Imidotungsten(vi) complexes with chelating amino and imino phenolates

Article abstract of DOI:10.1039/c0dt01438a

The reaction of WOCl₄ with 2,4-di-tert-butyl-6-((isopropylamino) methyl)phenol followed by the reaction with phenyl isocyanate leads to the formation of imidotungsten(vi) complex [W(NPh)Cl₃(OC ₆H₃(CH₂NH-i-Pr)-2-t-Bu₂-4,6)] 4 with a chelating aminophenolate ligand. When the same procedure was applied using aminophenols with bulkier substituents in the amino group, the final product was an unexpected Schiff-base complex [W(NPh)Cl₃(OC₆H ₃(CHNPh)-2-t-Bu₂-4,6)] 5, where the ligand is derived from 2,4-di-tert-butyl-6-((phenylimino)methyl)phenol. Complex 5 is also formed in the thermal degradation of 4. On the whole, 5 appears to be formed by a disproportionation of intermediate compounds, which are analogous to complex 4. The solid-state structures of 4 and 5 have been determined by X-ray crystallography whereas the solution structures were studied by ¹H and ¹³C NMR. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c0dt01438a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 2868
www.rsc.org/dalton
PAPER
Imidotungsten(VI) complexes with chelating amino and imino phenolates†
Mikko M. Ha¨nninen,a Reijo Sillanpa¨a¨,a Henri Kivela¨b and Ari Lehtonen*b
Received 21st October 2010, Accepted 17th December 2010
DOI: 10.1039/c0dt01438a
The reaction of WOCl4 with 2,4-di-tert-butyl-6-((isopropylamino)methyl)phenol followed by the
reaction with phenyl isocyanate leads to the formation of imidotungsten(VI) complex
[W(NPh)Cl3(OC6H3(CH2NH-i-Pr)-2-t-Bu2-4,6)] 4 with a chelating aminophenolate ligand. When the
same procedure was applied using aminophenols with bulkier substituents in the amino group, the final
product was an unexpected Schiff-base complex [W(NPh)Cl3(OC6H3(CH NPh)-2-t-Bu2-4,6)] 5, where
the ligand is derived from 2,4-di-tert-butyl-6-((phenylimino)methyl)phenol. Complex 5 is also formed in
the thermal degradation of 4. On the whole, 5 appears to be formed by a disproportionation of
intermediate compounds, which are analogous to complex 4. The solid-state structures of 4 and 5 have
been determined by X-ray crystallography whereas the solution structures were studied by 1H and 13
NMR.
C
Introduction
Aryloxide derivatives of tungsten(VI) are traditional catalyst
precursors for the metathesis of alkenes when activated by main-
group organometallic co-catalysts.1 Well-known examples of these
precursors include chloride derivatives such as WOCl4-x(OAr)x and
WCl6-x(OAr)x. Although the metathesis reactions can be catalysed
by well-known and commercially available Mo or Ru based
catalysts, there is still use for some simple two-component catalyst
systems.2 Especially, the ring-opening metathesis polymerization
(ROMP) of dicyclopentadiene and other norbornene derivatives
is catalysed by simple tungsten-based catalyst systems to produce
elastomers and other valuable materials.3 As typical for high-
valent metal chlorides, these catalysts are rather sensitive towards
water and other nucleophiles. Especially, if the coordination
number of the metal centre is low, these compounds must
be handled under an inert atmosphere using rather tedious
practices. The stability of high-valent tungsten complexes can
be improved by coordinative protection, i.e. by using chelating
ligands, which carry a potentially coordinating neutral group
within the molecule (Fig. 1).4,5 For example, the phenylimido
tungsten(VI) aryloxides [W(NPh)Cl3(OC6H2(CH2NMe2)2-2,6-Me-
4)] (1)4a and [W(NPh)Cl3(OC6H3(CH2NMe2)-2-t-Bu2-4,6)] (2)5
(Scheme 1) are stable under ambient atmosphere and can be melted
in air without decomposition. We may assume that the stability
of the tungsten imido complexes with the O,N-bidentate amino
Fig. 1 Stable aminophenol complexes.
Scheme 1 Preparation of aminophenols.
phenolate ligand depends on the strength of the coordinative W–N
bond.
Previously, we prepared compound 2 and its molybdenum
analogue and used them as catalyst precursors for ROMP of
norbornene derivatives.5 It was found that the use of aluminium
co-catalyst Et2AlCl was necessary to obtain any activity. Although
these catalysts showed good activity for ROMP of norbornene,
the activity for the reactions of other monomers was poor.
Thus, it seems that the rather strong coordinative M–N bond
deteriorates the catalytic activity of activated complexes, as a
high activity requires a low coordination number of the reacting
metal centre. To weaken the M–N bond and thus improve
aLaboratory of Inorganic Chemistry, Department of Chemistry, University
of Jyva¨skyla¨, FI-40351, Jyva¨skyla¨, Finland
bLaboratory of Materials Chemistry and Chemical Analysis, Depart-
ment of Chemistry, University of Turku, FI-20014, Turku, Finland.
