHCH(4-CF3-C6H4)O] (11f). 11f was
prepared following the procedure described for 11a, starting with 2b
and enone 10f. Brown solid, 88% yield. 1H NMR (500 MHz, benzene-
d6): δ 7.68 (d, 3JHH = 8.5 Hz, 2H), 7.57 (s, 4H), 7.49 (d, 3JHH = 8.5 Hz,
2H), 6.46 (s, br, 1H), 5.32 (d, 3JHH = 2.1 Hz, 1H), 1.83 (s, 14H), 1.72
(s, 14H). 13C NMR (126 MHz, benzene-d6): δ 159.81, 152.62, 145.96,
128.95 (q, 2JCF = 32.0 Hz), 128.76 (q, 2JCF = 32.0 Hz), 126.94, 125.79,
125.65, 125.39 (q, 1JCF = 272.4 Hz), 125.34 (q, 3JCF = 3.7 Hz), 125.19
(q, 3JCF = 3.8 Hz), 125.04, 101.34, 81.89, 12.32, 11.81 (note: only see
one of the CF3 quarter). 19F NMR (470 MHz, benzene-d6): δ −62.16,
−62.24. HRMS (ESI): calcd for C37H40F6O2Ti 679.2490 [MH]+,
found 679.2485. Anal. Calcd for C37H40F6O2Ti: C, 65.49; H, 5.94.
Found: C, 65.26; H, 6.29.
3JHH = 8.3, 7.1 Hz, 2H), 7.16 (tt, JHH = 7.4 Hz, JHH = 1.4 Hz, 1H),
3
4
one multiplet under solvent peak, 6.44 (d, 3JHH = 1.9 Hz, 1H), 5.97 (s,
3
1H), 5.78 (s, 1H), 5.47 (d, JHH = 1.9 Hz, 1H), 1.88 (s, 3H), 1.82 (s,
3H), 1.81 (s, 3H), 1.80 (s, 3H), 1.73 (s, 3H), 1.69 (s, 3H), 1.68 (s,
3H), 1.62 (s, 3H). 13C NMR (126 MHz, benzene-d6): δ 159.57,
148.75, 142.18, 133.75, 133.20, 128.37, 128.15, 127.74, 127.43, 127.11,
126.91, 125.66, 122.29, 121.64, 120.83, 118.68, 112.65, 112.30, 110.16,
101.29, 84.98, 14.34, 13.86, 13.48, 13.30, 12.55, 12.04, 11.87, 11.82.
HRMS (ESI): calcd for C33H38O2Ti 515.2430 [MH]+, found
515.2420. Anal. Calcd for C33H38O2Ti: C, 77.03; H, 7.44. Found: C,
76.72; H, 7.26.
Cp′2Ti[OC(4-NMe2-C6H4)CHCH(4-CF3-C6H4)O] (12g). 12g was
Cp*2Ti[OC(4-NMe2-C6H4)CHCH(4-CF3-C6H4)O] (11g). 11g was
prepared following the procedure described for 11a, starting with 4a
1
prepared following the procedure described for 11a, starting with 2b
and chalcone 10g. Dark green solid, 75% yield. H NMR (500 MHz,
1
3
3
and enone 10g. Dark green solid, 81% yield. H NMR (500 MHz,
benzene-d6): δ 7.81 (d, JHH = 8.5 Hz, 2H), 7.55 (d, JHH = 8.3 Hz,
2H), 7.52 (d, 3JHH = 8.5 Hz, 2H), 6.70 (d, 3JHH = 8.5 Hz, 2H), 6.41 (d,
3JHH = 1.9 Hz, 1H), 6.05 (s, 1H), 5.86 (s, 1H), 5.30 (d, 3JHH = 1.9 Hz,
1H), 2.56 (s, 6H), 1.97 (s, 3H), 1.92 (s, 3H), 1.87 (s, 3H), 1.82 (s,
3H), 1.81 (s, 3H), 1.72 (m, 9H). 13C NMR (126 MHz, benzene-d6): δ
160.86, 153.21, 150.28, 133.58, 133.53, 130.59, 128.80, 128.76 (q, 2JCF
3
3
benzene-d6): δ 7.86 (d, JHH = 8.8 Hz, 2H), 7.65 (d, JHH = 8.3 Hz,
2H), 7.56 (d, 3JHH = 8.5 Hz, 2H), 6.73 (d, 3JHH = 8.9 Hz, 2H), 6.57 (d,
3JHH = 1.9 Hz, 1H), 5.35 (d, 3JHH = 2.0 Hz, 1H), 2.57 (s, 6H), 1.91 (s,
15H), 1.83 (s, 15H). 13C NMR (126 MHz, benzene-d6): δ 161.94,
2
153.50, 150.25, 131.37, 128.65 (q, JCF = 29.4 Hz), 127.13, 126.57,
L
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
= 32.0 Hz), 127.77, 127.70, 126.56, 125.44 (q, JCF = 272.7 Hz),
(12) Polse, J. L.; Andemen, R. A.; Bergman, R. G. J. Am. Chem. Soc.
