Understanding the role of an easy-to-prepare aldimine-alkyne carboamination catalyst, [Ti(NMe2)3(NHMe2)][B(C 6F5)4]

Article abstract of DOI:10.1016/j.jorganchem.2010.09.017

A series of reactivity studies of the carboamination pre-catalyst [Ti(NMe₂)₃(NHMe₂)][B(C₆F ₅)₄] as well as the preparation of other catalysts are reported in this work. Treatment of [Ti(NMe

Full text of DOI:10.1016/j.jorganchem.2010.09.017

Journal of Organometallic Chemistry 696 (2011) 235e243
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Understanding the role of an easy-to-prepare aldimineealkyne carboamination
catalyst, [Ti(NMe2)3(NHMe2)][B(C6F5)4]
*
Falguni Basuli, Benjamin Wicker, John C. Huffman, Daniel J. Mindiola
Department of Chemistry and Molecular Structure Center, Indiana University, Bloomington, IN 47405, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of reactivity studies of the carboamination pre-catalyst [Ti(NMe2)3(NHMe2)][B(C6F5)4] as well as
the preparation of other catalysts are reported in this work. Treatment of [Ti(NMe2)3(NHMe2)][B(C6F5)4]
with the aldimines Ar0N]CHtol (Ar0 ¼ 2,6-Me2C6H3, tol ¼ 4-MeC6H4), and depending on the reaction
conditions, results in isolation of [Me2N]CHR0][B(C6F5)4] (1) or (Me2N)2CHtol, as well as the asymmetric
titanium dimer [(Me2N)2(HNMe2)Ti(m2-N[2,6-Me2C6H3])2Ti(NHMe2)(NMe2)][B(C6F5)4] (2). Protonation of
CpTi(NMe2)3 and Cp*Ti(NMe2)3 results in isolation of the salts, [CpTi(NMe2)2(NHMe2)][B(C6F5)4] (3) and
[Cp*Ti(NMe2)2(NHMe2)][B(C6F5)4] (4), respectively. Treatment of compounds 3 or 4 with H2N[2,6-iPr2C6H3]
results in formation of the imido salts [CpTi(]N[2,6-iPr2C6H3])(NHMe2)2][B(C6F5)4] (5) (58% yield) or
[Cp*Ti(]N[2,6-iPr2C6H3])(NHMe2)2][B(C6F5)4] (6). When Ti(NMe2)4 is treated with [Et3Si][B(C6F5)4], the
salt [Ti(NMe2)3(N[SiEt3]Me2)][B(C6F5)4] (7) is obtained, and treatment of the latter with [2,6-iPr2C6H3]N]
Received 4 August 2010
Received in revised form
2 September 2010
Accepted 3 September 2010
Available online 21 September 2010
Keywords:
Carboamination
Titanium
Imido
CHtol produces the imine adduct [Ti(NMe2)3(
tion catalytic activity of complexes 2e7 was investigated and compared to [Ti(NMe2)3(NHMe2)][B(C6F5)4].
Likewise, proposed mechanism to the active carboamination catalyst stemming from
[Ti(NMe2)3(NHMe2)][B(C6F5)4] is described.
k
1-[2,6-iPr2C6H3]N]CHtol)][B(C6F5)4] (8). The carboamina-
Nitrogen
a
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
catalytic activity. Despite this attractive entry to these new organic
frameworks, typical reaction conditions involving the metallocene
system required extensive time frames, high temperatures using
10e20 mol% of catalyst, as well as the system had also limited range
of functional group tolerance at both the aldimine and alkyne
substrates [2]. The reaction also suffered from not being atom-
economical when the imido group on zirconiumwas inequivalent to
the aldimine N-aryl group given the ability of [Cp2Zr]NAr] to
undergo imine metathesis with the aldimine and hence afford
In 2004 the Bergman group discovered a catalytic reactionwhere
an aldimine can be added across an internal alkyne to form an
a,b-unsaturated imine (denoted as the carboamination of alkynes
with aldimines, Scheme 1) [1,2]. This interesting transformationwas
made possible via the use of a transient, mononuclear zirconocene
imido precursor, [Cp2Zr]NAr] (Ar ¼ p-substituted aryl where the
p-group is H, OMe, or Me), which could reversibly [2 þ 2]-cycloadd
the alkyne, then insert the aldimine into the ZreC bond of the aza-
metallacyclobutene species (Scheme 1) [1,2]. A [4 þ 2] retro-
cycloaddition stemming from the six-membered metallacycle
a mixture of a,b
-unsaturated imines when R2 s R1 (Scheme 1) [3].
This was the consequence of having a mixture of [Cp2Zr]NAr]
catalysts which would unavoidably result in two separate cycles
therefore lowering the yield of the unsaturated organic product.
In an attempt to circumvent some of these limitations, and given
our interest in the reactivity and the catalytic ability of 3d early-
transition metal complexes having metal-ligand multiple bonds
[4,5], we explored the role of various titanium imidos [6e9] in this
interesting reaction given the fact that NeC and CeC bonds are
catalytically formed and broken. In this contribution we report some
investigations toward understanding the role of an easy-to-prepare
carboamination pre-catalyst, [Ti(NMe2)3(NHMe2)][B(C6F5)4] [9], by
examining the source and role of the imide, variance of the dime-
thylamide ancillary ligand, and anion or activator being used in the
catalytic process.
resulted in extrusion of
a,b-unsaturated imine concurrent with
regeneration of the imido thereby closing the cycle for carboami-
nation of alkynes with aldimines (Scheme 1). Cleverly, the Bergman
system applied a zirconium pre-catalyst Cp2Zr(NHAr)(CH3) (or the
corresponding metallacyclobutene complex resulting from
[2 þ 2]-cycloaddition of the alkyne across the Zr]N bond of
[Cp2Zr]NAr]), which underwent thermolytic a-hydrogen abstrac-
tion to generate the transient imide, [Cp2Zr]NAr]. In the absence of
the alkyne such an intermediate would dimerize thereby killing its
* Corresponding author. Fax: þ1 812 855 8300.
E-mail address: mindiola@indiana.edu (D.J. Mindiola).
0022-328X/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.09.017

236
F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
CH3
a mixture of products after 3 days at 25 ꢀC, we were able to isolate, in
low yield, the iminium salt, [Me2N]CHR0][B(C6F5)4] (1) (Scheme 2).
A single crystal X-ray structure, NMR spectroscopic data and
combustion analysis confirmed the proposed connectivity of 1
(Fig. 1)-see Supplementary material. This result therefore suggested
that the “NAr” motif of the imine serves as the imido source via the
substitution of one eNMe2 ligand in [Ti(NMe2)3(NHMe2)][B(C6F5)4].
The most notable structural feature of 1 (Fig. 1) is formation of an
iminium motif (N]C, 1.307(6) Å versus NeC bonds of 1.466(5) and
1.470(6) Å), consequently resulting in formation of a planar nitrogen
[Zr]
NHR1
-CH4
NR1
Ph
NR1
Ph
R2
Ph
Ph
[Zr]
H
R1
R1
N
Ph
(sum of CeNeC angles
¼
360ꢀ). The reaction depicted in
N
Scheme 2 implies that a neutral, four-coordinate titanium imido,
[ArN]Ti(NMe2)2(NHMe2)] (A), is likely generated in the reaction
mixture along with other side products (Scheme 2), although we
have no spectroscopic evidence for formation of A given the
complexity of the mixture. Likewise, attempted synthesis of putative
A from reactions such as Ti(NMe2)4 and various anilines failed to
produce any isolable product(s) given the lipophilic nature of the
complexes being formed.
[Zr]
Ph
[Zr]
Ph
R2
N
R1
Ph
H
R1N=CHR2
Scheme 1. Proposed Carboamination Cycle Using a Zirconium-Imide Precursor [Zr] ¼
Cp2Zr.
2. Results and discussion
We, however, argue against A being the active catalyst in the
carboamination cycle since Ti(NMe2)4 fails to catalyze the carboa-
mination of alkynes with aldimines in the presence or absence of
aniline [9]. Therefore, we propose compound 1 and putative A to be
formed early in the reaction, but then react to formwhat we speculate
to be the active form of the catalyst. Our hypothesis is in part,
corroborated by an independent reaction using similar conditions to
those of the catalytic reaction (thermolyzing the mixture).