E-mail: ari.lehtonen@utu.fi; Tel: +35823336733
† Electronic supplementary information (ESI) available. CCDC reference
numbers 797991 and 797992. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0dt01438a
2868 | Dalton Trans., 2011, 40, 2868–2874
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Calculated W–N bond lengths, MPA and NPA charges for the
N donors and f angles of model complexes
the catalytic performance, we decided to prepare a series of
corresponding tungsten compounds with varying substituents in
the amino group. Particularly, we assumed that the elongated
W–N bond may transform the coordination mode of the ligand
into monodentate in solution, consequently erasing the stability
of the complex derived from the chelate effect and inducing the
5-coordination favourable for the catalytic application. To keep
the synthetic procedure as simple as possible, a straightforward
two-step reaction was chosen for the preparation of a family of
these new complexes. However, this procedure ended up with some
unexpected results, as it was found that the reaction outcome
depends strongly on the substituents in the amino group of
the phenolate ligand. Depending on these substituents, the final
product may carry either amino or imino nitrogen as the neutral
donor atom.
˚
Entry
R1
R2
r(W–N)A
Q (MPA)
Q (NPA)
f/◦
1
2
3
4
5
6
7
8
9
H
H
H
H
Me
Et
iPr
Me
Ph
H
2.38
2.44
2.46
2.47
2.60
2.68
3.06
2.38
2.41
-0.800
-0.601
-0.609
-0.619
-0.407
-0.411
-0.416
-0.424
-0.442
-0.459
-0.335
-0.306
-0.317
-0.232
-0.205
-0.184
-0.155
-0.210
38.9
38.2
39.1
37.9
37.2
35.8
34.7
—
Me
Et
iPr
Me
Et
iPr
—
—
—
complexes with secondary amines compared to the more common
tertiary amine complexes. In general, the calculated W–N bond
˚
lengths for secondary amine complexes are from 0.2 to 0.6 A
shorter than with corresponding tertiary amines. The calculated
bond lengths are in good agreement with the experimental data
where available. The small differences between the theoretical and
Results and discussion
Computational studies on the iminotungsten(VI) complexes with
aminophenolates
˚
experimental results (approximately 0.1 A) in W–N bond distances
are most likely due to the relativistic and solid-state packing effects
which are not counted in calculations.
In our earlier studies, we have experienced some difficulties
isolating the aminophenolate complexes of W(VI) ion, hence
we studied the stability of the complexes computationally in
addition to experimental studies. The coordinative bond between
tungsten(VI) ion and the amine nitrogen is most likely the weakest
bond in these compounds thus defining the stability of the
complex. Generally, the factors contributing to the strength of
the above mentioned bond are the sterical hindrance induced
by alkyl substituents, inductive effect and resonance effect. To
study the influence of the preceding effects on the W–N bond, we
performed the geometry optimizations (at the DFT BP86/def2-
TZVP level) and partial atomic charge calculations to the group
of the simplified model complexes of iminotungsten(VI) chloride
with different aminophenolate ligands (Fig. 2).
In order to explain this trend we calculated the atomic charges
of the complexes. Although the partial atomic charges are
rather arbitrary models they provide some useful and chemically
interpretable information on the properties of the molecules.
Hence, we performed Mulliken Population Analysis (MPA) and
Natural Population Analysis (NPA) for the preceding model
compounds (see Table 1). According to the extensive study by
Hadad et al. onthe competence of differentatomic chargeschemes,
NPA gives truthful charges on the nitrogen atoms in substituted
anilines.6 Consequently, we assume that the NPA analyses could
provide reasonable results on the partial atomic charges in amine
complexes as well.
The electron releasing ability (inductive effect) of alkyl sub-
stituents follows the order H > CMe3 > CHMe2 > CH2Me >
CH3. Accordingly, the calculated MPA and NPA charges for the
nitrogen donors in the primary and secondary amines are more
negative than in the tertiary amines. This trend is in line with the
W–N bond lengths as the bond length decreases when the negative
partial charge of nitrogen increases. Since the partial atomic charge
of tungsten remains in practice at the same level through whole
series the charge difference between tungsten and nitrogen atoms
(see Table 1 and ESI†) increases thus adding to the attractive ionic
contribution of the bond.
Fig. 2 Schematic presentation of simplified model complexes.
In addition to the partial charge of the nitrogen donor, there
exists a correlation of the calculated W–N bond lengths with the
pyramidality (dihedral angle CH2–R1–R2–N, f) of the nitrogen
atom. The pyramidality of a nitrogen atom indicates the steric
hindrance of the donor group. Hence, the more pyramidalized
(i.e. larger f angle) the nitrogen atom the better donor it can be.
Keeping these results in mind, we decided to prepare a series of
imidotungsten(VI) complexes with substituted aminophenols.
The amine ligands in the model complexes can be divided into
primary (entry 1 in Table 1), secondary (entries 2–4) and tertiary
amines (entries 5–7). Furthermore, the geometries of two related
Schiff base complexes (entries 8–9) were optimized although the
bonding mode is different compared with the aminophenolate
complexes. In these models, the C N double bond may provide
some p-contribution to the W–N bond thus shortening the
bonding distance and increasing the stability of the Schiff base
complexes.
The effect of the alkyl group on the W–N bond can be clearly
seen from the calculated data, i.e. the bond length increases upon
the size of the substituents. The most significant result from the
theoretical study is the notable shortening of the W–N bond in the
Synthesis and structural characterization of imidotungsten(VI)
aminophenolate 4
The aminophenols 3a–3e (Scheme 1) were prepared apply-
ing a simple solvent-free Mannich reaction starting from
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2868–2874 | 2869
©

View Article Online
◦
˚
Table 2 Selected bond lengths (A) and angles ( ) for 4 and 5
2,4-di-tert-butylphenol, primary or secondary amine and
paraformaldehyde.7 All new aminophenols are colourless, crys-
talline solids at room temperature and can be isolated in 50–80%
yields.