1995, 117, 5393−5394.
3
125.28 (q, JCF = 3.7 Hz), 122.41, 121.19, 120.91, 118.69, 112.98,
112.59, 112.38, 97.14, 83.90, 40.34, 14.39, 13.97, 13.50, 13.27, 12.56,
11.98, 11.95, 11.87. 19F NMR (470 MHz, benzene-d6): δ −62.14.
HRMS (ESI): calcd for C36H42F3NO2Ti 626.2725 [MH]+, found
626.2723. Anal. Calcd for C36H42F3NO2Ti·1/6(H2O): C, 68.79; H,
6.79; N, 2.23. Found: C, 68.65; H, 6.68; N, 2.6.
Cp′2Ti[OC(4-OMe-C6H4)CHCH(4-CF3-C6H4)O] (12h). 12h was
prepared following the procedure described for 11a, starting with 4a
and chalcone 10h. Brownish-red solid, 80% yield. 1H NMR (500 MHz,
benzene-d6): δ 7.75 (d, 3JHH = 8.8 Hz, 2H), 7.53 (s, 4H), 6.89 (d, 3JHH
= 8.7 Hz, 2H), 6.37 (s, br, 1H), 6.03 (s, 1H), 5.85 (s, 1H), 5.25 (d,
3JHH = 1.9 Hz, 1H), 3.36 (s, 3H), 1.93 (s, 3H), 1.88 (s, 3H), 1.83 (s,
3H), 1.79 (s, 3H), 1.78 (s, 3H), 1.71 (s, 3H), 1.69 (s, 3H), 1.65 (s,
3H). 13C NMR (126 MHz, benzene-d6): δ 160.21, 159.61, 152.95,
(13) Schwartz, D. J.; Smith, M. R., III; Andersen, R. A.
Organometallics 1996, 15, 1446−1450.
(14) Hazari, N.; Mountford, P. Acc. Chem. Res. 2005, 38, 839−849.
(15) Odom, A. L. Dalton Trans. 2006, 225−233.
(16) Basuli, F.; Aneetha, H.; Huffman, J. C.; Mindiola, D. J. J. Am.
Chem. Soc. 2005, 127, 17992−17993.
(17) Aneetha, H.; Basuli, F.; Bollinger, J.; Huffman, J. C.; Mindiola,
D. J. Organometallics 2006, 25, 2402−2404.
(18) Doxsee, K. M.; Mouser, J. K. M. Tetrahedron Lett. 1991, 32,
1687−1690.
(19) Nguyen, T. T.; Bertke, J. A.; Gray, D. L.; Hull, K. L.
Organometallics 2015, 34, 4190−4193.
(20) Backvall, J.-E. Modern Oxidation Methods; Wiley, 2005.
̈
2
134.69, 133.64, 133.62, 128.97, 128.87 (q, JCF = 32.0 Hz), 127.79,
(21) Hagadorn, J. R.; Arnold, J. Organometallics 1998, 17, 1355−
127.71, 126.80, 125.40 (q, 1JCF = 272.4 Hz), 125.31 (q, 3JCF = 3.7 Hz),
122.48, 121.30, 121.14, 118.86, 113.65, 113.11, 112.68, 98.60, 83.76,
54.83, 14.37, 13.93, 13.45, 13.26, 12.52, 11.96, 11.90, 11.83. 19F NMR
(470 MHz, benzene-d6): δ −62.17. HRMS (ESI): calcd for
C35H39F3O3Ti 613.2409 [MH]+, found 613.2412. Anal. Calcd for
C35H39F3O3Ti·1/2(H2O): C, 67.63; H, 6.49. Found: C, 67.52; H, 6.22.
1368.
