Heterogeneously catalyzed asymmetric hydrogenation of α-arylenamides over immobilized RhBPE and RhDUPHOS complexes

Article abstract of DOI:10.1016/j.jcat.2009.05.006

Optically active α-1-arylalkylamine derivatives were successfully synthesized through heterogeneous asymmetric hydrogenation of α-arylenamides over immobilized RhBPE and RhDUPHOS complexes on aluminum-containing M41S and SBA-15 type materials. The heterog

Full text of DOI:10.1016/j.jcat.2009.05.006

Journal of Catalysis 265 (2009) 229–237
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Heterogeneously catalyzed asymmetric hydrogenation of
immobilized RhBPE and RhDUPHOS complexes
a-arylenamides over
*
Adrian Crosman, Wolfgang F. Hoelderich
Department of Chemical Technology and Heterogeneous Catalysis, University of RWTH-Aachen, Worringerweg 1, 52074 Aachen, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Optically active
asymmetric hydrogenation of
a
-1-arylalkylamine derivatives were successfully synthesized through heterogeneous
-arylenamides over immobilized RhBPE and RhDUPHOS complexes on
Received 7 January 2009
Revised 12 February 2009
Accepted 6 May 2009
Available online 31 May 2009
a
aluminum-containing M41S and SBA-15 type materials. The heterogeneous chiral catalysts were pre-
pared by immobilization of rhodium diphosphine complexes on Al-MCM-41, Al-MCM-48, and Al-SBA-
15. The complexes were bonded to the carrier by the interaction of the cationic rhodium of the organo-
metallic complex with the anionic host framework, as well as between Al Lewis acid sites and P Lewis
basic sites. The catalysts were characterized with XRD, FT-IR and MAS-NMR, as well as thermoprogramed
desorption of ammonia, thermogravimetric analysis, and nitrogen sorption experiments. The immobi-
lized catalysts showed high activities and excellent chemo- and enantioselectivities. Up to 99% e.e. at
Keywords:
Hydrogenation
Enantioselectivity
Arylenamide
99% conversion and >99% chemoselectivity were observed in the case of studied a-arylenamides. The cat-
Ruthenium
alysts could be reused up to four times without loss of catalytic activity or enantioselectivity.
Rhodium
Heterogeneous chiral catalysts
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
production of chiral chemicals [11,12]. Due to the cost of sophisti-
cated chiral ligands, often exceeding that of the noble metal em-
Optically active
a
-1-arylalkylamines constitute an important
ployed, catalyst recovery is of paramount importance for the
application of enantioselective metal catalysis to large-scale pro-
cesses, eventually in continuous flow reactors. Furthermore, even
if the activity and selectivity of homogeneous catalysts is excep-
tionally high, toxicological and environmental problems should
also be taken into account.
Since the late 1960s, many approaches have been published by
academia and industry to ‘‘heterogenize”, ‘‘immobilize” or ‘‘an-
chor” a homogeneous catalyst [13–16]. In addition, many excellent
reviews have emerged in the recent years, which describe in detail
the synthesis and (academic) use of polymer-supported catalysts
[17–27] and catalysts on inorganic carriers [28–32], or both [33–
35]. Covalent binding is by far the most frequently used strategy.
It can be effected either by copolymerization of functionalized li-
gands with a suitable monomer, or by grafting functionalized li-
class of compounds that have been employed extensively as
resolving agents, chiral auxiliaries, and intermediates in the syn-
thesis of a wide range of biologically active molecules [1,2]. Many
reliable synthetic methods have been described in the literature.
Generally, these methods involve optical resolution procedures,
biocatalytic methods, or stoichiometric use of chiral precursors
or chiral auxiliaries [3–5]. Asymmetric catalytic reduction of C@N
or C@C double bonds potentially could provide a very efficient
and a convenient route to many chiral amine derivatives. Recently,
success along these lines of research was achieved exclusively in
homogeneous catalysis [6,7].
Homogeneous catalysts are known to exhibit high selectivity
and activity in a variety of asymmetric transformations under rel-
atively mild conditions. In 2001, the significant achievements in
the design and application of asymmetric homogeneous catalysts
were recognized by the award of the Chemistry Nobel Prize to
Knowles and Noyori for enantioselective hydrogenation catalysis
and to Sharpless for enantioselective oxidation catalysis [8–10].
However, despite the huge amount of work devoted to this subject
in both academic and industrial fields, homogeneous asymmetric
catalysis is most probably not used at its full extent in the overall
gands or metal complexes with reactive groups on to
preformed support.
a
Another common and simple immobilization technique of cata-
lytically active metal complexes is ionic binding, which is particu-
larly useful for cationic rhodium or palladium catalysts. Various
supports with ion-exchange capabilities can be used, including
standard organic or inorganic ion-exchange resins, inorganic mate-
rials with polarized groups, and zeolites [28,29,32]. The advantage
of immobilization by adsorption is the ease of preparation of the
heterogenized catalyst by simple procedures, very often even with-
out the need to previously functionalize the ligand. In this respect,
* Corresponding author. Fax: +49 241 8022291.
E-mail address: hoelderich@rwth-aachen.de (W.F. Hoelderich).
0021-9517/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2009.05.006

230
A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
immobilization by ionic binding can be seen as a special case of
heterogenization via adsorption [36,37].
ethylammonium hydroxide (TEAOH, 40 wt.%, aqueous), 0.21 g
NaAlO2, and 50 g H2O were mixed together and stirred at room
temperature for 1 h, followed by the addition of 10 g tetradecylt-
rimethylammonium bromide. The resulting mixture was stirred
for 4 h. Then, 15.23 g Ludox-HS 40 was added dropwise over a per-
iod of 1 h, followed by vigorous stirring at ambient temperature for
4 h. The crystallization took place at 105 °C for 6 days. After the
third and fifth day, pH was adjusted at 10.2 using CH3COOH
(10 wt.%, aqueous). After crystallization, the solid phase was recov-
ered by filtration and washed well with water. The white material
was dried at 120 °C overnight, followed by calcination in static air
at 540 °C for 6 h (heating rate 1 °C/min).