Accordingly, treating tolN]CHtol or [2,6-Me2C6H3]N]CHtol with
1 eq. of [Ti(NMe2)3(NHMe2)][B(C6F5)4] in the absence of the alkyne,
and then heating the mixture for 5 days at 25 ꢀC afforded a mixture of
productsfromwhichtheorganicsideproduct, (Me2N)2CHtolcould be
observed by 1H NMR spectroscopy and MS analysis as well as
comparison with an independently prepared sample [16]. Although
we were unable to characterize any titanium products from the
reaction involving tolN]CHtol and [Ti(NMe2)3(NHMe2)][B(C6F5)4],
the use of a more hindered aldimine, [2,6-Me2C6H3]N]CHtol,
allowed for isolation of one titanium product. In this case, small
quantities of a titanium complex were obtained by fractional crys-
tallization of the reaction mixture from a solution of C6H5F layered
with hexane. The 1H NMR spectrum of these crystals revealed
formation of an asymmetric complex due to the observation of broad
and inequivalent eNMe2 and eNHMe2 resonances. Single crystal
structuralanalysis confirmed thiscompound to bethe imidotitanium
dimer, [(Me2N)2(HNMe2)Ti(m2-N[2,6-Me2C6H3])2Ti(NHMe2)(NMe2)]
[B(C6F5)4] (2) (Fig. 2)-see Supplementary material.
2.1. Understanding the source of the imide using the
carboamination pre-catalyst [Ti(NMe2)3(NHMe2)][B(C6F5)4]
We reported previously that mixing of the commercially avail-
able reagents Ti(NMe2)4 with [NHMe2Ph][B(C6F5)4] gives rise to the
salt, [Ti(NMe2)3(NHMe2)][B(C6F5)4], which has been demonstrated
to be a very reactive pre-catalyst for the carboamination of alkynes
with aldimines [9]. It was found that neither the anilinium proton
source nor the neutral titanium complex, alone, were active cata-
lysts for this transformation, which was surprising given that Ti
(NMe2)4 has been shown by Odom and Schafer to be a highly active
hydroamination catalyst [10e12]. One peculiar feature of this
reaction that we found intriguing was the source of the imide
ligand during the carboamination process. It was determined that
the absence of aniline, which was presumed to be the source of
imido, had no effect in the catalytic formation of the a,b-unsatu-
rated imine [9]. This was puzzling since it is generally accepted that
a terminal imido ligand is being transferred in the catalytic cycle
(Scheme 1, vide supra) [4,13,14]. The fact that a catalytic carboa-
mination process could occur in the absence of aniline suggested
that the imide group is derived from the aldimine as opposed to the
aniline. Therefore, we investigated another mechanism to forma-
tion of the imide which did not involve the more common route,
such as transimination (imide formation derived from aniline
deprotonation) [4].
It has been reported that aldimines can serve as an imido source in
the presence of suitable nucleophiles [15]. Accordingly, we explored
the reactivity of [Ti(NMe2)3(NHMe2)][B(C6F5)4] with aldimines such
as Ar0N]CHtol (Ar0 ¼ 2,6-Me2C6H3, tol ¼ 4-MeC6H4) (Scheme 2) in
the absence of diphenylacetylene. Although the reaction yielded
The molecular structure of 2 displays several interesting features
such as formation of an asymmetric Ti2N2 core (Ti1eN3, 1.800(3);
Ti1eN12,1.877(3); N3eTi1eN12, 92.93(12); N3eTi2eN12, 78.49(10);
Ti1eN3eTi2, 92.60(11); Ti1eN12eTi2, 95.77(11)) due to the bridging
of two imido ligandsto inequivalent metal centers. Theseimidegroups
are most likely derived from the aldimine [2,6-Me2C6H3]N]CHtol. As
Scheme 2. Proposed route to formation of compound 1 and the titanium complex 2.

F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
237
processes taking effect during the carboamination cycle: i) hypo-
thetical three-coordinate (or other masked forms of low-coordi-
nates species) titanium imides and bridging imide salts playing
a critical role in the carboamination cycle, ii) aldimine is an effective
imide transfer source to [Ti(NMe2)3(NHMe2)][B(C6F5)4], and iii) the
dimethylamides are “non-innocent” supporting ligands. Unfortu-
nately, the loss of dimethylamide in the form of (Me2N)2CHtol
implies a non-economical carboamination reaction, and since the
supporting dimethylamide ligands are sacrificed during the forma-
tion of the active species, this catalyst could die over time.
2.2. Exploring other ancillary ligands
Fig. 1. Molecular structure of compound 1 with thermal ellipsoids at the 50% proba-
bility level and with hydrogen atoms (except for the hydrogen on C8), and B(C6F5)4
counter anion omitted for the purpose of clarity. Distances are reported in angstroms
(Å) and angles in (ꢀ). C8eN9, 1.307(6); N9eC10, 1.466(5); N9eC11, 1.470(6); C8eC5,
1.435(4); C8eN9eC10, 120.6(4); C8eN9eC11, 125.4(4); C10eN9eC11, 114.0(4);
N9eC8eC5, 129.6(4).
In addition to Ti(NMe2)4 we have expanded this study to
the monocyclopentadienyl family, namely the known complexes
CpTi(NMe2)3 [18], and Cp*Ti(NMe2)3 [19]. Since we have observed
that eNMe2 is necessary for incorporation of the imide group, we
inquired whether a robust ligand, such as Cpꢁ or Cp*ꢁ, would block
this decomposition route and/or result in formation of a more active
catalyst. Stoichiometric reactions between the precursors CpTi
(NMe2)3 and Cp*Ti(NMe2)3 and [NHMe2Ph][B(C6F5)4] resulted in
formation of the discrete salts [CpTi(NMe2)2(NHMe2)][B(C6F5)4] (3)
and [Cp*Ti(NMe2)2(NHMe2)][B(C6F5)4] (4), respectively (Scheme 4).
We were able to obtain single crystal X-ray diffraction data for each
complex (Fig. 3), and we propose that these complexes are formed
analogously to the original pre-catalyst, [Ti(NMe2)3(NHMe2)]
[B(C6F5)4] [9]- protonation of one of the amide ligands without amine
loss.
noted, the Ti2N2 core is asymmetric due to one titanium center being
5-coordinate, while the other metal center is 4-coordinate and
cationic. Formation of a monocation complex having two inequivalent
titanium centers, 2, suggests that the reaction shown in Scheme 2
is
likely
taking
place-
transient
(Me2N)2Ti]NAr
(A)
(Ar0 ¼ [2,6-Me2C6H3]) dimerizes with an unsaturated titanium imide
salt [(Me2N)(Me2NH)Ti]N[2,6-Me2C6H3]][B(C6F5)4] (B). Formation of
transient B, or a saturated version coordinated by neutral ligands such
as NHMe2 or aldimine, would be consistent with the reaction of 1 with
A to generate B and (Me2N)2CHtol, which we have observed.
The molecular structure of complex 3 reveals a three-legged
piano stool species with one of the amides being protonated see
Supplementary material. As a result, the distances between the two
amide ligands and titanium (1.8917(17) and 1.8773(17) Å) versus
the amine and titanium (2.1992(17) Å) are distinct. The structure of
3 also reveals formation of a discrete salt because of the protonation
of only one of the amide ligands. The molecular structure of
complex 4 is akin to 3 (see Supplementary material), although it
displays a more sterically congested titanium center due to the
bulkier Cp*ꢁ ligand which is manifested by the longer TieCp
(centroid) distance inasmuch as TieN distances and more acute
NeTieN angles (100.90(7), 104.26(7) and 107.75(8)ꢀ for 3 versus
98.75(8),100.41(8) and 105.09(8)ꢀ for 4). The molecular structure of
4 was solved with two independent molecules per asymmetric
unit. Instead of isolating complexes 3 and 4, they can also be
Notably, complex 2 is a highly active carboamination catalyst
(125 C for 24 h, 90% isolated yield of product when using tolN]
CHtol and PhCCPh as substrates), and comparable to the parent
precursor complex [Ti(NMe2)3(NHMe2)][B(C6F5)4]. These results
and observations lend credence to us proposing the following
ꢀ
Fig. 2. Molecular structure of complex 2 with thermal ellipsoids at the 50% probability
level and with hydrogen atoms (except for the hydrogen on N30 and N24), and one B
(C6F5)4 counter anion omitted for the purpose of clarity. Four disordered C6H5F solvent
molecules (two of which lie on center of inversion) have been also omitted. All non-
solvent non-hydrogen atoms were refined with anisotropic displacement parameters.
