Hepta-coordinate halocarbonyl molybdenum(II) and tungsten(II) complexes as heterogeneous polymerization catalysts

Article abstract of DOI:10.1016/j.molcata.2006.04.023

[MX₂(CO)₃(DAB)] (M = Mo, W; X = I, Br) (5-8) complexes bearing the 1,4-diazobutadiene (DAB) ligand RN{double bond, long}C(Ph){single bond}C(Ph){double bond, long}NR [R = (CH₂)₃Si(OEt)₃] were immobilized in MCM-41 mesoporous silica. The tethering, stepwise procedure started with treatment of the MCM-41 mesoporous material with a toluene solution of the 1,4-diazobutadiene ligand, under reflux. The molybdenum and tungsten organometallic cores were subsequently introduced into the ligand-silicas by pore volume impregnation of a solution of the complexes [MX₂(CO)₃(NCMe)₂] (M = Mo, W; X = Br, I). The modified materials were extensively characterized by several techniques, such as FTIR, solid-state MAS and CP MAS NMR (¹³C, ²⁹Si), powder XRD, and nitrogen adsorption-desorption measurements. These new materials (containing 2.6-2.9 wt.% Mo or 0.4-0.6 wt.% W) catalyze the ring-opening metathesis polymerization (ROMP) of norbornene (NBE) and norbornadiene at 328 K, in contrast with the very low activity exhibited by the precursor complexes and with their behavior at lower temperature. Addition of AlCl₃ as a co-catalyst enhanced the catalytic performance of the material MCM-DAB-MoBr₂ (12) in the ROMP of NBE.

Full text of DOI:10.1016/j.molcata.2006.04.023

Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
Hepta-coordinate halocarbonyl molybdenum(II) and tungsten(II)
complexes as heterogeneous polymerization catalysts
Jorge Gimeneza, Carla D. Nunesa, Pedro D. Vaza, Anabela A. Valenteb,
c
a,∗
´
Paula Ferreira , Maria Jose Calhorda
a Department of Chemistry and Biochemistry, CQB, Faculty of Science, University of Lisbon, C8 Campo Grande, 1749-016 Lisboa, Portugal
b Department of Chemistry, CICECO, University of Aveiro, Campus de Santiago, 3810-193 Aveiro, Portugal
c Department of Ceramics and Glass Engineering, CICECO, University of Aveiro, Campus de Santiago, 3810-193 Aveiro, Portugal
Received 27 February 2006; received in revised form 6 April 2006; accepted 8 April 2006
Available online 5 June 2006
Abstract
[MX2(CO)3(DAB)] (M = Mo, W; X = I, Br) (5–8) complexes bearing the 1,4-diazobutadiene (DAB) ligand RN C(Ph) C(Ph) NR
[R = (CH2)3Si(OEt)3] were immobilized in MCM-41 mesoporous silica. The tethering, stepwise procedure started with treatment of the MCM-41
mesoporous material with a toluene solution of the 1,4-diazobutadiene ligand, under reflux. The molybdenum and tungsten organometallic cores
were subsequently introduced into the ligand-silicas by pore volume impregnation of a solution of the complexes [MX2(CO)3(NCMe)2] (M = Mo,
W; X = Br, I). The modified materials were extensively characterized by several techniques, such as FTIR, solid-state MAS and CP MAS NMR
(13C, 29Si), powder XRD, and nitrogen adsorption–desorption measurements. These new materials (containing 2.6–2.9 wt.% Mo or 0.4–0.6 wt.%
W) catalyze the ring-opening metathesis polymerization (ROMP) of norbornene (NBE) and norbornadiene at 328 K, in contrast with the very
low activity exhibited by the precursor complexes and with their behavior at lower temperature. Addition of AlCl3 as a co-catalyst enhanced the
catalytic performance of the material MCM-DAB-MoBr2 (12) in the ROMP of NBE.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Molybdenum; Tungsten; MCM-41; Polymerization; DAB ligands; Hepta-coordination; Catalysis
1. Introduction
[M(CO)3X2(NCMe)2] have been successfully used by Baker
and co-workers in the ring-opening metathesis polymeriza-
tion (ROMP) of norbornene (NBE) and norbornadiene (NBD)
[15,16] with satisfactory results, though some of the complexes
were only active upon the addition of a Lewis acid, such as
Organometallic complexes can efficiently and selectively cat-
alyze many reactions and are widely applied in industrial pro-
cesses. One drawback of such homogeneous catalysts is the
difficulty of separating the products from the reaction solu-
tion and recovering and recycling the catalyst. Therefore, the
possibility of incorporating complexes in materials to yield
heterogeneous catalysts has opened new possibilities in recent
years [1,2]. Seven-coordinate Mo(II) complexes [MX2(CO)3L2]
(M = Mo, W; X = halogen) and their derivatives, obtained by
replacing the labile nitrile ligands, have been extensively inves-
tigated [3–5]. Their synthesis [6–9], structures [10,11] and
properties have been studied since they were reported for the
first time by Nigam and Nyholm in 1957 [12], and they have
been shown to exhibit catalytic activity [13,14]. Complexes
´
AlCl3 or ZrCl4. Szymanska-Buzar et al. carried out the study of
ROMP reactions with seven-coordinate Mo(II) and W(II) com-
plexes [17–19].
The micelle templated silicas MCM-41 (hereafter denoted
as MCM) belong to the M41S family developed by Mobil Cor-
poration in 1992 [20,21], and display a set of properties, such
as a stable and ordered mesoporous structure, large surface
area (usually >1000 m2 g−1), and narrow pore size distribu-
tion, which make them ideal for hosting molecules of various
sizes, shapes and functionalities. These features are suitable
for the inclusion of organometallic complexes and have led to
significant developments, among others, in the fields of catal-
ysis, adsorption/sorption, separation, sensors, optically active
materials [1,22]. The number of papers describing this type of
chemistry has been growing each year [1,2,23].
∗
Corresponding author. Tel.: +351 217500196; fax: +351 217500088.
E-mail address: mjc@fc.ul.pt (M.J. Calhorda).