(22) Virginia, W.; Cyclopentadienyl, P. O.; Formation, T. C.;
Characterization, S.; Tetramer, T.; Petersen, J. L. Inorg. Chem. 1980,
19, 181−185.
(23) See Supporting Information.
(24) Sweeney, Z. K.; Polse, J. L.; Andersen, R. A.; Bergman, R. G. J.
Am. Chem. Soc. 1998, 120, 7825−7834.
ASSOCIATED CONTENT
* Supporting Information
(25) Carney, M. J.; Walsh, P. J.; Hollander, F. J.; Bergman, R. G.
Organometallics 1992, 11, 761−777.
■
S
(26) Pappas, I.; Chirik, P. J. Angew. Chem., Int. Ed. 2014, 53, 6241−
6244.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00111.
(27) Hanna, T. A.; Baranger, A. M.; Walsh, P. J.; Bergman, R. G. J.
Am. Chem. Soc. 1995, 117, 3292−3293.
1H NMR spectra of reactions monitored by NMR,
(28) Vaughan, G. A.; Hillhouse, G. L.; Rheingold, A. L. J. Am. Chem.
Soc. 1990, 112, 7994−8001.
kinetic experiments, H, 13C, and 19F NMR spectra of
1
(29) Jones, N. A.; Liddle, S. T.; Wilson, C.; Arnold, P. L.
Organometallics 2007, 26, 755−757.
titanocene complexes (PDF)
Crystallographic data for 4b, 8b, and 11a (CIF)
(30) Demele, O.; Trapp, N.; Diederich, F. Angew. Chem., Int. Ed.
2015, 54, 12339−12344.
AUTHOR INFORMATION
Corresponding Author
*E-mail: kamihull@illinois.edu.
(31) Mujkic, M.; Lentz, D. Dalton Trans. 2012, 41, 839−849.
(32) Kocyigit-Kaymakcıoglu, B.; Beyhan, N.; Tabanca, N.; Abbas Ali,
A.; Wedge, D. E.; Duke, S. O.; Ulrich, R.; Bernier, U. R.; Khan, I. A.
Med. Chem. Res. 2015, 24, 3632−3644.
■
Notes
(33) Smith, J. A.; Brintzinger, H. H. J. Organomet. Chem. 1981, 218,
159.
The authors declare no competing financial interest.
(34) Bercaw, J. E.; Marvich, R. H.; Bell, L. G.; Brintzinger, H. H. J.
Am. Chem. Soc. 1972, 94, 1219.
ACKNOWLEDGMENTS
■
(35) Mashima, K.; Sakai, N.; Takaya, H. Bull. Chem. Soc. Jpn. 1991,
64, 2475−2483.
The authors thank the University of Illinois, Urbana−
Champaign, and the Petroleum Research Fund for their
generous support.
REFERENCES
■
(1) Amslinger, S. ChemMedChem 2010, 5, 351−356.
(2) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry Part B:
Reaction and Synthesis, 4th ed.; Kluwer Academic/Plenum Publishers:
New York, 2001.
(3) Allen, N.; Edge, M. Fundamentals of Polymer Degradation and
Stabilisation; Elsevier Applied Science, 1992.
(4) Takeda, T. Modern Carbonyl Olefination; Wiley-VCH, 2004.
(5) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 259−281.
(6) Kortman, G. D.; Orr, M. J.; Hull, K. L. Organometallics 2015, 34,
1013−1016.
(7) Dwyer, P. N.; Puppe, L.; Buchler, J. W.; Scheidt, W. R. Inorg.
Chem. 1975, 14, 1782−1785.
(8) Goedken, V. L.; Ladd, J. A. J. Chem. Soc., Chem. Commun. 1982,
142, 142−144.
(9) Smith, M. R.; Matsunaga, P. T.; Andersen, R. A. J. Am. Chem. Soc.
1993, 115, 7049−7050.
(10) Hanna, T. E.; Lobkovsky, E.; Chirik, P. J. Inorg. Chem. 2007, 46,
2359−2361.
(11) Housmekerides, E.; Ramage, D. L.; Kretz, C. M.; Shontz, J. T.;
Pilato, R. S.; Geoffroy, G. L.; Rheingold, J. A. L.; Haggerty, B. S. Inorg.
Chem. 1992, 31, 4453−4468.
M
DOI: 10.1021/acs.organomet.6b00111
Organometallics XXXX, XXX, XXX−XXX