Distances are reported in angstroms (Å) and angles in (ꢀ). Ti1eTi2, 2.8942(9); Ti1eN3,
1.800(3); Ti1eN12, 1.877(3); Ti1eN21, 1.900(3); Ti1eN24, 2.157(3); Ti2eN3, 2.186(3);
Ti2eN12, 2.023(3); Ti2eN27, 1.884(3); Ti2eN30, 2.274(3); Ti2eN33, 1.908(3);
Ti1eN3eTi2, 92.60(11); Ti1eN12eTi2, 95.77(11); N3eTi1eN12, 92.93(12);
N3eTi2eN12, 78.49(10); N21eTi1eN24, 98.86(14); N3eTi1eN21, 121.85(14);
N3eTi1eN24, 108.82(12); N3eTi2eN30, 161.66(11); N3eTi2eN33, 98.49(12);
N3eTi2eN27, 99.85(12); N24eTi1eN12, 110.06(12); N21eTi1eN12, 124.11(13);
N12eTi2eN27, 128.96(12); N12eTi2eN30, 83.19(11); N12eTi2eN33, 117.31(12).
Fig. 3. Molecular structures of complexes 3 (left) and 4 with thermal ellipsoids at the
50% probability level and with hydrogen atoms (except for the hydrogen on N2)
omitted for the purpose of clarity. The B(C6F5)4 counter anion, one solvent molecule
(for 3), and one of the two crystallographically independent molecules shown for the
structure of 4 have been also omitted for the purpose of clarity. Distances are reported
in angstroms (Å) and angles in (ꢀ). For 3: Ti1eN2, 2.1992(17); Ti1eN5, 1.8917(17);
Ti1eN8, 1.8773(17); Ti1eC(Cp*), (22.341(2), 2.343(2), 2.365(2), 2.391(2), 2.392(2));
N2eTi1eN5, 100.90(7); N2eTi1eN8, 104.26(7); N5eTi1eN8, 107.75(8). For 4: Ti1eN2,
2.204(2); Ti1eN5, 1.8969(17); Ti1eN8, 1.8983(18); Ti1eC(Cp*), (2.340(2), 2.378(2),
2.379(2), 2.411(2), 2.432(2)); N2eTi1eN5, 98.75(8); N2eTi1eN8, 100.41(8);
N5eTi1eN8, 105.09(8).

238
F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
ligand as well as a single environment for the cyclopentadienyl
ligand. To conclusively establish the connectivity of these
complexes, we collected single crystal structural data on complex 6
(Fig. 4). The molecular structure of 6 reveals a cationic, monomeric
titanium imido complex. The Ti]N distance of 1.7309(15) Å and T]
NeCipso angle (162.61(14)ꢀ) is consistent with metal-imide forma-
tion and is much shorter than the TieNamine (2.2011(17) and 2.2178
(17) Å) distances. To avoid clashing with the Cp* methyl groups the
aryl-imide group is almost oriented co-parallel to the Cp* plane
(Fig. 4). The molecular structure of complex 6 resembles previously
reported monocyclopentadienyl imido derivatives of titanium [20].
Performing catalytic reactions with 5 or 6 using identical conditions
to that reported for [Ti(NMe2)3(NHMe2)][B(C6F5)4] yielded similar
results to those observed for precursors Cp*Ti(NMe2)3 or Cp*Ti
(NMe2)3 (vide supra). The fact that the substitution of eNMe2 for
a more robust Cpꢁ or Cp*ꢁ ligand lowers the activity of the catalysts
is surprising given that monocyclopentadienyl complexes of tita-
nium have been demonstrated to be efficient hydroamination
catalysts [20]. Consequently, we propose that the more saturated
and sterically protected nature of the 3 and 4 presumably disfavors
the cycloaddition or the subsequent insertion step during the car-
boamination cycle. We argue that the difficult step is not imide
formation, as neither the addition of aniline nor the use of the imido
complexes improved the catalytic reaction.
CpTi(NMe2)3
NCHtol
Ph
or Cp*Ti(NMe2)3
[HNMe2Ph][B(C6F5)4]
N
H
+
PhC CPh
Ph
C6H6 or toluene
Scheme 3. Catalytic carboamination reactions using monocyclopentadienyl based
precursors.
generated in situ during the catalytic reaction. Accordingly,
protonation of CpTi(NMe2)3 (10 mol%) with [NHMe2Ph][B(C6F5)4]
(10 mol%) in the presence of diphenylacetylene and the aldimine
tolN]CHtol affords the corresponding
a,b-unsaturated imine
(<40% yields) after 16 h at 130 ꢀC (Scheme 3). As observed with
Ti(NMe2)4, the use of aniline (H2Ntol) does not affect the outcome
of the reaction, implying that the imide, again, is derived from the
aldimine. Using the more sterically encumbered precursor Cp*Ti
(NMe2)3 under the same reaction conditions did not improve
reactivity, indicating that neither reagent is as active as the original
pre-catalyst [Ti(NMe2)3(NHMe2)][B(C6F5)4] [9].
To establish if the aldimine also undergoes imide group transfer
to complexes 3 or 4, stoichiometric reactions were performed
between these complexes and various aldimines, such as ArN]
CHtol (Ar ¼ 2,6-Me2C6H3, 2,6-iPr2C6H3, tol), over several days at
25 ꢀC. Unfortunately, no products could be isolated from the
complicated mixture, so an independent route to the imido was
developed. Previous work by our group demonstrated that
Ti(NMe2)4 in the presence of stoichiometric [NHMe2Ph][B(C6F5)4]
and aniline, H2N[2,6-iPr2C6H3], can yield the corresponding imido
salt [Ti(]N2,6-iPr2C6H3)(NHMe2)3(NMe2)][B(C6F5)4] [9]. Analo-
gously, treating 3 or 4 with H2N[2,6-iPr2C6H3] afforded the corre-
sponding imido complexes, [CpTi(]N2,6-iPr2C6H3)(NHMe2)2]
[B(C6F5)4] (5) (58% yield) or [Cp*Ti(]N2,6-iPr2C6H3)(NHMe2)2]
[B(C6F5)4] (6) (35% yield), respectively (Scheme 4). The 1H NMR
spectra of 5 and 6 reveals a freely rotating aryl group on the imido
2.3. Understanding the role of the activator, [NHMe2Ph][B(C6F5)4]
Our work has established that [NHMe2Ph][B(C6F5)4] protonates
one of the eNMe2 ligands of Ti(NMe2)4 to yield the discrete salt,
[Ti(NMe2)3(NHMe2)][B(C6F5)4] [9]. Very likely, the generation of an
amine ligand, HNMe2, is necessary to allow the binding of the
substrate, such as an aldimine (vide supra). As a result, we inquired
if other “activators” would enhance the reactivity of Ti(NMe2)4 for
the carboamination of alkynes with aldimines. Accordingly, treat-
ment of Ti(NMe2)4 with [Et3Si][B(C6F5)4] [21] resulted in the
formation of a new complex, [Ti(NMe2)3(N[SiEt3]Me2)][B(C6F5)4]
(7) (Scheme 5). Single crystal X-ray diffraction studies of complex 7
confirmed the proposed connectivity as a discrete salt (Fig. 5),
R
R
R
R
R
R
R
R
[HNMe2Ph][B(C6F5)4]
-NMe2Ph
R
R
+
other
Ti
Ti
by-products
NHMe2
NMe2
NMe2
Me2N
Me2N
NMe2
B(C6F5)4
R = H, 3; R = CH3, 4
[HNMe2Ph][B(C6F5)4]
H2N[2,6-iPr2C6H3]
-HNMe2
-NMe2Ph
H2N[2,6-iPr2C6H3]
-HNMe2
R
R
R
R
iPr2
R
Ti
N
Me2HN
NHMe2
iPr2
B(C6F5)4
R = H, 5; R = CH3, 6
Scheme 4. Synthesis of the salt complexes 3 and 4 and imido complexes 5 and 6.

F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
239
Fig. 4. Molecular structures of complexes 6 with thermal ellipsoids at the 50% prob-
ability level and with hydrogen atoms (except for the hydrogen on N15 and N18)
omitted for the purpose of clarity. The B(C6F5)4 counter anion has been also omitted for
the purpose of clarity. Distances are reported in angstroms (Å) and angles in (ꢀ).
Ti1eN2, 1.7309(15); Ti1eN15, 2.2011(17); Ti1eN18, 2.2178(17); Ti1eC(Cp*), (2.3349
(19), 2.353(2), 2.3669(18), 2.429(2), 2.4530(19)); N2eTi1eN15, 103.86(7);
N2eTi1eN18, 97.92(7); N15eTi1eN18, 100.08(7), Ti1eN2eC3, 162.61(14).