4
5
W1–N1
W1–N8
W1–O1
W1–Cl1
W1–Cl2
W1–Cl3
N1–W1–N8
W1–O1–C1
N1–W1–O1
1.723(7)
2.339(8)
1.870(6)
2.347(3)
2.371(2)
2.370(3)
171.9(4)
145.22(19)
98.6(3)
1.712(7)
2.329(8)
1.868(6)
2.359(2)
2.3232(9)
2.3630(10)
173.8(4)
143.1(5)
96.4(3)
To synthesize a new series of imidotungsten(VI) complexes with
the above mentioned aminophenolato ligands, we decided to use
a simple two-step one-pot reaction. The first step comprises a
standard procedure to prepare tungsten(VI) aryloxides, namely a
simple reaction of oxotungsten(VI) tetrachloride with a stoichio-
metric amount of phenolic ligand precursor in an appropriate
solvent.8 The second step involves the reaction of a tungsten oxo
compound with phenyl isocyanate in toluene, a reaction which
is known to produce corresponding metal imido complexes.9
As expected, the stoichiometric reaction of a phenolic ligand
precursor 3a with WOCl4 in CH2Cl2 led rapidly to the formation
of a dark red solution. When all the ligand had reacted (by TLC
analyses), the solvent and HCl co-product were evaporated to
afford the stable intermediate. This dark red compound reacted
further with phenyl isocyanate in PhMe at reflux temperature to
yield a purple imido complex 4 (see Scheme 2). Alternatively, 4
can be prepared starting with the known reaction10 of WOCl4
and phenyl isocyanate in heptane, followed by the phenolysis
reaction with 3a in CH2Cl2. However, the latter method is not very
convenient, as it involves a slow reaction of poorly soluble WOCl4.
Compound 4 is soluble in common organic solvents and it can be
easily isolated and purified by column chromatography as a deep
red solid material. As a solid, 4 is stable under ambient conditions,
although it decomposes slowly in solution. The compound was
observed to undergo a slow hydrolysis in CDCl3 due to the ambient
water vapor absorbed by the solvent, whereupon traces of the free
phenolic ligand appeared in the 1H spectrum after ca. 1 day from
dissolution and within a week were about 6% of the intensity of
the signals of the tungsten complex 4. This hydrolysis product
(i.e. aminophenol 3a) had equivalent chemical shifts for its two
methylene protons, as well as for two methyl groups in its isopropyl
moiety, since the amine side arm can rotate freely and its amino
nitrogen can undergo the usual umbrella inversion.
Fig. 3 The molecular structure of 4 with 30% probability.
As the solid-state structure of 4 suffers some conformational
disorder, the molecular structure was verified by careful NMR
measurements. The 1H and 13C NMR spectra of 4 in CDCl3
solution were consistent with the expected constitution and atom–
atom connectivity of the compound, and the NMR signals could
be straightforwardly assigned with the help of standard 2D NMR
experiments (COSY, HSQC, HMBC), excluding stereochemistry.
The NMR data are given in the Experimental section and Table 3
(cf. Fig. 4 for the atom numbering). The 1H chemical shifts of the
two methylene protons at C7, and also the 1H and 13C chemicals
shifts of the two methyl groups at C9, were seen to be non-
equivalent indicating that each pair is diastereotopic. Below, the
diastereotopic groups will be distinguished by labels x and y, label
x referring to the atom or group with a larger 1H chemical shift.
The coordinative bond between N8 and W1 seems to remain
intact in the CDCl3 solution. There is a NOE between o-H of the
phenylimido ligand and the protons of the 2-tBu group consistent
with the assumption of bidentate binding (which prevents rotation
of the tBu group away from the phenyl). Also, the proton–proton
coupling constants in the amino substituent are not averaged
but imply a reasonably fixed torsion angles about the bonds as
expected if the N8 is coordinated to tungsten. If the substituent
was not bound to W1, and free to rotate about the various bonds,
not only would the J-coupling constants be averaged but the
CH2 hydrogens and the two 9-methyl groups would be chemically
equivalent (assuming fast nitrogen inversion) and display identical
chemical shifts, or nearly so, contrary to observation.
Scheme 2 Formation of aminophenolate complex 4.
Compound 4 was crystallized from a toluene–hexane mixture
as a toluene hemi solvate 4·0.5PhMe. The molecular structure
consists of neutral mononuclear units, in which the central W(VI)
ion is surrounded by one terminal phenyl imido group, one biden-
tate aminophenolate and three chlorides (Fig. 3). As expected due
to the strong structural trans-effect of the imido nitrogen,11 the
neutral donor atom of an aminophenolate is situated trans to the
imido group. The coordination geometry and bonding parameters
(see Table 2) around the metal ion are comparable to those
found earlier for corresponding aminophenolate compounds.4a,5
However, as expected by theoretical calculations, the coordinative
˚
W–N bond length in 4 is rather short, 2.339 A. Related bond
˚
lengths are 2.451 and 2.452 A in 1 and 2, respectively.