1381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.04.023

J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
91
In the present work, bis(acetonitrile) Mo(II) and W(II)
complexes were used as precursors for the immobilization of
organometallic species in MCM, by a tethering approach. In
thefirststep, thetriethoxysilyl-N-substituted1,4-diazobutadiene
(DAB) ligand was grafted on the host material MCM-41, and
was subsequently allowed to react with the organometallic
complexes [MX2(CO)3(NCMe)2] (M = Mo, W; X = Br, I). The
derivatized materials were fully characterized and tested as cat-
alysts for the ring-opening metathesis polymerization (ROMP)
of norbornadiene (NBD) and norbornene (NBE).
of benzil [C6H5(CO)(CO)C6H5] with two equivalents of (3-
aminopropyl)triethoxysilane, according to a well-known pro-
cedure [24]. The precursor complexes [MX2(CO)3(NCMe)2]
(M = Mo, W; X = I or Br) (1–4) [25] were allowed to react with
one equivalent of the DAB ligand, affording the complexes
[MX2(CO)3{Ph-DAB-CH2Si(OEt)3}2] (M = Mo, W; X = I or
Br) (5–8) by displacement of both labile acetonitrile ligands and
in good yield. These complexes were subsequently immobilized
on the MCM-41 host material, as outlined in Scheme 1 [24,26].
Both the ligand DAB [24] and the precursor complexes
1–4 [25] have been characterized elsewhere and therefore
will not be discussed here. The 1H NMR spectrum of
[MoI2(CO)3{Ph-DAB-CH2Si(OEt)3}2] (5) shows a multiplet
in the range 7.36 < δ < 7.95 ppm assigned to the phenyl groups,
as well as multiplet signals at δ = 3.47–3.52 ppm (NCH2),
1.93–1.98 ppm (CH2CH2CH2) and a broad singlet at 0.71 ppm
(SiCH2) for the propyl chains. The presence of ethoxy groups
2. Results and discussion
2.1. The organometallic complexes
The functionalized triethoxysilyl ligand RN C(Ph) C(Ph)
NR (DAB) [R (CH2)3Si(OEt)3] was prepared by the reaction
Scheme 1.

92
J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
Table 1
FTIR wavenumbers of νC≡O, νC≡N and νC modes for complexes 1, 2, 5, and 6
N
Complex
νC≡O (cm−1
)
νC≡N (cm−1
)
νC (cm−1
)
N
[MoI2(CO)3(NCMe)2] (1)
[MoI2(CO)3(DAB)] (5)
[MoBr2(CO)3(NCMe)2] (2)
[MoBr2(CO)3(DAB)] (6)
1921 (vs), 2016 (vs), 2072 (m)
1946 (m), 2017 (m), 2067 (vw)
1943 (vs), 2021 (vs), 2091 (vw)
1930 (vs), 2011 (s), 2063 (vw)
2276 (vs), 2304 (vs)
–
2279 (s), 2309 (m)
–
–
1637 (s)
–
1635 (s)
is reflected by signals at δ = 3.47–3.52 ppm (OCH2, convo-
luted with NCH2) and 1.04 ppm (CH3). These signals are
slightly shifted relative to those of the free ligand DAB, which
exhibits three multiplets at δ = 7.32–7.50 ppm, 7.57–7.75 ppm,
and 7.88–7.93 ppm (aromatic protons), and resonances at
δ = 3.40–3.44 ppm, 1.76–1.86 ppm and 0.60–0.68 ppm assigned
to the NCH2, CH2CH2CH2, and SiCH2, respectively; the ethoxy
groups peaks are observed at δ = 3.73–3.84 ppm (OCH2CH3)
and 1.15–1.24 ppm (OCH2CH3) [24].
The absence of resonances from the acetonitrile ligands indi-
cates the coordination of the ligand to the molybdenum centre.
This is confirmed by the 13C NMR spectrum, displaying one
resonance at δ = 166.7 ppm, due to the N C atoms, and signals
at δ = 128.9 ppm, 128.0 ppm and 127.2 ppm, assigned to the aro-
matic carbon atoms. Signals from the propyl chains appear at
δ = 68.7 ppm (NCH2), 21.7 ppm (CH2CH2CH2) and 13.1 ppm
(SiCH2). The ethoxy groups have resonances at δ = 58.1 ppm
(OCH2) and 20.4 ppm (CH3).
The FTIR spectrum of [MoI2(CO)3{Ph-DAB-CH2Si
(OEt)3}2] (5) shows the displacement of the labile acetonitrile
ligands during the synthesis, as evidenced by the absence of
the νC≡N stretching modes at ca. 2300 cm−1 (Table 1). A new
band at 1637 cm−1 is clearly observed arising from the new
νC N mode, and two broad and intense bands assigned to νsi o
stretching modes appear at 1116 cm−1 and 1024 cm−1. In the
free ligand DAB, the band corresponding to the νC N mode is
observed at 1647 cm−1. The shift of the νC N band to lower
frequency is indicative of the coordination of DAB to the com-
plex [MoI2(CO)3(MeCN)2] (1), with weakening of the C N
bond. For complex 6 [MoBr2(CO)3{Ph-DAB-CH2Si(OEt)3}2],
comparable changes in the vibrational spectrum occur, but the
νC N mode is shifted to lower wavenumbers than in 5. This
suggests a stronger M N bond of the metal centre to DAB in
the dibromo complex 6 than in the diiodo one 5. The bands
assigned to the νsi o stretching modes in complex 6 are observed
at 1167 cm−1 and 1075 cm−1. Some small changes are also
observed for the νC≡O bands upon chelation of the DAB ligand
to the bis(acetonitrile) Mo(II) complexes 1 and 2 and for the
corresponding W(II) complexes, as shown in Table 1. This
indicates that the formation of Mo N bonds affects to some
extent the strength of the Mo C bonds by changing the amount
of ␲-backdonation to the carbonyl ligands.
The spectroscopic characterization of all the complexes 5–8
by 1H, 13C NMR and FTIR led to similar results (see Section 4
for remaining complexes).
