C-H activation of pendant alkoxides by tungsten imide complexes

Article abstract of DOI:10.1016/S0022-328X(99)00423-4

The tungsten complex W(=NAr)(O-t-Bu)₂Cl₂(THF) (1, Ar=2.6-diisopropylamine) reacts with an imine substrate at 80°C in C₆D₆ to give an N-aryl imine product. A Chauvin [2+2] type mechanism was excluded when 1 was found to be thermally unstable. Decomposition of 1 yielded isobutylene, t-butyl chloride, 2,6-diisopropylphenylamine and precipitates that exhibited an IR absorbance for M=O. There was no evidence for the intermediacy of radical or cationic species. Imide-mediated C-H activation of the t-butoxide ligand was proposed as the most likely pathway. The tetraalkoxide W(=NAr)(O-t-Bu)₄ was also found to decompose at 80°C, but at a slower rate. The fluorinated alkoxide complexes, W(=NPh)((OC(CF₃)₂(CH₃))₂Cl ₂(THF) and W(=N-t-Bu)₂((OC(CF₃)₂(CH₃)) ₂ did not decompose under similar conditions.

Full text of DOI:10.1016/S0022-328X(99)00423-4

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 591 (1999) 104–113
CꢀH activation of pendant alkoxides by tungsten imide complexes
Amy Y. Jordan, Tara Y. Meyer *
Department of Chemistry, Uni6ersity of Pittsburgh, Pittsburgh, PA 15260, USA
Received 2 June 1999; accepted 12 July 1999
Abstract
The tungsten complex W(ꢁNAr)(O-t-Bu)2Cl2(THF) (1, Ar=2.6-diisopropylamine) reacts with an imine substrate at 80°C in
C6D6 to give an N-aryl imine product. A Chauvin [2+2] type mechanism was excluded when 1 was found to be thermally
unstable. Decomposition of 1 yielded isobutylene, t-butyl chloride, 2,6-diisopropylphenylamine and precipitates that exhibited an
IR absorbance for MꢁO. There was no evidence for the intermediacy of radical or cationic species. Imide-mediated CꢀH
activation of the t-butoxide ligand was proposed as the most likely pathway. The tetraalkoxide W(ꢁNAr)(O-t-Bu)4 was also found
to decompose at 80°C, but at a slower rate. The fluorinated alkoxide complexes, W(ꢁNPh)((OC(CF3)2(CH3))2Cl2(THF) and
W(ꢁN-t-Bu)2((OC(CF3)2(CH3))2 did not decompose under similar conditions. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Tungsten; Imide; Alkoxide; C–H activation
1. Introduction
Imide complexes with supporting alkoxide ligands are
good candidates for imine metathesis catalysts because
competitive p-donation from the alkoxide can poten-
tially activate the extremely stable metal–nitrogen bond
[1,2]. This hypothesis, combined with the electronic
tuneability of the OR group, motivated us to study the
imine reactivity of alkoxide and mixed alkoxide/chloride
derivatives of the type Wꢁ(NR)(OR)nCl4−n
.
W(ꢁNAr)(O-t-Bu)2Cl2(THF) [3] (Ar=2,6-diiso-
propylphenyl) (1), previously prepared by Schrock et al.
reacted with imine to give the predicted organic ex-
change product. However, in contrast to the previously
reported behavior of (DME)Cl2Mo(ꢁNR)2 [4–6], there
was an initiation period, side products, and no evidence
for the expected tungsten imido complex (Scheme 1). In
this account, we describe evidence that the observed
ꢁNR exchange product does not arise from a [2+2]
imide/imine metathesis, like that observed by Bergman
and co-workers for Cp*CpZr(ꢁN-t-Bu)(THF) [7–9],
but rather from a reactive intermediate or intermediates
produced from the alkoxide-mediated decomposition of
1.
Scheme 1.
2. Results
Initially, reactions of 1 with imine appeared to be
imide/imine metathesis reactions. In an NMR tube, 1
was heated in the presence of one (p-Tol)NꢁCH(t-Bu)
(Tol=4-MeC6H4), in C6D6 for 15 days at 55–75°C.
The expected product imine ArNꢁCH(t-Bu) was first
observed by 1H-NMR spectroscopy after 5 days. In
1
addition, however, the H- and 13C-NMR spectra con-
* Corresponding author. Fax: +1-412-6248611.
E-mail address: tmeyer@pitt.edu (T.Y. Meyer)
tained resonances for free THF, t-butyl chloride, and
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00423-4

A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
105
of isobutylene, t-butylchloride, THF, and the standard,
anisole.
No evidence in the reaction mixtures for the presence
of free t-butyl cation or for the presence of free radicals
was observed. Free cation could result in the produc-
tion of the Friedel–Crafts C6D6 alkylation product,
C6D5C(CH3)3 [11]. This alkylation product was not
present nor did the addition of excess benzyltriethylam-
monium chloride significantly decrease the amount of
isobutylene produced as might be expected if such a
cation were trapped by chloride anion [12]. A free
radical pathway should give coupling products such as
2,2,3,3-tetramethylbutane and 2-methylpropane. These
products were not observed nor did addition of dihy-
droanthracene (DHA), a radical trap, change the
course of the reaction [13].
Scheme 2.
Trace amine did not affect the decomposition of 1.
Although the n-butylamine reacted slowly with imine
(p-Tol)NꢁCH(t-Bu) to give (n-Bu)NꢁCH(t-Bu), upon
heating, 1 decomposed as previously described yielding
imine ArNꢁCH(t-Bu), with no further incorporation of
the n-butylamine into any product. The isopropyl
groups on the imide are apparently not directly in-
volved in the reaction as is shown by the fact that
W(ꢁNPh)(O-t-Bu)2Cl2(THF) (2) decomposed (3 days,
ꢀ70°C) to give the same distribution of product types
as 1.
The decomposition of the tetraalkoxide derivative
W(ꢁNAr)(O-t-Bu)4 (Ar=2,6-diisopropylphenyl) (3)
proceeded more slowly than 1 (Scheme 3). Imide com-
plex 3 required approximately 1 month at 80°C in C6D6
to give significant quantities of precipitate, isobutylene,
t-butyl alcohol, and 2,6-diisopropylphenylamine. De-
composition of 3 in the presence of imine PhNꢁCH(t-
Bu) gave imine ArNꢁCH(t-Bu). The decomposition
rate was dramatically accelerated from 1 month to 4
days, the time previously associated with 1, by the
addition of trace amounts of W(ꢁNAr)Cl4(THF) (Ar=
2,6-diisopropylphenyl) (4) to 3.
isobutylene (Scheme 1). Moreover, there were precipi-
tated solids and no resonances consistent with the
expected product imide.
