A new heterogeneous hybrid ruthenium catalyst being an eco-friendly option for the production of polymers and organic intermediates

Article abstract of DOI:10.1039/b202872g

We succeeded in synthesising and characterising a new heterogeneous hybrid ruthenium catalyst that exhibits excellent stability, reusability and leaching characteristics. The host-guest interaction is studied by XRD, XRF, ICP/MS, BET, FT-Raman and solid state NMR analysis. Moreover, we tested this catalytic system in ring-closing metathesis (RCM), ring-opening metathesis polymerization (ROMP), Kharasch addition, atom transfer radical polymerization (ATRP) and vinylation reactions. The results obtained from these tests show that for ROMP, RCM, Kharasch addition and vinylation reactions the heterogeneous catalyst possesses important advantages in comparison with its homogeneous analogue.

Full text of DOI:10.1039/b202872g

View Article Online / Journal Homepage / Table of Contents for this issue
A new heterogeneous hybrid ruthenium catalyst being an eco-friendly
option for the production of polymers and organic intermediates
a
Bob De Clercq, Frederic Lefebvre and Francis Verpoort*
b
a
´ ´
a
Department of Inorganic and Physical Chemistry, Division Organometallic Chemistry
and Catalysis, Ghent University, Krijgslaan 281 (S-3), 9000, Ghent, Belgium.
E-mail: Francis.Verpoort@rug.ac.be; Fax: +32-9-264.49.83; Tel: +32-9-264.44.36
Laboratoire LCOMS, UMR CNRS-CPE 9986, 43 bd avenue du 11 novembre 1918,
69626, Villeurbanne Cedex, France
b
Received 21st March 2002, Accepted 20th June 2002
First published as an Advance Article on the web 8th August 2002
We succeeded in synthesising and characterising a new heterogeneous hybrid ruthenium catalyst that exhibits
excellent stability, reusability and leaching characteristics. The host–guest interaction is studied by XRD, XRF,
ICP/MS, BET, FT-Raman and solid state NMR analysis. Moreover, we tested this catalytic system in ring-
closing metathesis (RCM), ring-opening metathesis polymerization (ROMP), Kharasch addition, atom transfer
radical polymerization (ATRP) and vinylation reactions. The results obtained from these tests show that for
ROMP, RCM, Kharasch addition and vinylation reactions the heterogeneous catalyst possesses important
advantages in comparison with its homogeneous analogue.
Olefin metathesis is a catalytic reaction in which alkenes are
1. Introduction
converted into new products via the rupture and reformation of
C–C double bonds. Depending on the starting material (cyclic
or acyclic alkenes) and the reaction parameters, RCM, acyclic
diene metathesis (ADMET) or ROMP proceed (Scheme 1, A).5
ATRP is based on a dynamic equilibration between the pro-
pagating radicals and the dormant species which is established
through the reversible transition metal-catalysed cleavage of
the covalent carbon–halogen bond in the dormant species
(Scheme 1, B).6
The Kharasch reaction consists of the addition of a polyha-
logenated alkane across an olefin through a radical mechanism
(Scheme 1, C).7
In the enol ester synthesis the ruthenium(II) complexes pro-
mote the electrophilic activation of alkynes to give Markov-
nikov or anti-Markovnikov (E and Z) addition reactions of
carboxylic acids to the triple bond (Scheme 1, D).8
Besides the formation of the (E)-alk-1-enyl ester, also a
small percentage of (Z)-alk-1-enyl ester, Markovnikov addi-
tion products and disubstituted enol esters are obtained.
It is known from the literature that a limitation of ruthe-
nium(II) complexes containing an electron-donating bidentate
nitrogen ligand is the ability of these complexes to dimerize
terminal alkynes into enynes (Scheme 1, E).9
The use of a reaction system in which the catalyst is in a dif-
ferent phase to that of the substrate and the products allows
easy removal of the catalyst from the reaction mixture via
filtration, centrifugation or decantation. The potential techno-
logical advantages in converting a process catalysed by a
homogeneous metal complex into one involving a heteroge-
neous analogue have been well rehearsed.10 Suffice to say that
on a laboratory scale supported metal complex catalysts con-
siderably facilitate product work-up and isolation, while on a
large scale such heterogeneous species allow processes to be
run continuously using packed or fluidised bed columns with
considerable financial advantages both in terms of capital
expenditure on plant and with regard to recurrent costs.
A heterogeneous catalytically active material can be formed
by immobilising active species on or in a support material such
Awareness of environmental issues has proven to be a potent
driving force in the development of environment friendly pro-
cesses and technologies in the chemical industry.1 This has
manifested in a determined effort, on the part of the industry,
to develop catalytic processes which obviate waste production,
i.e. solving the problem at the beginning and not dealing with
the waste disposal later.2 As a consequence of the inherent
advantages of heterogeneous reagents, much effort has been
directed towards the development of heterogeneous processes
and the heterogenization of known active homogeneous cata-
lysts.3 One of the major goals of green chemistry is to develop
environmentally acceptable routes to important organic pro-
ducts. Methods by which this can be achieved are increasing
product selectivity, aiming for 100% atom efficiency and repla-
cing stoichiometric reagents with heterogeneous catalysts.
Increasingly tighter guidelines are being issued concerning
the disposal of waste materials, with increasing public and cor-
porate pressure to comply with them. In this respect it was our
goal to apply the principles of making chemistry more environ-
mentally friendly, being applied to the homogeneous catalyst 1
(Fig. 1) for which the synthesis and activity in Ring Closing
Metathesis (RCM), Ring Opening Metathesis Polymerization
(ROMP), Atom Transfer Radical Polymerization (ATRP),
Kharasch addition and vinylation reactions, were recently
reported by us.4
Fig. 1 Homogeneous catalyst 1.