Fig. 5. Molecular structures of complex 7 with thermal ellipsoids at the 50% proba-
bility level and with hydrogen atoms omitted for the purpose of clarity. Distances are
reported in angstroms (Å) and angles in (ꢀ). Ti1eN2, 2.032(5); Ti1eN12, 2.018(9);
Ti1eN15, 1.842(7); Ti1eN18, 1.796(9); Si5eN2, 1.844(4); N2eTi1eN12, 102.1(3);
N2eTi1eN18, 113.4(3); N2eTi1eN15, 101.2(4); Si5eN2eTi1, 122.2(2); N12eTi1eN18,
107.2(3); N12eTi1eN15, 123.4(3); N18eTi1eN15, 101.2(4).
formed via the coordination of the SiEt3 cation to the lone pair of
electrons in one of the eNMe2 ligand - see Supplementary material.
In the structure of complex 7, there is considerable disorder with
the eNMe2 groups and the final model used two Ti positions and
several poorly defined fragments. Regardless, all non-hydrogen
atoms were refined with anisotropic displacement parameters and
all hydrogen atoms were placed in ideal positions and refined as
riding atoms with relative isotropic displacement parameters so
the overall connectivity is without question.
Coordination of the electrophile to one of the amides results in
elongation of the TieN bond for the amine (Ti1eN2, 2.032(5) Å),
when compared to the other three amide ligands (Ti1eN12, 2.018(9);
Ti1eN15, 1.842(7); Ti1eN18, 1.796(9) Å). Surprisingly, complex 7 is
not as active as a catalyst as [Ti(NMe2)3(NHMe2)][B(C6F5)4] for the
carboamination of internal alkynes with aldimines. Despite this
discouraging result, such data suggests that the “activator” must
coordinate to one of the eNMe2 ligands in order to render the system
active at all. In fact, changing the activator source to [Et3Si][[B(C6F5)4]
or [Ph3C][B(C6F5)4] (product not characterized) in combination with
Ti(NMe2)4, aldimine, and alkyne, did result in formation of a better
played in the pre-catalyst complex [Ti(NMe2)3(NHMe2)][B(C6F5)4].
To investigate this, we treated the latter complex with the bulky
aldimine, [2,6-iPr2C6H3]N]CHtol, under mild conditions
(25 ꢀC, 24 h) and in the absence of the alkyne. Unfortunately the
reaction resulted in a mixture of products. However, treating
complex 7 with [2,6-iPr2C6H3]N]CHtol results in mixture of
products but from which were able to isolate the a metal-based
complex, namely the the N-bound aldimine adduct
[Ti(NMe2)3(
Fig. 6 displays the molecular structure of complex 8, which
possesses a cationic titanium(IV) center having an
1-bound imine
k
1-[2,6-iPr2C6H3]N]CHtol)][B(C6F5)4] (8) (Scheme 5).
k
ligand (Ti1eN2, 2.1465(13) Å) while the short N2eC3 distance of
1.289(2) Å is consistent with a relatively unperturbed imine moiety
when compared to structurally similar aldimines [17]. Overall,
carboamination catalyst since conversions to the
a,b-unsaturated
imine product were at 42% and 37% yields, respectively, after 72 h at
130ꢀC. Using these activators provided lower yield of product when
compared to [Ti(NMe2)3(NHMe2)][B(C6F5)4] (Table 1). The electro-
philes alone, in the absence of Ti(NMe2)4, were not effective for car-
boamination, even when aniline was used in the reaction mixture.
the gross structural features of complex 8 are similar to
[Ti(NMe2)3(NHMe2)][B(C6F5)4] [9], having the basal three eNMe2
ligands oriented in a propeller-type fashion, but where the amine,
HNMe2, has been replaced with an aldimine. This suggests that the
dimethylsilylamine in complex 7 most likely serves as a labile
ligand, opening a coordination site for binding of the aldimine
(Scheme 3). This property could also apply to the HNMe2 ligand in
complex [Ti(NMe2)3(NHMe2)][B(C6F5)4], although we were unable
to isolate 8 from the complicated reaction mixture. The lability of
the amine ligand explains why complex Ti(NMe2)4 must first be
activated (with [NHMe2Ph][B(C6F5)4] or another electrophile) in
order to provide entry for the aldimine.
2.4. Exploring the role of the labile ligand in [Ti(NMe2)3(NHMe2)]
[B(C6F5)4] and 7
From the previous results, it was found that the dimethylamide
ligands can undergo substitution with the ArN group of the
aldimine. However, it was not clear what role the amine ligand
Pr
NCHtol
iPr
iPr
Et3Si
tolHC=N
Ti
NMe2
Ti
NMe2
Ti
[Et3Si][B(C6F5)4]
toluene
iPr
NMe2
NMe2
+
+
Et3SiNMe2
NMe2
Me2N
Me2N
other
products
Me2N
NMe2
NMe2
NMe2
B(C6F5)4
B(C6F5)4
8
7
Scheme 5. Activation of Ti(NMe2)4 with [Et3Si][B(C6F5)4] to form complex 7, and preparation of aldimine 8 from 7.

240
F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
Table 1
intermediate species in the carboamination reaction has been
corroborated by our ability to isolate dinuclear and catalytically
active species, 2, as well as side products such as [Me2N]CHtol]
[B(C6F5)4] and (Me2N)2CHtol. However, we cannot refute the
possibility of a monomer, dimer or oligomer being the active species
in the catalytic cycle, due to the inherent ability of the amine to
protect the titanium center as well as mask low-coordinate frag-
ments. Finally, we have established that replacing one eNMe2
group for Cp or Cp*, or replacing [NHMe2Ph][B(C6F5)4] with other
activators such as Et3Siþ and Ph3Cþ result in a less active carboa-
mination catalysts when compared to [Ti(NMe2)3(NHMe2)]
[B(C6F5)4].
Comparison of catalytic activities of Ti(NMe2)4 with various electrophiles for the
carboamination reaction below. Reactions were carried out on a 0.25 mmol scale of
aldimine using 10 mol% catalyst in C6D6 at 125 ꢀC and %yield was assayed by 1H NMR
spectroscopy.
NCHtol
Ph
10 mol% Ti(NMe2)4
10 mol% Electrophile
N
H
+
PhC CPh
125 oC, 16 h, C6D6
Ph
4. Experimental section
Electrophile
%yield of product
General Considerations. Unless otherwise stated, all operations
were performed in an M. Braun Lab Master box under an atmo-
sphere of purified argon or using high vacuum standard Schlenk
techniques under an argon atmosphere [22]. Anhydrous n-hexane
was purchased from Aldrich in sure-sealed reservoir (18 L) and
dried by passage through two columns of activated alumina and
a Q-5 column [23]. Fluorobenzene was purchased from Aldrich,
degassed, and filtered through activated alumina. Deuterated
benzene was purchased from Cambridge Isotope Laboratory (CIL),
degassed and dried over CaH2 then vacuum transferred to 4 Å
molecular sieves. Deuterated bromobenzene was purchased from
Aldrich and was dried over alumina. Celite, alumina, and 4 Å
molecular sieves were activated under vacuum overnight at 200 ꢀC.
CpTi(NMe2)3 [18], Cp*Ti(NMe2)3 [19], and [Et3Si][B(C6F5)4] [21]
were prepared according to the literature. [HNMe2Ph][B(C6F5)4]
and [Ph3C][B(C6F5)4] was purchased from Boulder Scientific and
dried at 60 ꢀC for 48 h under reduced pressure to eliminate any
traces of CH2Cl2. All the aldimines were synthesized by heating an
equimolar mixture of aldehyde and corresponding aniline at 120 ꢀC
for 15e20 min. The crude aldimine product was extracted into
hexanes or ether (25e50 mL). Concentration to a small volume and
cooling gave crystalline solids which were filtered and dried under
vacuum. All other chemicals were used as received. 1H, 13C, 19F, and
11B NMR spectra were recorded on Varian 400 or 300 MHz NMR
spectrometers. 1H and 13C NMR are reported with reference to
solvent resonances (residual C6D5H in C6D6, 7.16 ppm and
128.0 ppm; C6D4HBr in C6D5Br 7.33, 7.05, 6.97 and 131.1, 129.6,
126.4, 122.4 ppm). 19F NMR chemical shifts are reported with
respect to external HOCOCF3 (ꢁ78.5 ppm). 11B NMR spectra were
reported with respect to external BF3$(OEt2) (0.0 ppm). X-ray
diffraction data were collected on a SMART 6000 (Bruker) system
under a stream of N2 (g) at low temperatures [24,25]. Mass spectra
[NHMe2Ph][B(C6F5)4]
[Et3Si][B(C6F5)4]
[Ph3C][B(C6F5)4]
90
42
37
3. Conclusions
We have examined, reactivity wise, how the easy-to-prepare
[Ti(NMe2)3(NHMe2)][B(C6F5)4] [9] most likely performs the car-
boamination of alkynes with aldimines. Despite unsuccessful
attempts to trap what we speculate to be the “truly active” state of
the catalyst along the carboamination cycle, side products
originating from stoichiometric reactions have shed some clues
to reasonable intermediates formed from the pre-catalyst,
[Ti(NMe2)3(NHMe2)][B(C6F5)4]. Our study suggests that the labile
HNMe2 ligand in [Ti(NMe2)3(NHMe2)][B(C6F5)4] allows for coordi-
nation of the aldimine, and that formation of an imido species
occurs via metathesis of the “NAr” group for two non-innocent
eNMe2 groups on the metal center, establishing the aldimine as the
imide source. However, even though complex [Ti(NMe2)3(NHMe2)]
[B(C6F5)4] is overall more active than the “Cp2Zr]NAr” catalyst
originally discovered by Bergman, such a reagent is also not atom-
economical. Evidence for a neutral or cationic titanium imidos as
were recorded on
spectrometer.