2.2. The functionalized materials
Two lots of samples of MCM-41 host material 9 or 9*
(* denotes a different lot) were prepared by a template approach
as described previously [24]. This material was then deriva-
tized by grafting of the phenyl-substituted 1,4-diazobutadiene
ligand DAB onto the silica-matrix mesoporous host material, in
toluene, overnight. After filtration, the solids were washed thor-
oughly with dichloromethane, and dried in vacuum at 100 ◦C
for several hours, originating materials 10 or 10*, respectively.
This procedure has been reported by I.S. Gonc¸alves and co-
workers to immobilize dioxomolybdenum(VI) species using the
ligand-silica system 10, which originated active catalysts for
the epoxidation of olefins [24,26]. In this work, 10 or 10* were
allowed to react with an excess of the bis(acetonitrile)complexes
(1–4) in methylene chloride, overnight, at room temperature, and
Fig. 1. Powder XRD patterns for materials MCM (9 and 9*), MCM-DAB (10 and 10*), MCM-DAB-MoI2 (11), MCM-DAB-MoBr2 (12), MCM-DAB-WI2 (13) and
MCM-DAB-WBr2 (14).

J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
93
Fig. 2. Nitrogen adsorption (ꢀ) (9 and 9*); (♦) (10 and 10*); (ꢀ) (11 and 13); (ꢀ) (12 and 14) and desorption (×) (9 and 9*); (*) (10 and 10*); (−) (11 and 13); (+)
(12 and 14) isotherms at 77 K.
after filtering and drying, afforded the composite materials 11,
12and13, 14, respectively. Inthecaseof 11and12, themolybde-
num loadings were 2.6 wt.% and 2.9 wt.% (0.27 mmol g−1 and
0.30 mmol g−1), respectively. For the tungsten containing mate-
rials 13 and 14, the results show metal loadings of only 0.4 wt.%
and 0.6 wt.% (0.05 mmol g−1 and 0.09 mmol g−1), respectively.
These results indicate that the immobilization of the tungsten
complexes was not very successful, although several attempts
were pursued.
The new materials were characterized by powder XRD
diffraction, adsorption studies, FTIR and solid-state 13C and
29Si NMR.
The powder XRD patterns for the process-control are shown
in Fig. 1. The powder pattern of the MCM parent, calcined
material 9, clearly shows four reflections in the 2θ range 2–10◦,
indexed to a hexagonal cell as (1 0 0), (1 1 0), (2 0 0) and (2 1 0).
tering contrast between the silica walls and pore-filling material,
a situation well described in the literature [27,28].
Nitrogen adsorption studies at 77 K revealed that the pristine
MCM samples (9 and 9*) exhibit reversible type IV isotherms
(Fig. 2), characteristicofmesoporoussolids(porewidthbetween
2 nm and 50 nm, according to the IUPAC) [29]. The calcu-
lated textural parameters (SBET and Vt) of these materials agree
with literature data (Table 2) [30,31]. The capillary condensa-
tion/evaporation steps of pristine MCM samples appear in the
relative pressures range 0.26–0.4. The sharpness of this step
reflects the uniform pore size.
The isotherms of the functionalized materials revealed much
lower N2 uptake, accounting for decreases in SBET (52–63%)
and Vt (67–76%). These results suggest that immobilization
of the complexes on the internal silica surface was accom-
plished (Fig. 2, Table 2). This conclusion is also supported by
the decrease of the p/p0 coordinates of the inflection points of
the isotherms upon post-synthesis treatments [32]. The height of
the capillary condensation step, which is related to the volume
of pore space confined by absorbate film on the pore walls, is
much smaller in the case of the modified MCM materials. Fur-
thermore, the maximum of the PSD curve for MCM determined
by the BJH method, dBJH, decreases from 3.7 nm to less than
3 nm (Table 2).
˚
˚
The d-value of the (1 0 0) reflection was 34.6 A (34.8 A for
*
˚
material 9 ), corresponding to a lattice constant of a = 40.0 A
(=2d100/ 3) (40.2 A for material 9 ). Upon functionalization
√
*
˚
of the walls of the parent host material 9 or 9* with the diazobu-
tadiene ligand DAB and subsequent inclusion of the metallic
complexes (1–4), the powder patterns remain unchanged in what
concerns the positions of the peaks assigned to the characteristic
reflections, suggesting the retention of the long range hexagonal
symmetry of the host material. However, a drastic reduction of
the peaks intensities is clearly observed. This is not interpreted as
a loss of crystallinity, but rather to a reduction in the X-ray scat-
The FTIR spectrum of the parent host material MCM 9
(Fig. 3) is similar to that of other mesoporous siliceous matrices
[27,28]. The bands at 1235 cm−1 and 1087 cm−1 are assigned
Table 2
Textural parameters for host and composite materials from N2 isotherms at 77 K
a
b
c
Sample
SBET (m2 g−1
)
ꢀSBET (%)
VP (cm3 g−1
)
ꢀVP (%)
dBJH (nm)
9
1106
526
411
467
981
438
494
420
–
52
63
58
–
55
50
57
0.83
0.27
0.20
0.23
0.76
0.20
0.23
0.19
–
67
76
72
–
74
70
75
3.7
2.9
2.9
2.9
3.7
2.7
2.5
2.6
10
11
12
9*
10*
13
14
a
Variation of surface area in relation to parent MCM.
b
c
*
Variation of total pore volume in relation to parent MCM.
Median pore width determined by the BJH method.
MCM from a different lot.

94
J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
Fig. 3. FTIR spectra of [MoI2(CO)3(NCMe)2] (1), MCM (9), MCM-DAB (10)
and MCM-DAB-MoI2 (11).
to the νasym mode of the Si O Si framework, the correspond-
ing νsym mode appearing at ca. 800 cm−1. An absorption at
963 cm−1 is assigned to a δSi O mode of the Si OH groups
[27,28]. The treatment of MCM 9 with DAB led to the deriva-
tized material 10, which presents new bands in the vibra-
Fig. 4. 29Si MAS and CP MAS NMR for materials MCM (9), MCM-DAB (10),
MCM-DAB-MoI2 (11) and MCM-DAB-MoBr2 (12).
tional spectrum relative to that of 9. The bands at 3067 cm−1
,
with absence of the νC≡N bands and the presence of bands
assigned to the carbonyls, suggesting that the W(CO)3 frag-
ments were also successfully immobilized inside the pores of
the MCM host material.