2870 | Dalton Trans., 2011, 40, 2868–2874
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 3 Selected 1H chemical shifts, JH,H coupling constants and Nuclear Overhauser Effects (NOEs) for compound 4 in CDCl3 at 298 K. The NOE
enhancements (in %) were obtained by setting the integral of the inverted 1H signal to -100 (or -300 in case where the methyl signals were selected)
Proton
d (ppm)
J/Hz (coupl. partner)
NOEs (%)
5
7x
7y
NH
9
CH3,x
CH3,y
7.11
4.79
4.32
3.24
4.46
1.42
1.30
2.4 (3)
4-tBu(1.5), 7y(1.7), 7x(0.9)
-14.3 (7y), 9.0 (NH)
-14.3 (7x), 2.5 (NH)
9.0 (7x), 2.5 (7y), 1.4 (9)
6.7 (CH3,y), 6.6 (CH3,x), 1.4 (NH)
6.6 (9)
CH3,y(2.2), CH3,x(1.0), NH(1.1), 7y(13.1), 5(1.3)
CH3,y(1.4), CH3,x(2.1), NH (2.6), 7x(13.8), 5(2.7)
CH3,y(0.4), CH3,x(2.3), 7y(2.1), 9(2.4), 7x(1.0)
CH3,y(1.9), CH3,x(2.2), NH(2.2)
CH3,y(1.4), NH(1.7), 7y(1.4), 9(2.0), 7x(0.5)
CH3,x(1.5), 7y(1.2), 9(2.4), 7x(1.7)
6.7 (9)
Formation and structural characterization of a Schiff-base
complex 5
In order to prepare aminophenolate complexes with elongated
coordinative W–N bonds, we used selected ligand precursors with
bulky substituents in amino groups (Scheme 3).
Fig. 4 Conformations A and B of compound 4. The former corresponds
to the X-ray structure of 4 while the latter, obtained by ring-inversion of
the C1–C6–C7–N8–W1–O1 ring, represents the preferred conformation
in CDCl3 solution as deduced by NMR spectroscopy. Relevant NOE
correlations, indicating spatial proximity, are shown as double-headed
arrows.
Scheme 3 Formation of iminophenol complex 5.
The two-step reaction of WOCl4 with a diethylaminophenol 3b
was carried out by following the same procedure outlined above
for the synthesis of 4. However, these experimental conditions did
not produce the expected aminophenol complex, but the reaction
led to the formation of a deep purple solution, from which an
intense purple compound 5 was obtained as the only isolable
product. The solid compound is stable under ambient atmosphere.
It is also stable in a CDCl3 solution, but decomposes slowly in
coordinating solvents. Compound 5 was characterized by NMR
spectroscopy and X-ray diffraction to be an unexpected Schiff-
As the first assumption, the structure of 4 in a CDCl3 solution
might be expected to resemble the solid-state structure (confor-
mation A in Fig. 4); however, the observed J-coupling constants
and NOEs involving the diastereotopic protons do not fit well to
conformation A regardless of which way the groups x and y are
assigned to the stereopositions. For example, the large observed J-
coupling constant between the NH and 7x protons (9.0 Hz) implies
them to be either in syn- or antiperiplanar positions with respect
to each other (according to the Karplus equation).12 Of these,
only the latter is consistent with the NOEs between NH and the
C7 protons (Table 3), but there is no C7-hydrogen antiperiplanar
to NH in conformation A. On the other hand, these and the
other J and NOE parameters of 4 fit well to the conformation
B (see Fig. 4 for graphical representation of the NOEs), which is
obtained from A by a ring-inversion of the six-membered C1–C6–
C7–N8–W1–O1 ring. It is likely that in CDCl3 solution several
conformers are populated, including some close to A, but the
preferred conformation(s) should resemble the structure B.
1
base complex. The H NMR spectrum showed resonances for
12 aromatic protons as well as signals for two tert-butyl groups
in the aliphatic region. A sharp one-proton singlet was found at
8.50 ppm, which is typical for imine N CH group. Also, the
13C NMR resonance at 167.60 ppm indicates the presence of
a sp2 hybridized imine carbon. X-Ray crystallography verified
that compound 5 is a rare example of an imidotungsten(VI)
complex with Schiff base imino ligands.13,14 The structure is made
up of mononuclear units, in which the central W(VI) ion is
surrounded by one terminal phenyl imido group, one bidentate
iminophenolate ligand and three chlorides. In general, complex 5
has a rather similar coordination sphere around the metal centre
than the aminophenolate 4 (see Table 2 and Fig. 5). Yet again,
the neutral imine nitrogen is bonded trans to the phenylimido
The 15N NMR chemical shifts were obtained indirectly from
15
1H{ N}-HSQC and -HMBC 2D spectra; the former, being a
group, the N–W N angle being 175.91(9) A. The NMR data
˚
single-bond correlation experiment, only showed the expected
amino-type NH nitrogen at -316.2 ppm whereas in the (long-
range) HMBC spectrum also the imino nitrogen at 31.3 ppm
was seen. These 15N chemical shifts, relative to nitromethane, are
consistent with sp3 and sp2 hybridizations, respectively.
indicate that the solid state structure is retained in CDCl3 solution
as the coordinative bond between N8 and W1 remains intact.
As estimated by computational results (entry 9 in Table 1), the
˚
1
coordinative W–N bond length in 5 is relatively long, 2.327(3) A,
t
˚
whereas related bond lengths are 2.308(11) A in [W(N Bu)2L ]
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2868–2874 | 2871
©

View Article Online
found in this complex is formed in the reaction of a coordinated
aminophenolate with an imidotungsten group.