DOI: 10.1039/b202872g
New J. Chem., 2002, 26, 1201–1208
1201
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2002

View Article Online
Scheme 1 Schematic representation of olefin metathesis (A), ATRP (B), Kharasch addition (C), vinylation (D) and dimerization (E).
as porous silica, titania or alumina.11 The porosity of such
materials can lead to increased reaction selectivity by giving
high selectivity to the desired product. This is the ideal situa-
tion and can be achieved by the synthesis of chemically modi-
fied porous solids or supported reagent catalysts.12 The present
paper reports on the application of hybrid heterogeneous cat-
alyst 2 (Fig. 2) prepared by the covalent anchoring of catalytic
complex 1 via a spacer molecule onto mesoporous silica.
Furthermore, the activity of this new material was compared
to that of the homogeneous analogue.
TMS was used as reference compound. Solid-state NMR spec-
tra were acquired on a Bruker DSX-300 spectrometer operat-
1
ing at 300.18 MHz for H-NMR, 75.49 MHz for 13C-NMR,
121.51 MHz for 31P-NMR and 59.595 MHz for 29Si-NMR.
The spectra were recorded under MAS conditions with a clas-
sical 4 mm probe head allowing spinning frequencies up to 12
kHz. The anchoring of the homogeneous catalyst was con-
firmed by a Raman spectrometer Bruker Equinox 55 with a
FRA 106 module. The loading of the heterogeneous hybrid
catalyst was determined with a Varian Liberty ICP/MS spec-
trometer and an ARL 9400 Sequential XRF spectrometer.
XRD spectra were recorded on a Siemens diffractometer
D5000. Elemental analyses were performed with Carlo Erba
EA 1110 equipment. The BET analysis was done on a Gemini
Micromeritics 2360 surface area analyser with Flow prep 060
degasser. The samples were dried overnight at 423 K and
cooled to room temperature prior to adsorption. Extra care
with the functionalised materials was necessary due to the pos-
sibility of aerial oxidation, therefore transfer to the balance
and outgassing of the system was fast. Nitrogen isotherms
were recorded at 77 K. Specific surface areas were determined
from the linear part of the BET plot (P/P0 ¼ 0.05–0.3). The
number- and weight-average molecular weights (Mn and Mw)
and polydispersity (Mw/Mn) of the polymers were determined
by gel permeation chromatography (CHCl3 , 25 ꢀC) using poly-
styrene (for styrene) or polyMMA (for the acrylates and
methacrylates) standards. The GPC instrument used was a
Waters Maxima 820 system equipped with a PL gel column.
For ATRP and Kharasch addition reactions, all reagents
and solvents were dried, distilled and stored under nitrogen
at ꢁ20 ꢀC with conventional methods. ATRP of methacrylates
[methyl methacrylate (MMA), isobutyl methacrylate (IBMA)],
2. Experimental
2.1. General
All reactions and manipulations were performed under an
argon atmosphere by using conventional Schlenk-tube techni-
ques. Argon gas was dried by passage through P2O5 (Aldrich
97%).1H-NMR spectra (500 MHz) were recorded on a Bruker
AM spectrometer. The chemical shifts are reported in ppm and
Fig. 2 Cartoon type representation of the hybrid catalytic system 4.
1202
New J. Chem., 2002, 26, 1201–1208

View Article Online
acrylates [methyl acrylate (MA), butyl acrylate (BA)] and styr-
ene (Styr) was carried out with respectively ethyl 2-methyl-2-
bromopropionate, methyl 2-bromopropionate and (1-bro-
moethyl)benzene as initiator. GC analysis was performed
using the following conditions: column: SPB2-5; 30 m ꢂ 0.25
mm ꢂ 0.25 mm film thickness; carrier gas: He, 100 kPa; detec-
tor: FID; gas chromatograph: Varian 3300; integrator: Vista
401. GC/MS measurements were performed using the follow-
ing conditions: column: SPB2-5; 30 m ꢂ 0.25 mm ꢂ 0.25 mm
film thickness; carrier gas: He, 100 kPa; detector: FID; gas
chromatograph: Varian 4600; MS: Finnigan MAT ITD. The
homogeneous catalyst was prepared according to literature
procedures.4 Cyclooctene and norbornene were purchased
from Aldrich and distilled from CaH2 under nitrogen prior
to use. Commercial grade solvents were dried and deoxyge-
nated for at least 24 h over appropriate drying agents under
nitrogen atmosphere distilled prior to use. Unless otherwise
noted, all other compounds were purchased from Aldrich
Chemical Co., and used as received.
2 mmol of bromopropyltriethoxysilane in 1 ml THF was
added dropwise. Thereafter the reaction mixture was stirred
for 3 h at RT.
Synthesis of 3. The organomagnesium derivative 2 was
transferred quantitatively to a solution of 2 mmol of 1 in
THF (15 ml) and stirred for 6 h at RT to afford the spacer-
modified catalyst 3. The MgBr2 (by-product of the reaction)
was filtered off under nitrogen atmosphere. After evaporation
of the solvent, 3 appeared as a brown solid. 1H-NMR (CDCl3)
d 8.22 (s, 1H), 7.24 (s, 2H), 6.98–6.81 (m, 4H), 5.47 (d, 2H),
5.34 (d, 2H), 3.81 (q, 6H), 2.81 (sp, 1H), 2.48 (s, 6H), 2.19
(s, 3H), 1.65 (t, 2H), 1.29 (m, 4H), 1.24 (d, 6H), 1.18 (t, 9H);
elemental analysis calcd. (%) for RuC34H48O4NClSi (699.29):
C 58.39, H 6.92, N 2.00; found: C 58.28, H 6.87, N 2.07%.