2978 cm−1, 2927 cm−1 and 2891 cm−1 are assigned to νC H
modes of the ligand. The band at 1652 cm−1 can be assigned
to the νC=N mode of the ligand (1647 cm−1 in free DAB),
suggesting that the ligand was immobilized inside the pores
of the host material without its structure being affected. The
βC H and νC C modes of the ligand are responsible for a set
of three bands at 1494 cm−1, 1447 cm−1 and 1390 cm−1. The
presence of the carbonyl ligands in the immobilized complexes
allows valuable information to be extracted from the vibrational
All the materials were also characterized by 29Si MAS and
CP MAS NMR and 13C CP MAS solid-state NMR spectroscopy.
As shown in Fig. 4, the 29Si MAS and CP MAS spectra present
some changes reflecting the evolution of the tethering procedure
during the preparation of the composite materials.
The spectra of material 9 show the presence of two convo-
spectra of the new materials, as the νC stretching mode is a
luted resonances assigned to Q3 and Q4 (Qn = Si(OSi)n(OH)4−n
)
O
very strong one. Chelation of the immobilized ligand 10 to the
metal fragments MX2(CO)3 (M = Mo; X = I, Br) is also deduced
from the vibrational spectra. The appearance of the bands at
2073 cm−1, 2022 cm−1 and 1954 cm−1, assigned as the νC≡O
modes for 11, denotes the presence of three coordinated car-
bonyl groups, and the retention of the organometallic M(CO)3
core during the immobilization procedure. The absence of the
bands at 2276 cm−1 and 2304 cm−1, corresponding to the νC≡N
modes of the acetonitrile in the precursors, reflects their substi-
tution by other ligands. Moreover, the νC=N mode of the DAB
ligand is observed at 1631 cm−1, a value lower than that of the
free ligand, indicating chelation to the metal centre. In the case
at δ = −102.0 ppm and −109.0 ppm, respectively. Also a min-
imal amount of Q2 species can be observed at δ = −92.5 ppm.
Upon grafting of the chelating ligand to the walls of the par-
ent host material, the spectra of the new material 10 shows a
reduction of the intensities of the resonances assigned to Q2
and Q3 species, owing to the silylation of the surface by nucle-
ophilic substitution. Although the Q3 sites are the most reac-
tive, some remain unreacted at the surface, suggesting that they
can be blocked to functionalization due to hydrogen bonding
(SiO)3Si OH. . .OH Si(SiO)3 [29]. The spectra of materials 10,
11 and 12 show three new convoluted signals in the −50 ppm to
−58 ppm range. These resonances are associated with organosil-
ica species T1, T2 and T3, where Tm = RSi(OSi)m(OEt)3−m. The
presence of these signals in the spectra is indicative of the suc-
cessful introduction of the ligand DAB, and also shows that the
ligand is covalently bound to the surface of the host material.
The 29Si NMR spectra of materials 11 and 12 are similar to
that of 10. In fact, since the ligand binds to the metal centres by
of 12, the corresponding νC≡O bands are observed at 2082 cm−1
2018 cm−1 and 1944 cm−1, and the νC N mode of the ligand at
1627 cm−1
,
.
The composite materials 13 and 14 containing W(II) (pre-
pared from 10*) were found to exhibit similar spectroscopical
features to those of the Mo(II) analogues (materials 11 and 12),

J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
95
6), described earlier in this work, made possible the observation
of all signals. The resonances corresponding to the C N and
C≡O carbon atoms are not clear in the 13C CP MAS NMR of
materials 11 and 12, owing to the overlapping of the spinside
band and of the high relaxation time of the carbonyl groups.
Despite this, the presence of such groups is confirmed by the
corresponding FTIR spectrum (see above).
The solid-state NMR studies for the W(II) containing mate-
rials 13 and 14 led to the same conclusions found for composite
materials 11 and 12, as in the case of the FTIR results: a full
description of spectral data can be found in Section 4.
2.3. Polymerization catalytic studies
The study of olefin polymerization catalyzed by Mo(II)
and W(II) systems has received much attention over the last
years [15–19]. Complexes 1–4 have already been tested for
ROMP catalysis of NBE and NBD (polyNBE and polyNBD
denote the corresponding polymer product). Metathesis poly-
merization of NBD is successfully carried out by complexes
[MX2(CO)3(NCMe)2] at room temperature for X = Br, whereas
the diiodo requires a co-catalyst, such as ZrCl4 or AlCl3, in order
to initiate polymerization [15 and references therein].
Fig. 5. 13C CP MAS NMR for materials MCM-DAB (10), MCM-DAB-MoI2
(11) and MCM-DAB-MoBr2 (12).
the nitrogen atoms, no significant alterations of the silica atoms
chemical shifts would be expected.
Complexes 5–8 did not catalyze the ROMP reactions of NBE
and NBD at 298 K (see other conditions in Section 4). At 328 K,
however, some catalytic activity was observed, yielding 1–8%
polymer (Table 3). Complex 6 (M = Mo, X = Br) presents a
slightly higher activity than the remaining ones. The composite
materials 11–14 are more active catalysts for this reaction than
the homogeneous counterparts 5–8. Probably, the Si(OEt)3
groups present in the complexes, lead to the formation of less
active or inactive oligomeric catalytic species [24].
The 13C CP MAS NMR of the functionalized material 10 is
very similar to that of the free DAB ligand (liquid 13C NMR),
Fig. 5. Only a slight decrease in the intensity of resonances
assigned to the SiO CH2CH3 groups is observed, which is
expected, since the binding of the ligand to the surface of the
host material takes place by the hydrolysis of these groups.
