Immobilization of rhodium diphosphine complexes on mesoporous Al-MCM-41 materials: Catalysts for enantioselective hydrogenation

Article abstract of DOI:10.1006/jcat.2001.3296

Chiral heterogeneous catalysts for the enantioselective hydrogenation of prochiral alkenes were prepared from chiral rhodium diphosphine complexes and Al-MCM-41. TPD of ammonia was carried out at 1 atm. The ionic interaction of the cationic complex with the anionic host framework may account for the bonding forces observed. A decrease of the weak acidic sites of Al-MCM-41 was also noted. These catalysts are suitable for the hydrogenation of functionalized olefins. The organometallic complexes remained stable within the mesopores of the carrier under the reaction conditions. The catalyst can be recycled by filtration or centrifugation. There was no leaching of the homogeneous complex observed.

Full text of DOI:10.1006/jcat.2001.3296

Journal of Catalysis 203, 150–156 (2001)
doi:10.1006/jcat.2001.3296, available online at http://www.idealibrary.com on
Immobilization of Rhodium Diphosphine Complexes on Mesoporous
Al-MCM-41 Materials: Catalysts for Enantioselective Hydrogenation
H. H. Wagner, H. Hausmann, and W. F. Ho¨lderich1
Department of Chemical Technology and Heterogeneous Catalysis, University of Technology RWTH Aachen,
Worringer Weg 1, 52074 Aachen, Germany
Received February 27, 2001; revised May 21, 2001; accepted May 23, 2001
ported liquid-phase catalysts (10). Recent reviews reveal
the potential of heterogeneous chiral catalysts created by
the methods mentioned (11–13).
Heterogeneous chiral catalysts for the enantioselective hydro-
genation of prochiral alkenes are prepared from chiral rhodium
diphosphine complexes and Al-MCM-41. Impregnation of the
mesoporous carrier Al-MCM-41 with the organometallic complexes
in dichloromethane leads to strong and not leaching hydrogenation
catalysts. The catalysts are characterized by spectroscopic meth-
ods such as FT-IR and MAS-NMR as well as thermoprogrammed
desorption of ammonia, thermogravimetric analysis, and nitrogen
sorption experiments. The hydrogenation of dimethylitaconate was
studied as a test reaction. The immobilized catalysts showed high
activity and excellent regio- and enantioselectivity. Up to 92% ee,
100% conversion, and 99% regioselectivity were observed for the hy-
drogenation of dimethyl (R)-methylsuccinate. The catalysts could
be reused without any regeneration procedure and with no loss of
catalytic acitivity. Leaching of the homogeneous complex out of the
c
Nanosized channels of ordered mesoporous materials
related to the M41S family can be tailored to pore sizes
ranging from 1.5 nm to ca. 100 nm (14). MCM-41-type
molecular sieves are suitable as carriers for transition metal
complexes, offering new opportunities for the encapsula-
tion of large catalyst species and for the catalytic conversion
of substrates much larger than in common zeolites (15). The
application of such host/guest compounds as catalysts for
oxidative reactions has been successfully tested (16–21).
Homogeneous rhodium diphosphine complexes belong
to the most important catalysts in enantioselective hy-
drogenation reactions. We previously prepared surface-
bonded rhodium phosphine complexes in Al-MCM-41.
In a solution of dichloromethane, [Rh(acac)(chiraphos)]
and Al-MCM-41 react to form a surface-bonded Rh-
chiraphos complex due to an exchange reaction of the
acetylacetonato ligand and surface oxygen of the acidic
support (22). However, the occupation of the coordina-
tion sphere of the catalytic active Rh site led to the com-
plete loss of chiral induction of this complex. Still it was
our aim to immobilize such complexes without the in-
troduction of a linker to the ligand. This method would
then allow us to successfully heterogenize homogeneous
catalysts without modification of their chemical structure.