Anchoring of 3 on MCM-41. 2 mmol of 3 was dissolved in 15
ml THF. This solution was quantitatively transferred to 2 g
MCM-41 that was dried overnight at 150 ꢀC under vacuum.
After 24 h stirring at RT the heterogeneous catalyst 4 was fil-
tered off under nitrogen atmosphere and washed with cold
THF (5 ꢂ 15 ml). ICP-MS measurements reveal that the Ru
amount in the washing phase (unreacted part of the initial 2
mmol) is compatible with the 0.89 wt% found for 4.
2.2. MCM-41
MCM-41 was synthesised as described previously.13 After cal-
cination the mesoporous material was characterised by XRD,
N2 adsorption and Raman spectroscopy. MCM-41 is dried
overnight in vacuo at 423 K to achieve complete thermodesorp-
tion of physically adsorbed water from the silica surface.
2.4. ROMP experiments
2.3. Synthesis of the heterogeneous catalyst 4 (Scheme 2)
In a typical ROMP experiment 0.005 mmol of the catalyst sus-
pension in toluene was transferred into a 15 ml vessel followed
by the addition of a catalytic amount (2 equiv.) of trimethylsi-
lyldiazomethane (TMSD) diluted in 1 ml toluene via a preci-
sion syringe over 0.5 h to allow the formation of the
initiating metal carbene species. Then the monomer solution
in toluene (800 equivalents for norbornene, 200 equivalents
for cyclooctene) was added and the reaction mixture was kept
stirring at 85 ꢀC for 3 hours. To stop the polymerization reac-
tion, 2–3 ml of an ethyl vinyl ether/2,6-di-tert-butyl-4-methyl-
phenol (BHT) solution is added and the solution is stirred till
the deactivation of the active species is completed. The solu-
tion is poured into 50 ml methanol (containing 0.1% BHT)
and the polymers are precipitated and filtered off. To remove
the remaining heterogeneous catalyst from the polymers, they
are dissolved in CHCl3 so that the catalyst can be filtered off.
The CHCl3 is then removed in vacuo from the polymer solution
until a high viscosity is reached after which the polymers are
precipitated by adding 100 ml methanol. The as-obtained
white polymers are then filtered off and dried in vacuum over-
night.
Synthesis of the organomagnesium compound 2. To a suspen-
sion of 2 mmol Mg powder in THF (10 ml), a solution of
2.5. RCM experiments
All reactions were performed on the bench top in air by weigh-
ing 5 mol% of the catalyst into a dry 10 ml vessel and suspend-
ing the solid in 2 ml toluene. Then we added a catalytic amount
(2 equiv.) of trimethylsilyldiazomethane (TMSD) diluted in
1 ml toluene via a precision syringe over 0.5 h to allow the for-
mation of the initiating metal carbene species. A solution of
the appropriate substrate (0.1 mmol) in toluene (2 ml) was
added, together with the internal standard dodecane. The reac-
tion mixture was stirred for 17 hours at 85 ꢀC. Product forma-
tion (all reaction products were unambiguously identified
previously4b and diene disappearance were monitored by GC
1
analysis and confirmed in reproducibility experiments by H-
NMR spectroscopy through integration of the allylic methy-
lene peaks (here the solvent was deuterated toluene and the
internal standard 1,3,5-mesitylene). GC analysis of the reac-
tion mixture also ruled out the formation of cycloisomers,
oligomers or telomers.
Scheme 2 Synthesis of the heterogeneous catalyst 4.
New J. Chem., 2002, 26, 1201–1208
1203

View Article Online
2.6. ATRP experiments
In a typical ATRP experiment 0.0117 mmol of catalyst was
placed in a glass tube (in which the air was expelled by three
vacuum–nitrogen cycles) containing a magnet barr and capped
by a three-way stopcock. Then the monomer and initiator were
added so that the molar ratios [catalyst]/[initiator]/[monomer]
were 1/2/800. All liquids were handled under argon with dried
syringes. The reaction mixture was then heated for 17 h at the
reaction temperature, which was 85 ꢀC for the acrylates and
methacrylates and 110 ꢀC for styrene. After cooling, it was
diluted in THF and poured in 50 ml n-heptane (for the acry-
lates and methacrylates) or 50 ml methanol (for styrene) under
vigorous stirring after which the precipitated polymer was fil-
tered with suction. To remove the remaining heterogeneous
catalyst from the polymers, they were dissolved in CHCl3 so
that the catalyst could be filtered off. The CHCl3 was then
removed in vacuo from the polymer solution until a high visc-
osity was reached after which the polymers were precipitated
by adding 100 ml n-heptane (for the acrylates and methacry-
lates) or 100 ml methanol (for styrene). The as-obtained white
polymers were then filtered off and dried in vacuum overnight.
Fig. 3 XRD pattern of the pristine MCM-41.
3.1.2. N2 adsorption analysis. The pore size distributions
(PSD) of both the pristine MCM-41 and the heterogeneous
catalyst were calculated using the desorption branches of the
N2 adsorption isotherms and the Barrett–Joyner–Halenda
(BJH) algorithm14 (Fig. 4).
The BJH algorithm is a classical method that treats
adsorbed molecules as a separate, homogeneous, compressible
phase in equilibrium with the gas phase. The method uses Har-
kins and Jura’s thickness equation (which gives the thickness
of the adsorbed layer of molecules on a flat surface at a given
pressure), and the Kelvin equation to account for capillary
effects in the porous solid. Since the Kelvin equation is not
valid for pores below 1.8 nm, the BJH approach can only be
used for mesoporous materials.