The anomalous nature of this reaction was confirmed
by the subsequent observation that all of the organic
compounds, except ArNꢁCH(t-Bu), were present when
tungsten imide 1 was heated in the absence of imine
(Scheme 2). After 4 days at 70°C in C6D6, 1 decom-
posed to give a precipitate and solution color change
that paralleled that observed in the presence of imine.
1
As observed by H-NMR, the first step in the process
was the release of coordinated THF. Intermediate spe-
cies then appeared and disappeared as t-butyl chloride
and isobutylene were released. Concurrently, for high
initial concentrations of 1, broad singlets appeared ca.
l 9 and 12.
The reaction appeared to be independent of concen-
tration, both of imine and 1. Decompositions per-
formed in the absence of imine, in the presence of
stoichiometric imine, and in the presence of excess
imine, all proceeded on a similar time scale (3–4 days)
and gave t-butyl chloride and isobutylene as products.
Nor were significant rate differences noted when the
concentration of 1 was increased fivefold.
By NMR spectroscopy, the non-volatile components
of the decomposition, isolated by removal of volatiles
in vacuo, comprised of 2–3 different NAr-containing
species (Scheme 2). Diisopropylphenylamine and/or the
ammonium salt were amongst these compounds [10]. In
addition, these residues exhibited the higher frequency
singlets ca. l 9 and 12 that were observed in the
original reaction mixture. IR spectra of the nonvolatile
compounds exhibited an absorbance at 750 cm−1 (Nu-
jol), consistent with a WꢁO functionality. Repeated
attempts to obtain a full crystal structure of the precip-
itate were unsuccessful. Treatment of the nonvolatile
compounds from the decomposition with imine (p-
Tol)NꢁCH(t-Bu) in C6D6 gave imine ArNꢁCH(t-Bu).
Volatile components were also collected and consisted
The decomposition of the tetrachloride complexes, 4
and W(ꢁNPh)Cl4(THF) [14] (5), was also investigated.
Upon heating in C6D6 at 80 and 60°C, respectively, for
more than 2 weeks, a small amount of precipitate
formed, but the total concentration of the imides re-
mained nearly constant and no new resonances were
observed.
Scheme 3.

106
A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
The fluorinated alkoxide derivatives W(ꢁNPh)-
((OC(CF3)2(CH3))2Cl2(THF) [15] (6) and W(ꢁN-t-
Bu)2((OC(CF3)2(CH3))2 (7) did not decompose after
weeks at elevated temperatures nor did they react sig-
nificantly (B5%) with added imine.
Decomposition behavior similar to that observed for
the tungsten alkoxide 1 was also observed for a related
molybdenum compound. The complex Mo(ꢁNAr)2(O-
t-Bu)2 (8) decomposed to yield isobutylene and diiso-
propylphenylamine when heated for prolonged periods
(100°C, 45 days). Moreover, when the non-volatile
components of the decomposition reaction mixture
were heated in the presence of (n-Pr)NꢁCHPh and
PhNꢁCH(t-Bu) both imine/imine and imine/imide ꢁNR
exchange products were observed.
3. Discussion
Metal complexes used for the catalysis of polymers
and the formation of metal oxide compounds often
utilize alkoxide supporting ligands [16]. However,
alkoxide ligands are not always innocent ancillary
groups in transition metal complexes [16]. Several
mechanisms have been proposed for their decomposi-
tion including heterolytic cleavage of the CꢀO bond
(Eq. (1)) [11], homolytic cleavage of the CꢀO bond (Eq.
(2)) [17], and CꢀH activation (Eq. (3) [18], Eq. (4) [19])
(Scheme 4).
Heterolytic cleavage was described by Wolczanski
and co-workers in zirconium tritox systems that form
the trityl cation in solution along with the metal oxide
[11]. Cummins and co-workers established homolytic
cleavage as the mechanism of decomposition for (t-
Scheme 5.
BuO)3Mo[m-N]Ti(NRAr)3 [17]. Chisholm and co-work-
ers invoked CꢀH activation to explain the conversion of
a ditungsten t-butoxide complex to a hydrido-oxo com-
pound [18]. Similarly, Sen and co-workers proposed
CꢀH activation as an explanation for the products
obtained in the flash vacuum pyrolysis of Ti(OR)4
compounds [19,20]. In both of these last two systems
the alkoxide’s g-hydrogen was accessible for removal.
We propose that intramolecular CꢀH activation rep-
resents the most probable of these pathways for the
decomposition of the tungsten imide alkoxide com-
plexes (Scheme 5) [21]. The precedent is strong. The
imide functional group is known to mediate CꢀH acti-
vation [22]. The mechanism is plausible. The six-mem-
bered transition state should be energetically accessible
and relatively low in energy by analogy to organic
ene-type reactions. The pathway does not require
cations or radicals. None of the experiments or controls
yielded evidence to support the presence of these high
energy species in the reaction mixtures.
Imide-mediated CꢀH activation, along with the fur-
ther reactions of the first-formed products, also ex-
plains all of the spectroscopic observations. After the
loss of THF, the initial CꢀH activation should produce
isobutylene and a metal complex with both oxo and
amido groups. A subsequent a-hydride elimination
from the amide, a reaction that is well-known for
complexes of this class, should give HCl and reform the
imido linkage. Amongst other reactions, we can postu-
late that the HCl can react with isobutylene to give the
observed t-butyl chloride. It is, of course, possible to
Scheme 4.

A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
107
decomposition followed by addition of imines con-
firmed that the observed reactions paralleled those that
we have described for the tungsten alkoxides in this
report.
propose direct abstraction of the t-butoxide hydrogen
by chloride or alkoxide, but these reactions are less
likely given the relative basicities of the ligands.
After both metal amide and HCl are present in the
reaction mixture many new pathways open. For exam-
ple, the amine can be produced either by CꢀH activa-
tion of another t-butoxide ligand or by reaction of HCl
with amide. Ammonium salts could also be expected to
form under these conditions. Amido groups and/or
ammonium salts are the likely species responsible for
4. Experimental
4.1. General
1
Manipulations were performed under an inert atmo-
sphere of nitrogen using standard Schlenk techniques
or a Vacuum Atmospheres glovebox. The solvents used
were dry and oxygen free. Reagent grade ether and
toluene were prepared by distillation from sodium ben-
zophenone ketyl under nitrogen. Reagent grade hexane
was washed with sulfuric acid, passed through a
column of activated alumina, and distilled from sodium
benzophenone ketyl under nitrogen. Benzene-d6 was
distilled from sodium benzophenone ketyl and stored in
a glass vessel equipped with a Teflon stopcock. Chloro-
form-d and methylene chloride-d2 were prepared by
distillation from P2O5 and stored in glass vessels
equipped with Teflon stopcocks. Methylene chloride
was prepared by distillation from P2O5. Reagents were
purified by standard methods unless otherwise noted.