Recyclable catalysts have been prepared in such a man-
ner by anchoring rhodium complexes on heteropolyacids
(23).
MCM-41 framework was not observed.
2001 Academic Press
Key Words: immobilization; rhodium; phosphine; MCM-41;
asymmetric; hydrogenation.
1. INTRODUCTION
The immobilization of homogeneous catalyst systems is
an attractive challenge because it opens access to the prepa-
ration of new, environmentally friendly means of chemical
synthesis. Although homogeneous catalysts show remark-
able performance for a large variety of reactions, technical
problems such as separation, recovery, and recyclization
of the soluble catalysts create the demand for heteroge-
neous catalysts with comparable performance. Often enan-
tioselectively catalyzed reactions are exclusively performed
using organometallic transition metal complexes. Various
attempts toward the immobilization of organometallic
complexes have been investigated previously (1) such as
attaching to supporting materials by chemisorption (2, 3),
immobilization by steric hindrance in zeolitic micro- or
mesopores (“ship-in-the bottle” concept) (4–9), or sup-
In this paper we present new heterogeneous rhodium
diphosphine catalysts and their application in the enan-
tioselective hydrogenation of dimethylitaconate. The
results indicate the localization of the complex inside
the mesoporous channel system. Chiral heterogeneous
rhodium diphosphine catalysts without any linker group
on the complexes are prepared and successfully applied
in enantioselective hydrogenation. A similar method has
been published recently by R. D. Broene et al. (24).
1 To whom correspondence should be addressed. Fax: +49-241-8888291.
E-mail: hoelderich@rwth-aachen.de.
150
0021-9517/01 $35.00
c
Copyright
2001 by Academic Press
All rights of reproduction in any form reserved.

IMMOBILIZATION OF RHODIUM DIPHOSPHINE COMPLEXES
2. EXPERIMENTAL
151
Characterization
X-ray diffraction patterns of powdered samples were ob-
tained with a Siemens D5000 diffractometer equipped with
a rotating anode and Cu K radiation.
Nitrogen adsorption/desorption isotherms at liquid ni-
trogen temperature were measured on a Micromeritics
ASAP 2010 instrument. The samples were pre-outgassed
at 150 C. Pore diameter and specific pore volume were cal-
culated according to the Barrett–Joyner–Halenda (BJH)
theory. The specific surface area was obtained using the
Brunauer–Emmett–Teller (BET) equation.
All experiments were performed under purified inert gas
by Schlenk techniques unless otherwise stated. Solvents
were dried, and oxygen and moisture were removed by
standard methods. Due to those facts the new systems are
sometimes hard to reproduce in loading of the complex and
catalytic performance. This could also be due to the prop-
erties of the Al-MCM-41 carrier materials. Due to the high
air sensitivity of the organometallic complexes combined
with the high surface area of the heterogeneous catalysts,
the best results can be achieved only with freshly prepared
materials.
Chemical analysis was performed with inductively
coupled plasma atomic emission spectroscopy (ICP-AES)
(Spectroflame D, Spectro Analytic Instrument).
Synthesis of Al-MCM-41
For thermogravimetric analysis a Netzsch 209/2/E eq-
uipped with a STA 409 controller was used. The heating rate
was 5 C/min, and -Al2O3 was used as reference material.
The temperature-programmed desorption (TPD) of am-
monia was performed at atmospheric pressure using a flow-
type fixed-bed adsorber. Prior to the TPD experiments the
materials were calcined at 200 C in dry nitrogen for 8 h,
loaded with ammonia at 100 C, and purged with nitrogen
until a stable baseline formed. The desorption was done
in the temperature range 100–400 C at a heating rate of
5 C/min under nitrogen flow.
The solid-sate NMR spectra were recorded on a Bruker
DSX 500 spectrometer, equipped with a 2.5-mm CP/MAS
probehead. In the case of the 31P-MAS NMR spectra, 5000–
10,000 transients gave satisfactory signal-to-noise ratios.