a
ThermoFinnigan MAT 95 XP mass
NOTE: It is imperative that both substrates and inert atmo-
spheres be free of moisture, oxygen, coordinating solvents (Et2O,
THF, pyridine, etc), and Cl sources (CH2Cl2, CHCl3, CCl4, etc). Traces
of these elements will reduce the catalytic activity. It is highly
recommended that substrates, solvents, and titanium precursors be
freed of coordinating solvents prior to usage in catalytic reactions.
4.1. Catalytic reactions
4.1.1. Catalytic reactions using [Ti(NMe2)3(NHMe2)][B(C6F5)4] with
variance of electrophile
Fig. 6. Molecular structures of complexes
8 with thermal ellipsoids at the 50%
probability level and with hydrogen atoms (except for the hydrogen on C3), and B
(C6F5)4 counter anion omitted for the purpose of clarity. Distances are reported in
angstroms (Å) and angles in (ꢀ). Ti1eN2, 2.1465(13); Ti1eN23, 1.8670(13); Ti1eN26,
1.8550(14); Ti1eN29, 1.8666(14); N23eTi1eN2, 110.34(5); N2eTi1eN26, 105.29(6);
N2eTi1eN29, 110.77(6); N23eTi1eN26, 108.62(6); N23eTi1eN29, 110.52(6);
N26eTi1eN29, 111.16(6).
In a typical experiment, alkyne (PhCCPh), (0.275 mmol), tolyl
substituted aldimine at C and N (0.25 mmol), dimethylanilinium
tetrakis(pentafluorophenyl)borate (0.025 mmol, 10%) and Ti(NMe)4
(0.025 mmol, 10%) were mixed in a J. Young NMR tube in C6D6
(0.8 mL) inside the glove box. The NMR tube was removed from the

F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
241
glove box and was heated at 125 ꢀC. The reaction progress was
4.4. Spectroscopic identification of (Me2N)2CHtol
monitored by 1H NMR spectroscopy. (Generally all the reactions
were complete in 24 h). After the reaction was completed, the NMR
tube was cooled to room temperature, the product purified by silica
To a solution of [Ti(NMe2)3(NHMe2)][B(C6F5)4] (50e100 mg) in
C6H5F was added an equal molar amount of [2,6-Me2C6H3]N]CHtol.
Along with several other products compound (Me2N)2CHtol was also
generated and identified by 1H NMR spectroscopy as well as
comparison to reported values [16]. The reaction mixture was also
analyzed by high resolution chemical ionization mass spectrometry
to confirm the presence of (Me2N)2CHtol. For MS-CI
(Me2N)2CHtol þ Hþ: Calculated, 192.1621; Found, 191.1543.
gel column chromatography to afford yellow a,b-unsaturated imine
product. The same experiment was performed using the same
amount of other electrophiles such as [Et3Si][B(C6F5)4] and [Ph3C]
[B(C6F5)4].
4.1.2. Catalytic reactions using complex 2, CpTi(NMe2)3 and Cp*Ti
(NMe2)3
In a typical experiment, alkyne (PhCCPh), (0.275 mmol), tolyl
substituted aldimine at C and N (0.25 mmol), dimethylanilinium
tetrakis(pentafluorophenyl)borate (0.025 mmol, 10%) and 2,
CpTi(NMe2)3 or Cp*Ti(NMe2)3 (0.025 mmol, 10%) were mixed in a J.
Young NMR tube in C6D6 (0.8 mL) inside the glove box. The NMR
tube was removed from the glove box and was heated at 125 ꢀC. The
reaction progress was monitored by 1H NMR spectroscopy.
(Generally all the reactions were complete in 24 h). After the
reaction was completed, the NMR tube was cooled to room
temperature, the product purified by silica gel column chroma-
4.5. Preparation of complex 2
To
a solution of [Ti(NMe2)3(HNMe2)][B(C6F5)4] (200 mg,
0.221 mmol)in C6H5Fwasaddedasolutionof[2,6-Me2C6H3]N]CHtol
(49.39 mg, 0.221 mmol) in C6H5F. The solution was stirred for 5 days
and filtered over Celite. The solution was layered with pentane and
cooled at ꢁ35 ꢀC to afford crystals overnight. The product was
collectedinlowyield(10e20%)bydecantingthesolutionandthesolid
dried under vacuum.
1H NMR (C6D6, 400.12 MHz):
6H), 2.12 (s, 12H), 2.00 (br, 12H). Aryl resonances were obscured by
solvent. 19F NMR (C6D5Br, 376.43 MHz):
Elemental Analysis. Calcd. (for loss of one HNMe2): C 48.43, H 3.64,
N 7.06. Found: C 48.23, H 3.80, N 6.37.
d 3.25 (s, 6H), 2.66 (s, 6H), 2.64 (s,
tography to afford yellow
a,b-unsaturated imine product.
d
ꢁ132.4, ꢁ163.2, ꢁ167.1.
4.2. Isolation of the -unsaturated by column chromatography
a,b
The reaction was carried out at 125 ꢀC for 24 h as indicated
above. The -unsaturated imine product (Scheme 6) was purified
4.6. Preparation of complex 3
a,b
by column chromatography (5% ether:hexanes) as a yellow solid.
Recrystallization of the residue from hexane at ꢁ20 ꢀC afforded
yellow needles.
A solution of CpTi(NMe2)3 (122.2 mg, 0.498 mmol) in C6H5F was
cooled to ꢁ35 ꢀC and to this was added a solution of [HNMe2Ph]
[B(C6F5)3] in C6H5F. The solution was then stirred for 30 min and
filtered over Celite. The solution was layered with pentane and
cooled at ꢁ35 ꢀC, and crystals formed overnight. The product was
collected by filtration and dried under vacuum to yield 325.6 mg
(0.350 mmol, 70.3%).
1H NMR (C6D6, 400.12 MHz):
d
8.36 (d, J ¼ 7.8 Hz, 2H), 7.37 (m,
4H), 7.12e7.06 (m, 4H), 7.02e6.94 (m, 3H), 6.89 (d, J ¼ 8.2 Hz, 2H),
6.75 (d, J ¼ 8.0 Hz, 4H), 1.96 (s, 3H), 1.94 (s, 3H). 13C {H} NMR (C6D6,
100.62 MHz): d 167.45, 149.34, 140.01, 138.58, 138.01, 135.91, 133.88,
133.07, 130.98,130.84,129.53,129.38, 129.12,128.93, 128.79,128.32,
126.55, 119.72, 21.06, 20.80. One resonance is obscured by C6D6.
HRMS CI: Calcd. for C29H25N (Mþ) 387.1982, found 387.1964.
1H NMR (C6D6, 400.12 MHz):
d 5.85 (s, 5H), 2.89, (s, 12H), 2.25
(br, 1H), 6.65 (d, J ¼ 5.6 Hz, 6H). 13C {H} NMR (C6D5Br, 100.62 MHz):
d
148.6 (d, J ¼ 242.8 Hz), 138.4 (d, 258.9 Hz), 136.6 (d, 248.8 Hz),
115.3, 47.6, 40.5. One resonance was obscured by C6D5Br. 19F NMR
(C6D5Br, 282.32 MHz):
4.3. Preparation of compound 1
d
ꢁ133.6, ꢁ163.6, ꢁ167.6. Elemental Anal-
ysis. Calcd: C 44.81, H 3.86, N 5.50. Found: C 44.55, H 3.87, N 5.72.