Experimental
Materials and methods
WOCl4 and W(NPh)Cl4 were prepared according to known
procedures.17,10 Solvents used in syntheses were freshly distilled
over CaH2 prior to use. Other chemicals were obtained from
commercial sources and used without further purification. The
NMR spectra were recorded in CDCl3 solutions at 25 ◦C
with a Bruker Avance 400 spectrometer (1H: 399.75 MHz, 13C:
100.52 MHz) (expect for 4). The NMR spectra of 4 were recorded
on a Bruker Avance 500 spectrometer equipped with a broad-band
observe probe (BBO-5 mm-Zgrad), with the probe temperature set
to 298 K (without calibration). The proton chemical shifts were
referenced to internal tetramethylsilane (dTMS = 0.00 ppm), while
the reference frequencies for the 13C and 15N chemical shifts were
1
calculated from the TMS H frequency by using N13C(TMS) =
25.145020 and N15N(CH3NO2) = 10.136767.18 The NOE spectra
Fig. 5 The molecular structure of 5 with 30% probability.
were measured with the 1D DPFGSE-NOE sequence19 by using a
13
mixing time of 0.3 s. The 1H{ C} HSQC and HMBC experiments
(L1 is a dianion of N,N-bis(3,5-di-tert-butylsalicylidene)ethane-
1,2-diamine, the neutral imine nitrogen is trans to the imido
1
were optimized for JCH = 145 Hz and JCH(long-range) = 10 Hz,
15
2
2
and the corresponding { N} analogues for 90 Hz and 8 Hz. The
˚
dianion) and 2.234(8) A in [WO2(L )(MeOH)]·MeOH (L = 2-
(salicylideneamino)-5-t-butylphenolate dianion, the neutral imine
nitrogen is trans to the oxo group).13,15
1H chemical shifts and J-coupling constants were extracted by
simulation and iteration of the proton spectrum with the PERCH
NMR software.20 All samples were kept in vacuum for four hours
prior to the NMR analyses.
When the above mentioned reaction was repeated using
aminophenols 3c–3e as ligand precursors, complex 5 was formed in
30–35% yield (based on tungsten) in all individual experiments.16
The Schiff-base complex 5 is also formed in the reaction of
W(NPh)Cl4 with ligand precursors 3b–3e in refluxing toluene,
which points out that phenyl isocyanate is not required in the
formation of the new Schiff-base ligand, but the imino group
in this new ligand comes from a W NPh functionality. The
mechanism of this reaction is so far unclear, though formally
it appears a disproportionation of an intermediate compound,
which is analogous to complex 4. The weak coordination ability
of the amino group in phenols 3b–3e has obviously some role in
this reactivity. To have more evidence on this disproportionation,
we investigated a thermal decomposition of well-characterized
aminophenolato complexes. Isolated compounds 2 and 4 are stable
in toluene under reflux temperature, but if they are heated in
Preparation of the ligands. The ligand precursors 3a–3e were
prepared applying a simple solvent-free Mannich reaction.9 The
starting materials 2,4-di-tert-butylphenol (20 mmol), substituted
amine (22 mmol) and paraformaldehyde (22 mmol) were mixed in
an open vessel and the reaction mixtures were heated at 120 ◦C for
16 h, cooled to the room temperature and treated with methanol
to obtain white crystalline solids. Secondary amine 3a was isolated
from an etheral solution as a hydrochloride, neutralized by
aqueous NaHCO3 solution and re-crystallized from hexane prior
to use. All synthesised aminophenols were characterized by NMR.
2,4-Di-tert-butyl-6-((isopropylamino)methyl)phenol (3a). 3.0 g
(54%). 1H NMR (CDCl3): d 7.23 (s, 1H, ArH), 6.88 (s, 1H, ArH),
3.95 (s, 2H, ArCH2), 2.92 (septet, J = 6.3 Hz, 1H, CH(CH3)2),
1.44 (s, 9H, 2-tBu), 1.30 (s, 9H, 4-tBu), 1.17 (d, J = 6.3 Hz, 6H,
CH(CH3)2). 13C NMR (CDCl3): d 154.2 (C1), 140.2 (C4), 137.2
(C2), 123.2 (C5), 122.2 (C3), 121.0 (C6), 50.2 (C9), 46.1 (C7), 34.8
(1C, 2-tBu), 34.7 (1C, 4-tBu), 31.8 (3C, 4-tBu), 30.0 (3C, 2-tBu),
20.7 (2C, 9-CH3).
◦
1,2-dichlorobenzene at 150 C, the above mentioned Schiff-base
complex 5 is formed in 35% isolated yield. The higher stability
of these complexes is consistent with the observed and calculated
coordinative W–N distances.
Conclusions
2,4-Di-tert-butyl-6-((diethylamino)methyl)phenol (3b). 3.6 g
(62%). 1H NMR (CDCl3): d 7.20 (s, 1H, ArH), 6.96 (s, 1H, ArH),
3.65 (s, 2H, ArCH2), 2.52 (q, 4H, m, CH2CH3), 1.44 (s, 9H, 2-tBu),
1.31 (s, 9H, 4-tBu), 1.02 (t, 6H, CH2CH3). 13C NMR (CDCl3): d
154.7, 141.1, 137.0, 123.5, 122.5, 120.2, 49.0, 45.9, 34.8, 34.2, 34.2,
33.1, 17.3.
