Hybrid organic - Inorganic catalysts: A cooperative effect between support, and palladium and nickel salen complexes on catalytic hydrogenation of imines

Article abstract of DOI:10.1016/j.jcat.2004.02.017

A method for immobilizing chiral salen palladium and nickel complexes (salen = (R,R)-N′,N′-bis-(3,5-di-tert-butylsalicylidene)-1,2- cyclohexanediamine) on ordered mesoporous silica supports (MCM-41), delaminated ITQ-2 and ITQ-6 zeolites, and amorphous silica has been developed. Hydrogenation of imines was studied with the homogeneous as well as with the heterogenized counterpart catalysts. The high accessibility introduced by the structure of the supports allows the preparation of highly efficient immobilized catalysts with TOFs of 200×10³ h^-1. A moderate acidity in the support increases the catalytic activity by stabilizing the charged transition state and can duplicate the activity of the homogeneous analogues. No deactivation of the catalysts was observed after repeated recycling.

Full text of DOI:10.1016/j.jcat.2004.02.017

Journal of Catalysis 224 (2004) 170–177
www.elsevier.com/locate/jcat
Hybrid organic—inorganic catalysts:
a cooperative effect between support, and palladium and nickel
salen complexes on catalytic hydrogenation of imines
V. Ayala,a A. Corma,b,∗ M. Iglesias,c J.A. Rincón,a and F. Sánchez a
a
Instituto de Química Orgánica General, CSIC, Juan de la Cierva 3, 28006 Madrid, Spain
b
Instituto de Tecnología Química, UPV-CSIC, Universidad Politécnica de Valencia, Avda. de los Naranjos s/n, 46002 Valencia, Spain
c
Instituto de Ciencia de Materiales de Madrid, CSIC, Cantoblanco, 28049 Madrid, Spain
Received 15 December 2003; revised 12 February 2004; accepted 12 February 2004
Abstract
ꢀ
ꢀ
A method for immobilizing chiral salen palladium and nickel complexes (salen = (R,R)-N ,N -bis-(3,5-di-tert-butylsalicylidene)-1,2-
cyclohexanediamine) on ordered mesoporous silica supports (MCM-41), delaminated ITQ-2 and ITQ-6 zeolites, and amorphous silica has
been developed. Hydrogenation of imines was studied with the homogeneous as well as with the heterogenized counterpart catalysts. The
high accessibility introduced by the structure of the supports allows the preparation of highly efficient immobilized catalysts with TOFs of
3
−1
200 × 10 h . A moderate acidity in the support increases the catalytic activity by stabilizing the charged transition state and can duplicate
the activity of the homogeneous analogues. No deactivation of the catalysts was observed after repeated recycling.
2004 Elsevier Inc. All rights reserved.
Keywords: Immobilized salen; Heterogenization; Imine hydrogenation; Nickel; Palladium
1. Introduction
condensation [2h], selective hydrogenation [5g,5h]. In many
cases, metal–salen complexes have been immobilized on dif-
ferent materials, such as polymers [3], or inorganic solids,
such as silica [4] and zeolites [5]. Their interest is related
to the fact that they perform highly selective processes in a
multiphase system, in which the catalysts can be easily re-
covered.
The selective hydrogenation of imines is a reaction of in-
terest in the production of fine chemicals, and the catalysts
of choice have traditionally been Ir and Rh complexes [6],
even if they require relatively high H2 pressures (∼ 80 bar)
for carrying out the reaction. There is incentive to design
highly active catalysts that can work under much milder
reaction pressure and, if possible, to use less expensive met-
als such as palladium and nickel. We show here that salen
Pd and Ni complexes are active catalysts for hydrogena-
tion of imines at 4 bar H2 pressure. Moreover by prop-
erly choosing the support it is possible to achieve a two
fold increase in activity with respect to the homogeneous
counterpart, by means of a cooperative effect of the sup-
port.
One drawback of homogeneous catalysis is the separation
and recovering of the catalyst when the reaction is com-
pleted. Binding the active metal complexes to insoluble or
water-soluble carriers is an attractive strategy allowing sep-
aration by filtration or extraction [1]. However, immobiliza-
tion usually affects negatively the catalytic performance and
efforts should be made to design new systems in which a co-
operative effect between the solid support and the transition
metal complex is established that can increase the activity of
the homogeneous catalyst.