The resonances due to the aromatic nuclei in materials 11
and 12 are also well described, despite the large width of the
signal. Nevertheless, the solution 13C NMR of complex 5 (and
NeitherthepristineMCMhostmaterial9northeDABderiva-
tized material 10 possess catalytic activity towards the ROMP
Table 3
Results for ROMP reaction of NBE and NBD initiated by isolated and immobilized [MX2(CO)3(L)2] complexes
Catalyst
T (K)
PolyNBE
Yield (%)
PolyNBD
Yield (%)
TOFa (g mol−M1 h−1
)
TOFa (g mol−M1 h−1
)
[MoI2(CO)3(DAB)] (5)
[MoBr2(CO)3(DAB)] (6)
[WI2(CO)3(DAB)] (7)
[WBr2(CO)3(DAB)] (8)
328
328
328
328
5
8
2
3
11
18
5
2
6
1
2
4.7
14
2.3
4.7
7
298
328
4
15
31
117
1
6
8.0
48.0
MCM-DAB-MoI2 (11)
298
12
36
24
59
94
280
211
476
6
10
–
48.0
79.9
328
328b
328c
MCM-DAB-MoBr2 (12)
MCM-DAB-WI2 (13)
–
298
328
0.4
1
22
55
–
–
MCM-DAB-WBr2 (14)
298
328
1.6
3
89
166
–
–
a
Turnover frequency calculated at 72 h.
Second run.
In the presence of AlCl3.
b
c

96
J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
of the olefins, indicating that the active species must contain Mo
or W. For all metal-containing catalysts, the ROMP of NBE is
easier to accomplish than that of NBD. This is in agreement
with literature results, where some authors reported that NBD
is less readily polymerized than NBE [19]. The Mo-containing
catalysts are more active than the tungsten analogues. The high-
est catalytic activity was observed for the derivatized mate-
rial 12 (M = Mo, X = Br), giving TOF of 280 g mol Mo−1 h−1
and 80 g mol Mo−1 h−1 for ROMP of NBE and NBD, respec-
tively. These results reflect the highest activity of complex
6 (M = Mo, X = Br) in homogeneous phase. In general, poly-
mer yields obtained in the presence of the supported cata-
lysts are better than those found for the corresponding unsup-
ported complexes 5–8, although the values are not very high
(Table 3). They may be improved by increasing the reaction
temperature. For example, in the case of 12, a rise in tem-
perature of 30 K leads to a three-fold increase in polyNBE
yield.
It has been reported that the incorporation of aluminium in
MCM, used as co-catalyst in metallocene systems, enhances the
catalytic activity in ethene polymerization [33]. Accordingly, the
performance of these catalysts may be improved by changing the
support properties, by doping the silica-matrix with aluminium
or by the addition of a co-catalyst, such as AlCl3. In order to
check the effect of added aluminium, a catalytic reaction using
material 12, AlCl3, and NBE in the molar ration of 1:3:200 at
328 K, was carried out. After 72 h, the reaction was stopped and
the yield of polyNBE was estimated to be 59% with TOF of
476 g mol−M1o h−1. This result is better than that obtained without
AlCl3, indicating that the co-catalyst has a beneficial effect. It
should, however, be mentioned that AlCl3 alone did not show
any catalytic activity.
phase complexes 5–8 exhibit poor catalytic activity (complex
6 (M = Mo, X = Br) being the most active), the catalytic perfor-
mance of the corresponding heterogenized catalysts (11–14) in
the ROMP of olefins is far superior. In general, the conversions
are higher for NBE than for NBD. Nevertheless, the polymer
yields achieved with the solid catalysts described in this work
are, in general, relatively low. Polymer yield tends to increase
with reaction temperature. The addition of AlCl3 as a co-catalyst
enhanced the catalytic performance of the material MCM-DAB-
MoBr2 (12) in the ROMP of NBE.
4. Experimental
4.1. General
All preparations and manipulations were performed using
standard Schlenk techniques under nitrogen. Commercial grade
solvents were dried and deoxygenated by standard procedures
(Et2O over Na/benzophenone ketyl; CH2Cl2 and CH3CN over
˚
CaH2), distilled under nitrogen and kept over 4 A molecular
˚
sieves (3 A for CH3CN). Purely siliceous MCM was synthe-
sized as described previously using [(C16H33)N(CH3)3]Br as
the templating agent [24]. After calcinations (540 ◦C for 6 h), the
material was characterized by powder XRD, N2 adsorption and
IR spectroscopy. Prior to the grafting experiment, physisorbed
water was removed from calcined MCM by heating at 180 ◦C
in vacuum (10−2 Pa) for 2 h. Microanalyses were performed at
the University of Aveiro. BET specific surface areas (SBET, p/p0
from 0.03 to 0.13) and specific total pore volume, Vt, were esti-
mated from N2 adsorption isotherms measured at 77 K. The pore
size distributions (PSD) were calculated by the BJH method
using the modified Kelvin equation with correction for the sta-
tistical film thickness on the pore walls [34,35]. The statistical
film thickness was calculated using Harkins–Jura equation in
the p/p0 range from 0.1 to 0.95.
Powder XRD data were collected on a Phillips PW1710
diffractometer using Cu K␣ radiation filtered by graphite.
FTIR spectra in transmission mode were measured with a
Mattson Satellite FTIR spectrometer using KBr pellets. 1H
and 13C solution NMR spectra were obtained with a Bruker
Avance-400 spectrometer. 29Si and 13C solid-state NMR spec-
tra were recorded at 79.49 MHz and 100.62 MHz, respectively,
on a (9.4 T) Bruker Avance 400P spectrometer. 29Si MAS
NMR spectra were recorded with 40◦ pulses, spinning rates
5.0–5.5 kHz, and 60 s recycle delays. 29Si CP MAS NMR
spectra were recorded with 5.5 ␮s 1H 90◦ pulses, 8 ms con-
tact time, a spinning rate of 4.5 kHz, and 4 s recycle delays.
13C CP MAS NMR spectra were recorded with a 4.5 ␮s 1H
90◦ pulse, 2 ms contact time, a spinning rate of 8 kHz, and
4 s recycle delays. Chemical shifts are quoted in ppm from
TMS. 13C spectra were also recorded in the solid-state at
125.76 MHz on a Bruker Avance 500 spectrometer. Phenyl-
substituted 1,4-diazobutadiene ligand (DAB) was prepared as
described previously [24]. The precursor organometallic com-
plexes MoI2(CO)3(NCCH3)2 (1), MoBr2(CO)3(NCCH3)2 (2),
WI2(CO)3(NCCH3)2 (3) and WBr2(CO)3(NCCH3)2 (4) were
also prepared according to literature methods [25].