2.7. Kharasch addition experiments
All reactions were performed on the bench top in air by weigh-
ing 0.01 mmol of the catalyst into a dry 10 ml vessel and sus-
pending the solid in 2 ml toluene. Then the solution of alkene
(3 mmol), CCl4 (4.33 mmol) and dodecane (0.083 ml) in
toluene (1 ml) were added after which the reaction mixture
was heated at 85 ꢀC for 17 h. The yields were obtained by
GC analysis of the reaction mixture using dodecane as internal
standard.
2.8. Vinylation experiments
In a typical vinylation experiment, 4.4 mmol of carboxylic
acid, 4.4 mmol of alkyne and 0.04 mmol of catalyst were trans-
ferred in a 15 ml glass vessel containing 3 ml toluene. Then the
reaction mixture was heated for 4 hours at 100 ꢀC under an
inert atmosphere. The total yield was determined with Raman
spectroscopy by following the diminishing intensity of the nC
of phenylacetylene and octadiyne and using a calibration curve
(Conformation was obtained by GC/MS). The selectivities
were determined by GC/MS making use of the different frag-
mentations of the isomers. GC/MS measurements excluded
also the formation of other products than those reported here.
Specific surface areas were calculated using the BET (Bru-
nauer, Emmett and Teller) equation. The BET equation was
developed to estimate the specific surface area of non-porous
solids.15 Adsorption in non-porous solids can be modelled as
adsorption on a flat surface. In a cylindrical mesopore of
ꢃ3.0 nm in diameter, the formation of the monolayer is only
negligibly affected by the wall curvature and by the presence
of the opposite wall surface. Since in the case of nitrogen
adsorption within MCM-41, the size (ꢃ0.3 nm) of the adsorp-
tive molecule is small compared with the size of the pore dia-
meter (3.0 nm), the BET equation can be justifiably used to
estimate the specific surface area of this solid.
=
C
The data obtained from the N2 adsorption measurements
and the XRD analyses are summarized in Table 1. The
observed specific surface area for the pristine MCM-41 is
1303 m2 gꢁ1 and the pore volume is 1.0213 cm3 gꢁ1. The meso-
pore diameter of the pristine MCM-41, which has a value of
2.53 nm, was obtained from the PSD curve (Fig. 4, A). The
3. Results and discussion
3.1. Characterisation
3.1.1. X-Ray diffraction (XRD). The XRD pattern of the
pristine MCM-41 sample (Fig. 3) has a sharp d100 peak as well
as secondary diffractions observed at larger angles (d110 and
d200), indicating the presence of hexagonally arranged pore
structures, which are characteristic for MCM-41 solids. The
d100 spacing amounts to 3.714 nm. The hexagonal unit cell
parameter a0 , a measure of the spacing between the hexagonal
p
layers, is calculated as 2d100
/
3. This parameter is related to
the average distance between pores in a two-dimensional fra-
mework, and for an ideal structure the parameter is equal to
the internal pore diameter plus the thickness of the pore walls.
For the pristine MCM-41 we find that a0 ¼ 4.289 nm. As the
pattern of the XRD spectrum is the same for the heteroge-
neous catalyst and the pristine MCM-41, we can conclude that
the anchoring of the catalyst via a spacer molecule did not
affect the hexagonal long-range-ordered structure of the meso-
porous MCM-41. For the heterogeneous catalyst the d100 spa-
cing and a0 amount to respectively 3.692 nm and 4.26 nm.
Fig. 4 Pore size distributions of both the pristine MCM-41 (A) and
the heterogeneous catalyst 4 (B).
1204
New J. Chem., 2002, 26, 1201–1208

View Article Online
the presence of silanol groups and water. In the 29Si CP
MAS NMR spectrum of the pristine MCM-41 three different
peaks at 90 ppm, 100 ppm and 110 ppm are observed.
These values can be attributed to respectively Si(OH)2(OSi)2 ,
Si(OH)(OSi)3 and Si(OSi)4 . The proton spectrum of the
MCM-41 + the spacer molecule (bromopropyltriethoxysilane)
only reveals the presence of –CH2 and –CH3 groups. However,
the 13C CP MAS NMR spectrum of this sample reveals some
interesting features. Two peaks at 50 ppm and 36 ppm can be
attributed to respectively a –OCH2– and a –CH2Br– configura-
tion. Furthermore, at 11 ppm there is an overlapping of the
bands assigned to –CH3 and –SiCH2– groups. The 29Si CP
MAS NMR spectrum of the MCM-41 + spacer reveals unam-
biguously the presence of a (SiO)3Si*C– species at ꢁ58.19 ppm
and a (SiO)2(OEt)Si*C– species at ꢁ106.5 ppm. The proton
spectrum of the heterogeneous hybrid catalyst only reveals
the presence of aromatic and aliphatic protons as broad unre-
solved peaks. At 9.80 ppm a small peak of the imine proton is
present. The 13C CP MAS NMR spectrum of the hetero-
geneous catalyst reveals around 166 ppm the carbon of the
Table 1 Comparison of the properties of pristine MCM-41 and the
heterogeneous catalyst 4
Specific
surface
area/m2 gꢁ1
Pore
volume/
cm3 gꢁ1
Pore
diameter/
nm
Wall
thickness/
nm
MCM-41
Heterogeneous
catalyst 4
1303
625
1.0213
0.6216
2.53
2.43
1.76
1.83
wall thickness is derived from the position of the [100] peak
(d100 and the specific mesopore diameter. Wall thick-
)
ness ¼ [a0 ꢁ dMCM-41] ¼ 1.76 nm. The PSD of the heteroge-
neous catalyst (Fig. 4, B) clearly shows that the pore size of
the MCM-41 is slightly decreased from 2.53 nm to 2.43 nm.