In recent years, reactions catalyzed by chiral–salen-based
transition metal complexes have become a matter of inter-
est [2]. Chiral–salen complexes have been applied in differ-
ent kind of reactions such as asymmetric epoxidation [2a–c,
2g], cyclopropanation [2e], aziridination [2f], Knoevenagel
*
Corresponding author. Fax: 34(96)3877809.
E-mail address: acorma@itq.upv.es (A. Corma).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2004.02.017
2004 Elsevier Inc. All rights reserved.

V. Ayala et al. / Journal of Catalysis 224 (2004) 170–177
171
Table 1
CH3CH2O–); 0.56 (t, 2H, CH2CH2Si); 13C NMR (CDCl3):
δ = 197.07 (–CHO); 160.40 (–NHCONH–); 158.10
(=C–OH); 138.55 (=C–CH2NH–); 133.65 (=CH, arom.);
130.46 (=CH, arom.); 130.11 (=C–C(CH3)3); 120.33
(=C–CHO); 58.43 (CH2O–); 43.78 (CONHCH2–); 42.88
(PhCH2NHCO–); 34.83 (C(CH3)3); 29.12 (C(CH3)3);
23.55 (CH2CH2CH2); 18.26 (CH3CH2O–); 7.50 (CH2Si–);
MS+ (EI) (m/z, %): 454 (M+, 5); 408 (M+–EtOH, 68);
206 (M+–CONH(CH2)3Si(OEt)3, 100); 191 (M+–
NHCONH(CH2)3Si(OEt)3, 40); 163 (M+-291, 33).
Imines I and II used as substrates for the catalytic tests
were synthesized according with the method described in the
literature [10].
Textural characteristics of the solid supports
Support
Si/Al
BET surface
area
(m /g)
Micropore
surface
(t plot m /g)
External
(or mesoporous)
suface area
2
2
2
(m /g)
Silica
MCM-41
ITQ-2
ITQ-6
MCM-41
ITQ-2
–
540
1030
830
618
900
810
546
–
540
1030
700
608
900
720
520
∞
∞
∞
44
25
30
0
130
10
0
90
26
ITQ-6
2. Experimental
2.3. Heterogenization of hydroxyaldehydes
2.1. Materials
Typically, silica or the corresponding MCM-41, ITQ-2, or
ITQ-6 (2 g) materials were suspended in toluene (25 ml) and
stirred at room temperature for 30 min. Then, a solution of
2 mmol of the triethoxysilylaldehydes 2a or 2b in toluene
(5 ml) and 400 µl of water were added. The slurry was
heated at 100 ◦C for 16 h and the yellow solid was filtered
off and washed successively with toluene, petroleum ether,
ethanol, and ether. The solid was dried under high vacuum
for 24 h to afford the anchored hydroxyaldehydesSil-3a, Sil-
3b, MCM41-3a, MCM41-3b, ITQ2-3b, and ITQ6-3b.1
MCM41-3b: % C: 13.90, % H: 2.63, % N: 1.91 cor-
responding to 0.78 mmol/g; IR (KBr): ν = 3421, 1642,
1564 cm−1; UV-vis (λ, nm) 425, and 360; 13C NMR (solid):
δ = 199.29 (–CHO); 160.58 (–NHCONH–); 147.84 (–C
OH); 143.84; 138.13; 130.98; 120.18; 57.45 (CH3CH2O–);
54.29 (–NHCH2Ph–); 42.04 (–NHCH2CH2–); 27.60
(–CH2CH2CH2–); 18.99 (CH3CH2O–); 9.64 (–CH2Si–).
The supports were prepared following [7] for MCM-41
and [8,9] for ITQ-2 and ITQ-6. The textural characteristics
of the solids are given in Table 1.
2.2. Reactants and methods
The silylating agent 3-triethoxysilylpropylisocyanatewas
obtained from ABCR. The aldehydes (5-azidomethyl-2-
hydroxybenzaldehyde (1a) and 5-azidomethyl-3-tert-butyl-
2-hydroxybenzaldehyde (1b)) were prepared as follows:
35.9 mmol of chloroaldehyde precursor was dissolved
in 10 ml dimethylformamide at room temperature. Then,
sodium azide (37.7 mmol) was slowly added while stirring
and maintained until the reaction was completed (6–10 h).