Al-MCM-48 was prepared as follows: in a 500 mL PE flask, TEA-
OH (40 wt. %, aqueous), NaAlO2, and water were stirred at room
temperature for 1 h, followed by the addition of cetyltrimethylam-
monium bromide. Then, Ludox HS-40 was added dropwise over a
period of 1 h, followed by vigorous stirring for 4 h. The molar com-
position of the gel was as follows: Si/Al = 40, TEAOH/Si = 0.3, sur-
factant/Si = 0.45, and water/Si = 60. The crystallization was
performed at 120 °C for 8 days using a Teflon-lined autoclave. After
the second, fourth, and sixth day, pH was adjusted at 10.8 using
CH3COOH (10 wt.%, aqueous). The final solid reaction product
was extracted from the autoclave, filtered, washed with distilled
water, and dried at 120 °C overnight. Finally, the solid was calcined
at 540 °C for 6 h (heating rate 1 °C/min.).
For SBA-15 synthesis, 20 g of amphiphilic triblock copolymer,
poly(ethylene glycol)-block-poly(propylene glycol)-block-poly-
(ethylene glycol) (average molecular weight 5800, Aldrich), was
dispersed in 150 g of water and 600 mL of 2 M HCl solution while
stirring, followed by the addition of 42 g of tetraethyl orthosilicate
to the homogeneous solution with stirring. This gel mixture was
continuously stirred at 40 °C for 6 h, and finally crystallized in a
Teflon-lined autoclave at 90 °C for 3 days. After crystallization,
the solid product was filtered, washed with deionized water, and
dried in air at room temperature. The material was calcined in sta-
tic air at 550 °C for 24 h to decompose the triblock copolymer and
obtain a white powder (SBA-15) [48]. This white powder is used as
the parent material to produce aluminum-containing material de-
noted as Al-SBA-15.
Silica SBA-15 (10 g, 166.5 mmol Si) was dispersed in 250 mL of
dry hexane containing 0.85 g (4.1 mmol Al) aluminum isopropoxide.
The resulting suspension was stirred at room temperature for 24 h,
and afterwards the powder was filtered, washed with dry hexane,
and dried at 120 °C in air. This solid Al-SBA-15 was then calcined in
static air at 550 °C for 6 h (heating rate 1 °C/min) [49].
Using entrapment as a method for heterogenization, the size of
the metal complex is more important than the specific adsorptive
interaction. There are two different preparation strategies: One is
based on building up catalysts in well-defined cages of porous sup-
ports. This approach is also called ‘‘ship in a bottle” [38]. The other
approach is to build up a polymer network around a preformed
catalyst.
Supported aqueous-phase catalysts (SAPCs) can be seen as a
special case of adsorption, whereby a water-soluble catalyst dis-
solved in a very polar solvent is adsorbed on a hydrophilic support
forming a water film on the inner surface of the support [39,40]. In
the case of supported liquid-phase catalysis (SLPC), the water film
on the inner surface is replaced by a solvent of low vapor pressure
(e.g., phthalic acid esters) [41]. The reaction itself takes place in the
supported liquid or at the interface of the supported liquid film, or
in the gas phase or organic phase when dealing with SLPC or SAPC,
respectively. The use of SLPC catalysts is generally restricted to the
synthesis of low-boiling compounds.
Mesoporous molecular sieves have received much attention in
the field of catalysis, especially for their use as supports. Our group
has previously reported the successful immobilization of different
homogeneous catalysts on such molecular sieves (e.g., Al-MCM-41,
Al-MCM-48, and Al-SBA-15) and their use in asymmetric hydroge-
nation, epoxidation, and ring opening of epoxides [42–45]. Re-
cently, Hutchings et al. reported the immobilization of RhDuphos
on Al-MCM-41 and its use for asymmetric hydrogenation of di-
methyl itaconate and methyl-2-acetamidoacrylate [46]. Herein,
we present a novel method for the synthesis of optically active
a-1-arylalkylamines derivatives through asymmetric hydrogena-
tion of -arylenamides over immobilized RhBPE and RhDUPHOS
a
complexes on mesoporous materials. Ion exchange, catalytic, and
adsorptive properties of molecular sieve materials are based on
the existence of acid sites which arise from the presence of acces-
sible hydroxyl groups associated with tetrahedral aluminum in the
silica–alumina framework. Our research was focused on the use of
M41S and SBA-15 type materials as carriers, which are character-
ized by a well-defined pore structure and high surface area, offer-
ing new opportunities for the immobilization of large
homogeneous catalyst species without any modification of their
chemical structure. The heterogenization is based on an ionic
interaction between the negatively charged Al-M41S or Al-SBA-
15 framework and the cationic rhodium of the organometallic
complex. Furthermore, activity of the obtained catalysts was
investigated in the enantioselective hydrogenation of different
a-1-arylalkylenamides.
2.2. Preparation and immobilization of the rhodium diphosphine
complexes
2. Experimental
In a Schlenk tube under Ar, [(COD)2Rh]OTf (0.150 mmol) was
dissolved in 20 mL of CH2Cl2, and then a solution containing the
corresponding diphosphine (0.165 mmol diphosphine; (R,R)-Me-
BPE or (R,R)-Me-DUPHOS) was added. The resulting mixture was
stirred at room temperature for 6 h, and then the solvent was elim-
inated under reduced pressure. The obtained solid was dried under
vacuum and at 50 °C overnight (see Fig. 1).
In order to immobilize Rh–diphosphine-complexes, 1 g meso-
structured material (Al-MCM-41, Al-MCM-48, or Al-SBA-15) was
suspended in 9 mL CH2Cl2 and stirred at room temperature for
1 h. The Rh–diphosphine-complex (0.15 mmol) was dissolved in
1 mL CH2Cl2 and added to the solid/CH2Cl2 suspension. The reac-
tion mixture was stirred at room temperature for 24 h followed
by solvent evaporation under reduced pressure. The recovered so-
lid was Soxhlet extracted with MeOH for 24 h, in order to remove
any residual free Rh-complex (see Fig. 2). The absence of the
homogeneous complex in the upper aliquot was checked by ICP-
It is well known that Rh-complexes containing diphosphines
are very sensitive to moisture and oxygen. Phosphines can be eas-
ily oxidized under such conditions, and then lose their catalytic
activity. Considering these reasons, all the experiments involving
diphosphines, rhodium diphosphine complexes, and immobilized
complexes were carried out in argon-filled MBRAUN LabStar glove-
box or using standard Schlenk-type techniques. In order to remove
all adsorbed water molecules, the solid supports were calcined
overnight at 300 °C prior to the immobilization of organometallic
complexes. Moreover, all used solvents were dried and degassed
using well-known standard methods.