WOCl4 reacts with aminophenol ligand 2,4-di-tert-butyl-6-
((isopropylamino)methyl)phenol and phenyl isocyanate in a
two-step reaction to produce an imidotungsten(VI) complex
[W(NPh)Cl3(OC6H3(CH2NH-i-Pr)-2-t-Bu2-4,6)] 4. When the
same procedure was applied using tertiary aminophenols with
bulkier substituents in the amino group, the final product is
a Schiff-base complex [W(NPh)Cl3(OC6H3(CH NPh)-2-t-Bu2-
4,6)] 5. Complex 5 is also formed in the thermal degradation of
4. The 2,4-di-tert-butyl-6-((phenylimino)methyl)phenolate ligand
2,4-Di-tert-butyl-6-((diisopropylamino)methyl)phenol
(3c).
4.8 g (75%). 1H NMR (CDCl3): d 7.20 (s, 1H, ArH), 6.86
(s, 1H, ArH), 3.86 (s, 2H, ArCH2), 3.18 ((septet, J = 6.5 Hz,
2H, CH(CH3)2), 1.43 (s, 9H, 2-tBu), 1.39 (s, 9H, 4-tBu), 1.14
2872 | Dalton Trans., 2011, 40, 2868–2874
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
(d, J = 5.4 Hz, 12H, CH(CH3)2). 13C NMR (CDCl3): d 155.1,
140.0, 135.2, 123.2, 122.3, 121.7, 49.3, 47.5, 34.9, 34.1, 31.8, 29.5,
19.8.
130.5, 130.0, 128.2, 127.3, 124.1, 123.5, 35.5 (1C, 2-tBu), 34.9
(1C, 4-tBu), 31.4 (3C, 4-tBu), 30.1 (3C, 2-tBu). Anal. calcd for
C27H31Cl3N2OW: C 47.02, H 4.53, N 4.06; found C 46.92, H 4.41,
N 4.10.
2,4-Di-tert-butyl-6-((methyl(phenyl)amino)methyl)phenol (3d).
4.5 g (69%). 1H NMR (CDCl3): d 7.36 (t, J = 7.2 Hz, 2H, ArH),
7.28 (s, 1H, ArH), 7.21 (d, J = 6.4 Hz, 2H, ArH), 7.06 (t, J =
7.2 Hz, 1H, ArH), 6.94 (s, 1H, ArH), 4.35 (s, 2H, ArCH2), 2.78
(s, 3H, NCH3) 1.45 (s, 9H, 2-tBu), 1.32 (s, 9H, 4-tBu). 13C NMR
(CDCl3): d 153.9, 151.2, 141.2, 135.9, 129.2, 123.6, 123.2, 122.5,
121.3, 119.2, 60.3, 39.8, 34.9, 34.2, 31.7, 29.7.
Degradation of aminophenolate complexes 2 and 4
0.15 mmol (100 mg) samples of 2 and 4 were dissolved in 5 ml of
1,2-dichlorobentsene in thick-walled screw cap vials and heated
at 150 ◦C for 16 h. Reactions produced corresponding Schiff-base
complex 5, which was isolated by column chromatography in ca.
35% yields.
2,4-Di-tert-butyl-6-(piperidin-1-ylmethyl)phenol (3e). 4.2
g
(69%). 1H NMR (CDCl3): d 7.23 (s, 1H, ArH), 6.84 (s, 1H, ArH),
3.65 (s, 2H, ArCH2), 2.60 (s, br, 4H, ring-CH2N), 1.63 (s, br, 4H,
ring-CH2), 1.46 (s, br, 2H, ring-CH2), 1.43 (s, 9H, 2-tBu), 1.30 (s,
9H, 4-tBu). 13C NMR (CDCl3): d 154.5, 140.2, 135.4, 123.3, 122.7,
121.1, 62.9, 53.7, 34.9, 34.1, 31.7, 29.6, 25.8, 24.1. Spectra were
identical to those reported earlier for 3e.21
Crystal structure determinations†. The single crystals of 4
were grown from a toluene–hexane mixture and the crystals of
5 were grown from acetonitrile. The crystallographic data of both
compounds were collected at 123 K on an Enraf Nonius Kappa
CCD area-detector diffractometer using graphite monochroma-
˚
tised Mo-Ka radiation (l = 0.71073 A). Data collection was
Preparation of 4. Method A: WOCl4 (340 mg, 1.0 mmol)
was treated with a solution of a ligand precursor 3a (277 mg,
1.0 mmol) in CH2Cl2 (20 ml) under a N2 atmosphere. The intense
red reaction mixture was allowed to reflux while the reaction was
monitored by TLC (PhMe as an eluent). After two hours, the TLC
analysis showed a single purple spot (Rf 0.9), which discolours in
a few minutes in air. The volatiles were then removed in vacuum
and remaining dark solid was dissolved in 10 ml of toluene.