The mixture was diluted with 15 ml of water and extracted
with ethyl acetate (3 × 10 ml). The combined organic ex-
tracts were washed successively with water (10 ml) and
brine (10 ml). The organic layer was dried over anhydrous
magnesium sulfate and the solvent evaporated under reduced
pressure. The crude was purified by chromatography in sil-
ica (hexane–ethyl acetate 6:1) to afford 1a or 1b in high
yields (90–95%).
2.4. Preparation of immobilized salen ligands
Typically, stirred slurry of 1 mmol of the correspond-
ing anchored hydroxyaldehydes Sil-3a, Sil-3b, MCM41-
3a, MCM41-3b, ITQ2-3b, and ITQ6-3b was prepared in
ethanol (10 ml) under argon at room temperature. After
10 min 125 mg (1.1 mmol) of (1S,2S)-1,2-diaminecyclohex-
ane was added and stirred for 16 h. To this slurry, a solution
of 211 mg (1.1 mmol) of 3-tert-butyl-5-methylsalicylalde-
hyde in ethanol (5 ml) was added. The resulting mixture
was stirred for 4 h at room temperature and then refluxed
for an additional 2 h. Then, the mixture was allowed to cool
and the deep yellow solid was filtered off and washed thor-
oughly with ethanol and then with ether. The solid was dried
under vacuum to afford the corresponding anchored salen
ligands Sil-4a, Sil-4b, MCM41-4a, MCM41-4b, ITQ2-4b,
and ITQ6-4b. The loadings were calculated from the ele-
mental analysis data. The filtered solution was concentrated
N-[(5-Formyl-4-hydroxyphenyl)methyl]-Nꢀ-(3-triethox-
ysilylpropyl)urea (2a) aꢀnd N-[(3-tert-butyl-5-formyl-4-hy-
droxyphenyl)methyl]-N -(3-triethoxysilylpropyl) urea (2b)
were prepared as follows: Solutions of the aldehydes 1a
(2 mmol) or 1b (2 mmol) and 3-triethoxysilylpropylisocy-
anate (2.1 mmol) were prepared in 40 ml of ethyl acetate at
room temperature. Then, slurry of 50 mg of Pd/C (10%) in
2 ml of ethyl acetate was carefully added and the mixture
was vigorously stirred overnight under positive H2 pressure.
The slurry was then filtered and the filtrate evaporated in
vacuum to afford 2a or 2b in almost quantitative yields.
2b: IR (KBr): ν = 3350, 1650, 1570 cm−1; H NMR
1
(CDCl3): δ = 11.65 (s, –OH –); 9.78 (s, 1H, –CH O); 7.35
(s, 1H, arom.); 7.26 (s, 1H, arom.); 4.72 (br. s, –NH –);
4.58 (br. s, –NH –); 4.24 (d, 2H, PhCH2NH–); 3.72 (q, 6H,
CH3CH2O–); 3.11 (m, 2H, CH2CH2NH–); 1.53 (2H, m,
NHCH2CH2CH2Si); 1.32 (9H, s, (CH3)3C–); 1.15 (t, 9H,
1
Representative data for new materials were included. Experimental
and spectroscopic data for all materials could be obtained directly from au-
thors.

172
V. Ayala et al. / Journal of Catalysis 224 (2004) 170–177
Scheme 1
and purified by flash chromatographyin silica (hexane–ethyl
acetate, 30:1) to afford the free salen ligand.
Table 2
Metal loading in the support
ITQ2-4b: % C: 10.50, % H: 2.00, % N: 1.96 corre-
LIGAND
Loadings (mmol-metal/g-support)
sponding to 0.35 mmol/g; IR (KBr): ν = 1636, 1564 cm−1
;
Pd
Ni
UV-vis (λ, nm): 374, 268 and 238; 13C NMR (solid):
δ = 172.01; 160.35; 138.26; 129.61; 118.39; 72.84; 58.54;
42.35 (NHCH2CH2–); 33.87; 28.04 ((CH3)3C–); 15.20;
8.07 (CH2Si–).
a
Free
1.52
0.21
0.13
0.12
0.17
0.19
0.18
1.79
0.30
0.21
0.19
0.10
0.16
0.25
Sil-4a
MCM41-4a
Sil-4b
MCM41-4b
ITQ2-4b
ITQ6-4b
2.5. Synthesis of Pd and Ni complexes
a
Found for 5Pd · 2EtOH and 5Ni · 2H O.