2.1. Preparation of the mesostructured materials
Al-MCM-41 was prepared according to a slightly altered meth-
od reported by van Hooff [47]. In a 250 mL PE flask, 10.5 g tetra-

A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
231
4-chlorobenzonitrile, and 2-methylbenzonitrile) was added drop-
wise to the round-bottomed flask, and then the solution was al-
lowed to stir until solidified. The obtained solid was crushed
with a spatula. Then, the system was cooled down with the help
of dry ice/isopropanol bath, and acetic anhydride (2 equivalents)
was added and the mixture was stirred for 1 h. Further, THF was
added, and the mixture was allowed to warm at the room temper-
ature under vigorous stirring. The slow addition of excess metha-
nol was followed. The solution was transferred to a separatory
funnel, water was added, and then the aqueous mix was extracted
three times with chloroform. The organic layers were dried with
sodium sulfate and removed under reduced pressure. The crude
compound was purified via flash chromatography with an appro-
priate mixture of hexanes and ethyl acetate as eluent (usually ethyl
acetate/hexane = 2/1). The overall yields were usually low, be-
tween 5% and 20%. Characterization data of the synthesized mole-
cules were close to the ones reported in the literature.
P
P
P
P
R,R-Me-BPE
R,R-Me-Duphos
Fig. 1. Diphosphine ligands used in the present study.
AES analysis as well as by FT-IR spectroscopy. The final material
was vacuum dried at room temperature overnight.
2.5. Catalytic tests
2.3. Catalyst characterization
2.5.1. Homogeneous hydrogenations
In a glove box, a Fischer–Porter bottle was charged with the cor-
responding amount of substrate, anhydrous, degassed MeOH, and
the rhodium diphosphine complex (0.2 mol%). After five vacuum/
H2 cycles, the tube was pressurized to a pressure of 3 bar H2. The
solution was stirred at 20–25 °C for 16 h. The products of reaction
were concentrated on a rotavap, and the residue was passed through
a short SiO2 column (EtOAc/hexane: 1/1) to remove the catalyst.
Without further purification, the enantiomeric excesses were deter-
mined directly on an aliquot of the crude products thus obtained.
XRD powder diffraction patterns were collected on a Siemens
Diffractometer D5000 equipped with a secondary monochromator,
a variable diaphragm V 20, and a nickel filter using Cu K radiation
a
(wavelength 1.5406 Å), the angle speed was 0.02° minꢀ1. Bulk ele-
mental chemical analyses were done with inductive couple plasma
atomic emission spectroscopy (ICP-AES) on a Spectroflame D
(Spectro Analytic Instrument). Nitrogen adsorption/desorption iso-
therms at liquid nitrogen temperature were measured on a
Micromeritics ASAP 2010 instrument. The samples were pre-out-
gassed at 150 °C. Pore diameter and specific pore volume were cal-
culated based on the Barrett–Joyner–Halenda (BJH) theory. The
specific surface area was obtained using the Brunauer–Emmett–
Teller (BET) equation.
2.5.2. Heterogeneous hydrogenations
In a glove box, a Fischer–Porter bottle was charged with the cor-
responding amounts of substrate, anhydrous, degassed MeOH, and
immobilized rhodium disphosphine complex. After five vacuum/
H2 cycles, the tube was pressurized to a pressure of 3 bar H2. The
reactions were allowed to stir at room temperature for 16 h. The ali-
FT-IR analyses were done on a Nicolet Protégé 460 equipped
with an evacuable furnace cell with KBr windows containing the
sample wafer. Without using a binder, the samples were pressed
into self-supporting wafers, which, after being mounted onto the
sample cell, were dried at 160 °C and 10ꢀ3 mbar for 16 h.
For thermogravimetric analysis, a Netzsch 209/2/E equipped
with a STA 409 controller was used. The heating rate was 5 °C/
quots were passed through a 20-lm PTFE filter and submitted to the
analysis without any further treatment. Conversion and selectivity
were monitored via HPLC (column: RP 18 400 mm), whereas enanti-
oselectivity was determined via a chiral capillary GC using the
appropriate chiral column (Chirasil-Dex-CB 25 m or Lipodex G-6).
min, and
a-Al2O3 was used as a reference material.
2.5.3. Recycling
2.4. -1-Arylalkylenamides synthesis
a
The heterogeneous catalysts were recycled using standard
methods (the catalysts were recovered by filtration or centrifuga-
tion, washed with 5 mL MeOH, dried overnight under vacuum at
room temperature, and reused in the next cycle at the same sub-
strate/catalyst ratio). In order to prove that the reaction was heter-
ogeneously catalyzed and to exclude the possibility of leaching and
homogeneous catalysis, the hot filtration test was performed
according to Lempers and Sheldon [50].
All enamides were synthesized according to a slightly altered
method reported by Burk et al. [6]. To argon flushed 2-necked
round-bottomed flask, the Grignard reagent (1.3 equivalents; i.e.,
methyl magnesium iodide, ethyl magnesium iodide, and isopropyl
magnesium iodide) in diethyl ether was added. The arylnitrile
(30 mmol; i.e., benzonitrile, 4-methylbenzonitrile, 4-trifluoro-
methylbenzonitrile, 4-methoxybenzonitrile, 4-fluorobenzonitrile,
+
P
P
P
+
1.Al-SBA-15 or M41S, CH2Cl2, RT, 24h
*
Rh
Rh
OTf-
*
P
2.Soxhlet extraction, MeOH, 24h
Al-
Fig. 2. Immobilization of chiral rhodium complex.

232
A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
Table 1
Elemental analysis for parent and loaded materials.