NMR spectra were recorded on a Bruker AF 300 MHz
spectrometer at 25°C.
Tungsten hexachloride (Strem) and hexamethyldis-
iloxane (Acros) were used as received. Benzyltriethyl-
ammonium chloride salt was dried overnight under
vacuum before using. 9,10-Dihydroanthracene was re-
crystallized twice from ethanol and dried under vacuum
before using. N-butylamine was purified by distillation
after stirring over calcium hydride.
Imines were prepared using an analogous procedure
to that described in Ref [25]. In general, the procedure
consisted of condensing the parent aldehyde and an
amine in benzene over molecular sieves. Purification by
vacuum distillation gave pure imine.
the l 9 and 12 resonances in the H-NMR spectra [23].
Moreover, the now sterically unsaturated oxo com-
pounds can dimerize or exchange ligands to give species
that would precipitate from solution. It is probable that
the reaction represents a mixture containing some or all
of these species.
The cascade of reactions initiated by the CꢀH activa-
tion can account for the exchange of ꢁNR groups with
added imine since amine, ammonium salts, amides or
imides are all potentially capable of mediating this type
of reaction. Finally, the lack of dependence of the
initiation period on added imine, amine or substrate
concentration is consistent with a monomolecular rate
determining step.
The role of the ancillary chloride ligand is intriguing.
There are several possible explanations for the dramatic
difference in decomposition rate between the te-
traalkoxide 5 and the dichlorodialkoxide 1. Electronic
arguments are not persuasive since any activation of the
alkoxide induced by the removal of electron-density
from the metal center would be accompanied by a
concomitant decrease in imide basicity. Steric argu-
ments are slightly more satisfactory in that the te-
traalkoxide may not be able to assume the appropriate
geometry for facile imide CꢀH activation. There is a
third, more probable, explanation for the role of Cl,
however. HCl produced from amide elimination, which
is not trapped by isobutylene, can act as a catalyst for
further decomposition of both starting material and
products. The latter explanation seems particularly ap-
pealing since the decompositions proceed rapidly after
the initiation period. In addition, the initiation period
for tetraalkoxide 3 decreases dramatically in the pres-
ence of trace amounts of tetrachloride 4 [24].
Finally, the findings of this study allow us to explain
the anomalous reactivity observed for the analogous
t-butoxide supported molybdenum complex 8. As we
established in a prior communication [5], this com-
pound reacts with imine to produce the ꢁNR exchange
product that would be predicted based on an imine
metathesis scheme. After further study, however, it
became clear that the reaction did not belong to the
general class of imide/imine exchange reactions in
which it was originally grouped. In particular, reso-
nances for isobutylene were always present. Controlled
The following compounds were prepared as de-
scribed in the literature procedures (Ar=2,6-diiso-
propylphenyl): W(ꢁNAr)Cl4(THF) [3], W(ꢁNAr)(O-
t-Bu)2Cl2(THF) [3], W(ꢁNPh)Cl4(THF) [14], W(ꢁNPh)-
(OC(CH3)(CF3)2)2Cl2(THF) [15], and Mo(ꢁNAr)2(O-t-
Bu)2 [26].
Significant 1H-NMR resonances for commonly en-
1
countered compounds. t-Butyl chloride H (C6D6): l
1.32 (s, 9, (CH3)3CCl), 13C{1H} (C6D6):
l 67.1
((CH3)3CCl) and 34.3 ((CH3)3CCl); isobutylene 1H
(C6D6): l 4.74 (s, 2, H2CꢁC(CH3)2) and 1.58 (s, 6,
H2CꢁC(CH3)2), 13C{1H} (C6D6): l 111.3 (H2CꢁC-
1
(CH3)2) and 24.1 (H2CꢁC(CH3)2); THF, H (C6D6): l
3.55 (s, 4, (OCH2)2) and 1.4 (s, 4, (CH2)2); t-butyl
alcohol 1H (C6D6): l 4.3 (s, 1, (CH3)3COH) and 1.04 (s,

108
A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
9, (CH3)3COH)). In some cases the presence of t-butyl
chloride and t-butyl alcohol was verified by the addi-
tion of an authentic sample to the NMR experiment.
4.2. NMR experiments
Reactions that produced equilibrium mixtures of
reagents and multiple products were performed in
NMR tubes. In most cases the isolation of individual
organometallic products was not possible. NMR as-
signments are reported for significant species, but aro-
matic region overlap prevents complete assignments in
many cases. Although hexamethylbenzene and anisole
were used as internal standards, reliable quantification
of the products after decomposition was inhibited in
most cases by product volatility and by unfortunate
overlap of resonances. Standard solutions of imine were
typically prepared by weighing the imine (ꢀ100–200
mg) into a 2 ml volumetric flask and diluting to 2 ml
with C6D6. Standard solutions of internal standards
were prepared analogously.
4.1.1. W(ꢁN-t-Bu)2(OC(CH3)(CF3)2)2
Bis(t-butylimido)bis(t-butylamido)tungsten (1.05 g,
2.238 mmol) was stirred with hexafluro-2-methyliso-
propanol (0.80 g, 4.40 mmol) overnight to give a red–
orange mixture. The solvent was removed in vacuo to
give a red oil. The oil was purified by trap-to-trap
distillation to yield colorless crystals and a light orange
liquid. Both the crystals and the oil were produced by
1H-NMR spectroscopy. Unoptimized yield, 4.3%.
1H-NMR (C6D6): l 1.43 (s, 3, (OC(CH3)(CF3)2)2) and
1.23 (s, 18, (NC(CH3)3)2). 13C{1H}-NMR (C6D6) tenta-
tive assignments: l 124.1 (q, (OC(CH3)(CF3)2)2), 68.2
(s, (NC(CH3)3)2), 33.2 (s, (NC(CH3)3)2), and 18.8 (s,
(OC(CH3)(CF3)2)2). 19F-NMR (C6D6): l −78.8 (s, 12,
(OC(CH3)(CF3)2)2).