The samples were rotated at 20,000 Hz. The spectra were
recorded at room temperature (296 K) and aqueous 85%
H3PO4 was used as the external standard.
For the synthesis of Al-MCM-41 the quartenary am-
monium surfactant (C14H29)N(CH3)Br was added under
constant stirring to a solution of tetraethylammoniumhy-
droxide (TEAOH) and NaAlO2 at ambient temperature
(14, 25). After 4 h colloidal silica (Ludox HS-40) was added
dropwise over a period of 1 h, followed by vigorous stirring
for 4 h. The Si/Al ratio of the gel was 40. Crystallization
took place over a 7-days period at 105 C. The resulting
white solid was washed with water and dried at 120 C. The
template was removed by heating the material to 540 C for
several hours.
Immobilization of Chiral Rhodium
Diphosphine Complexes
All solvents were purified, degassed, and saturated with
argon, and the preparation of the Rh complexes was car-
ried out carefully under argon by standard Schlenk tech-
niques. The complexes were prepared from [CODRhCl]2
and 1.1 equiv of chiral diphosphine in dichloromethane as
solvent. The carrier material Al-MCM-41 was then added.
After the mixture was stirred for 24 h, the solid was fil-
tered and washed thoroughly with dichloromethane un-
til the washings remained colorless in order to remove
any residual free Rh complex. After being dried in a vac-
uum the catalysts were Soxhlet extracted with methanol
for 24 h. The absence of the homogeneous complex in
the mixture was checked by ICP-AES analysis as well as
by FT-IR spectroscopy. The heterogeneous catalyst has a
pale yellow color similar to the color of the homogeneous
complex.
The reaction products were analyzed using a Siemens
RGC202 gas chromatograph equipped with a 25-m Lipodex
E capillary column.
3. RESULTS AND DISCUSSION
In order to get a first impression of whether the cho-
sen carrier system would be a suitable host for the ho-
mogeneous catalysts, a computer simulation of the struc-
tures of several Rh diphosphine complexes was carried
out whereby the Rh complexes were placed in a simulated
tube of Al-MCM-41 (Fig. 1). The graphical illustration of
CODRhDuphos fixed in Al-MCM-41 shows that the com-
plex fits easily in a tube of 2.2 nm.
Impregnation of Al-MCM-41 with chiral rhodium dip-
hosphine complexes leads to strong interaction of the com-
Hydrogenation of Dimethylitaconate
Hydrogenation of dimethylitaconate was carried out in plex in the mesoporous system of the carrier material
glass autoclaves at ambient temperature and 3 bar hydro- (Fig. 2). Dichloromethane was found to be an effective
gen pressure for 24 h. The catalyst was added to a solution solvent for the impregnation. After being stirred for 24 h
of 5 mmol of olefin in 10 ml of methanol. The molar ra- the yellow solution of 0.15 mmol of [(1,5-cyclooctadiene)
tio of olefin to rhodium (determined by atomic absorption Rh(1, 2-bis((2S, 5S)-2,5-dimethylphospholano) benzene)]+
spectroscopy) was 1000.
Cl (CODRhDuphos) complex per g of Al-MCM-41 in

¨
152
WAGNER, HAUSMANN, AND HOLDERICH
FIG. 1. Graphical illustration of CODRhDuphos fixed in Al-MCM-41.