The catalytic performance of catalyst 12 was also evaluated
in a second reaction cycle. After reaction, the solid catalyst was
separated from the reaction solution by filtration, and the poly-
mer was precipitated by the addition of methanol. Both the
conversion and polymer yield decreased from the first to the
second run (Table 3). ICP-AES of the used and recovered solids
showed that from the first to the second runs there is a 0.5% loss
of metal, indicating that metal leaching does not occur to a great
extent. The catalyst recovery procedure may be inefficient to
remove all of the organic matter entrapped in the pore system,
accounting for partial loss of catalytic activity observed upon
recycling.
3. Conclusions
The immobilization of hepta-coordinate halocarbonyl Mo(II)
and W(II) complexes on MCM was successfully accomplished
by the tethering approach, using a phenyl-substituted 1,4-
diazobutadienespacer, andwasconfirmedbyseveraltechniques.
For instance, the bands assigned to νC≡O modes observed by
FTIR showed the incorporation of the organometallic M(CO)3
cores and the retention of their integrity. The immobilization of
W(II) complexes inside the pores of the MCM-41 host material
was not so successful as that of the Mo(II) analogues, accord-
ing to the results of the metal loadings. While in homogeneous

J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
97
4.2. Catalytic studies
4.3.3. [WI2(CO)3(DAB)] (7)
Yield: 48%. C35H52N2Si2O9I2W (1138.62) calcd: C 36.92,
H 4.60, N 2.46; found: C 37.06, H 4.71, N 2.58. IR (KBr, ν
cm−1): 3056 (vw), 2976 (w), 2935 (w), 2025 (vw), 1980 (s),
1934 (s), 1633 (s), 1451 (w), 1246 (vs), 1081 (vs), 950 (s), 796
(m), 693 (w). 1H NMR (400.10 MHz, CDCl3, r.t., δ ppm): 0.66 (s
bd, 4H, SiCH2), 1.15–1.31 (m, 18H, OCH2CH3), 1.72–1.90 (m,
4H, CH2CH2CH2), 3.32–3.34 (m, 4H, NCH2), 3.69–3.81 (m,
12H, OCH2CH3), 7.31–7.75(m, 5H, Ph), 7.57–7.75(m, 3H, Ph),
7.88–7.93 (m, 2H, Ph). 13C NMR (100.25 MHz, CDCl3, r.t., δ
ppm): 9.2 (SiCH2), 19.8 (OCH2CH3), 24.8 (CH2CH2CH2), 55.9
(CH2N C), 57.8 (OCH2CH3), 127.6, 128.9, 129.4 (Ph), 165.6
(CH2N C).
The complexes and materials reported herein were tested in
the ROMP of NBE and NBD under N2 atmosphere at 298 K
and 328 K and using toluene as solvent. The polymers were
precipitated by addition of methanol at the end of the reac-
tion, i.e., after 72 h and upon catalyst separation. The catalytic
reactions using complexes 5–8 were carried out using a cata-
lyst/olefin molar ratio of 1:200. In the case of the studies using
the composite materials 11–14 a metal/olefin molar ratio of
1:200 (based on the metal loadings as determined by ICP-AES)
was used. For the catalytic reaction using AlCl3 as co-catalyst
the metal/Al/olefin molar ratio was 1:3:200. In a typical experi-
ment, a certain amount of the materials with 18.5 mmol of olefin
was mixed in toluene (10 ml) at 298 K or 328 K. All the reactions
were stopped after 72 h. This was accomplished by separating
the catalysts by filtration, followed by the addition of methanol
to the toluene solution in order to precipitate the polymer. The
solid polymer was separated by filtration and the polymers dried
in vacuum before being weighed. The yields were calculated
based on the initial weight of olefin used. The identification of
the polymers was done by 1H NMR [15–19].
4.3.4. [WBr2(CO)3(DAB)] (8)
Yield: 58%. C35H52N2Si2O9Br2W (1044.61) calcd: C 40.24,
H 5.02, N 2.68; found: C 40.11, H 5.11, N 2.79. IR (KBr, ν
cm−1): 3067 (vw), 2986 (w), 2930 (vw), 2032 (vw), 1985 (s),
1940 (m), 1649 (s), 1234 (vs), 1076 (vs), 978 (vs), 791 (m),
700 (w). 1H NMR (400.10 MHz, CDCl3, r.t., δ ppm): 0.74–0.85
(s bd, 4H, SiCH2), 1.19 (t, 18H, OCH2CH3), 1.78–1.85 (m,
4H, CH2CH2CH2), 3.20–3.30 (t, 4H, NCH2), 3.61–3.73 (m,
12H, OCH2CH3), 7.48 (s bd, 2H, Ph), 7.58–7.62 (m, 5H, Ph),
7.73–7.87 (m, 3H, Ph). 13C NMR (100.25 MHz, CDCl3, r.t., δ
ppm): 8.7 (SiCH2), 19.1 (OCH2CH3), 23.6 (CH2CH2CH2), 57.3
(CH2N C), 58.4 (OCH2CH3), 126.9, 128.9, 129.6 (Ph), 166.5
(CH2N C).
4.3. Preparation of the complexes of type
[MX2(CO)3(DAB)] (M = Mo or W, and X = I or Br) (5–8)
A solution of 1–4 (0.50 mmol) in acetonitrile (10 ml) was
treated with 1 equivalent of ligand DAB in acetonitrile (5 ml).
The resulting turbid solution was stirred for 4 h at room temper-
ature. The solvent was evaporated, and the solid product washed
with hexane and dried in vacuum.