Sample
Si mmol/g
Al mmol/g
Na mmol/g
Rh mmol/g
Si/Al
Al/Na
Al/Rh
Al-MCM-41
RhDUPHOS/Al-MCM-41
Al-MCM-48
14.52
13.74
14.70
13.96
0.383
0.361
0.396
0.368
0.386
0.301
0.404
0.308
–
0.07
–
37.90
38.16
37.12
37.93
0.99
1.19
0.98
1.19
–
5.14
–
RhDUPHOS/Al-MCM-48
0.08
4.81
Table 2
solution to the white solid during the first immobilization step
showed a complete immobilization of the complex onto the solid
surface. For this step, the impregnation solvent of choice was anhy-
drous dichloromethane because of its aprotic nature and its capa-
bility to dissolve appreciable amounts of complex. Alcohols cannot
be used as they compete with the Rh complexes for interaction
with the solid surface. Upon extraction with methanol, the second
immobilization step, which in contrast to the nonpolar dichloro-
methane, adsorbs strongly on the carrier surface, the organometal-
lic complex desorbs from the terminal silanol groups due to a
competitive reaction with the polar alcohol.
Dependence of the Rh content in case of CODRhDUPHOS immobilized on Al-MCM-41
with different Si/Al ratios.
Sample
Si/Al
Rh Mmol/g
Theoretical Rh mmol/g
RhDUPHOS/Al-MCM-41
RhDUPHOS/Al-MCM-41
RhDUPHOS/Al-MCM-41
80
40
20
0.02
0.07
0.10
0.15
0.15
0.15
3. Results and discussion
The X-ray powder diffraction of the Al-MCM-41, Al-SBA-15, and
Al-MCM-48 showed the characteristic hexagonal or cubic struc-
ture, respectively. Al-MCM-41 showed a sharp peak ascribed to
(100) reflection of the hexagonal structure of mesopores at
2h = 2.44°, corresponding to d100 = 3.045 nm. Besides the strong
peak, weak ones ascribed to (110) and (200) reflections at
2h = 4.2° and 4.9°, d = 1.68 and 1.48 nm, respectively, were ob-
served. XRD pattern of Al-MCM-48 sample revealed the presence
of a sharp peak ascribed to the (211) reflection of the cubic struc-
ture at 2h = 2.28°, corresponding to d211 = 3.327 nm. A second peak
corresponding to the (220) reflection was observed at 2h = 2.76°.
Al-SBA-15 showed a well-resolved pattern with a prominent peak
at 0.8°, and two peaks at 1.4° and 1.6° 2h which match well with
the pattern reported for SBA-15 [51]. The XRD peaks can be in-
dexed to a hexagonal lattice with a d(100) spacing of 110 Å, corre-
The dependence of the immobilization on the aluminum con-
tent of the mesoporous materials was studied in case of Al-
MCM-41 by using materials with Si/Al ratios of 20, 40 and 80 (Ta-
ble 2). These complementary experiments showed an almost linear
dependence of the immobilized rhodium with the aluminum con-
tent of Al-MCM-41. The attempt to immobilize the complexes on
all silica M41S and SBA-15 failed due to a complete removal of
the complex from the solid surface during the Soxhlet extraction.
These results clearly showed the influence of the network tetrahe-
dral aluminum. This tetrahedral aluminum is negatively charged,
and a counter cation is present for charge neutralization. In case
of Al-MCM-41 and Al-MCM-48, the negative charge is neutralized
by a sodium cation. Table 1 presents the aluminum/sodium molar
ratio before and after immobilization. As expected, after immobili-
zation this ratio is increasing due to the ionic exchange of sodium
with the cationic rhodium complex.
N2 sorption isotherms were measured for the pure carrier and
for all immobilized complexes, respectively. Fig. 3 and Table 3
present a comparison of sorption isotherms and textural character-
istics for pure and loaded samples. In all the cases, these isotherms
presented the characteristic form of mesoporous materials. It can
be easily remarked that the N2-adsorbed/desorbed volume was
higher in the case of pure carrier than in the case of immobilized
complexes.
sponding to a large unit cell parameter a0 = 127 Å (a0 = 2d(100)/
p
3). However, for carrier materials the intensity of the reflections
essentially did not change upon loading the carrier with the orga-
nometallic complexes, nor after a catalytic cycle, showing that the
mesoporous structures were not affected by the incorporation of
the catalyst.
The elemental analysis showed a Rh content between 0.07 and
0.08 mmol/g, whereas the theoretical content was 0.15 mmol/g
(Table 1). The color transfer from the transition metal complex
700
600
500
400
700
600
500
400
300
200
100
0
Al-MCM-41
RhBPE/Al-MCM-41
300
Al-SBA15
200
100
RhDuphos/Al-SBA15
0
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P )
Relative pressure (P/P )
0
0
Fig. 3. N2 sorption isotherms.

A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
233
Table 3
Lewis or Brønsted acidity. After incorporation of aluminum, the
FTIR spectra of adsorbed Py showed that both Lewis and Brønsted
acid sites were created. Al-MCM-41, Al-MCM-48 and Al-SBA-15
exhibitedꢀ1several peaks due to strong Lewis-bound pyridine
(1577 cmꢀ1), and pyridinium ion on Brønsted acid sites
(1546 cmꢀ1 and 1641 cmꢀ1), while all silica SBA-15 showed only
small peaks due to hydrogen-bonded pyridine (1446 cmꢀ1 and
1596 cmꢀ1) [52,53]. Moreover, after the immobilization of the
organometallic complex a slight decrease in the Brønsted and Le-
wis acidity was observed. This decrease is due to guest/host inter-
action between the organometallic complex and the solid
framework.
Temperature-programed desorption of ammonia was used to
characterize the acidity of all silica and aluminated SBA-15, as well
as Al-MCM-41 and Al-MCM-48. Fig. 6 presents the results obtained
for SBA-15 and Al-SBA-15. Comparison of these TPD curves showed
an increase of the total acidity after alumination. The peaks with
maxima at about 300 °C, found for both samples, were due to
weakly bonded ammonia (physisorbed ammonia and ammonia ad-
sorbed on weak acid sites). The peaks centered at 480 °C were not
well defined, and an interpretation could be risky. These experi-
ments showed that weak acid sites were formed after alumination
on the surface of the solid.