CH2Cl2 solvent underwent complete decolorization. The The binding energy of this adsorption varies on the dif-
CODRhDuphos complex is prepared by the cleavage ferent reactive sites like Brønsted and Lewis acid sites
of the CODRhchloride dimer with the diphosphine lig- and silanol groups. Upon the extraction with methanol,
and Duphos. The carrier material turned yellow, indi- which in contrast to nonpolar dichloromethane adsorbs
cating the homogeneous catalyst loaded onto the sup- strongly on the Al-MCM-41 surface, the organometallic
port. Furthermore, the FT-IR spectra of the mother liquor complex desorbs from silanol groups due to a competi-
showed no signals corresponding to the free rhodium tive reaction with the polar alcohol. The amount of com-
phosphine complex. Soxhlet extraction of the impreg- plex immobilized on the carrier was determined by el-
nated Al-MCM-41 with methanol led to a partially leach- emental analysis. Rhodium contents vary between 0.02
ing of the complex from the carrier material. It is as- and 0.07 mmol per g of Al-MCM-41 depending on the
sumed that the organometallic complex is adsorbed on amount of complex offered and on the batch of Al-MCM-41
the inner and outer surfaces of the Al-MCM-41 structure. used.
FIG. 2. Impregnation of Al-MCM-41 with rhodium diphosphine complex.

IMMOBILIZATION OF RHODIUM DIPHOSPHINE COMPLEXES
153
FIG. 3. N2 adsorption and desorption isotherms of (a) Al-MCM-41
(carrier material) and (b) CODRhDuphos immobilized on Al-MCM-41.
FIG. 4. DSC spectra of (a) mechanical mixture of Al-MCM-41 and
CODRhDuphos and (b) CODRhDuphos immobilized on Al-MCM-41.
The X-ray diffraction pattern of the isolated yellow ma-
terial shows a strong peak corresponding to the (100) reflex,
indicating that the structure of the carrier material remains
unchanged during the immobilization procedure.
the complex is consistent with the content determined by
chemical analysis.
Temperature-programmed desorption (TPD) of ammo-
nia shows weak acidic centers on the Al-MCM-41 car-
rier material. In the case of the immobilized complex the
amount of desorbed ammonia decreases compared to the
pure carrier material (Fig. 5). This indicates an interac-
tion of the immobilized complex with acidic sites on the
Al-MCM-41. This result is also supported by FT-IR spec-
troscopy.
N2 adsorption isotherm of the supported catalyst shows
a decrease of ca. 10% of mesoporous pore volume com-
pared to the corresponding carrier material (Fig. 3 and
Table 1). The BET surface was found to decrease from
1430 m2/g in the unfunctionalized carrier material to
1250 m2/g for the loading material with a corresponding de-
crease in mesopore volume from 1.25 cm3/g to 1.07 cm3/g
(Table 1). These results indicate that the complex is de-
posited on the inner surface of the Al-MCM-41. The spe-
cific surface of the carrier was reduced by loading it with
the complex. This is consistent with partial mesopore filling
by the organometallic catalyst.
Quantitantive loadingofthe organometalliccomplexwas
demonstrated by thermal gravimetric analysis. Thermo-
gravimetric and differential scanning calorimetric (DSC)
measurements show that the immobilized complex is sta-
ble up to 250 C (Fig. 4). Oxidative decomposition of the
fixed complex took place in two steps at 295 C and 420 C
whereas a mechanical mixture of the homogeneous com-
plex with the carrier material exhibited no distinct peaks.
The loss of weight of ca. 4.5 wt% caused by the burning of
The infrared spectra show no change of wavenumber
1
but a decrease of intensity for the signal at ca. 3740 cm
which is assigned to the stretching vibration of terminal
silanol groups (Fig. 6). The vibration bands of immobi-
lized CODRhDuphos are similar to those in the solution
of dichloromethane. However, these signals resulting from
organic compounds are very weak and not characteristic
enough to surely identify or resolve a structure.
In order to elucidate the chemical structure of the im-
mobilized rhodium diphosphine complexes MAS-NMR
TABLE 1
Surface Areas and Pore Volumes of Al-MCM-41 (Carrier
Material) and of CODRhDuphos Immobilized on Al-MCM-41
Surface areaa
(m2/g)
Pore volumeb
(cm3/g)
Al-MCM-41
CODRhDuphos
1430
1250
1.25
1.07
immobilized on Al-MCM-41
FIG. 5. Temperature-programmed desorption (TPD) of ammonia of
(a) Al-MCM-41 (carrier material) and (b) CODRhDuphos immobilized
on Al-MCM-41.
a Calculated according to the BET method.
b According to the BJH theory.