4.2.1. Decomposition of W(ꢁNAr)(O-t-Bu)2Cl2(THF)
(1) in the presence of one equi6alent of
(p-Tol)NꢁCH(t-Bu)
A solution of 1 (6.2 mg, 0.0096 mmol, one equiva-
lent) in C6D6 (480 ml) was prepared in an NMR tube
equipped with a Teflon stopcock. Anisole was added as
an internal standard (40 ml of a 0.3 M soln) along with
(p-Tol)NꢁCH(t-Bu) (ca. one equivalent, 13.5 ml of a
0.71 M soln). The sample was maintained at 25°C for 1
day to verify the stoichiometry and the stability of the
reaction mixture. The sample was heated and main-
tained at 55–65°C for approximately 16 h and at
65–75°C for a total of 14 days. The progress of the
4.1.2. W(ꢁNAr)(O-t-Bu)4
A −40°C solution of lithium t-butoxide (0.26 g, 4.4
equivalents) in ether (10 ml) was added dropwise to a
−40°C
solution
of
tetrachloro(2,6-diisopropyl-
imido)tungsten (0.42 g, 0.74 mmol, one equivalent) in
THF (15 ml) and ether (10 ml). As the reaction warmed
the color changed from dark green, to red, to orange,
to a cloudy yellow color. After 2 days at 25°C, the
mixture was filtered through Celite and the solvent was
removed in vacuo leaving a light yellow solid. The solid
was recrystallized from hexanes at −40°C. Unopti-
1
reaction was monitored by H-NMR spectroscopy. De-
composition was observed after 4 days at elevated
temperatures. While heating, solids precipitated and the
color changed from yellow to red–brown. Isobutylene,
t-butyl chloride, and dissociated THF were observed.
After 5 days at elevated temperatures the product imine
ArNꢁCH(t-Bu) was first observed. There was a 60%
conversion of starting material imine (p-Tol)NꢁCH-
(t-Bu) to product imine ArNꢁCH(t-Bu), but resonances
consistent with the new imide W(ꢁN-p-Tol)(O-t-
1
mized yield, 11.8%. H-NMR (C6D6): l 4.36 (sept, 2,
((2,6-((CH3)2CH)2C6H3)), 1.46 (s, 36, (OC(CH3)3)4),
and 1.39 (d, 12, ((2,6-((CH3)2CH)2C6H3))).
4.1.3. W(ꢁNPh)(O-t-Bu)2Cl2(THF)
A −40°C solution of lithium t-butoxide (0.23 g, 2.87
mmol) in ether (10 ml) was added dropwise to a
−40°C solution of tetrachloro(phenylimido)tungsten
(0.64 g, 1.3 mmol, one equivalent) in THF (15 ml) and
ether (20 ml). As the reaction warmed the color
changed from dark green, to red, to orange, to finally a
cloudy yellow color. After several hours at 25°C, the
mixture was filtered through Celite and the solvent was
removed in vacuo leaving a light yellow solid. The solid
was dissolved in hexanes and filtered again through
Celite. The solid was recrystallized from hexanes at
−40°C. The preparation of this compound was at-
tempted many times, but the solid was never pure by
1H-NMR. The lack of purity of the dissolved sample
1
Bu)2Cl2(THF) were not observed. H-NMR (C6D6) of
(2,6-((CH3)2CH)2C6H3)NꢁCH(t-Bu) (unisolated): l 7.26
(s, 1, (CH(t-Bu)), 3.05 (m, 2, (2,6-((CH3)2CH)2-
C6H3)N), 1.16 (d, 12, (2,6-((CH3)2CH)2C6H3)N), and
1.05 (s, 9, (CH(t-Bu)). Aromatic resonances could not
be definitively assigned due to the complexity of the
spectrum in this region.
4.2.2. Decomposition of W(ꢁNAr)(O-t-Bu)2Cl2(THF)
(1)
A solution of 1 (8.7 mg, 0.013 mmol) in C6D6 (500
ml) was prepared in an NMR tube. Anisole was added
as an internal standard (40 ml of a 0.3 M soln) and the
tube was flame sealed under vacuum. The sample was
maintained at 65–75°C for a total of 7 days, and was
1
was due to Cl/OR ligand exchange. H-NMR (C6D6): l
7.28 (d, 2, o-H, NPh), 7.01 (t, 2, m-H, NPh), 6.58 (t, 1,
p-H, NPh), 4.23 (br s, 4, (OCH2)2), 1.48 (s, 18,
(OC(CH3)3)2), and 1.41 (s, 4, (CH2)2).

A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
109
1
monitored by H-NMR spectroscopy. Decomposition
spectra of both samples contained resonances for the
organic products t-butyl chloride, isobutylene, and dis-
sociated THF. The spectra were complex but a promi-
nent resonance at l 6.9 (s) was observed in both
samples. Sample B also contained significant resonances
at l 12.88 (s) and 8.13 (br s).
was observed after 4 days at elevated temperatures.
During which time solids precipitated and the color
changed from yellow to red–brown. Several resonances
grew in and disappeared, the most prominent one at l
1.04 (s) indicating that an intermediate, with additional
resonances unassigned, could be present. Dissociated
THF and two organic products, isobutylene and t-butyl
chloride, were identified in the reaction mixture. Signifi-
cant resonances from unidentified species were also
4.2.5. Reaction of (p-Tol)NꢁCH(t-Bu) with the
non6olatile components after decomposition of
W(ꢁNAr)(O-t-Bu)2Cl2(THF) (1)
1
observed: H-NMR (C6D6): l 6.92 (s), 6.75 (d), and
A solution of 1 (30.5 mg, 0.047 mmol) in C6D6 (260
ml) was prepared in an NMR tube equipped with a
Teflon stopcock. Anisole was added as an internal
standard (300 ml of a 0.3 M soln). The sample was
heated and maintained at 75–85°C for approximately 1
day during which time 1 decomposed, solids precipi-
tated, and the color changed from yellow to brown–
yellow. The reaction was monitored by 1H-NMR
spectroscopy. Several resonances grew in and disap-
peared the most prominent one at l 1.07 (s) indicating
that an intermediate, with additional resonances unas-
signed, could be present. The volatile compounds of the
sample were vacuum transferred to another NMR tube.
1H-NMR of the volatiles showed the presence of
isobutylene, t-butyl chloride, and free THF. The non-
volatile compounds were redissolved (undissolved solids
1.13 (br s).