Ti(NMe2)4 (104 mg, 0.463 mmol) was dissolved in C6H5F at room
temperature. To this was added solutions of [HNMe2Ph][B(C6F5)3]
(371.67 mg, 0.463 mmol) and [2,6-iMe2C6H3]N]CHtol (207.17 mg,
0.926 mmol) in C6H5F. The solution was stirred for 3 days and
filtered over Celite. The solution was layered with pentane and
cooled at ꢁ35 ꢀC, and crystals formed overnight. The product was
collected in low yield (5e20%) by decanting the solution and drying
the solid under vacuum.
4.7. Preparation of complex 4
A solution of Cp*Ti(NMe2)3 (100 mg, 0.317 mmol) in C6H5F was
cooled to ꢁ35 ꢀC and to this solution was added a solution of
[HNMe2Ph][B(C6F5)3] in C6H5F. The solution was stirred for 30 min
and then filtered over Celite. To the solution was added a few drops
of hexane and red crystals formed after 12 h at box temperature
(e27 ꢀC). The product was collected by decanting the solution and
the solid dried under vacuum to yield 297 mg (0.298 mmol, 94.1%).
d 2.78 (s, 12H), 1.89, (s, 7H), 1.63
1H NMR (C6D6, 300.07 MHz):
2.16 (s, 3H). Aryl protons were obscured by solvent resonances.
19F NMR (C6D5Br, 282.32 MHz):
d 7.53 (s,1H), 2.94(s, 3H), 2.91 (s, 3H),
1H NMR (C6D5Br, 400.12 MHz):
d
ꢁ132.6, ꢁ162.1, ꢁ166.2.
(s, 15H). 13C {H} NMR (C6D5Br, 100.62 MHz):
d
148.7 (d, J ¼ 243.1 Hz),
Elemental Analysis. Calcd: C 49.36, H 1.70, N 1.69. Found: C
49.46, H 1.92, N 1.68.
138.4 (d, J ¼ 241.14 Hz), 136.6 (d, J ¼ 241.37 Hz), 124.4, 47.0, 40.3, 11.0.
One resonance was obscured by C6D5Br. 19F NMR (C6D5Br,
282.32 MHz):
48.27, H 3.44, N 4.22. Found: C 48.33, H 3.53, N 4.09.
d
ꢁ133.3, ꢁ165.9, ꢁ168.2. Elemental Analysis. Calcd: C
Ph
N
H
4.8. Preparation of complex 5
Ph
A solution of 3 (67.4 mg, 0.073 mmol) in C6H5F was added to
a J-Young tube and to this solution was added 12.9 mg (0.073 mmol)
of 2,6-diisopropyl aniline. The reaction was heated at 110 ꢀC for 72 h
and the mixture filtered through Celite. The solution was layered
with pentane and after 12 h at ꢁ35 ꢀC crystals were grown and
Scheme 6.
aldimine substrates.
a,b-Unsaturated imine using diphenylacetylene and tolyl substituted

242
F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
collected by decanting the solution and drying the solid under
vacuum to yield 44.5 mg (0.041 mmol, 58%) of the desired product.
(11)ꢀ,
(MoK
a
¼
g
¼ 90ꢀ, U ¼ 3125(2) Å3, Z ¼ 4, Dc ¼ 1.759 g cmꢁ3
, l
a
) ¼ 0.71073 mmꢁ1, T ¼ 173(2) K, Bruker SMART 6000, total
1H NMR (C6D5Br, 400.12 MHz):
d
6.08 (s, 5H), 3.04 (sept, J ¼ 6.8 Hz,
reflections 38,916, unique reflections F > 4s(F), 7172, observed
2H), 2.99, (br, 2H), 2.30 (br, 6H), 2.22 (br, 6H), 1.14 (6, J ¼ 7.1 Hz, 12 H).
reflections 3194 (Rint ¼ 0.1108). The structure was solved using
SHELXS-97 and refined with SHELXL-97. A direct-method solution
was calculated which provided most non-hydrogen atoms from the
E-map. Full-matrix least squares/difference Fourier cycles were per-
formed which located the remaining non-hydrogen atoms. All non-
hydrogen atoms were refined with anisotropic displacement
parameters. Although the anion was well defined and ordered it was
apparent that there was a disorder with the C10H14N cation. Exami-
nation of residuals indicated that the molecule is “flipped” about the
long axis about 20% or the time. The SHELX “SAME” command was
used to refine the secondary structure. All hydrogen atoms were
placed in ideal positions and refined as riding atoms with relative
isotropic displacement parameters. GOF ¼ 0.732 and the final
refinement converged at R(F) ¼ 0.0366 (observed data) and wR
(F2) ¼ 0.0617 (refinement data).
Aryl protons were obscured by solvent.19F NMR (C6D5Br, 282.32 MHz):
d
ꢁ132.7, ꢁ162.2, ꢁ166.3. Elemental Analysis Calcd. (with one C6H5F
included): C 53.10, H 3.58, N 3.64. Found: C 52.99, H 3.50, N 4.08.
4.9. Preparation of complex 6
A solution of 4 (65.1 mg, 0.065 mmol) in C6H5F was added to
a J-Young tube. To this was added 11.5 mg (0.098 mmol) of
2,6-diisopropyl aniline. The reaction was heated at 110 ꢀC for 72 h.
The solution was then filtered through Celite and layered with
pentane. After 12 h at ꢁ35 ꢀC crystals were grown and then
collected by filtration and dried under vacuum to yield 25.7 mg
(0.023 mmol, 35%) of the desired product.
1H NMR (C6D5Br, 400.12 MHz):
d
6.99 (d, J ¼ 6.7 Hz, 2H), 6.87 (t,
J ¼ 6.7 Hz, 1H), 2.67 (sept, J ¼ 6.8 Hz, 2H), 2.61 (br, 6H), 2.49 (br, 2H),
Single crystals of 2 were grown from a fluorobenzene/hexane-
layered solution at room temperature under an N2 atmosphere.
Crystal data for C68H65BF23N7Ti2, 2: M ¼ 1523.88, Triclinic, space
group Pe1, a ¼ 12.2372(13) Å, b ¼ 16.6642(18) Å, c ¼ 17.4838(19) Å,
2.33 (br, 6h), 1.74 (s, 15H), 1.12 (d, J ¼ 6.8 Hz, 12H). 13C {H} NMR
(C6D5Br, 100.62 MHz):
d
138.2 (d, J ¼ 239.8 Hz), 128.0 (d,
J ¼ 247.6 Hz), 126.1 (d, J ¼ 245.0 Hz), 114.1, 110.9, 110.4, 104.8, 104.6,
36.6, 29.8, 25.8, 0.5. 19F NMR (C6D5Br, 282.32 MHz):
d
ꢁ131.9,
a
¼ 83.185(3)ꢀ,
b
¼ 71.840(4)ꢀ,
g
¼ 83.243(3)ꢀ, U ¼ 3351.5(6) Å3,
) ¼ 0.71073 mmꢁ1, T ¼ 123(2) K,
ꢁ162.1, ꢁ166.1. Elemental Analysis. Calculated: C 53.26, H 4.11, N
Z ¼ 2, Dc ¼ 1.510 g cmꢁ3
, l(MoKa
3.72. Found: C 52.36, H 3.92, N 4.06.
Bruker SMART 6000, total reflections 76,283, unique reflections
F > 4s(F), 15,458 observed reflections 9935 (Rint ¼ 0.0861). The
4.10. Preparation of complex 7
structure was solved using SHELXS-97 and refined with SHELXL-97.
A direct-method solution was calculated which provided most non-
hydrogen atoms from the E-map. Full-matrix least squares/differ-
ence Fourier cycles were performed which located the remaining
non-hydrogen atoms. All non-hydrogen atoms were refined with
anisotropic displacement parameters. In addition to the Ti dimer
and anion, there are four C6H5F solvent molecules present, all
disordered and two lying on centers of inversion. All hydrogen
atoms were placed in ideal positions and refined as riding atoms
with relative isotropic displacement parameters. GOF ¼ 1.035 and
the final refinement converged at R(F) ¼ 0.0669 (observed data)
and wR(F2) ¼ 0.1919 (refinement data).
A solution of Ti(NMe2)4 (100 mg, 0.446 mmol) in C6H5F was
cooled to ꢁ35 ꢀC. To this was added a solution of [Et3Si][B(C6F5)3] in
C6H5F. The mixture was stirred for 15 min and filtered over Celite.