¨
WAGNER, HAUSMANN, AND HOLDERICH
154
nuclei (25). This indicates that no new Al or Si species have
been formed upon reaction with the rhodium complexes. It
is also possible that a signal created by such a reaction is
too weak to be observed in the spectra of the immobilized
complexes.
31P-MAS NMR spectra of the solid homogeneous
CODRhDuphos complex and the same complex immobi-
lized on Al-MCM-41 are depicted in Fig. 7. The homoge-
neous complex gives a strong signal at 68 ppm and two
weaker ones at 78 and 90 ppm due to trace amounts of
oxidic impurities. A difference in the chemical shift was
observed between the solution spectrum of the homoge-
neous catalyst (76.3 ppm) (26) and the corresponding solid-
state spectrum (68 ppm relative to external H3PO4). This
difference could result from crystal-packing forces in the
solid state which could dictate a conformational arrange-
ment in the solid state not present in solution. The breadth
of the signals prevented an accurate determination of the
Rh–P coupling. The MAS NMR spectra of the immobilized
CODRhDuphos complex show only one signal at 90 ppm
(27). Since there is no signal of the homogeneous complex
and no signal of the free phosphine ligand visible in the
MAS spectra, we conclude that the phosphine ligand is
completely coordinated to the rhodium. The immobiliza-
tion of this rhodium-Duphos complex leads to a shift of
the 31P signal of 22 ppm to lower magnetic field compared
FIG. 6. IR spectra of (a) CODRhDuphos (solution in dichlorome-
thane), (b) CODRhDuphos immobilized on Al-MCM-41 (self-supported
wafer), and (c) Al-MCM-41 carrier material (self-supported wafer).
spectra of the pure Al-MCM-41 carrier, the homogeneous
complexes, and the immobilized complexes have been
recorded. Due to the low content of immobilized complex
(4 wt% ) the 13C and 1H resonance signals are too weak for
quantitative interpretation. The pure carrier and the immo-
bilized complexes show similar spectra for the 27Al and 29Si
FIG. 7. 31P-MAS-NMR of (a) the homogeneous CODRhDuphos complex and (b) CODRhDuphos immobilized on Al-MCM-41.

IMMOBILIZATION OF RHODIUM DIPHOSPHINE COMPLEXES
TABLE 2
155
to the homogeneous complex. This shift can be attributed
to an interaction of the guest complex with the surface
of the Al-MCM-41 host, especially due to the interaction
with a Lewis acidic center on the surface which withdraws
electron density from the rhodium metal or the phosphine
ligand.
The Catalytic Activity and Enantioselectivity of Immobi-
lized Rhodium Diphosphine Complexes in the Hydrogenation of
Dimethylitaconate
Immobilized complex,
used ligand
Conversion (% )
TON
ee (% )
Among the catalysts presented here, the rhodium com-
plex is formed by a chiral diphosphine and a cyclooc-
tadiene ligand. Several forces could involved in the
bonding of the complex on Al-MCM-41. Electrostatic in-
teraction of the cationic complexes occurs with the an-
ionic framework of the Al-MCM-41 structure. A similar
mechanism was reported for the immobilization of man-
ganese complexes on Al-MCM-41 (20). Direct bridging of
the rhodium to surface oxygen of the mesoporous walls
has also been observed and could occur after cleavage of
the diene complex during the hydrogenation reaction (22).
However, no evolution of cyclooctadiene during the immo-
bilization reaction is observed. The FT-IR spectra of the fil-
trate obtained after the impregnation show no excess of free
cyclooctadiene.