Textural characteristics of parent and loaded materials.
Sample
Surface area
Pore volume
Pore diameter
(Å)
(m2 gꢀ1
)
(cm3 gꢀ1
)
(1623 cm
and 1456 cmꢀ1), weak Lewis-bound pyridine
Al-MCM-41
RhDUPHOS/Al-
MCM-41
Al-SBA-15
RhDUPHOS/Al-SBA-
15
1190
1092
0.97
0.77
34
29
467
397
0.92
0.78
94
89
The infrared spectra of the loaded materials revealed the pres-
ence of bands attributable to the characteristic organic structure
of the diphosphine complexes. Although these bands are weak
and not characteristic enough to resolve a structure, several typical
bands could be distinguished, as for example the @C–H band
centered at 2962 cmꢀ1 and 2857 cmꢀ1. Furthermore, a medium
strength band at 1460 cmꢀ1 was found, which could be assigned
to the CH2 bending vibration or P–C–H group (see Fig. 4).
The acidity of the samples was characterized by FTIR spectro-
scopy investigating the spectra of adsorbed pyridine as probe
molecule. The spectra of adsorbed pyridine (Py) after degassing
at 200 °C are presented in Fig. 5. Pure silica materials have an
electrically neutral framework and consequently showed no
1.8
1040
Al-SBA-15
1.6
RhDuphos/Al-SBA-15
1.4
1020
Al-SBA-15
1.2
1.0
0.8
0.6
0.4
0.2
1000
SBA-15
980
4000
3500
3000
2500
2000
1500
200
400
600
800
Temperature (oC)
-1
Wavenumbers (cm
)
Fig. 4. IR spectra of Al-SBA-15 and RhDuphos/Al-SBA-15.
Fig. 6. Temperature-programed desorption of ammonia of SBA-15 and Al-SBA-15.
A
B
Al-SBA-15
RhDUPHOS/Al-SBA-15
Al-MCM-41
Al-MCM-48
Al-SBA-15
SBA-15
1650
1550
Wavenumbers (cm )
1450
1650
1550
Wavenumbers (cm )
1450
-1
-1
Fig. 5. IR spectra of adsorbed pyridine: (A) pure carriers and (B) Al-SBA-15 and RhDUPHOS/Al-SBA-15.

234
A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
For Al-MCM-41, temperature-programed desorption of ammo-
nia showed the presence of weak acidic centers on this type of car-
rier material. In the case of the immobilized complex, the amount
of desorbed ammonia decreases compared to the pure carrier
material (Fig. 7). This indicates an interaction of the immobilized
complex with acidic sites on the Al-MCM-41. This result was also
supported by FT-IR spectroscopy.
Quantitative loading of the organometallic complex on the con-
sidered supports was demonstrated with thermogravimetric anal-
yses. For all the immobilized complexes, thermogravimetric and
differential scanning calorimetric measurements showed a ther-
mal stability up to 200 °C. In general, the loss of weight for the het-
erogenized catalysts as a function of temperature was about 3–
5 wt%. Since the oxidation of the pure carriers showed no charac-
teristic peak, we attribute the weight loss to the decomposition
of the Rh-complex. The results were consistent with the Rh and P
content determined by ICP AES analyses. For all studied samples,
the oxidation of the organic groups took place in two steps, corre-
sponding in DSC spectra to two peaks with maxima at about 250 or
400 °C. Fig. 8 evidences these observations on the example of
RhBPE/Al-SBA-15.
In order to elucidate the chemical structure of the immobilized
rhodium diphosphine complexes, MAS-NMR of the pure Al-SBA-15
carrier, the homogeneous complexes, and the immobilized com-
plexes have been recorded. Due to the low content of the immobi-
lized complex, the 13C and 1H resonance signals were too weak for
quantitative interpretation. The pure carrier and immobilized com-
plexes showed not significantly different spectra for the 27Al and
29Si nuclei. Therefore, the interpretation of a chemical shift is risky.
31P MAS NMR spectra of the solid homogeneous CODRhDuhos
complex and the same complex immobilized on Al-MCM-41 were
recorded (Fig. 9). The homogeneous complex gave a strong signal
at 68 ppm. The MAS NMR spectrum of the immobilized CODRhDU-
PHOS complex showed only a single signal at 90 ppm. The absence
of signals of the homogeneous complex and of the free phosphine
ligand was an indication that the phosphine ligand was completely
coordinated to the rhodium. Similar results were obtained for
CODRhDUPHOS immobilized on Al-MCM-48 or Al-SBA-15. The
immobilization of this rhodium DUPHOS complex led to a shift of
the 31P signal of 22 ppm to lower magnetic field compared to that
of the homogeneous complex as a result of the interaction of the
guest complex with the surface of the Al-MCM-41 host, more spe-
cifically to the interaction with a Lewis acidic center on the surface
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
Al-SBA-15
RhBPE/Al-SBA-15
200
400
600
800
Temperature (oC)
Fig. 8. DSC spectra of Al-SBA-15 and RhBPE/Al-SBA-15.
CODRhDuphos
CODRhDuphos/Al-MCM-41
140
120
100
80
60
ppm
40
20
0
Fig. 9. 31P MAS NMR of homogeneous CODRhDuphos and CODRhDuphos immo-
bilized on Al-MCM-41.
Al-MCM-41
which withdraws electrons from the rhodium metal or the phos-
phine ligand, and thus lowers the electron density.
The rhodium complexes used consisted of a chiral diphosphine
and a cyclooctadiene ligand. There could be several forces involved
in the bonding of the complex to aluminum-containing mesostruc-
tured materials. For example, an electrostatic interaction of the
cationic complexes with the anionic framework of the mesoporous
material could occur. A similar mechanism was reported for the
immobilization of manganese complexes on Al-MCM-41 [54]. Fur-
thermore, 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
[55]. However, no evolution of cyclooctadiene during the immobi-
lization reaction could be observed, and FT-IR spectra of the filtrate
RhDUPHOS/Al-MCM-41
100
150
200
250
300
350
400
Temperature (°C)
Fig. 7. Temperature-programed desorption of ammonia of Al-MCM-41 and RhDU-
PHOS/Al-MCM-41.