4.4. MCM-DAB (10)
A solution of ligand DAB (0.70 g, 1.13 mmol) in toluene
(10 ml) was added to a suspension of MCM-41 (0.8 g) in toluene
(10 ml) and the mixture heated at 100 ◦C for 9 h. The resultant
solid was then filtered off and washed four times with CH2Cl2
(4 × 15 ml), and dried in vacuum at 50 ◦C for 3 h. Elemental
analysis found (%): C 16.20, N 1.50, H 2.80. IR (KBr, ν cm−1):
3067 (vw), 2978 (w), 2927 (vw), 2891 (vw), 1652 (s), 1494
(s), 1447 (s), 1390 (m), 1245 (vs), 1080 (vs), 951 (s), 800 (s),
702 (m). 13C CP/MAS NMR (δ ppm): 8.8 (SiCH2), 16.2, 20.5,
41.6, 57.5, 128.1 (Ph-C). 29Si MAS NMR (δ ppm): −55.4 (T1),
−109.5 (Q4). 29Si CP/MAS NMR (δ ppm): −54.9 (T1), −59.6
(T2), −67.0 (T3), −91.9 (Q2), −101.9 (Q3), −109.2 (Q4).
4.3.1. [MoI2(CO)3(DAB)] (5)
Yield: 85%. C35H52N2Si2O9I2Mo (1050.72) calcd: C 40.01,
H 4.99, N 2.67; found: C 39.90, H 5.11, N 2.78. IR (KBr, ν
cm−1): 3058 (vw), 2973 (w), 2923 (vw), 2887 (vw), 2067 (vw),
2017 (m), 1946 (m), 1679 (m), 1637 (s), 1593 (s), 1492 (m), 1446
(s), 1400 (s), 1116 (vs), 1024 (vs), 952 (s), 764 (s), 698 (s). 1H
NMR (400.10 MHz, CDCl3, r.t., δ ppm): 0.71 (s bd, 4H, SiCH2),
1.04 (t, 18H, OCH2CH3), 1.93–1.98 (m, 4H, CH2CH2CH2),
3.47–3.52 (m, 16H, NCH2 and OCH2CH3), 7.36–7.95 (m, 10H,
Ph). 13C NMR (100.25 MHz, CDCl3, r.t., δ ppm): 13.1 (SiCH2),
20.4 (OCH2CH3), 21.7 (CH2CH2CH2), 58.1 (OCH2CH3), 68.7
(CH2N C), 127.2, 128.0, 128.9 (Ph) 166.7 (CH2N C).
4.5. Preparation of the materials of the type
MCM-DAB-MX2 (11–14)
4.3.2. [MoBr2(CO)3(DAB)] (6)
Yield: 72%. C35H52N2Si2O9Br2Mo (956.71) calcd: C 43.94,
H 5.48, N 2.93; found: C 43.80, H 5.51, N 2.88. IR (KBr, ν
cm−1): 3057 (vw), 2978 (w), 2925 (vw), 2881 (vw), 2063 (w),
2011 (s), 1930 (vs), 1690 (m), 1635 (s), 1594 (s), 1444 (s), 1396
A solution of 1–4 (0.4 mmol) in CH2Cl2 (10 ml) was added
to a suspension of 10 (1.0 g) in CH2Cl2 (10 ml) and the mixture
stirred at room temperature for 24 h. The solution was filtered
off and the pale yellow powder washed repeatedly with CH2Cl2
(4 × 20 ml), before drying under vacuum at room temperature
for several hours.
1
(s), 1167 (m), 1075 (vs), 957 (vs), 766 (s), 707 (m). H NMR
(400.10 MHz, CDCl3, r.t., δ ppm): 0.71–0.86 (m, 4H, SiCH2),
1.05 (t, 18H, OCH2CH3), 1.93–1.98 (m, 4H, CH2CH2CH2),
3.48–3.53 (m, 16H, NCH2 and OCH2CH3), 7.31–7.97 (m, 10H,
Ph). 13C NMR (100.25 MHz, CDCl3, r.t., δ ppm): 10.5 (SiCH2),
18.8 (OCH2CH3), 23.3 (CH2CH2CH2), 57.5 (OCH2CH3), 66.4
(CH2N C), 126.9, 128.3. 129.1 (Ph), 166.3 (CH2N C).
4.5.1. MCM-DAB-MoI2 (11)
Elemental analysis found (%): C 13.60, N 1.45, H 2.74,
Mo 2.6. IR (KBr, ν cm−1): 3072 (vw), 2978 (w), 2932 (vw),
2896 (vw), 2073 (vw), 2022 (m), 1954 (m), 1631 (s), 1494

98
J. Gimenez et al. / Journal of Molecular Catalysis A: Chemical 256 (2006) 90–98
(s), 1447 (s), 1400 (m), 1241 (vs), 1080 (vs), 945 (m), 806
(s), 702 (m). 13C CP/MAS NMR (δ ppm): 10.0 (SiCH2),
16.8, 21.2, 42.9, 58.0, 129.5 (Ph-C). 29Si MAS NMR (δ
ppm): −55.8 (T1), −110.0 (Q4). 29Si CP/MAS NMR (δ ppm):
−53.9 (T1), −60.0 (T2), −67.7 (T3), −92.8 (Q2), −102.6 (Q3),
−109.7 (Q4).
References
[1] M.H. Valkenberg, W.F. Ho¨lderich, Catal. Rev. 44 (2002) 321.
[2] A. Taguchi, F. Schu¨th, Micropor. Mesopor. Mater. 77 (2005) 1.
[3] P.K. Baker, Chem. Soc. Rev. 27 (1998) 129.
[4] H. Liu, M.J. Calhorda, V. Fe´lix, M.G.B. Drew, J. Organomet. Chem.
632 (2001) 175.
[5] H. Liu, M.J. Calhorda, V. Fe´lix, M.G.B. Drew, L.F. Veiros, Inorg. Chim.
Acta 327 (2002) 169.
4.5.2. MCM-DAB-MoBr2 (12)
[6] R. Colton, I.B. Tomkins, Aust. J. Chem. 19 (1966) 1143.
[7] D.P. Tate, W.R. Knipple, J.M. Augl, Inorg. Chem. 1 (1962) 433.
[8] A.D. Westland, N. Muriithi, Inorg. Chem. 11 (1972) 2971.
[9] P.K. Baker, Adv. Organomet. Chem. 40 (1996) 45.
[10] M.G.B. Drew, Prog. Inorg. Chem. 23 (1977) 67.
[11] M. Meln´ık, P. Sharrock, Coord. Chem. Rev. 65 (1985) 49.