S,S-Me-Duphos
R,R-Diop
S,S-Chiraphos
(+)-Norphos
100
57
8
>4000
2280
320
92 (R)
35 (R)
47 (R)
48 (S)
16
640
The catalyst can easily be recovered and reused without
further treatment. The supported Rh-Me-Duphos catalyst
was recycled four times. The conversion and the enantios-
electivity of the catalyst remain at high levels after those
runs (Fig. 9). After 8 to 10 consecutive runs a decrease
in catalytic performance was observed. This phenomenon
goes along with the formation of lumps of the catalyst.
Organic deposits like polymerization products of the sub-
strate lead to such a deactivation. However, they could not
be identified by spectroscopic analysis.
Catalytic Tests
In order to prove that the reaction is catalyzed hetero-
geneously and to exclude the possibility of leaching and
homogeneous catalysis, the reaction mixture was separated
from the catalyst before complete conversion occurs. This
test is also known as the hot filtration test. Hydrogenation
of the reaction solution following filtration after 3.5 h does
not give any further reaction. After 24 h the conversion of
the filtered sample remains at 14% whereas the original
batch with catalyst goes to complete conversion of 100%
(Fig. 10). This test proves that no homogeneous catalysis
took place. ICP-AES analysis of the filtered reaction so-
lution showed traces of rhodium, phosphorus, silicon, and
aluminum. The relative amounts of this analysis correspond
Several diphosphine ligands have been applied and the
corresponding complexes have been tested for the immo-
bilization. Those new catalysts are used in the enantiose-
lective hydrogenation (Fig. 8). As a test reaction for the
catalytic activity the hydrogenation of dimethylitaconate
was employed. No reaction took place in the blank test
when the carrier Al-MCM-41 itself was used as catalyst.
The catalytic results of the immobilized rhodium com-
plexes are depicted in Table 2. The S,S-Me-Duphos ligand
reaches the best results with 92% ee at 100% conversion
of dimethyl (R)-methylsuccinate. In this single batch re-
action the substrate/rhodium ratio is 4000, resulting in a
turnover number of 4000 for the immobilized Me-Duphos-
rhodium complex. However, this catalysts could be recy-
cled at least four times without any loss in activity (see
Fig. 9); i.e., the TON is >16,000. The corresponding sup-
ported catalysts with R,R-Diop and S,S-Chiraphos ligands
lead to enantioselectivities of 34% ee and 47% ee with
lower activities. With the (+)-Norphos ligand the favored
enantiomer is dimethyl (S)-succinate which is formed with
48% ee.
FIG. 9. The catalytic activity and enantioselectivity of immobilized
CODRhDuphos complex during the recycling in the hydrogenation of
dimethylitaconate.
FIG. 8. Diphosphine ligands in the rhodium complexes.

¨
WAGNER, HAUSMANN, AND HOLDERICH
156
REFERENCES
1. Baiker, A., and Blaser, H. U., in “Handbook of Heterogeneous Cata-
lysis” (G. Ertl, H. Kno¨zinger, and J. Weitkamp, Eds.), Vol. 5. p. 2422.
VCH, Weinheim, 1997.
2. Corma, A., Iglesias, M., Mohino, F., and Sanchez, F., J. Organomet.
Chem. 544, 147 (1997).
3. Pugin, B., J. Mol. Catal. A 107, 273 (1996).
4. Heinrichs, C., and Ho¨lderich, W. F., Catal. Lett. 58, 75 (1999).
5. Kahlen, W., Wagner, H. H., and Ho¨lderich, W. F., Catal. Lett. 54, 58
(1998).
6. Ogunwumi, S. B., and Bein, T., Chem. Commun. 901 (1997).
7. Sabater, M. J., Corma, A., Domenech, A., Fornes, V., and Garcia, H.,
Chem. Commun. 1285 (1997).