4.2.3. Decomposition of W(ꢁNAr)(O-t-Bu)2Cl2(THF)
(1) in the presence of nine equi6alents of
(n-Pr)NꢁCHPh
A solution of 1 (5.5 mg, 0.0085 mmol, one equiva-
lent) in C6D6 (600 ml) was prepared in an NMR tube
equipped with a Teflon stopcock. Hexamethylbenzene
was added as an internal standard (50 ml of a 0.4 M
soln) along with imine (n-Pr)NꢁCHPh (ca. nine equiva-
lents, 68 ml of a 0.868 M soln). The sample was
maintained at 25°C for 1 day to verify the stoichiome-
try and the stability of the reaction mixture. The sample
was heated and maintained at 55–65°C for approxi-
mately 1 day and at 65–75°C for a total of 48 days.
The progress of the reaction was monitored by 1H-
NMR spectroscopy. Decomposition was observed after
3 days at elevated temperatures. During this time solids
precipitated and the color did not change to red–
brown, but remained yellow. Product imine
ArNꢁCHPh was observed, but due to overlapping sig-
nals it was not possible to quantify the amount of imine
produced. Isobutylene, t-butyl chloride, and dissociated
THF were observed in the H-NMR spectrum. Reso-
nances consistent with the new imide W(ꢁN-n-Pr)(O-t-
Bu)2Cl2(THF) were not observed. H-NMR (C6D6) of
(2,6-((CH3)2CH)2C6H3)NꢁCH(Ph) (unisolated): l 3.12
(m, 2, (2,6-((CH3)2CH)2C6H3)N) and 1.16 (d, 12, (2,6-
((CH3)2CH)2C6H3)N). Aromatic and methyne reso-
nances could not be definitively assigned due to the
complexity of the spectrum in these regions.
1
remained). Significant H-NMR (C6D6) of non-volatile
compounds: l 12.4 (s), 8.5 (br s), 1.38 (v br s), 1.05 (d),
and 1.00 (d). Partial 13C{1H}-NMR (C6D6) of non-
volatile compounds: l 141, 124.9, 29.1, and 24.3.
The imine, (p-Tol)NꢁCH(t-Bu) (66 ml of a 0.71 M
soln, one equivalent) was added to the non-volatile
components of the sample. After 2 h at 75–85°C, 35%
of starting material imine (p-Tol)NꢁCH(t-Bu) was con-
verted to product imine ArNꢁCH(t-Bu). There were no
resonances for imide W(ꢁN-p-Tol)(O-t-Bu)2Cl2(THF).
The progress of the reaction was monitored by
1H-NMR spectroscopy. 1H-NMR (C6D6) of (2,6-
((CH3)2CH)2C6H3)NꢁCH(t-Bu) (unisolated): l 7.25 (s,
1, CH(t-Bu)), 3.04 (m, 2, (2,6-((CH3)2CH)2C6H3)N),
1.16 (d, 12, (2,6-((CH3)2CH)2C6H3)N), and 1.06 (s, 9,
CH(t-Bu)). Aromatic resonances could not be defini-
tively assigned due to the complexity of the spectrum in
this region.
1
1
4.2.4. Concentration dependence of the decomposition
of W(ꢁNAr)(O-t-Bu)2Cl2(THF) (1)
Two solutions of different concentrations were pre-
pared as follows: sample A consisted of 1 (5.6 mg,
0.0086 mmol), hexamethylbenzene (50 ml of a 0.4 M
soln), and C6D6 (600 ml); sample B consisted of 1 (31.9
mg, 0.049 mmol), hexamethylbenzene (100 ml of a 0.4
M soln), and C6D6 (600 ml). Both samples were placed
in NMR tubes equipped with Teflon stopcocks and
maintained at 70°C for 9 days. The progress of the
reactions was monitored by 1H-NMR spectroscopy.
Sample A decomposed after 4 days at elevated temper-
atures and sample B decomposed after 5 days. The
4.2.6. Decomposition of W(ꢁNAr)(O-t-Bu)2Cl2(THF)
(1) in the presence of two equi6alents of
benzyltriethylammonium chloride salt
To a solution of complex 1 (6.7 mg, 0.0103 mmol),
hexamethylbenzene (50 ml of a 0.4 M soln), and C6D6
was added benzyltriethylammonium chloride (ca. two
equivalents, 4.7 mg, 0.021 mmol) in an NMR tube.
Although the salt did not completely dissolve, the sam-
ple immediately changed in color from yellow to or-
ange. The sample remained at 25°C for a total of 6 days
and progress of the reaction was monitored periodically

110
A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
1
by H-NMR spectroscopy. After 3 days 1 decomposed,
and although, the resonances are shifted from those
observed in pure C6D6, isobutylene and t-butyl chloride
were present. Dissociated THF was present along with
significant resonances from unidentified species 1H-
NMR (C6D6): l 10.3 (br s) and 1.27 (s).
W(ꢁN-p-Tol)(O-t-Bu)2Cl2(THF) were not observed.
1H-NMR (C6D6) of (2,6-((CH3)2CH)2C6H3)NꢁCH-
(t-Bu) (unisolated): l 7.25 (s, 1, CH(t-Bu)), 3.05 (m, 2,
(2,6-((CH3)2CH)2C6H3)N), 1.17 (d, 12, (2,6-((CH3)2-
CH)2C6H3)N), and 1.07 (s, 9, CH(t-Bu)). Aromatic
resonances could not be definitively assigned due to the
complexity of the spectrum in this region. In addition,
t-butyl chloride, isobutylene, and dissociated THF were
observed. The spectrum was complex but significant
resonances from unidentified species were also ob-
4.2.7. Decomposition of W(ꢁNAr)(O-t-Bu)2Cl2(THF)
(1) in the presence of nine equi6alents of
9,10-dihydroanthracene
1
A solution of 1 (6.1 mg, 0.0094 mmol, one equiva-
lent) in C6D6 (500 ml) was prepared in an NMR tube.
Anisole was added as an internal standard (100 ml of a
0.3 M soln) along with 9,10-dihydroanthracene (ca.
nine equivalents, 0.0164 g) and the tube was flame
sealed under vacuum. The sample was maintained at
25°C for 1 day to verify the stoichiometry and the
stability of the reaction mixture. The sample was heated
and maintained at 75–85°C for a total of 7 days. The
served. H-NMR (C6D6): l 8.52 (br s) and 0.663 (br s).
4.2.9. Decomposition of W(ꢁNPh)(O-t-Bu)2Cl2(THF)
(2)
A solution of 2 (14.9 mg, 0.0264 mmol) in C6D6 (700
ml) was prepared and sealed under vacuum in an NMR
tube. The sample was maintained at 65–75°C for 3
days, during which time 2 decomposed, solids precipi-
tated, and the color changed from yellow to orange.