A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
235
Table 4
obtained after the impregnation did not contain bands attributable
to free cyclooctadiene.
Hydrogenation of a
-arylenamides over free or immobilized RhBPE complex.a
d
Entry
R
Support
Conversionb (%)
e.e.c (%)
TOFFIRST HOUR (hꢀ1
)
1e
2
3
H
H
H
H
CH3
CH3
CH3
CH3
CF3
CF3
CF3
CF3
F
F
F
F
Cl
Cl
Cl
Cl
–
>99
99
99
99
>99
99
99
94
>99
92
99
86
>99
99
99
98
>99
99
99
99
>99
96
96.9
95.6
96.4
91.6
97.2
95.6
96.8
94.4
99.0
97.1
99.0
97.3
97.8
97.3
97.8
97.5
97.9
95.4
96.2
95.9
96.8
95.3
95.9
94.8
412
247
280
219
384
298
345
244
n.d.
n.d.
n.d.
n.d.
405
256
321
128
395
357
372
320
n.d.
n.d.
n.d.
n.d.
3.1. Catalytic results
Al-MCM-41
Al-MCM-48
Al-SBA-15
–
Al-MCM-41
Al-MCM-48
Al-SBA-15
–
Al-MCM-41
Al-MCM-48
Al-SBA-15
–
Al-MCM-41
Al-MCM-48
Al-SBA-15
–
Al-MCM-41
Al-MCM-48
Al-SBA-15
–
Al-MCM-41
Al-MCM-48
Al-SBA-15
[(COD)RhDUPHOS]OTf and [(COD)RhBPE]OTf have been tested
for immobilization on aluminum-containing mesoporous materi-
als. The activity of the free and immobilized complexes in enantio-
4
5e
6
selective hydrogenation of different
a-1-arylalkylenamides
7
8
synthesis was investigated (Fig. 10). In blank reactions using the
parent materials, no reactions took place, and over Rh supported
on mesostructured materials no enantiomeric excess was ob-
served, although conversions up to 99% percent were found. More-
over, for all hydrogenation experiments the chemical selectivity
was >99%. No side products were observed.
9e
10
11
12
13e
14
15
16
17e
18
19
20
21e
22
23
24
As presented in Table 4, the enantioselectivities obtained in the
presence of heterogenized RhBPE complex were with maximal 5%
lower than the ones obtained over the free complex. For heteroge-
neous catalysis, the reaction rates were higher for Al-MCM-48-
based catalyst. However, the reaction rates were higher for homo-
geneously performed reactions. Substitution at para position did
not greatly influence the enantioselectivity. No significant elec-
tronic effects on the enantioselectivity were observed. However,
it was found that substitution at the ortho-position has a big effect
both on the activity and on the enantioselectivity. For example,
hydrogenation of N-acetyl-1-(20-methylphenyl) etheneamine over
RhBPE/Al-MCM-41 occurred with 49% conversion and 55.4% e.e.
Table 5 presents the best results obtained with free and immo-
bilized RhDUPHOS complexes. In all cases, the enantioselectivity
obtained over the immobilized RhDUPHOS complex was slightly
lower compared to that obtained over the free complexes. The
enantioselectivity obtained in the hydrogenation of N-acetyl-1-
phenyletheneamine over RhDUPHOS was of 95.3 for the immobi-
lized complex and of 95.8% for the free complex, respectively.
Moreover, turn over frequency corresponding to the first hour of
reaction was lower in case of immobilized RhDUPHOS complex.
OCH3
OCH3
OCH3
OCH3
99
91
a
All reactions were conducted at room temperature and a H2 pressure of 3 bar,
using 0.05–0.1 M solutions of substrate and a S/C ratio of 500. Reaction time was
16 h, and >99% chemoselectivity was observed in all cases.
b
The catalytic activity was monitored by HPLC.
Enantiomeric excesses were determined chiral capillary GC using Chromopack
c
Chirasil column.
d
Turn over frequency measured at a reaction time of 1 h.
Reaction performed in the presence of [CODRhBPE]OTf.
e
Table 5
Hydrogenation of
a
-arylenamides over free or immobilized RhDUPHOS complex.a
d
Entry
R
Support
Conversionb (%)
e.e.c (%)
TOFFIRST HOUR (hꢀ1
)
1e
2
H
H
CH3
CH3
F
F
Cl
Cl
OCH3
OCH3
–
>99
99
>99
96
>99
98
>99
98
95.8
95.3
96.2
96.0
98.0
97.1
98.7
98.1
97.3
96.9
379
268
397
271
365
244
401
288
n.d.
n.d.
For example, the corresponding TOFFIRST
obtained for the
HOUR
Al-SBA-15
–
Al-MCM-48
–
Al-MCM-48
–
Al-MCM-41
–
Al-MCM-41
hydrogenation of N-acetyl-1-phenyletheneamine was 268 hꢀ1 for
3e
4
the immobilized complex and was 379 hꢀ1 for the free complex,
5e
6
respectively. However, the hydrogenation of a-arylenamides over
immobilized RhDUPHOS occurred with more than 94% conversion
and 95.3% enantioselectivity.
The dependence of reaction speed and enantioselectivity with
the reaction temperature was investigated. As expected, the hydro-
7e
8
9e
10
>99
94
genation of a-arylenamides at higher temperatures was performed
a
All reactions were conducted at room temperature and a H2 pressure of 3 bar,
using 0.05–0.1 M solutions of substrate and a S/C ratio of 500. Reaction time was
16 h and >99% chemeoselectivity was observed in all cases.
faster but with slightly lower enantioselectivities. For example,
hydrogenation of N-acetyl-1-phenyletheneamine over RhBPE/Al-
MCM-41 occurred with 94.3 or 92.4% e.e. at 45 or 60 °C, with cor-
responding TOFFIRST HOUR of 311 or 405 hꢀ1 respectively. Moreover,
it was found that hydrogenation of such substrates both over free
and heterogenized RhBPE complex is less sensitive to the solvent.
By use of different solvents such as MeOH, EtOH, i-PrOH, CH2Cl2,
THF, EtOAc, and toluene, the enantioselectivity varied with <5%.
However, the reaction rates were slightly lower when aprotic sol-
vent is used.
b
The catalytic activity was monitored by HPLC.