8. Schuster, C., and Ho¨lderich, W. F., Catal. Today 60, 193 (2000).
9. Schuster, C., Tompos, A., and Ho¨lderich, W. F., submitted.
10. Wan, K. T., and Davis, M. E., Nature 370, 449 (1994).
11. Baiker, A., Curr. Opin. Solid State Mater. Sci. 3, 86 (1998).
12. Bein, T., Curr. Opin. Solid State Mater. Sci. 4, 85 (1999).
13. Thomas, J. M., Maschmeyer, T., Johnson, B. F. G., and Shephard,
D. S., J. Mol. Catal. A 141, 139 (1999).
14. Beck, J. S., Vartuli, J. C., Roth, W. J., Leonowicz, M. E., Kresge, C. T.,
Schmitt, K. D., Chu, C. T.-W., Olson, D. H., Sheppard, E. W., McCullen,
S. B., Higgins, J. B., and Schlenker, J. L., J. Am. Chem. Soc. 114, 10834
(1992).
FIG. 10. Hot filtration test: (a) reaction after 24 h in the presence of
immobilized CODRhDuphos; (b) reaction after catalyst has been sepa-
rated after 3.5 h, the reaction does not proceed.
to the composition of the heterogeneous catalyst used. This
indicates that this loss occurs by attrition of the Al-MCM-41
and not by leaching of the complex.
15. Huber, C., Moller, K., and Bein, T., J. Chem. Soc., Chem. Commun.
2619 (1994).
4. CONCLUSION
16. Armengol, E., Corma, A., Vicente, F., Garc´ıa, H., and Primo, J., Appl.
Catal. A 181, 305 (1999).
17. Ernst, S., and Selle, M., Microporous Mater. 27, 355 (1999).
18. Frunza, L., Kosslick, H., Landesser, H., Ho¨ft, E., and Fricke, R., J. Mol.
Catal. A 123, 179 (1997).
19. Kim, G.-L., and Kim, S.-H., Catal. Lett. 57, 139 (1999).
20. Kim, S.-S., Zhang, W., and Pinnavaia, T. J., Catal. Lett. 43, 149 (1997).
21. Piaggio, P., McMorn, P., Langham, C., Bethell, D., Bulman-Page, P. C.,
Hancock, F. E., and Hutchings, G. J., New J. Chem. 1167 (1998).
22. Janssen, A., Niederer, J. P. M., and Ho¨lderich, W. F., Catal. Lett. 48,
165 (1997).
Chiral heterogeneous catalysts are prepared from chi-
ral rhodium diphosphine complexes and Al-MCM-41. The
bonding forces could be due to the ionic interaction of the
cationic complex with the anionic host framework. A re-
duction of the weak acidic sites of Al-MCM-41 has also
been observed.
These catalysts are suitable for the hydrogenation of
functionalized olefins. The organometallic complexes re-
main stable within the mesopores of the carrier under the
reaction conditions. The catalyst can be recycled by filtra-
tion or centrifugation and no leaching of the homogeneous
complex was observed.
23. Augustine, R., Tanielyan, S., Anderon, S., and Yang, H., Chem. Com-
mun. 1257 (1999).
24. de Rege, F. M., Morita, D. K., Ott, K. C., Tumas, W., and Broene,
R. D., Chem. Commun. 1797 (2000).
25. Luan, Z., Cheng, C.-F., Zhou, W., and Klinowski, J., J. Phys. Chem 99,
1018 (1995).
ACKNOWLEDGMENT
26. Burk, M. J., Feaster, J. E., Nugent, W. A., and Harlow, R. L., J. Am.
Chem. Soc. 115, 10125 (1993).
27. Bemi, L., Clark, H. C., Davies, J. A., Fyfe, C. A., and Wasylishen,
R. E., J. Am. Chem. Soc. 104, 438 (1982).
The authors are very grateful for the financial support of this work
by the Sonderforschungsbereich SFB 380 of the Deutsche Forschungsge-
meinschaft (DFG).