2
Typically, palladium(II) and nickel(II) complexes were
prepared by adding ethanolic solutions of Pd(OAc)2 or
Ni(OAc)2 (0.5 mmol/15 ml) to vigorously stirred slurries
of the corresponding anchored salen ligands Sil-4a, Sil-4b,
MCM41-4b, ITQ2-4b, and ITQ6-4b (over 0.5 mmol/g) or
free salen ligand (0.5 mmol) in CH2Cl2 (15 ml) at room tem-
perature. The resulting mixture was refluxed for 4 h and then
concentrated under reduced pressure. The residue was thor-
oughly washed with ethanol and then with ether and dried
in vacuum to afford the heterogenized complexes in almost
quantitative yields (Scheme 1). The catalysts prepared in this
way were digested in acid (NO3H/HF, 1/1) and the result-
ing aqueous solutions analyzed by atomic absorption. The
materials present metal loadings of 0.1–0.30 mmol-metal/g-
support (Table 2).
ITQ6-4b-Pd: % C: 12.69, % H: 1.83, % N: 2.04, %
Pd: 1.91 gives 0.18 mmol/g; IR (KBr): 1626, 1555 and
573 (νPd–O) cm−1. UV-vis (λ, nm): 400 and 350 13C NMR
(solid): δ = 160.02 (–CH=N–); 137.87; 129.41; 118.30
(–C–CH=N); 72.47 (–CH2CH–N); 58.67 (–NHCH2Ph–);
41.65 (NHCH2CH2–); 33.42; 27.64 ((CH3)3C–); 15.20;
8.52 (CH2Si–).
1562, and 576 (νNi–O) cm−1; UV-vis (λ, nm): 570, 420,
and 365; 13C NMR (solid): δ = 159.97 (–NHCONH–);
156.87 (–CH=N); 153.54 (=C–O–Ni); 136.92; 127.45;
121.75; 119.57; 117.63 (–C–CH=N); 68.73 (–CH2CH–N);
57.32 (–NHCH2Ph–); 34.27 (–NHCH2CH2–); 27.84, 13.04
(–CH2CH2CH2–); 6.00 (–CH2Si–).
2.6. Analytical techniques
C, H, and N analyses were carried out with a Heraeus and
a Perkin–Elmer 240C apparatus. Metal contents were ana-
lyzed by atomic absorption using a Perkin Elmer AAnalyst
300 atomic absorption spectrometer and ICP Perkin–Elmer
40. IR spectra were recorded with a Perkin–Elmer spec-
trum one using KBr wafers. 1H and 13C NMR spectra were
taken with a Varian Gemini-200 and Varian XR300 spec-
trometers operating at 200 and 300 MHz for 1H, and 50 and
75 MHz for 13C, respectively. Diffuse reflectance UV-vis
spectra were recorded on a Shimadzu UV-2401 PC spec-
trophotometer. Mass spectra were recorded on a VG 12-250
spectrometer using EI techniques (70 eV). High-resolution
13C NMR or CP/MAS spectra of powdered samples, in some
MCM41-4b-Ni: % C: 16.98, % H: 2.91, % N: 2.53,
% Ni: 0.56 gives 0.10 mmol/g; IR (KBr): ν = 1632,

V. Ayala et al. / Journal of Catalysis 224 (2004) 170–177
173
cases also with a Toss sequence, in order to eliminate the
spinning side bands, were recorded at 100.63 MHz, 6.5 µs,
90◦ pulse width, 1 to 3-ms contact time and 5 to 10-ms re-
cycle delay, using a Bruker MSL 400 spectrometer equipped
with an FT unit. The spinning frequency at the magic angle
(54◦44ꢀ) was 10–12 KHz. Gas chromatography analysis was
performed using a Hewlett–Packard G1800A with an elec-
tron ionization detector.
and ITQ-6 delaminated zeolites, and amorphous silica (see
characteristics in Table 1). With respect to the supports,
both silica and MCM-41 are short-range amorphous mate-
rials containing a large number of silanol groups available
for grafting. In the case of MCM-41, however, the material
presents a long-range ordering of hexagonal symmetry with
regular monodirectional channels of 3.5 nm diameter. On the
other hand, ITQ-6 and ITQ-2 delaminated zeolites present
both short and long range order, together with a large well-
structured external surface (see Figs. 1and 2, respectively)
in which the silanol groups act as grafting centers. However,
there are structural differences between the two delaminated
zeolites. Indeed, in the case of the ITQ-2 there are “cup-
like” apertures to the external surface with ∼0.8 × 0.8-nm
dimensions, while in the case of the ITQ-6 these cups are
less deep (∼0.3 nm). In a previous work and by a molecular
dynamics study, we have shown [13] that there is a mini-
mum of potential energy in the void space of the cups, in
which the molecules have a high tendency to adsorb. Tak-
ing this into account and the fact that the structured silanols
are located at the borders of those cups (see Fig. 1), one can
envisage a structural model analogous to that presented by
Breslow et al. [14] for cyclodextrines, in which the grafted
organometallic site is located at the border of the cyclodex-
trine cup. In our case, the metal complex will be located at
the border of the inorganic silicate cups. Then, one may ex-
pect that the reactant adsorbed in the cup of the delaminated
zeolite will react there with the catalyst located at the border.