PhNCO (0.13 ml, 1.2 mmol) was added and the mixture was
refluxed for 16 h. Intense red product was isolated by silica column
chromatography using PhMe as an eluent and finally crystallized
from hot hexane to obtain 4 in a 55% (360 mg) total yield. Method
B: 125 mg (0.30 mmol) of W(NPh)Cl4 and 83 mg (0.30 mmol) of
3a were mixed with 5 ml of toluene and stirred under reflux for
four hou◦rs. Product 4 was isolated in a 175 mg (88%) yield, mp
145–147 C. 1H NMR (500.13 MHz, see Fig. 4 for the numbering
scheme): d 7.61 (m, 2H, m-H), 7.42 (d, J = 2.4 Hz, 1H, 3-H), 7.37
(m, 2H, o-H), 7.11 (d, J = 2.4 Hz, 1H, 5-H), 7.05 (m, 1H, p-H),
4.79 (dd, J = 9.0, 14.3 Hz, 1H, 7-H(x)), 4.46 (septet of doublets,
J = 1.4, 6.6(¥3), 6.7(¥3) Hz, 1H, 9-H), 4.32 (dd, J = 2.5, 14.3 Hz,
1H, 7-H(y)), 3.24 (br d, J ª 8.5 Hz, 1H, NH), 1.44 (s, 9H, 2-
tBu), 1.42 (d, J = 6.6 Hz, 3H, 9-CH3(x)), 1.33 (s, 9H, 4-tBu), 1.30
(d, J = 6.7 Hz, 3H, 9-CH3(y)). 13C NMR (125.76 MHz, CDCl3):
d 156.1 (C1), 150.9 (i-C), 150.6 (C4), 141.6 (C2), 131.6 (p-C),
130.1 (2C, o-C), 129.4 (C6), 127.6 (2C, m-C), 124.2 (C5), 123.7
(C3), 51.0 (C9), 44.0 (C7), 35.4 (1C, 2-tBu), 34.9 (1C, 4-tBu), 31.5
(3C, 4-tBu), 30.6 (3C, 2-tBu), 22.9 (9-CH3(x)), 18.3 (9-CH3(y)).
15N NMR (50.70 MHz, CDCl3): -316.2 (N2), 31.3 (W NPh).
Elemental analyses were run for the toluene hemi solvate crystals
used in X-ray measurements. Anal. calcd for C27.5H39Cl3N2OW: C
46.93, H 5.59, N 3.98; found C 47.26, H 5.42, N 4.07.
performed using j and w scans and the data were processed using
DENZO-SMN v0.97.638.22 SADABS absorption correction was
applied to the data of both compounds.23 The structures were
solved by direct methods using the SHELXS-97 program and
full-matrix least-squares refinements on F2 were performed using
the SHELXL-97 program.24 Structure figures were drawn using
Ortep-3.25
Crystal data for 4. C27.5H39Cl3N2OW, M = 703.8, monoclinic,
˚
a = 10.15330(40), b = 21.6753(10), c = 13.67660(50) A, b =
◦
3
˚
102.8943(25) , V = 2933.93(21) A , space group P21/n (no. 14),
Z = 4, m(Mo-Ka) = 0.423 mm-1, 13 972 reflections measured,
5083 unique (Rint = 0.0414), which were used in all calculations.
Multi-scan absorption correction made. Tmin and Tmax 0.5036 and
0.7457, respectively. Final R1 = 0.0546, wR2 = 0.1225, GOF =
1.36. Minimum and maximum residual electron density 2.189 and
-3
˚
-1.543 e A .
Crystal data for 5. C27H31Cl3N2OW, M = 689.74, monoclinic,
˚
a = 13.77980(20), b = 14.20330(29), c = 14.75650(20) A, b =
◦
3
˚
105.2619(10) , V = 2786.266(81) A , space group P21/n (no. 14),
Z = 4, m(Mo-Ka) = 0.446 mm-1, 18 167 reflections measured,
5464 unique (Rint = 0.0310), which were used in all calculations.
Multi-scan absorption correction made. Tmin and Tmax 0.5346 and
0.7458, respectively. Final R1 = 0.0228, wR2 = 0.0507, GOF =
1.067. Minimum and maximum residual electron density 0.568
-3
˚
and -0.594 e A .
Computational details
All calculations were performed with Turbomole 6.1 program
package.26 The geometries of the complexes were optimized at
the DFT level of theory using the exchange functional of Becke27
in conjunction with correlation functional of Perdew28 (BP86).
The def2-TZVP basis set29 was used in all calculations. The
nature of the located stationary points on the potential-energy
surface was confirmed by calculation of the two lowest eigenvalues
of hessian matrix. Partial atomic charges were calculated using
MPA30 and NPA31 analyses as implemented in Turbomole 6.1
program package.
Preparation of 5. The experimental procedures described
above for the synthesis and isolation of 4 was repeated using
aminophenols 3b–3e as ligand precursors. Dark purple 5 was
isolated in a 30–35% yield in all experiment◦s. Solid compound was
crystallized from acetonitrile, mp 195–198 C. 1H NMR (CDCl3):
d 8.50 (s, 1H, –N CH–), 7.82 (s, 1H, ArH), 7.63 (t, J = 8.3 Hz,
2H, ArH), 7.47 (s, 1H, ArH), 7.31–7.40 (overlapping signals, 7H,
ArH), 1.50 (s, 9H, 2-tBu), 1.38 (s, 9H, 4-tBu). 13C NMR (CDCl3): d
167.6 (–N CH–), 154.6, 152.2, 151.1, 150.0, 142.2, 131.6, 131.4,
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2868–2874 | 2873
©

View Article Online
11 B. J. Coe and S. J. Glenwright, Coord. Chem. Rev., 2000, 203, 5.
References
12 M. J. Minch, Concepts Magn. Reson., 1994, 6, 41.
13 M. Motevalli, B. C. Parkin, R. Ramnauth and A. C. Sullivan, J. Chem.
Soc., Dalton Trans., 2000, 2661.
1 K. J. Ivin and J. C. Mol, Olefin metathesis and metathesis polymerization,
Academic press, San Diego, 1997.