Moreover we also see a reduction of the pore volume and
the specific surface area (respectively from 1.0213 to 0.6216
cm3 gꢁ1 and from 1303 to 625 m2 gꢁ1). All these results indi-
cate that the internal pores of the MCM-41 are occupied by
the catalytic complex (compound 4, Scheme 2) and that the
accessibility of the mesopores is maintained after modification.
=
–C N– bond. Again the aromatic and aliphatic carbon atoms
can be revealed from the spectrum. Around 9.8 ppm the over-
lapping of the –CH3 and the –SiCH2– peaks is observed again.
The 29Si CP MAS NMR spectrum of the heterogeneous cata-
lyst also reveals the presence of a (SiO)3Si*C– species at
ꢁ59.69 ppm and a (SiO)2(OEt)Si*C– species at ꢁ106.0 ppm.
From all this we can conclude that the anchoring of the homo-
geneous catalyst via the spacer molecule onto the MCM-41 can
take place with two or three covalent bonds. Furthermore,
these results prove that it is the spacer molecule that links
the catalytic metal centre to the mesoporous carrier.
3.1.3. Raman spectroscopy. The anchoring of the spacer
modified homogeneous complex (compound 3, Scheme 2)
was also checked using Raman spectroscopy (Fig. 5).
The low intensity scattered by the mesoporous support
makes it possible to examine the grafting of the homogeneous
catalyst. The comparison of the Raman spectra of the pristine
MCM-41 (Fig. 5, A) and the heterogeneous catalyst 4 (Fig. 5,
C) clearly shows the grafting of the homogeneous species.
Comparison of the Raman spectrum of the spacer modified
homogeneous catalyst 3 (Fig. 5, B) and hybrid catalyst 4
(Fig. 5, C) was performed to eliminate any doubt concerning
the chemical attachment of the homogeneous catalyst. We
clearly see that every peak in the spectrum of the homogeneous
catalyst 3 is also present in the spectrum of the heterogeneous
catalyst 4. The small shifts of some peaks in Fig. 5, C com-
pared with Fig. 5, B indicate the change in chemical environ-
ment of the different functional groups originating from the
chemical attachment of the catalyst to the carrier.
3.2. Catalysis
3.2.1. Ring opening metathesis polymerization (ROMP).
The results of the ROMP experiments with catalytic systems
14a and 4 are summarized in Table 2. For details concerning
the reaction conditions see the Experimental section.
From the results gathered in Table 2 it is clear that the het-
erogeneous catalyst 4 reaches lower yields than its homoge-
neous analogue 1, and this is the case for both the strained
norbornene and the less-strained cyclooctene. However, look-
ing at the polydispersities (PDI) of the formed polymers
reveals that with the heterogeneous system a much more con-
trolled polymerization occurs (PDI 6.95 versus 2.34 for norbor-
nene and 4.62 versus 2.95 for cyclooctene). This is an
indication that only one heterogeneous catalytically active spe-
cies is formed, but still diffusion limitations occur in the chan-
nels of the MCM-41. The anchoring of the homogeneous
catalyst probably limits the appearance of backbiting reactions
because the channels induce a certain steric hindrance which
3.1.4. X-Ray fluorescence (XRF) and inductive coupled
plasma/mass spectrometry (ICP/MS). XRF measurements
reveal a loading of 0.088 mmol (0.89 wt% Ru) Ru complex/
g heterogeneous catalyst. This loading was confirmed by
ICP/MS analysis of the heterogeneous hybrid catalyst.
3.1.5. Solid state NMR analysis. The structure of the het-
erogeneous catalyst 4 was also studied by solid state NMR.
For the pristine MCM-41, the proton spectrum only reveals
Fig. 5 Raman spectra of the pristine MCM-41 (A), the spacer modified homogeneous catalyst 3 (B) and the hybrid catalyst 4 (C).
New J. Chem., 2002, 26, 1201–1208
1205

View Article Online
Table 2 ROMP of norbornene and cyclooctene using catalytic systems 1 and 4
1/4
Yield (%)
Mn (ꢂ103)
Mw (ꢂ103)
PDI
cis/trans
norbornene (800 eq., 85 ꢀC, 3 h)
cyclooctene (200 eq., 85 ꢀC, 3 h)
98/73
71/65
98/69
57/44
682/163
262/131
6.95/2.34
4.62/2.95
0.39/0.35
0.38/0.40
impedes the polymers from folding themselves into a position
suitable to perform backbiting reactions. Also the fact that dis-
mutation and coupling reactions are eliminated through the
linking of the homogeneous species to the mesoporous carrier
may be attributed to a narrowing of the PDI of the polymers
formed with the heterogeneous system.
both systems succeed in yielding the ring-closed tetra-substi-
tuted malonate derivative (entry 3) in moderate yield.
3.2.3. Atom transfer radical polymerization (ATRP). The
results of the ATRP experiments using catalytic systems 14a
and 4 are gathered in Table 4. For details concerning the reac-
tion conditions see the Experimental section.