If this model is correct, we should expect an increase in the
concentration of reactants close to the catalytic active site,
with the corresponding increase in reaction rate.
2.7. Reaction procedure
Reactions were carried out in a stirred batch reactor (Au-
toclave Engineers) of 100-ml capacity at 40 ◦C and 4 bar of
H2, in the presence of 40 ml of ethanol. The metal/substrate
molar ratio varies from 1/1000 to 1/100000. The evolu-
tion of the reaction was monitored by GC with a capil-
lary column with methylsilicone (DV-1701) and methyl-
silicone:heptakis-[2,3-dipentyl-6-(tert-butylmethylsilyl)-β-
cyclodextrine as stationary phase) [11].
3. Results and discussion
3.1. Preparation of immobilized ligands and complexes
In the last years we have developed a modular system
combining functionalized ligands with different supports
and linkers in order to have a systematic access to a vari-
ety of immobilized chiral catalysts [12]. We have applied
here this methodology (Scheme 1) to immobilize salen lig-
ands on mesoporous silica supports such as MCM-41, ITQ-2
Fig. 1. Structure of ITQ-6, as an ideal single layer of a delaminated ferrierite zeolite.

174
V. Ayala et al. / Journal of Catalysis 224 (2004) 170–177
zeolites could be suitable supports for immobilizing ho-
mogeneous imine hydrogenation catalysts, since they could
avoid the mass-transfer limitations of the previously reported
support. Moreover, the fact that we could control, up to a
certain point, the polarity and acidity of the catalyst sur-
face could introduce an additional benefit from the point
of view of increasing reactant concentration at the surface,
and/or stabilizing reaction transition states. Thus, for inves-
tigating these factors, we have selected two model imines
(anti-N-benzyl-(1-phenylethylidene)imine, I, and 5-phenyl-
3,4-dihydro-2H-pyrrole, II):
I
II
For carrying out the hydrogenation reaction, we have pre-
pared less conventional imine hydrogenation catalysts based
on Ni and Pd salen that can be economically attractive with
respect to those based on Rh, Ru, or Ir. The reaction con-
ditions chosen were also milder (P ꢀ 10 bar) than those
reported before.
The results presented in Table 3, for the complexes de-
rived from ligand 4b, show that the Pd and Ni complexes
described here are very active in the homogeneous phase
with TOF of ∼105 h−1. On the other hand, the activity of
catalysts 4a was much lower and consequently further work
will be only done with catalysts 4b. This value is much
higher than those reported before for Ir, Rh, and Ru com-
plexes. We have checked that the catalytic activity observed
is due to the organometallic complex and not to any metal
formed by decomposition of the complex in the reaction me-
dia. In order to do this, we have followed the methodology
reported in [17]. The presence of bulk metal was eliminated
as follows: the reaction mixtures were passed through a mi-
cropore filter, the reactor was washed with ethanol, and the
washings were filtered. The filter did not show traces of bulk
metal (microscope, 100× magnification). Another problem
that can mask the results may come from the unnoticed pres-
ence of colloidal metal in “homogeneous” catalysts. Such
colloidal particles can be catalytically active and may be re-
Fig. 2. Structure of ITQ-2, as an ideal single layer of a MWW zeolite struc-
ture.