The progress of the reaction was monitored by
1H-NMR. The organic products identified in the reac-
tion mixture were isobutylene, t-butyl chloride, and free
THF.
1
progress of the reaction was monitored by H-NMR
spectroscopy. After 1 day at elevated temperatures 1
decomposed, solids precipitated, and the color changed
from yellow–orange to red–orange. Several resonances
grew in and disappeared the most prominent one at l
1.04 (s) indicating that an intermediate, with additional
resonances unassigned, could be present. No an-
4.2.10. Decomposition of W(ꢁNAr)(O-t-Bu)4 (3)
A solution of 3 (9.4 mg, 0.014 mmol) in C6D6 (450
ml) was prepared in an NMR tube. Hexamethylbenzene
was added as an internal standard (25 ml of a 0.4 M
soln) and the tube was flame sealed under vacuum. The
sample was maintained at 75–85°C for 27 days, during
which time 3 decomposed and solids precipitated. The
1
thracene was observed in the H-NMR spectra. Two
organic products, t-butyl chloride and isobutylene were
identified in the reaction mixture along with dissociated
THF.
4.2.8. Decomposition of W(ꢁNAr)(O-t-Bu)2Cl2(THF)
(1) in the presence of (p-Tol)NꢁCH(t-Bu) and
n-butylamine
1
progress of the reaction was monitored by H-NMR.
Three organic products were obtained from the reac-
tion mixture, isobutylene, t-butyl alcohol, and 2,6-di-
isopropylphenylamine. 1H-NMR (C6D6) of 2,6-diiso-
propylphenylamine (unisolated): l 3.5 (br s, 2, NH2),
2.87 (m, 2, ((CH3)2CH)2), and 1.15 (d, 12,
((CH3)2CH)2). Partial 13C{1H}-NMR (C6D6) of t-butyl
alcohol (unisolated): l 31.6 ((CH3)3COH).
A solution of 1 (7.3 mg, 0.011 mmol, one equivalent)
in C6D6 (480 ml) was prepared in an NMR tube
equipped with a Teflon stopcock. Anisole was added as
an internal standard (40 ml of a 0.3 M soln) along with
imine (p-Tol)NꢁCH(t-Bu) (1.1 equivalents, 16 ml of a
0.71 M soln). The sample was maintained at 25°C for 1
day to verify the stoichiometry and the stability of the
reaction mixture. N-butylamine (one equivalent, 1.1 ml)
was added and the sample was maintained at room
temperature for 21 days. The progress of the reaction
was monitored by 1H-NMR spectroscopy. Although,
the resonances are shifted slightly from those observed
in pure C6D6, the resonances indicated that 2,6-diiso-
propylamine, p-tolylamine, and (n-Pr)NꢁCH(t-Bu)
were present. The sample was heated and maintained at
55–65°C for approximately 1 day, at 65–75°C for 2
days, and at 75–85°C for a total of 17 days. After 5
days at elevated temperatures 1 decomposed. During
this time, solids precipitated and the color changed
from yellow to yellow–orange. There was a 40% con-
version of starting material imine 2 to product imine
ArNꢁCH(t-Bu). Resonances consistent with the imide
4.2.11. Decomposition of W(ꢁNAr)(O-t-Bu)4 (3) in the
presence of PhNꢁCH(t-Bu)
A solution of 3 (7.9 mg, 0.012 mmol), C6D6 (450 ml),
anisole, and imine PhNꢁCH(t-Bu) (10 ml of a 1 M soln,
0.010 mmol) was prepared in an NMR tube equipped
with a Teflon stopcock. The progress of the reaction
1
was monitored by H-NMR spectroscopy. The sample
was maintained at room temperature for 5 days, during
which time there were some small changes in the NMR
spectra but no decomposition products or reaction with
imine was observed. The sample was maintained at
75–85°C for 35 days. During this time, yellow solids
formed in the tube and 3 decomposed and reacted with
imine PhNꢁCH(t-Bu). Approximately 100% of imine
PhNꢁCH(t-Bu) was converted to the product imine

A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
111
ArNꢁCH(t-Bu). In addition, isobutylene and t-butyl
alcohol were observed in the NMR spectrum. There
were no resonances consistent with the imide
soln) in an NMR tube, equipped with a Teflon stop-
cock. The sample was maintained at 25°C for 1 day to
verify the stoichiometry and the stability of the reaction
mixture. The sample was heated and maintained at
55–65°C for 1 day, at 65–75°C for 2 days, and at
75–85°C for 3 days. The progress of the reaction was
1
W(ꢁNPh)(O-t-Bu)4 observed in the spectrum. H-NMR
(C6D6) of (2,6-((CH3)2CH)2C6H3)NꢁCH(t-Bu) (uniso-
lated): l 8.00 (s, 1, CH(t-Bu)), 2.70 (sept, 2, (2,6-
((CH3)2CH)2C6H3)N), 1.26 (s, 9, CH(t-Bu)), and 1.16
(d, 12, (2,6-((CH3)2CH)2C6H3)N). Aromatic resonances
could not be definitively assigned due to the complexity
of the spectrum in this region.
1
monitored by H-NMR spectroscopy. No change was
1
observed in the sample or its H-NMR spectrum. The
imine (n-Pr)NꢁCHPh (7.5 ml of a 0.8 M soln, one
equivalent) was added to the NMR tube. The sample
was maintained at 75–85°C for 62 days at which time
product imine PhNꢁCHPh was present in B5% yield
and 35% of the THF dissociated. There were no reso-
nances consistent with the imide W(ꢁN-n-
Pr)(OC(CH3)(CF3)2)2Cl2(THF) in the spectrum, nor
were there any resonances consistent with the decompo-
sition of 6. 1H-NMR (C6D6) of PhNꢁCHPh (uniso-
lated): l 8.119 (s, 1, PhNꢁCHPh). Aromatic resonances
could not be definitively assigned due to the complexity
of the spectrum in this region.
4.2.12. Decomposition of W(ꢁNAr)(O-t-Bu)4 (3) in the
presence of W(ꢁNAr)Cl4(THF) (4)
A solution of 3 (5.2 mg, 0.00799 mmol), C6D6 (600
ml), 4 (1 mg, 0.002 mmol), and anisole (15 ml of a 1 M
soln) was prepared in an NMR tube equipped with a
Teflon stopcock. The sample was maintained at 25°C
for 1 day and at 75–85°C for 4 days, during which time
ligands exchanged, solids formed, and 3 decomposed.