The results gathered in Table 4 show a marked decrease in
% yield when the heterogeneous system is used for conducting
the ATRP experiments. Moreover, the decrease in initiation
efficiency and the increasing polydispersities of the formed
polystyrene and polymethyl methacrylate indicate a loss of
control over polymerization when the heterogeneous catalytic
system 4 is used. From these results it is obvious that for
ATRP reactions, the immobilisation of the homogeneous cat-
alyst 1 on a mesoporous silica carrier via a spacer molecule is
not advantageous for the polymer properties.
3.2.2. Ring closing metathesis (RCM). The RCM activities
of catalysts 14a and 4 are summarised in Table 3. For details
concerning the reaction conditions see the Experimental sec-
tion.
An important trend that can be observed from the results in
Table 3 is the comparable yields of the heterogeneous and the
homogeneous catalytic system. Both systems quantitatively
convert diethyldiallyl malonate (entry 1), 1,7-octadiene (entry
4) and diallyl ether (entry 5). Moreover both systems also
smoothly convert the tri-substituted malonate derivative (entry
2), diallylphthalate (entry 6) and linalool (entry 7). In addition,
3.2.4. Kharasch addition. In Table 5 the results of the
Kharasch addition experiments of CCl4 across some represen-
tative olefins and using catalytic systems 14a and 4 are sum-
marised. For details concerning the reaction conditions see
the Experimental section.
Table 3 RCM of some representative substrates with catalytic sys-
tems 1 and 4 (E ¼ COOEt)
It is clear that for Kharasch addition reactions of CCl4
across the olefins depicted in Table 5, the % yields obtained
with the heterogeneous system 4 are slightly lower in compar-
ison withthe homogeneous compound 1. However, the very
easy separation of the catalyst from the reaction mixture in
the case of the heterogeneous compound 4 (the separation
can be done by simple filtration) makes the latter system much
more suitable for performing Kharasch addition reactions
than its homogeneous counterpart 1.
Isolated yield (%)
Entry
1
Substrate
Product
1
4
100
100
2
3
71
23
73
21
3.2.5. Vinylation. In Table 6 the results of the vinylation
reactions using catalytic systems 14a and 4 are summarised.
For details concerning the reaction conditions see the Experi-
mental section.
4
5
100
100
100
100
When octadiyne is used as a substrate, the addition of both
carboxylic acids result in the selective formation of (E)-alk-1-
enyl esters corresponding to a regio- and stereoselective anti-
Markovnikov addition of the acid to the triple bond and this
is irrespective of the catalytic system used. The percentage of
(E)-alk-1-enyl ester is for the two catalytic systems and both
acids more or less the same and varies in the range of 74%–
79%. The total yield, however, is dependent on the type of cat-
alyst and acid used. Whereas for formic acid the two catalytic
compounds reach comparable total yields, for acetic acid the
heterogeneous system 4 shows a somewhat higher activity.
6
7
94
76
89
78
Table 4 ATRP of some representative substrates using catalytic systems 1 and 4
Yield (%)
Mn (ꢂ103)
Mw (ꢂ103)
PDI
fia
Methyl methacrylate
Isobutyl methacrylate
Methyl acrylate
n-Butyl acrylate
Styrene
84/76
62/53
54/38
35/26
95/86
48/60
—
—
84/126
—
—
1.75/2.11
—
0.71/0.51
—
—
—
—
—
—
54/63
—
98/134
1.81/2.15
0.70/0.57
a
fi ¼ Initiation efficiency ¼ Mn,theor./Mn,exp. with Mn,theor. ¼ ([monomer]0/[initiator]0) ꢂ MW(monomer) ꢂ conversion.
1206
New J. Chem., 2002, 26, 1201–1208

View Article Online
Table 5 Karasch addition reaction of CCl4 of some representative
olefins using catalytic systems 1 and 4
no diffusion limitations occur under these test conditions. To
test the leaching of the catalytic system we filtered off the cat-
alyst after terminating the reaction mixture as described in the
Experimental section (reaction conditions: monomer/Ru/
TMSD ¼ 50/1/2, 5 ml toluene, 85 ꢀC, 1 h). After drying the
catalyst in vacuo for 1 night a sample for XRF analysis was
taken. The XRF measurement reveals that the Ru content of
the MCM-41 was the same as before polymerization (the
variability on XRF analyses is around ꢄ6% of the measured
value), indicating that possible leaching was below the detec-
tion limit. The recycling capabilities of our heterogeneous
hybrid catalytic system were tested in RCM experiments. As
a test substrate we used diethyldiallyl malonate. The procedure
followed to study the reusability of the heterogeneous catalyst
was the following: after 2 hours of reaction the catalyst was fil-
tered off under nitrogen atmosphere, dried in vacuo for 2 hours
and tested again (the other reaction condition parameters are
the same as mentioned in the Experimental section). Whereas
the first run gives a conversion of 100%, the second, the third
and the fourth runs give conversions of respectively 98%, 96%
and 96% after 1 hour of reaction. In 1998 Sheldon et al.16 high-
lighted that the conventional practice of recycling a heteroge-
neous catalyst without observing any significant loss of
activity is in fact no proof of heterogeneity. It appears now
commonplace for true heterogeneity to be tested by filtering
the heterogeneous catalyst at the reaction temperature (before
completion of the reaction) and testing the filtrate for activity.
When we applied this procedure to a RCM experiment with
diallylphthalate (reaction conditions the same as mentioned
in the Experimental section, but the reaction was stopped after
4 hours), we observed that the filtrate showed no further RCM
activity, as no further increase of the conversion was observed,
even after keeping the filtrate for an additional 17 hours at
85 ꢀC.