It was recently reported that Ir–xyliphos complexes are
very active and enantioselective catalysts for the hydrogena-
tion of imines [15]. However, when they were supported
on silica the authors found that the activity of the hetero-
genized analogs was lower by a factor of 2 to 5. More-
over, they observed that the polystyrene-bound complexes
were much less active and, in contrast with the homoge-
neous complexes, the immobilized systems were susceptible
to deactivation. In their excellent work, Pugin et al. [15a]
show that the decrease in activity and catalyst life of their
imine hydrogenation homogeneous catalyst upon heteroge-
nization was probably due to limited mass transport in the
pores and to the proximity of the Ir complexes. in any case
and by working at 80 bars and 25–30 ◦C in the presence of
acetic acid, Pugin et al. obtained an unprecedented high ac-
tivity (TOF up to 20×103 h−1) and productivity (TON up to
195 × 103) of the silica-bound catalyst, that are much higher
Table 3
Hydrogenation of imines catalyzed by homogeneous and heterogenized Ni–
a
and Pd–salen complexes derived from ligand 4b
Catalyst
Imine I
Imine II
Pd
Ni
Pd
Ni
83
102
135
128
80
Homo
MCM-41
ITQ-2
ITQ-6
Silica
110
143
150
148
120
118
176.2
180
150
140
75
94
122
115
70
than previously reported on diphosphine Rh (2 × 103 h−1
or Ru (4 × 102 h−1) [1,16].
)
a
−1
3
◦
From the above work it appeared to us that mesostruc-
tured solids, and well-structured all-accessible delaminated
TOF, h
× 10 ; conditions, 4–5 atm., 40 C. S/C molar ratio,
100000/1.

V. Ayala et al. / Journal of Catalysis 224 (2004) 170–177
175
sponsible for the observed catalytic activity, since they may
or may not be subsequently precipitated. It is not sufficient,
therefore, to verify merely that the solutions are clear, nor
to show that any bulk metal that is precipitated is catalyti-
cally inactive. We have performed an additional test using
colloidal nickel as catalyst for the hydrogenation of imine I,
and no reaction was observed after 3 h. Furthermore, there
is another consideration that supports the view that the solu-
tions are homogeneous and this is the reproducibility of the
reactions.
It appears from the experiments described above, that
Ni and Pd salen complexes are stable under reaction con-
ditions and are very active for imine hydrogenation. When
they were supported on amorphous silica, the TOF did not
decrease but remains practically the same as that in the
homogeneous phase. Interestingly, the activity of the cata-
lyst increases when supported on the mesoporous molecular
sieve MCM-41. The increase in activity is even higher in the
case of the delaminated zeolitic materials whose surface is
formed by accessible cups in which the reactant will be con-
centrated by adsorption [9]. In this case, the activity when
grafted on ITQ-2 is almost double that of the unsupported
homogeneous catalyst. The results show the benefit of the
high surface area, accessibility, and adsorption properties of
MCM-41, ITQ-2, and ITQ-6. Furthermore these structured
materials allow modifications of the surface that can further
enhance the catalytic performance of the hybrid organic–
inorganic catalysts.
imines in the homogeneous phase. The reason for this was
claimed to be caused by the stabilization of the transi-
tion state by the H+ in the reaction media, as can be seen
from Scheme 2. If this is true, and since we have claimed
an important role of the adsorption characteristics of the
mesoporous and delaminated zeolite supports on the cat-
alytic activity, we should further increase the reaction rate
by introducing some acidity on the support. To do this,
we have prepared MCM-41 and ITQ-2 containing tetra-
hedrally coordinated framework Al. By doing this, Brøn-
sted acidity has been developed on the catalyst as can be
seen by IR spectroscopy of adsorbed pyridine (Fig. 3). In-
deed, after pyridine adsorption and desorption at 150 ◦C in
In a previous paper, Pugin et al. have shown that acetic
acid increases the reaction rate for the hydrogenation of
Scheme 2
Fig. 3. IR spectroscopy of pyridine adsorbed on acid ITQ-2 (a), ITQ-6 (b), and MCM-41 (c).

176
V. Ayala et al. / Journal of Catalysis 224 (2004) 170–177
Table 4
4. Conclusions
Influence of the acidity of the supports on the activity of catalysts 4bNi in
hydrogenation of imines
a
We have shown that it is possible to boost the activity
of an organometallic catalyst by using an adequate support.
We have seen that by using mesostructured silicates and de-
laminated zeolites as carriers, both with very high surface
areas and accessibility to reactants, together with high ad-
sorption capacity, the activity of the grafted Pd and Ni salen
complex for the hydrogenation of imines is higher than with
the homogeneouscounterpart or when grafted on amorphous
silica.