2 For recent reviews, see for example a special issue of Adv. Synth. Catal.,
2007, 349, 1.
14 K. Dreisch, C. Andersson and C. Sta˚lhandske, Polyhedron, 1993, 12,
1335.
3 V. Dragutan, I. Dragutan and M. Dimonie, in Green Metathe-
sis Chemistry, ed. V. Dragutan, A. Demonceau, I. Dragutan and
E. S. Finkelshtein, Springer, Netherlands, 2010, p. 369.
4 (a) P. A. van der Schaaf, J. Boersma, W. J. J. Smeets, A. L. Spek and G.
van Koten, Inorg. Chem., 1993, 32, 5108; (b) P. A. van der Schaaf, D. M.
Grove, W. J. J. Smeets, A. L. Spek and G. van Koten, Organometallics,
1993, 12, 3955; (c) P. A. van der Schaaf, R. A. T. M. Abbenhuis, W. P. A.
van der Noorth, R. de Graaf, D. M. Grove, W. J. J. Smeets, A. L. Spek
and G. van Koten, Organometallics, 1994, 13, 1433.
5 A. Lehtonen, H. Balcar, J. Sedla´cˇek and R. Sillanpa¨a¨, J. Organomet.
Chem., 2008, 693, 1171.
6 K. C. Gross, P. G. Seybold and C. M. Hadad, Int. J. Quantum Chem.,
2002, 90, 445.
15 S. Yu and R. H. Holm, Inorg. Chem., 1989, 28, 4385.
16 Complex 5 can be also prepared in practically quantitative yield by
the reaction of W(NPh)Cl4 with an appropriate Schiff base ligand,
2,4-di-tert-butyl-6-((phenylimino)methyl)phenol, in toluene at reflux
temperature.
17 R. H. Crabtree and G. G. Hlatky, Polyhedron, 1985, 4, 521.
18 R. K. Harris, E. D. Becker, S. M. Cabral, De Menezes, R. Goodfellow
and P. Granger, Pure Appl. Chem., 2001, 73, 1795.
19 K. Stott, J. Stonehouse, T. L. Hwang, J. Keeler and A. J. Shaka, J. Am.
Chem. Soc., 1995, 117, 4199.
20 R. Laatikainen, M. Niemitz, U. Weber, J. Sundelin, T. Hassinen and J.
Vepsa¨la¨inen, J. Magn. Reson., Ser. A, 1996, 120, 1 Software homepage:
http://www.perchsolutions.com.
7 A. Riisio¨, O. Wichmann and R. Sillanpa¨a¨, Lett. Org. Chem., 2010, 7,
298.
21 R. K. J. Bott, D. L. Hughes, M. Schormann, M. Bochmann and S. J.
Lancaster, J. Organomet. Chem., 2003, 665, 135.
8 For example, see: (a) A. Bell, J. Mol. Catal., 1992, 76, 165; (b) W.
Nugent, J. Feldman and J. C. Calabrese, J. Am. Chem. Soc., 1995,
117, 8992; (c) F. J. Go´mez, M. S. Manak, K. A. Abboud and K. B.
Wagener, J. Mol. Catal. A: Chem., 2000, 160, 145; (d) A. Lehtonen and
R. Sillanpa¨a¨, Polyhedron, 2002, 21, 349.
9 For example, see: (a) P. Legzdins, E. C. Phillips, S. J. Rettig, J. Trotter,
J. E. Veltheer and V. C. Yee, Organometallics, 1992, 11, 3104; (b) J. C.
Bryan, A. K. Burrell, M. M. Miller, W. H. Smith, C. J. Burns and A. P.
Sattelberger, Polyhedron, 1993, 12, 1769; (c) R. I. Michelman, R. A.
Andersen and R. G. Bergman, J. Am. Chem. Soc., 1991, 113, 5100;
(d) S. R. Dubberley, A. Friedrich, D. A. Willman, P. Mountford and U.
Radius, Chem.–Eur. J., 2003, 9, 3634.
22 Z. Otwinowski and W. Minor, Methods Enzymol., 276, 307.
23 G. M. Sheldrick, SADABS., University, of Go¨ttingen, Germany, 2002.
24 G. M. Sheldrick, SHELXS-97 and SHELXL-97. Programs for Refine-
ment of Crystal Structures, University of Go¨ttingen, Germany, 1997.
25 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
26 R. Ahlrichs, M. Ba¨r, M. Ha¨ser, H. Horn and C. Ko¨lmel, Chem. Phys.
Lett., 1989, 162, 165.
27 A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 1988, 38, 3098.
28 (a) J. P. Perdew, Phys. Rev. B, 1986, 34, 7406; (b) J. P. Perdew, Phys. Rev.
B, 1986, 33, 8822.
29 F. Weigend and R. Ahlrichs;, Phys. Chem. Chem. Phys., 2005, 7, 3297.
30 R. S. Mulliken, J. Chem. Phys., 1955, 23, 1833.
10 D. C. Bradley, M. B. Hursthouse, K. M. Abdul Malik, A. J. Nielson
and R. L. Short, J. Chem. Soc., Dalton Trans., 1983, 2651.
31 A. E. Reed, R. B. Weinstock and F. Weinhold, J. Chem. Phys., 1985,
83(2), 735.
2874 | Dalton Trans., 2011, 40, 2868–2874
This journal is
The Royal Society of Chemistry 2011
©