Yield (%) using 1/4
Methyl methacrylate
Isobutyl methacrylate
Methyl acrylate
n-Butyl acrylate
Styrene
84/78
52/50
33/29
18/15
91/85
For example, with formic acid a comparable total yield of
around 70% is obtained for both catalysts. With acetic acid
however, up to 82% of total conversion is reached with the het-
erogeneous system 4, whereas the homogeneous system 1 only
reaches 76% total yield for the addition of acetic acid to octa-
diyne. Besides the formation of the (E)-alk-1-enyl ester, also a
small percentage of (Z)-alk-1-enyl ester, Markovnikov addi-
tion products and disubstituted enol esters are obtained.
When phenylacetylene is used as the alkyne source, the
dimerization reaction that was mentioned in the Introduction
becomes very important. With compound 4, almost no nucleo-
philic addition of the carboxylic acid to the triple bond occurs
and the (E)-enyne is formed with a selectivity of at least 85%
reaching an almost quantitative total yield and this occurs
for both acids. Compound 1 shows totally different behaviour.
When using phenylacetylene, the outcome of the reaction
depends on the carboxylic acid used. With formic acid, this
system forms selectively the (E)-alk-1-enyl ester. Performing
the reaction with acetic acid leads to the stereoselective forma-
tion of the (E)-enyne. In this case, system 1 reaches an excellent
total yield of 96% and a selectivity of 78% for the formation of
the (E)-enyne.
To our knowledge, system 4 is the most active heteroge-
neous ruthenium catalyst reported so far for (i) the selective
formation of (E)-alk-1-enyl esters from alkadiynes and satu-
rated carboxylic acids and (ii) the selective dimerization of
mono-alkynes into the corresponding (E)-enyne.
4. Conclusion
In conclusion, we succeeded in synthesising a stable and recycl-
able heterogeneous catalyst that was also fully characterised.
Moreover, filtrate tests and XRF measurements reveal negligi-
ble leaching of the active species from the carrier. Despite the
somewhat lower conversions of the heterogeneous system in
comparison with the homogeneous system in ROMP reac-
tions, the hybrid catalyst provides polymers with narrower
polydispersities, indicating more controlled polymerization
and diminished sensitivity of the heterogeneous system
towards backbiting reactions. In RCM experiments and Khar-
asch addition reactions, the heterogeneous catalyst shows
respectively a comparable or slightly lower catalytic activity
than its homogeneous counterpart. The very easy separation
of the catalyst from the reaction products, however, certainly
turns the scale in favour of the hybrid heterogeneous system.
3.2.6. Stability and recycling of the heterogeneous hybrid
catalyst. For the heterogeneous catalyst we also tested the sta-
bility (test reaction is ROMP using norbornene as a substrate,
reaction conditions: monomer/Ru/TMSD ¼ 100/1/2, 2 ml of
toluene, 85 ꢀC, 1 h). Stability of a supported heterogeneous
catalyst under reaction conditions means (i) no leaching of
the active species into solution and (ii) no degradation of the
active form of the catalyst during reaction. To test this we
added a second portion of substrate to the completely poly-
merised system. This second polymerization was also quantita-
tive. Even after adding a fifth portion of NBE the conversion
was still 92%. These results show that the active site of the cat-
alyst remains intact during the above-mentioned test and that
Table 6 Vinylation reactions of some representative substrates using catalytic systems 1 and 4a
Total yield (%)
% M.
% anti-M.(Z)
% anti-M.(E)
% head-to-tail enyne
% (Z)-enyne
% (E)-enyne
1
4
ph.ac.
ph.ac.
f. ac.
ac.ac.
f. ac.
ac.ac.
74
96
99
93
12
3
—
2
79
9
—
3
4
5
5
2
5
78
85
92
2
—
—
—
8
—
—
6
% disubstituted enol ester
1
4
octad.
octad.
f. ac.
ac.ac.
f. ac.
ac.ac.
71
76
70
82
11
13
16
14
6
5
6
3
76
79
74
78
7
3
4
5
a
ph.ac. ¼ phenylacetylene, octad. ¼ octadiyne, f.ac. ¼ formic acid, ac.ac. ¼ acetic acid, % M. ¼ percentage Markovnikov addition product, %
anti-M. ¼ percentage anti-Markovnikov addition product.
New J. Chem., 2002, 26, 1201–1208
1207

View Article Online
F. Verpoort and Catal. Lett., 2001, 74, 155; ( j) K. Melis, D. De
Vos, P. Jacobs and F. Verpoort, J. Mol. Catal. A: Chem., 2001,
169, 47.
(a) B. D. Clereq and F. Verpoort, J. Mol. Catal. A: Chem., 2002,
180, 67; (b) B. D. Clereq and F. Verpoort, Tetrahedron Lett., 2001,
42, 8959.
(a) M. Randall and M. Snapper, J. Mol. Catal. A: Chem., 1998,
133, 29; (b) K. Ivin, J. Mol. Catal. A: Chem., 1998, 133, 1.
(a) M. Kato, M. Kamigaito, M. Sawomoto and T. Higashimura,
Macromolecules, 1995, 28, 1721; (b) M. Matsuyama, M. Kami-
gaito and M. Sawomoto, J. Polym. Sci. Part A: Polym. Chem.,
1996, 34, 3585; (c) J. S. Wang and K. Matyjaszewski, J. Am.
Chem. Soc., 1995, 117, 5614; (d ) J. S. Wang and K. Matyjas-
zewski, Macromolecules, 1995, 28, 7901.
J. Iqbal, B. Bhatia and N. K. Nayyar, Chem. Rev., 1994, 94, 519.
(a) H. Doucet, B. Martin-Vaca, C. Bruneau and P. H. Dixneuf, J.