Catalyst
Si/Al
Imine I
Imine II
MCM-41
MCM-41
ITQ-2
ITQ-2
ITQ-6
∞
93
∞
25
∞
30
176
228
180
211
150
194
102
124
135
164
128
156
ITQ-6
a
−1
−3
◦
TOF: h × 10 ; conditions: 4–5 atm., 40 C. S/C molar ratio,
100000/1.
An additional boost in activity is achieved with those
supported salen complexes by changing the acidity of the
support, as a consequence of the stabilization of the reaction
transition state.
The stability of the catalysts was excellent and no deacti-
vation was observed.
With the Pd and Ni salen catalysts the enantioselectivity
was low, but the possibility of using these supports with dif-
ferent transition metal complexes for other enantioselective
reactions is open.
vacuum, an IR band at 1545 cm−1 develops that is due
to the formation of pyridinium ions formed by pyridine
protonation on the surface Brønsted acid sites of the car-
rier.
When hydrogenation of the imines was performed with
Pd and Ni salen complexes grafted on the acidic MCM-41,
ITQ-2, and ITQ-6 an increase in the reaction rate is observed
(Table 4) with respect to the catalysts with nonacidic sup-
ports. At this point, it is worth noting that by grafting on an
acidic delaminated zeolite with well-structured cups at the
surface in which the substrate is adsorbed and concentrated,
TOFs as high as 2 × 105 h−1 can be obtained for the hy-
drogenation of imines. Contrary to what was observed with
Ir–xyliphos, the stability of the catalyst toward recycling is
good, and no decrease in activity was observed after four
times reuse.
Acknowledgments
The authors thanks Comisión Interministerial de Ciencia
y Tecnología (CICYT) (MAT2003-07945-C02-01 and -02,
MAT2000-1368-C02-02) for financial support.
On the basis of these experimental results a reaction path-
way based on the heterolytic hydrogen cleavage (different
from the oxidative addition of H2, generally accepted for
Rh and Ir complexes) is proposed in Scheme 2. The cat-
alyst adds hydrogen to give complex I which is formed
between the catalyst and the hydrogen, involving hydride
ion transfer to the palladium replacing acetate ion, leaving
acetic acid. In the second equilibrium step alkene forms a
π-complex, i.e., complex II with the palladium and simul-
taneous hydride ion transfer taking place from palladium
to alkene. The last step involves transfer of a proton to
the substrate, leading to the separation of the hydrogenated
product from the catalyst; thus a stable catalyst is regener-
ated.
References
[1] D.E. Vos, I.F.J. Vankelecom, P.A. Jacobs (Eds.), Chiral Catalyst Im-
mobilization and Recycling, Wiley-VCH, Weinheim, 2000.
[2] (a) G. Pozzi, F. Cinato, F. Montanari, S. Quici, J. Chem. Soc., Chem.
Commun. (1998) 877;
(b) H.-L. Shyu, H.-H. Wei, G.-H. Lee, Y. Wang, J. Chem. Soc., Dalton
Trans. (2000) 915;
(c) N.H. Lee, E.N. Jacobsen, Tetrahedron Lett. 32 (1991) 6533;
(d) A. Chellamani, N.M.I. Alhaji, S. Rajagopal, J. Chem. Soc., Perkin
Trans. 2 (1997) 229;
(e) T. Niimi, T. Uchida, R. Irie, T. Katsuki, Tetrahedron Lett. 41 (2000)
3647;
(f) H. Nishikori, T. Katsuki, Tetrahedron Lett. 37 (1996) 9245;
(g) P. Piaggio, P. McHorn, D. Murphy, D. Bethell, P.C. Bullman Page,
F.E. Hancock, C. Sly, O.J. Kerton, G.J. Hutchings, J. Chem. Soc.,
Perkin Trans. 2 (2000) 2008;
There is, however, an aspect that we have not achieved
with these catalysts and this is high enantioselectivity. In-
deed, the salen Pd and Ni catalysts, despite their very
high activity, give enantioselectivities in the order of 10–
15%. These are far away from the up to 80% obtained
with Ir–xyliphos. We believe that the low enantioselectiv-
ity of the salen complexes for hydrogenation of imines
can be due to the planarity of the metal complex. The
intermediate species also presents a square-planar struc-
ture and the approximation of the substrate to the cat-
alytic center by only one face is not favored enough in this
case.
(h) M.L. Kantam, B. Bharathi, Catal. Lett. 55 (1998) 235.