The progress of the reaction was monitored by
1H-NMR spectroscopy. The volatile components of the
sample were vacuum transferred to another NMR tube.
1H-NMR of the volatile material showed the presence
of isobutylene and t-butyl chloride. The non-volatile
components were left behind in the original NMR tube.
1H-NMR showed the presence of 2,6-diisopropylamine
and/or 2,6-diisopropylammonium chloride, these are
indistinguishable in a complex reaction mixture. Signifi-
4.2.16. Thermal stability of W(ꢁN-t-Bu)2(OC(CH3)-
(CF3)2)2 (7) in the presence of (n-Pr)NꢁCHPh
A solution of 7 (5.5 mg, 0.0421 mmol, one equiva-
lent) in C6D6 (500 ml) was prepared in an NMR tube
equipped with a Teflon stopcock. Hexamethylbenzene
was added as an internal standard (240 ml of a 0.4 M
soln) along with imine (n-Pr)NꢁCHPh (73 ml of a 0.8 M
soln, 1.7 equivalents). The sample was maintained at
25°C for 1 day to verify the stoichiometry and the
stability of the reaction mixture. The sample was heated
and maintained at 55–65°C for 1 day, at 65–75°C for
2 days, and at 75–85°C for 18 days. The progress of the
reaction was monitored by 1H-NMR spectroscopy.
Only B5% of imine (n-Pr)NꢁCHPh was converted to
product imine (t-Bu)NꢁCHPh. There were no reso-
nances consistent with (bis)imide W(ꢁN-t-Bu)(ꢁN-n-
Pr)(OC(CH3)(CF3)2)2 observed in the spectrum, nor
were there any resonances consistent with the decompo-
1
cant resonances H-NMR (C6D6) of non-volatile com-
ponents (unisolated): l 2.86 (br s), 1.25 (m), 1.16 (d).
4.2.13. Thermal stability of W(ꢁNAr)Cl4(THF) (4)
A solution was prepared of 4 (5.4 mg, 0.00943 mmol)
and C6D6 (700 ml) in an NMR tube that was flamed
sealed under vacuum. The sample was heated and
maintained at 75–85°C for over 1 month. A dark
precipitate formed, but the total concentration of imide
remained nearly constant. The progress of the reaction
1
was monitored by H-NMR spectroscopy and no new
1
signals appeared in the H-NMR spectrum.
1
sition of 7. H-NMR (C6D6) of (t-Bu)NꢁCHPh (uniso-
lated): l 8.12 (s, 1, (t-Bu)NꢁCHPh). Aromatic and
t-butyl resonances could not be definitively assigned
due to the complexity of the spectrum in these regions.
4.2.14. Thermal stability of W(ꢁNPh)Cl4(THF) (5)
A solution was prepared of 5 (15.7 mg, 0.0321 mmol)
and C6D6 (700 ml) in an NMR tube that was flame
sealed under vacuum. The sample was heated and
maintained at 55–65°C for about 1 month. A dark
precipitate formed, but the total concentration of imide
remained nearly constant. The progress of the reaction
4.2.17. Decomposition of Mo(ꢁNAr)2(O-t-Bu)2 (8)
A solution of 8 (33 mg, 0.059 mmol) in C6D6 (750 ml)
was prepared in an NMR tube equipped with a Teflon
stopcock. Hexamethylbenzene was added as an internal
standard and the sample was maintained at 100°C for
45 days at which time 8 decomposed. The progress of
1
was monitored by H-NMR spectroscopy and no new
1
signals appeared in the H-NMR spectrum.
1
4.2.15. Thermal stability of W(ꢁNPh)(OC(CH3)-
(CF3)2)2Cl2(THF) (6) in the presence of
(n-Pr)NꢁCHPh
A solution was prepared of 6 (5.1 mg, 0.00654
mmol), C6D6 (400 ml), and anisole (100 ml of a 0.3 M
the reaction was monitored by H-NMR spectroscopy.
Isobutylene and 2,6-diisopropylphenylamine were iden-
1
tified. H-NMR (C6D6) of 2,6-diisopropylphenylamine
(unisolated):
l 3.19 (s, 2, NH2), 2.63 (sept, 2
((CH3)2CH)2) and 1.13 (d, 12, ((CH3)2CH)2). Aromatic

112
A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
resonances could not be definitively assigned due to the
complexity of the spectrum in this region. Significant
resonances from unidentified species were also ob-
served: 1H-NMR (C6D6): l 4.47 (m), 4.30 (m), 3.96 (m),
1.46 (d), 1.37 (m) and 1.13 (m).
with the imide functional group, especially under condi-
tions of prolonged heating.
Acknowledgements
4.2.18. Reaction of (n-Pr)NꢁCHPh and
PhNꢁCH(t-Bu) with the non6olatile components from
the decomposition of Mo(ꢁNAr)2(O-t-Bu)2 (8)
Support for this work was generously provided by
the NSF CAREER (CHE-9624138) and Dupont ATE
programs. T.Y.M. is an Alfred P. Sloan Fellow. We
thank Gidget K. Cantrell for providing the molybde-
num experimental results and for helpful discussions.
A solution of 8 (33 mg, 0.059 mmol) in C6D6 (750 ml)
was prepared in an NMR tube equipped with a Teflon
stopcock. Hexamethylbenzene was added as an internal
standard and the sample was maintained at 100°C for
45 days at which time 8 decomposed. The progress of
References
1
the reaction was monitored by H-NMR spectroscopy.
Isobutylene and 2,6-diisopropylphenylamine were iden-
tified. H-NMR (C6D6) of 2,6-diisopropylphenylamine
[1] P.A. Bates, A.J. Nielson, J.M. Waters, Polyhedron 6 (1987) 163.
[2] Y.-W. Chao, P.M. Rodgers, D.E. Wigley, S.J. Alexander, A.L.
Rheingold, J. Am. Chem. Soc. 113 (1991) 6326.
[3] R.R. Schrock, R.T. DePue, J. Feldman, K.B. Yap, D.C. Yang,
W.M. Davis, L. Park, M. DiMare, M. Schofield, J. Anhaus, E.
Walborsky, E. Evitt, C. Krger, P. Betz, Organometallics 9 (1990)
2262.
[4] G.K. Cantrell, T.Y. Meyer, J. Am. Chem. Soc. 120 (1998) 8035.
[5] G.K. Cantrell, T.Y. Meyer, Chem. Commun. (Cambridge)
(1997) 1551.