The solution was layered with pentane and cooled at ꢁ35 ꢀC, and
crystals formed overnight. The product was collected by decanta-
tion and dried under vacuum to yield 393 mg (0.386 mmol, 86.5%).
1H NMR (C6D6, 300.07 MHz):
d 2.95 (s, 3H), 2.89 (s, 24H), 2.15 (s,
3H), 0.80 (t, J ¼ 7.7 Hz, 9H), 0.57 (q, J ¼ 7.9 Hz, 6H). 11B NMR (C6D5Br,
128.37 MHz):
d
ꢁ16.3. 19F NMR (C6D5Br, 282.32 MHz):
d
ꢁ131.9,
ꢁ162.2, ꢁ166.1. Elemental Analysis. Calcd: C 44.81, H 3.86, N 5.50.
Found: C 44.55, H 3.87, N 5.72.
Single crystals of 3 were grown from a fluorobenzene/hexane-
layered solution at room temperature under an N2 atmosphere.
Crystal data for C41H29BF21N3Ti, 3: M ¼ 1021.38, Monoclinic, space
group P2(1)/c, a ¼ 12.027(2) Å, b ¼ 21.591(4) Å, c ¼ 17.203(3) Å,
4.11. Preparation of complex 8
¼ 90ꢀ,
b
¼ 108.993(5)ꢀ,
g
¼ 90ꢀ, U ¼ 4223.9(14) Å3, Z ¼ 4,
To a solution of 7 (171 mg, 0.167 mmol) in C6H5F was added
a solution of [2,6-iPr2C6H3]N]CHtol (46.92 mg, 0.167 mmol) in
C6H5F. The solution was stirred for 20 h and filtered through Celite.
The solution was then layered with pentane and cooled at ꢁ35 ꢀC,
and crystals formed overnight. The product was collected by
decantation and dried under vacuum.
a
Dc ¼ 1.606 g cmꢁ3
,
l
(MoK
a
) ¼ 0.71073 mmꢁ1, T ¼ 123(2) K, Bruker
SMART 6000, total reflections 54,918, unique reflections F > 4s(F),
9741 observed reflections 5971 (Rint ¼ 0.0673). The structure was
solved using SHELXS-97 and refined with SHELXL-97. A direct-
method solution was calculated which provided most non-hydrogen
atoms from the E-map. Full-matrix least squares/difference Fourier
cycles were performed which located the remaining non-hydrogen
atoms. One FC6H5 solvent molecule was found in the asymmetric
unit. All non-hydrogen atoms were refined with anisotropic
displacement parameters. All hydrogen atoms were placed in ideal
positions and refined as riding atoms with relative isotropic
displacement parameters. GOF ¼ 0.870 and the final refinement
converged at R(F) ¼ 0.0352 (observed data) and wR(F2) ¼ 0.0799
(refinement data).
1H NMR (C6D6, 400.12 MHz):
d
8.12 (s, 1H), 7.07 (dd, J ¼ 8.42,
16.24 Hz, 1H), 6.94 (d, 7,79 Hz, 2H), 6.57 (q, 8.45 Hz, 4H), 2.79 (s, 18H),
2.20 (sept, 6.73 Hz, 2H). 1.77 (s, 3H), 0.81 (d, 6.80 Hz, 6H), 0.52 (d,
6.74 Hz, 6H). 11B NMR (C6D5Br, 128.37 MHz):
d
ꢁ16.3. 13C {H} NMR
(C6D5Br, 100.62 MHz): d 173.8, 149.6, 148.7 (d, 235.62 Hz), 138.8, 138.5
(d, 243.76 Hz), 138.3, 136.6 (d, 241.53 Hz), 134.0, 130.5, 126.5, 125.8,
43.7, 43.1, 28.8, 24.3, 23.1, 21.7. Two aryl resonances were obscured by
C6D5Br. 19F NMR (C6D5Br, 376.43 MHz):
Elemental Analysis. Calcd: C 52.75, H 3.81, N 4.92. Found: C 52.57, H
3.87, N 4.74.
d
ꢁ133.4, ꢁ163.9, ꢁ167.7.
Single crystals of 4 were grown from a fluorobenzene/hexane-
layered solution at room temperature under an N2 atmosphere.
Crystal data for C40H34BF20N3Ti, 4: M ¼ 995.41, Triclinic, space
group Pe1, a ¼ 14.7239(8) Å, b ¼ 15.7062(9) Å, c ¼ 19.0241(11) Å,
4.12. X-ray crystallographic data
Single crystals of 1 were grown from a fluorobenzene/hexane-
layered solution at room temperature under an N2 atmosphere.
Crystal data for C34H14BF20N, 1: M ¼ 827.27, Monoclinic, space group
a
¼ 105.504(2)ꢀ,
b
¼ 107.322(2)ꢀ,
g
¼ 91.870(2)ꢀ, U ¼ 4016.9(4) Å3,
Z ¼ 4, Dc ¼ 1.646 g cmꢁ3
,
l
(MoK
a
) ¼ 0.71073 mmꢁ1, T ¼ 133(2) K,
Bruker SMART 6000, total reflections 39,538, unique reflections
P2(1)/c, a ¼ 15.311(7) Å, b ¼ 10.607(5) Å, c ¼ 19.260(9) Å,
b
¼ 92.558
F > 4
s
(F), 23,272 observed reflections 11,556 (Rint ¼ 0.0512). The

F. Basuli et al. / Journal of Organometallic Chemistry 696 (2011) 235e243
243
structure was solved using SHELXS-97 and refined with SHELXL-97.
Acknowledgements
A direct-method solution was calculated which provided most non-
hydrogen atoms from the E-map. Full-matrix least squares/differ-
ence Fourier cycles were performed which located the remaining
non-hydrogen atoms. The structure was found as proposed with
two independent molecules per asymmetric unit. All non-hydrogen
atoms were refined with anisotropic displacement parameters. All
hydrogen atoms were placed in ideal positions and refined as riding
atoms with relative isotropic displacement parameters.
GOF ¼ 0.756 and the final refinement converged at R(F) ¼ 0.0447
(observed data) and wR(F2) ¼ 0.0848 (refinement data).
We thank Dr. Aneetha Halikhedkar for insightful discussions and
Indiana University-Bloomington, and the Chemical Sciences, Geo-
sciences and Biosciences Division, Office of Basic Energy Science, Office
of Science, U.S. Department of Energy (No. DE-FG02-07ER15893) for
financial support of this research.
Appendix A. Supplementary material
CCDC 786995e787001 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Center via www.
ccdc.cam.ac.uk/data_request/cif.
Single crystals of 6 were grown from a fluorobenzene/hexane-
layered solution at room temperature under an N2 atmosphere. Crystal
data for C50H46BF20N3Ti, 6: M ¼ 1127.61, Monoclinic, space group P2
(1)/n, a ¼ 17.654(2) Å, b ¼ 13.7175(17) Å, c ¼ 21.887(3) Å,
a
¼
g
¼ 90ꢀ,
b
¼ 109.354(3)ꢀ, U ¼ 5000.9(11) Å, Z ¼ 4, Dc ¼ 1.498 g cmꢁ3
, l
References
(MoK
a
) ¼ 0.71073 mmꢁ1, T ¼ 125(2) K, Bruker SMART 6000, total
[1] R.T. Ruck, R.G. Bergman, Organometallics 23 (2004) 2231.
[2] R.T. Ruck, R.L. Zuckermann, S.W. Krska, R.G. Bergman, Angew. Chem. Int. Ed.
43 (2004) 5372.
reflections 38,290, unique reflections F > 4s(F), 11,489 observed
reflections 6443 (Rint ¼ 0.0682). The structure was solved using
SHELXS-97 and refined with SHELXL-97. A direct-method solutionwas
calculated which provided most non-hydrogen atoms from the E-map.
Full-matrix least squares/difference Fourier cycles were performed
which located the remaining non-hydrogen atoms. All non-hydrogen
atoms were refined with anisotropic displacement parameters. All
hydrogen atoms were placed in ideal positions and refined as riding
atoms with relative isotropic displacement parameters. GOF ¼ 0.825
and the final refinement converged at R(F) ¼ 0.0351 (observed data)
and wR(F2) ¼ 0.0783 (refinement data).
[3] (a) S.W. Krska, R.L. Zuckerman, R.G. Bergman, J. Am. Chem. Soc. 120 (1998) 11828;
(b) K.E. Meyer, P.J. Walsh, R.G. Bergman, J. Am. Chem. Soc. 116 (1994) 2669.