Enantiomeric excesses were determined via a chiral capillary GC using Chro-
c
mopack Chirasil column.
d
Turn over frequency measured at a reaction time of 1 h.
Reaction was performed in the presence of [CODRhDUPHOS]OTf.
e
It was also examined whether the heterogenized catalysts could
effectively hydrogenate
a-arylenamides possessing b-substituents
N(H)Ac
N(H)Ac
0.2 mol% Catalyst
3 bar H2, MeOH, rt,16h
R
R
R = H, CH3, F, Cl, CF3, OCH3
Fig. 10. Hydrogenation of a-arylenamides.

236
A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
(i.e., N-acetyl-1-phenylpropeneamine and N-acetyl-1-(40-trifluoro-
methylphenyl)isobuteneamine; Fig. 11). Although e.e. higher than
95% was obtained for hydrogenation of such substrates over
RhBPE/Al-MCM-48, complete conversions were obtained only for
low substrate/catalyst ratios (S/C = 100). No investigation of the
substrates E/Z issues and their influence on catalytic activity was
carried out.
The recyclability of the immobilized rhodium diphosphine com-
plexes was tested using standard procedures. It was found that
these catalysts could be recycled four times without any loss of
activity and chemo- or enantioselectivity. Fig. 12 presents the re-
use of RhBPE/Al-MCM-48 in hydrogenation of N-acetyl-1-pheny-
letheneamine at S/C ratio of 500. For this reaction, turn over
number over the 4 cycles was >2000. Moreover, only 89% conver-
sion was obtained for the hydrogenation of N-acetyl-1-phenyleth-
eneamine over homogeneous RhBPE at S/C ratio of 1000, which
corresponds to a TON of 890. This fact demonstrates that by recy-
cling the immobilized catalyst the overall TON of the reaction can
be increased.
Al-MCM-48, no further reaction was found after filtration at 1 h
(Fig. 12). After 16 h, the conversion of the filtered sample remains
at 43%, whereas the original batch with catalyst was completely
converted. This test proved that no homogeneous catalysis took
place.
The deprotection of the amine group has not been investigated.
However, the literature reports showed that the synthesized
amides could be hydrolyzed using well-known methods to the cor-
responding amines without loss of chirality [6,7]. According to the
above-mentioned reports, no racemization was observed during
deprotection step.
4. Conclusion
Heterogeneous chiral catalysts were prepared from chiral
rhodium diphosphine complexes and aluminum-containing
MCM-41, MCM-48, and SBA-15. The complexes were bonded to
the carrier by the interaction of the cationic rhodium of the orga-
nometallic complex with the anionic host framework, as well as
between Al Lewis acid sites and P Lewis basic sites. This method
is effective and simpler than covalent bonding of guest molecules.
These heterogenized catalysts were applied for asymmetric
In order to prove that the reaction is catalyzed heterogeneously
and to exclude the possibility of leaching and homogeneous catal-
ysis, the reaction mixture was separated from the catalyst before
complete conversion occurred (hot filtration test) [49]. In the case
of hydrogenation of N-acetyl-1-phenyletheneamine over RhBPE/
hydrogenation of
a-arylenamides. The immobilized catalysts
showed high activities and excellent chemo- and enantioselec-
tivities, achieving up to 99% e.e. at 99% conversion and >99%
selectivity. To the best of our knowledge, it is for the first time
CH3
H3C
CH3
a-1-arylalkylamines derivatives are obtained with the help of
asymmetric heterogeneous catalysis.
The recyclability of rhodium diphosphine complexes immobi-
lized on aluminum-containing mesostructured materials was dem-
onstrated using standard procedures. The catalysts could be reused
at least four times without any activity loss, obtaining TON larger
than 2000. The high activities observed with these supported orga-
nometallic complexes, plus the fact that the high activity is main-
tained upon reuse of the catalysts, indicated that these are truly
heterogeneous counterparts of homogeneous transition metal
complexes.
N(H)Ac
N(H)Ac
F3C
N-acetyl-1-phenylpropeneamine
N-acetyl-1-(4'-trifluoromethylphenyl)
isobuteneamine
Fig. 11. b-substituted
a
-arylenamides.
Conversion
e.e.
100
80
100
80
60
40
20
0
Catalyst filtered off
after 1h of reaction
60
40
20
0
1
2
3
4
0
2
4
6
8
10
12
14
16
Runs
Time (h)
Fig. 12. Recycling and hot filtration test for hydrogenation of N-acetyl-1-phenyletheneamine over RhBPE/Al-MCM-48.

A. Crosman, W.F. Hoelderich / Journal of Catalysis 265 (2009) 229–237
237
[27] T. Frenzel, W. Solodenko, A. Kirschning, Polymeric materials in organic
synthesis and catalysis, in: E.R. Buchmeiser (Ed.), Wiley-VCH, Weinheim,
2003, p. 201.
[28] M.H. Valkenberg, W.F. Holderich, Catal. Rev. 44 (2002) 321.
[29] I.F.J. Vankelecom, P.A. Jacobs, Chiral catalyst immobilization and recycling, in:
I.F.J. Vankelecom, D. De Vos, P.A. Jacobs (Eds.), Wiley-VCH, Weinheim, 2000, p.
19.
Acknowledgment
The authors are very grateful for the financial support from the
Sonderforschungsbereich SFB 380 of the Deutsche Forschungs-
gemeinschaft (DFG).
[30] S. Anderson, H. Yang, S.K. Tanielyan, R.L. Augustine. Chem. Ind. 82 (2001) 557.
[31] C.E. Song, S.-G. Le, Chem. Rev. 102 (2002) 3495.
[32] D.E. De Vos, M. Dams, B.F. Sels, P.A. Jacobs, Chem. Rev. 102 (2002) 3615.
[33] D. De Vos, B.F. Sels, P.A. Jacobs, Adv. Catal. 46 (2001) 1.
[34] R. Haag, Angew. Chem., Int. Ed. 41 (2001) 520.
[35] W.F. Hoelderich, H.H. Wagner, M.H. Valkenberg, Roy. Soc. Chem. 266 (2001)
76. Special publication.
[36] R. Augustine, S. Tanielyan, S. Anderson, H. Yang, Chem. Commun. (1999) 1257.