[6] G.K. Cantrell, T.Y. Meyer, Organometallics 16 (1997) 5381.
[7] K.E. Meyer, P.J. Walsh, R.G. Bergman, J. Am. Chem. Soc. 117
(1995) 974.
[8] K.E. Meyer, P.J. Walsh, R.G. Bergman, J. Am. Chem. Soc. 116
(1994) 2669.
[9] S.W. Krska, R.L. Zuckerman, R.G. Bergman, J. Am. Chem.
Soc. 120 (1998) 11 828.
[10] The complexity of the NMR and the potential for facile inter-
conversion prevents unambiguous assignment of 2,6-diisopropyl-
phenylamine or its ammonium salt.
[11] T.V. Lubben, P.T. Wolczanski, G.D. Van Duyne, Organometal-
lics 3 (1984) 977.
1
(unisolated):
l 3.19 (s, 2, NH2), 2.63 (sept, 2
((CH3)2CH)2) and 1.13 (d, 12, ((CH3)2CH)2). Aromatic
resonances could not be definitively assigned due to the
complexity of the spectrum in this region. Significant
resonances from unidentified species were also ob-
served: 1H-NMR (C6D6): l 4.47 (m), 4.30 (m), 3.96 (m),
1.46 (d), 1.37 (m) and 1.13 (m).
The volatile compounds of the sample were vacuum
transferred to another NMR tube. The nonvolatile
components showed resonances for 2,6-diisopropy-
lphenylamine and the unidentified species listed above.
Imines (n-Pr)NꢁCHPh and PhNꢁCH(t-Bu) were added
to the nonvolatile compounds and the reaction was
maintained at 25°C for 24 h. The progress of the
reaction was monitored by 1H-NMR spectroscopy.
1H-NMR (C6D6) PhNꢁCHPh (unisolated): l 8.12 (s, 1,
CHPh). 1H-NMR (C6D6) (n-Pr)NꢁCH(t-Bu) (uniso-
lated): 7.33 (s, 1, CH(t-Bu)), 3.26 (t, 2, CH3CH2CH2N),
1.59 (m, 2, CH3CH2CH2N), 1.03 (s, 9, CH(t-Bu), and
0.86 (t, 3, CH3CH2CH2N). 1H-NMR (C6D6) (2,6-
((CH3)2CH)2C6H3)NꢁCHPh (unisolated): l 8.0 (s, 1,
CHPh), 3.12 (sept, 2, ((CH3)2CH)2), and 1.16 (d, 12,
((CH3)2CH)2).
[12] There was no net conversion of t-butyl chloride to isobutylene
under the reaction conditions.
[13] T.J. Crevier, J.M. Mayer, J. Am. Chem. Soc. 119 (1997) 8485.
[14] D.C. Bradley, M.B. Hursthouse, K.M. Abdul Makik, A.J. Niel-
son, R.L. Short, J. Chem. Soc. Dalton Trans. (1983) 2651.
[15] P.A. van der Schaaf, A. Hafner, A. Mhlebach, Angew. Chem.
Int. Ed. Engl. 35 (1996) 1845.
[16] J.M. Mayer, Polyhedron 14 (1995) 3273.
[17] J.C. Peters, A.R. Johnson, A.L. Odom, P.W. Wanandi, W.M.
Davis, C.C. Cummins, J. Am. Chem. Soc. 118 (1996) 10 175.
[18] T.A. Budzichowski, M.H. Chisholm, K. Folting, J.C. Huffman,
W.E. Streib, J. Am. Chem. Soc. 117 (1995) 7428.
[19] M. Nandi, D. Rhubright, A. Sen, J. Am. Chem. Soc. 29 (1990)
3066.
[20] The term CꢀH activation is commonly used to describe a variety
of CꢀH cleavages that occur near metal centers. We have
adopted this usage. However, it is clear that, especially when a
basic ligand is involved, the reactions also fall into the general
category of deprotonation.
[21] Although intermolecular CꢀH activation cannot be ruled out
entirely (the spectra were sufficiently complex to mask minor
products), it is unlikely to be a significant process since it is
inherently reversible, in contrast to the intramolecular process
that we have described. Neither did we observe any obvious H/D
scrambling.
5. Conclusions
The reactions of the t-butoxide-supported imide
complexes 1, 3, and 8 with imines do not proceed via a
Chauvin-type [2+2] metathesis pathway. Experimental
data establish that these complexes are unstable to the
reaction conditions and that the products from the
decomposition are the agents of ꢁNR exchange. Al-
though it is impossible to completely eliminate the
intermediacy of cationic or radical species, no evidence
for their presence was observed. We conclude that the
imide-mediated CꢀH activation is the most probable
first step in the decomposition reaction and that, gener-
ally, t-butoxide ligands are not inherently compatible
[22] For examples see: (a) J. de With, A.D. Horton, Angew. Chem.
Int. Ed. Engl. 32 (1993) 903. (b) P.J. Walsh, F.J. Hollander,

A.Y. Jordan, T.Y. Meyer / Journal of Organometallic Chemistry 591 (1999) 104–113
113
R.G. Bergman, J. Am. Chem. Soc. 110 (1988) 8729. (c) S.Y. Lee,
R.G. Bergman, J. Am. Chem. Soc. 117 (1995) 5877. (d) C.C.
Cummins, S.M. Baxter, P.T. Wolczanski, J. Am. Chem. Soc. 110
(1988) 8731. (e) C.P. Schaller, C.C. Cummins, P.T. Wolczanski,
J. Am. Chem. Soc. 118 (1996) 591. (f) J. Belgacem, J. Kress, J.A.
Osborn, J. Mol. Catal. 86 (1994) 267.
complexes with both chloride and alkoxide ligands. For more
information on alkoxide ligand exchange see: (a) A.J. Nielson,
G.R. Clark, C.E.F. Rickard, Aust. J. Chem. 50 (1997) 259. (b) L.
Wesemann, L. Waldmann, U. Ruschewitz, B. Ganter, T. Wag-
ner, J. Chem. Soc. Dalton Trans. (1997) 1063.
[25] S.R. Sandler, W. Karo, Organic Functional Group Preparations,
vol. II, Academic Press, New York, 1986, p. 302.
[26] M. Jolly, J.P. Mitchell, V.C. Gibson, J. Chem. Soc. Dalton
Trans. (1992) 1329.
[23] A.J. Nielson, G.R. Clark, C.E.F. Rickard, Aust. J. Chem. 50
(1997) 259.
[24] The ligand exchange of chlorides with alkoxides can form new
.