[12] H.L. Nigam, R.S. Nyholm, Proc. Chem. Soc. (1957) 321.
[13] L. Bencze, A. Kraut-Vass, J. Mol. Catal. 28 (1985) 369.
[14] L. Bencze, A. Kraut-Vass, L. Pro´kai, J. Chem. Soc. Chem. Commun.
(1985) 911.
[15] M. Al-Jahdali, P.K. Baker, A.J. Lavery, M. Meehan, D.J. Muldoon, J.
Mol. Catal. A: Chem. 159 (2000) 51.
[16] P.K. Baker, M.A. Beckett, B.M. Stiefvater-Thomas, J. Mol. Catal. A:
Chem. 193 (2003) 77.
[17] T. Szyman´ska-Buzar, T. Glowiak, I. Czelus´niak, J. Organomet. Chem.
640 (2001) 72.
[18] T. Szyman´ska-Buzar, T. Glowiak, I. Czelus´niak, Polyhedron 21 (2002)
2505.
[19] I. Czelus´niak, T. Szyman´ska-Buzar, Appl. Catal. A 277 (2004) 173.
[20] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature
359 (1992) 710.
[21] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D.
Schmitt, C.T.-W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B.
Higgins, J.L. Schlenker, J. Am. Chem. Soc. 114 (1992) 10834.
[22] B. Hatton, K. Landskron, W. Whitnall, D. Perovic, G.A. Ozin, Acc.
Chem. Res. 38 (2005) 305.
[23] A. Corma, Chem. Rev. 97 (1997) 2373.
[24] C.D. Nunes, M. Pillinger, A.A. Valente, J. Rocha, A.D. Lopes, I.S.
Gonc¸alves, Eur. J. Inorg. Chem. (2003) 3870.
[25] P.K. Baker, S.G. Fraser, E.M. Keys, J. Organomet. Chem. 309 (1986)
319.
[26] C.D. Nunes, M. Pillinger, A.A. Valente, A.D. Lopes, I.S. Gonc¸alves,
Inorg. Chem. Commun. 6 (2003) 1228.
Elemental analysis found (%): C 13.60, N 1.45, H 2.68, Mo
2.8. IR (KBr, ν cm−1): 3062 (vw), 2977 (w), 2082 (vw), 2018
(m), 1944 (m), 1627 (s), 1499 (m), 1451 (m), 1403 (m), 1234
(vs), 1034 (vs), 943 (s), 800 (m), 705 (w). 13C CP/MAS NMR
(δ ppm): 9.5 (SiCH2), 16.7, 21.0, 42.7, 57.9, 129.1 (Ph-C). 29Si
MAS NMR (δ ppm): −53.8 (T1), −109.7 (Q4). 29Si CP/MAS
NMR (δ ppm): −55.5 (T1), −60.1 (T2), −68.8 (T3), −92.8 (Q2),
−102.2 (Q3), −109.7 (Q4).
4.5.3. MCM-DAB-WI2 (13)
Elemental analysis found (%): C 15.88, N 1.53, H 2.91, W
0.4. IR (KBr, ν cm−1): 3062 (vw), 2977 (w), 2924 (vw), 2029
(vw), 1987 (w), 1959 (m), 1632 (s), 1447 (w), 1399 (w), 1235
(s), 1076 (vs), 948 (m), 800 (m), 700 (w). 13C CP/MAS NMR
(δ ppm): 9.6 (SiCH2), 16.8, 20.9, 42.3, 57.5, 129.4 (Ph-C). 29Si
MAS NMR (δ ppm): −53.4 (T1), −109.6 (Q4). 29Si CP/MAS
NMR (δ ppm): −53.9 (T1), −59.9 (T2), −64.5 (T3), −92.0 (Q2),
−101.9 (Q3), −109.6 (Q4).
4.5.4. MCM-DAB-WBr2 (14)
Elemental analysis found (%): C 18.23, N 1.53, H 3.22, W
0.7. IR (KBr, ν cm−1): 3067 (vw), 2980 (w), 2929 (vw), 2891
(vw), 2058 (vw), 2020 (w), 1958 (m), 1632 (s), 1450 (m), 1388
(w), 1244 (m), 1075 (vs), 800 (s), 700 (m). 13C CP/MAS NMR
(δ ppm): 9.8 (SiCH2), 17.1, 20.8, 42.5, 57.7, 128.6 (Ph-C). 29Si
MAS NMR (δ ppm): −55.2 (T1), −109.6 (Q4). 29Si CP/MAS
NMR (δ ppm): −54.7 (T1), −60.4 (T2), −67.5 (T3), −91.6 (Q2),
−102.0 (Q3), −109.7 (Q4).
[27] B. Marler, U. Oberhagemann, S. Voltmann, H. Gies, Micropor. Mater.
6 (1996) 375.
Acknowledgments
[28] W. Hammond, E. Prouzet, S.D. Mahanti, T.J. Pinnavaia, Micropor.
Mesopor. Mater. 27 (1999) 19.
[29] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, second
ed., Academic Press, London, 1982.
[30] M.D. Alba, A. Becerro, J. Klinowski, J. Chem. Soc., Faraday Trans. 92
(1996) 849.
CDN (SFRH/BPD/14512/2003) and PDV (SFRH/BPD/
14903/2004) thank FCT for research grants. JG thanks
the Socrates Student Exchange Program for a grant. Paula
˜
Brandao is acknowledged for assistance with the NMR experi-
[31] A.A. Romero, M.D. Alba, W. Zhou, J. Klinowski, J. Phys. Chem. B
101 (1997) 5294.
[32] M. Kruk, M. Jaroniec, Langmuir 15 (1999) 5410.
[33] H. Rahiala, I. Beurroies, T. Eklund, K. Hakala, R. Gougeon, P. Trens,
J.B. Rosenholm, J. Catal. 188 (1999) 14.
ments.
Appendix A. Supplementary data
[34] M. Kruk, M. Jaroniec, Langmuir 13 (1997) 6267.
[35] M. Kruk, V. Antochshuk, M. Jaroniec, J. Phys. Chem. B 103 (1999)
10670.
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.molcata.2006.04.023.