Org. Chem., 1995, 60, 7247; (b) K. Melis, P. Samulkiewicz, J. Ryn-
kowski and F. Verpoort, Tetrahedron Lett., 2002, 43, 2713.
(a) J. A. S. Howell, N. F. Ashford, D. T. Dixon, J. C. Kola, T. A.
Albright and S. K. Kang, Organometallics, 1991, 10, 1852; (b) C.
Bruneau and P. H. Dixneuf, Acc. Chem. Res., 1999, 32, 311.
For ATRP reactions the immobilisation of the homogeneous
system does not lead to any advantages for the obtained poly-
mers. To conclude, vinylation experiments reveal that system 4
is the most active heterogeneous ruthenium catalyst reported
so far for (i) the selective formation of (E)-alk-1-enyl esters
out of alkadiynes and saturated carboxylic acids and (ii) the
selective dimerization of mono-alkynes into the corresponding
(E)-enyne.
Taking into account economical and environmental consid-
erations, the handling of immobilised Ru complexes presents
many advantages, such as easier separation and recovery from
the reaction mixture thus enhancing recycling possibilities. In
this perspective the above-mentioned hybrid catalytic system
holds great promise for eco-friendly synthesis of organic inter-
mediates.
4
5
6
7
8
9
Acknowledgements
10 (a) C. U. Pittman, in Polymer-supported Reactions in Organic
Synthesis, eds. P. Hodge and D. C. Sherrington, Wiley, Chiche-
ster, 1981, ch. 5, p. 249; (b) F. R. Hartley, Supported Metal Com-
plexes – a New Generation of Catalysts, Reidel, Dordrecht, 1985;
(c) S. J. Shuttleworth, S. M. Allin and P. K. Sharma, Synthesis,
1997, 1217; (d ) C. Bolm and A. Gerlach, Eur. J. Org. Chem.,
1998, 63, 21.
11 (a) J. H. Clark and D. J. Macquarrie, Org. Proc. Res. Dev., 1997,
1, 149; (b) S. Ernst, R. Gla¨ser and M. Selle, Stud. Surf. Sci. Catal.,
1997, 105, 1021; (c) J. H. Clark and D. J. Macquarrie, Chem. Soc.
Rev., 1996, 25, 303; (d ) A. Heckel and D. Seebach, Angew. Chem.,
Int. Ed., 2000, 39, 1.
B.D.C. is indebted to the IWT (Vlaams instituut voor de bev-
ordering van het wetenschappelijk-technologisch onderzoek in
de industrie) for a research grant. F.V. is indebted to the
FWO-Flanders (Fonds voor wetenschappelijk onderzoek-
Vlaanderen) for financial support and to the Research funds
of Ghent University.
References
12 (a) J. H. Clark and D. J. Macquarrie, Chem. Commun., 1998, 853;
(b) H. U. Blaser, B. Pugin, in Chiral Reactions in Heterogeneous
Catalysis, eds. G. Jannes and V. Dubois, Plenum, New York,
1995, pp. 33–57; (c) D. Seebach, R. E. Marti and T. Hintermann,
Helv. Chim. Acta, 1996, 79, 1710; (d ) H. Sellner and D. Seebach,
Angew. Chem., Int. Ed., 1999, 38, 1918.
13 I. Vankelecom, N. Moens, K. Vercruysse, R. Parton and P.
Jacobs, Studies in Surface Science and Catalysis: Heterogeneous
Catalysis and Fine Chemicals IV, eds. H. U. Blaser, A. Baiker
and R. Prins, Elsevier, Amsterdam, 1997, p. 724.
14 E. P. Barrett, I. S. Joyner and P. Halenda, J. Am. Chem. Soc.,
1951, 73, 373.
15 S. J. Kregg and K. S. W. Sing, Adsorption, Surface area and Por-
osity, 2nd edn., Academic Press, New York, 1982.
16 R. A. Sheldon, M. Wallau, I. W. C. E. Arends and U. Schuchardt,
Acc. Chem. Res., 1998, 31, 485.
1
(a) Pei-Shing and E. Dai, Catal. Today, 1995, 26, 3; (b) R. A. Shel-
don, J. Mol. Catal. A, 1996, 107, 75.
2
3
J. H. Clark and D. J. Macquarrie, Chem. Soc. Rev., 1996, 25, 303.
(a) E. R. Hartley, Catalysis by Metal Complexes, Supported Metal
Reagents, Reidel, Dordrecht, 1984; (b) A. J. Sandee, L. A. Van
Der Veen, J. N. H. Reek, P. C. J. Kamer, M. Lutz, A. L. Spek
and P. W. N. M. Van Leeuwen, Angew. Chem., Int. Ed., 1999,
38, 3231; (c) S. T. Nguyen and R. H. Grubbs, J. Organomet.
Chem., 1995, 497, 195; (d ) M. Ahmed, A. G. M. Barrett, D. C.
Braddock, S. M. Cramp and P. A. Procopiou, Tetrahedron Lett.,
1999, 40, 8657; (e) M. K. Dalal, M. J. Upadhyay and R. N. Ram,
J. Mol. Catal. A: Chem., 1999, 142, 325; ( f ) H. C. L. Abbenhuis,
Angew. Chem., Int. Ed., 1999, 38, 1058; (g) C.-J. Liu, W.-Y. Yu,
S.-G. Li and C.-M. Che, J. Org. Chem., 1998, 63, 7364; (h)
S. B. Garber, J. S. Kingsbury, B. L. Gray and A. H. Hoveyda,
J. Am. Chem. Soc., 2000, 122, 8168; (i) T. Opstal, K. Melis and
1208
New J. Chem., 2002, 26, 1201–1208