[4] A.R. Fout, U.J. Kilgore, D.J. Mindiola, Chem. Eur. J. 13 (2007) 9428.
[5] D.J. Mindiola, Comments Inorg. Chem. 29 (2008) 73.
[6] (a) F. Basuli, J.C. Huffman, D.J. Mindiola, Inorg. Chem. 42 (2003) 8003;
(b) F. Basuli, B.C. Bailey, J.C. Huffman, D.J. Mindiola, Chem. Commun. (2003) 1554.
[7] F. Basuli, H. Annetha, J.C. Huffman, D.J. Mindiola, J. Am. Chem. Soc. 127 (2005) 17992.
[8] F. Basuli, R.L. Clark, B.C. Bailey, D. Brown, J.C. Huffman, D.J. Mindiola, Chem.
Commun. (2005) 2250.
[9] H. Aneetha, F. Basuli, J. Bollinger, J.C. Huffman, D.J. Mindiola, Organometallics
25 (2006) 2402.
[10] A.L. Odom, Dalton Trans. (2005) 225.
Single crystals of 7 were grown from a fluorobenzene/hexane-
layered solution at room temperature under an N2 atmosphere.
Crystal data for C38H39BF20N4SiTi, 7: M ¼ 1018.54, Orthorhombic,
space group Pbca, a ¼ 16.802(8) Å, b ¼ 20.973(10) Å, c ¼ 24.050(11) Å,
[11] (a) Y. Shi, J.T. Ciszewski, A.L. Odom, Organometallics 20 (2001) 3967;
(b) Y. Shi, J.T. Ciszewski, A.L. Odom, Organometallics 21 (2002) 5148.
[12] J.A. Bexrud, J.D. Beard, D.C. Leitch, L.L. Schafer, Org. Lett. 7 (2005) 1959.
[13] N. Hazari, P. Mountford, Acc. Chem. Res. 38 (2005) 839.
[14] (a) D.E. Wigley, Prog. Inorg. Chem. 42 (1994) 239;
(b) W.A. Nugent, J.M. Mayer, Metal-Ligand Multiple Bonds. John Wiley & Sons,
New York, 1988;
(c) L.H. Gade, P. Mountford, Coord. Chem. Rev. 216 (2001) 65;
(d) A.P. Duncan, R.G. Bergman, Chem. Rec. 2 (2002) 431;
(e) W.A. Nugent, B.L. Haymore, Coord. Chem. Rev. 31 (1980) 123.
[15]. P. Gomez-Sal, A. Martin, M. Mena, M.C. Morales, C. Santamaria, Chem. Com-
mun. (1999) 1839.
a
¼
b
¼
g
¼ 90ꢀ, U ¼ 8475(7) Å3, Z ¼ 8, Dc ¼ 1.598 gꢂcmꢁ3
, l
(MoK
a
) ¼ 0.71073 mmꢁ1, T ¼ 123(2) K, Bruker SMART 6000, total
reflections 38,089, unique reflections F > 4s(F), 9745 observed
reflections 4703 (Rint ¼ 0.0822). The structure was solved using
SHELXS-97 and refined with SHELXL-97. A direct-method solution
was calculated which provided most non-hydrogen atoms from the
E-map. Full-matrix least squares/difference Fourier cycles were
performed which located the remaining non-hydrogen atoms. There
is considerable disorder with the eNMe2 groups. The final model
used twoTi positions and several poorly defined fragments. All non-
hydrogen atoms were refined with anisotropic displacement
parameters. All hydrogen atoms were placed in ideal positions and
refined as riding atoms with relative isotropic displacement
parameters. GOF ¼ 0.905 and the final refinement converged at R
(F) ¼ 0.0653 (observed data) and wR(F2) ¼ 0.2124 (refinement data).
Single crystals of 8 were grown from a fluorobenzene/hexane-
layered solution at room temperature under an N2 atmosphere.
Crystal data for C50H43BF20N4Ti, 8: M ¼ 1138.59, Triclinic, space
group Pe1, a ¼ 10.0164(14) Å, b ¼ 15.180(2) Å, c ¼ 16.942(3) Å,
[16] H. Boehme, M. Haake, G. Auterhoff, Archiv der Pharmazie 305 (1972) 10.
[17] (a) 1.250(15) Å has been reported for dialdimine, 1,2-[[2,6-iPr2C6H3]N¼CH]2
C6H4. J.M. Chitanda, D.E. Prokopchuk, J.W. Quail, S.R. Foley Organometallics 27
(2008) 2337;
(b) 1.274
Organometallics 19 (2000) 3305
Å
for 2,6-[[2,6-iPr2C6H3]N¼C(CH3)]2C6H4. S. Nuckel, P. Burger
[18] G. Chandra, M.F. Lappart, J. Chem. Soc. A (1968) 1940.
[19] A. Martin, M. Mena, C. Yélamos, R. Sarrano, J. Organomet. Chem. 467 (1994) 79.
[20] For some recent reviews and manuscripts. (a) K. Weitershaus, B.D. Ward,
R. Kubiak, C. Müller, H. Wadepohl, S. Doye, L.H. Gade, Dalton Trans. (2009)
4586;
(b) P.W. Roesky, T.E. Müller, Angew. Chem., Int. Ed. 42 (2003) 2708;
(c) F. Pohlki, S. Doye, Chem. Soc. Rev. 32 (2003) 104;
(d) I. Bytschkov, S. Doye, Eur. J. Org. Chem. (2003) 935;
(e) J.F. Hartwig, Pure Appl. Chem. 76 (2004) 507;
(f) F. Alonso, I.P. Beletskaya, M. Yus, Chem. Rev. 104 (2004) 3079;
(g) S. Doye, Synlett (2004) 1653;
(h) K.C. Hultzsch, Adv. Synth. Catal. 347 (2005) 367;
a
¼ 95.485(4)ꢀ,
b
¼ 99.275(4)ꢀ,
g
¼ 94.871(4)ꢀ, U ¼ 2517.4(7) Å3,
(i) K.C. Hultzsch, Org. Biomol. Chem. 3 (2005) 1819;
Z ¼ 2, Dc ¼ 1.502 g cmꢁ3
,
l
(MoK
a
) ¼ 0.71073 mmꢁ1, T ¼ 122(2) K,
(j) R. Severin, S. Doye, Chem. Soc. Rev. 36 (2007) 1407;
(k) T.E. Müller, K.C. Hultzsch, M. Yus, F. Foubelo, M. Tada, Chem. Rev. 108
(2008) 3795;
(l) J.-J. Brunet, N.-C. Rodriguez-Zubiri, Eur. J. Inorg. Chem. (2007) 4711;
(m) P.L. McGrane, M. Jensen, T. Livinghouse, J. Am. Chem. Soc. 114 (1992)
5459;
Bruker SMART 6000, total reflections 69,434, unique reflections
F > 4
s
(F), 14,742 observed reflections 9778 (Rint ¼ 0.0588). The
structure was solved using SHELXS-97 and refined with SHELXL-97.
A direct-method solution was calculated which provided most
non-hydrogen atoms from the E-map. Full-matrix least squares/
difference Fourier cycles were performed which located the
remaining non-hydrogen atoms. All non-hydrogen atoms were
refined with anisotropic displacement parameters. All hydrogen
atoms were placed in ideal positions and refined as riding atoms
with relative isotropic displacement parameters. GOF ¼ 0.966 and
the final refinement converged at R(F) ¼ 0.0447 (observed data)
and wR(F2) ¼ 0.1286 (refinement data).
(n) J.S. Johnson, R.G. Bergman, J. Am. Chem. Soc. 123 (2001) 2923.
[21] (a) J.B. Lambert, S. Zhang, S.M. Ciro, Organometallics 13 (1994) 2430;
(b) V.J. Scott, R. Celenligil-Cetin, O.V. Ozerov, J. Am. Chem. Soc. 127 (2005) 2852.
[22] For a general description of the equipment and techniques used in carrying out
this chemistry see: B.J. Burger, J.E. Bercaw, in: A.L. Wayda, M.Y. Darensbourg
(Eds.), Experimental Organometallic Chemistry, ACS Symposium Series, 357,
American Chemical Society, Washington D.C., 1987, pp. 79e98.
[23] A.B. Pangborn, M.A. Giardello, R.H. Grubbs, R.K. Rosen, F.J. Timmers, Organo-
metallics 15 (1996) 1518.
[24] SAINT 6.1, Bruker Analytical X-Ray Systems, Madison, WI.
[25] SHELXTL-Plus V5.10, Bruker Analytical X-Ray Systems, Madison, WI.