[37] F. M de Rege, D.K. Morita, K.C. Ott, W. Tumas, R.D. Broene, Chem. Commun.
(2000) 1797.
[38] D. De Vos, F. Thibault-Starzyk, P.P. Knops-Gerrits, R.F. Parton, P.A. Jacobs.
Macromol. Symp. 80 (1994) 157.
[39] J.P. Arhancet, M.E. Davis, J.S. Merola, B.E. Hanson, Nature 339 (1989) 454.
[40] M.E. Davis, Chemtech 22 (1992) 498.
[41] B. Cornils, W.A. Herrmann, Applied Homogeneous Catalysis with
Organometallic Compounds, Developments, vol. 2, Wiley-VCH, Weinheim,
2002.
[42] A. Crosman, W.F. Holderich, in: Proceedings of the 14th International Zeolite
Conference, Cape Town, South Africa, 2004, p. 2839.
[43] A. Crosman, W.F. Holderich, J. Catal. 232 (2005) 43.
[44] A. Crosman, W.F. Hoelderich, Catal. Today 121 (2007) 130.
[45] A. Crosman, C. Schuster, H. Wagner, M. Batorfi, J. Cubilos, W.F. Hoelderich, in:
D. Enders, K.-E. Jaeger (Eds.), Asymmetric Synthesis with Chemical and
Biological Methods, WILEY-VCH Verlag, Weinheim, 2007, p. 277.
[46] W.P. Hems, P. McMorn, S. Riddel, S. Watson, F.E. Hancock, G.J. Hutchings, Org.
Biomol. Chem. 3 (2005) 1547.
[47] J.H.C. van Hooff, Micropor. Mater. 5 (1995) 211.
[48] D. Zhao, J. Am. Chem. Soc. 120 (1998) 6024.
[49] Z. Luan, Chem. Mater. 11 (1999) 1621.
References
[1] J. Jacques, A. Collet, S.H. Wilen, Enantiomers, Racemates, and Resolutions, John
Wiley and Sons, New York, 1981.
[2] M. Nogradi, Stereoselective Synthesis, second ed., VCH, Weinheim, 1995.
[3] S. Itsuno, M. Sasaki, S. Kuroda, K. Ito, Tetrahedron: Asymmetry 6 (1995) 1507.
[4] M.M. Gharpure, A.S. Rao, Synthesis (1988) 410.
[5] D. Rossi, A. Calcagni, A. Romeo, J. Org. Chem. 44 (1979) 2222.
[6] M.J. Burk, Y. Wang, J. Lee, J. Am. Chem. Soc. 118 (1996) 5142.
[7] M. van den Berg, R.M. Haak, A.J. Minnaard, A.H.M. de Vries, J.G. de Vries, B.L.
Feringa, Adv. Syn. Catal. (2003) 1003.
[8] W.S. Knowless, Angew. Chem., Int. Ed. 41 (2002) 1998.
[9] R. Noyori, Angew. Chem., Int. Ed. 41 (2002) 2008.
[10] K.B. Sharpless, Angew. Chem., Int. Ed. 41 (2002) 2024.
[11] H.U. Blaser, E. Schmidt, Asymmetric Catalysis on Industrial Scale, Wiley,
2003.
[12] K. Mikami, M. Lautens, New Frontiers in Asymmetric Catalysis, Willey, 2007.
[13] R.A. Sheldon, Curr. Opin. Solid State Mater. Sci. 1 (1996) 101.
[14] B. Pugin, H.U. Blaser, Comprehensive asymmetric catalysis, in: E.N. Jacobsen,
A. Pfaltz, H. Yamamoto (Eds.), Springer, Berlin, Heidelberg, New York, 1999.
[15] H.U. Blaser, B. Pugin, M. Studer, Chiral catalyst immobilization and recycling,
in: I.F.J. Vankelecom, D.E. De Vos, P.A. Jacobs (Eds.), Wiley-VCH, Weinheim,
2000.
[16] F.J. Waller, Chem. Ind. 89 (2003) 1.
[17] R. Arshady, Microspheres Microcapsules Liposomes 1 (1999) 197.
[18] D. Bergbreiter, Chiral catalyst immobilization and recycling, in: I.F.J.
Vankelecom, D. De Vos, P.A. Jacobs (Eds.), Wiley-VCH, Weinheim, 2000, p. 43.
[19] D.E. Bergbreiter, Chem. Rev. 102 (2002) 3345.
[50] H.E.B. Lempers, R.A. Sheldon, J. Catal. 175 (1998) 62.
[51] A. Sayari, B.-H. Han, Y. Yang, J. Am. Chem. Soc. 126 (2004) 14348.
[52] I. Kiricsi, C. Flego, G. Pazzuconi, W.O. Parker Jr., R. Millini, C. Perego, G. Bellussi,
J. Phys. Chem. 98 (1994) 4627.
[53] T. Barzetti, E. Selli, D. Moscotti, L. Forni, J. Chem. Soc., Faraday Trans. 92 (1996)
1401.
[20] A.K. Kakkar, Chem. Rev. 102 (2002) 3579.
[21] C.A. Mc Namara, M.J. Dixon, M. Bradley, Chem. Rev. 102 (2002) 3275.
[22] N.E. Leadbeater, M. Marco, Chem. Rev. 102 (2002) 3217.
[23] Q.-H. Fan, Y.-M. Li, A.S.C. Chan, Chem. Rev. 102 (2002) 3385.
[24] R. Van Heerbeek, P.C.J. Kamer, P.W.N.M. Van Leeuwen, J.N.H. Reek, Chem. Rev.
102 (2002) 3717.
[54] S.S. Kim, W. Zhang, T.J. Pinnavaia, Catal. Lett. 43 (1997) 149.
[55] A. Janssen, J.P.M. Niederer, W.F. Hölderich, Catal. Lett. 48 (1997) 165.
[25] M. Benaglia, A. Puglisi, F. Cozzi, Chem. Rev. 103 (2003) 3401.
[26] S. Braese, F. Lauterwasser, R.E. Ziegert, Adv. Syn. Catal. 345 (2003) 869.