[3] (a) D.A. Annis, E.N. Jacobsen, J. Am. Chem. Soc. 121 (1999) 4147;
(b) Ch.E. Song, E.J. Roh, B.M. Yu, D.-Y. Chi, S.Ch. Kim, K.-J. Lee,
J. Chem. Soc., Chem. Commun. (2000) 615;
(c) B.B. De, B.B. Lohray, S. Sivaram, P.K. Dhal, Macromolecules 27
(1994) 1291;
(d) L. Canali, E. Cowan, H. Deleuze, C.L. Gibson, D.C. Sherrington,
J. Chem. Soc., Chem. Commun. (1998) 2561;
(e) L. Canali, E. Cowan, H. Deleuze, C.L. Gibson, D.C. Sherrington,
J. Chem. Soc., Perkin Trans. I (2000) 2055.
[4] (a) D. Pini, A. Mandoli, S. Orlandi, P. Salvadori, Tetrahedron: Asym-
metry 10 (1999) 3883;
(b) E.F. Murphy, L. Schmid, T. Bürgi, M. Maciejewski, A. Baiker, D.
Günther, M. Schneider, Chem. Mater. 13 (2001) 1296.

V. Ayala et al. / Journal of Catalysis 224 (2004) 170–177
177
[5] (a) M.J. Sabater, A. Corma, A. Doménech, V. Fornés, H. García,
J. Chem. Soc., Chem. Commun. (1997) 1285;
[9] (a) A. Corma, V. Fornés, S.B. Pergher, Nature 396 (1998) 353;
(b) A. Corma, V. Fornés, J. Martinez-Triguero, S.B. Pergher, J. Catal.
186 (1999) 57;
(b) S.B. Ogunwumi, T. Bein, Chem. Commun. (1997) 901;
(c) G.-J. Kim, J.-H. Shin, Catal. Lett. 63 (1999) 205;
(d) P. Sutra, D. Brunel, J. Chem. Soc., Chem. Commun. (1996) 2485;
(c) A. Corma, U. Diaz, M.E. Domine, V. Fornés, Angew. Chem. Int.
Ed. 39 (2000) 1499;
(e) X.-G. Zhou, X.-Q. Yu, J.-S. Huang, S.-G. Li, L.-S. Li, C.-M. Che,
J. Chem. Soc., Chem. Commun. (1999) 1789;
(f) E. Möllmann, P. Tomlinson, W.F. Hölderich, J. Mol. Catal. 206
(2003) 137;
(g) S. Ernst, H. Disteldorf, X. Yang, Micropor. Mesopor. Mater. 22
(1998) 457;
(h) S. Ernst, E. Fuchs, X. Yang, Micropor. Mesopor. Mater. 35–36
(2000) 137;
(i) W. Kahlen, H.H. Wagner, W.F. Hölderich, Catal. Lett. 54 (1998)
85.
(d) A. Corma, U. Diaz, M.E. Domine, V. Fornés, J. Chem. Soc., Chem.
Commun. (2000) 137.
[10] (a) A.H. Vetter, A. Berkessel, Synthesis (1995) 419;
(b) M.L. Haslego, C.A. Maryanoff, S. Lorraine, K.L. Sorgi, Heterocy-
cles 35 (1993) 643.
[11] E. Miranda, F. Sánchez, J. Sanz, M.I. Jiménez, I. Martínez Castro,
J. High Resolution Chromatogr. 21 (1998) 225.
[12] (a) A. Corma, M. Iglesias, M.V. Martín, J. Rubio, F. Sánchez, Tetrahe-
dron: Asymmetry 3 (1992) 845;
(b) A. Corma, A. Fuerte, M. Iglesias, F. Sánchez, J. Mol. Catal. A:
Chem. 107 (1996) 225;
(c) M.J. Alcón, A. Corma, M. Iglesias, F. Sánchez, J. Mol. Catal. A:
Chem. 194 (2003) 137.
[6] (a) H.U. Blaser, B. Pugin, F. Spindler, A. Togni, C.R. Chimie 5 (2002)
379;
(b) M. Studer, H.U. Blaser, C. Exner, Adv. Synth. Catal. 345 (2003)
45;
(c) R. Margalef-Catalá, C. Claver, P. Salagre, E. Fernández, Tetrahe-
dron Lett. 41 (2000) 6583;
(d) R. Margalef-Catalá, C. Claver, P. Salagre, E. Fernández, Tetrahe-
dron: Asymmetry 11 (2000) 1469.
[13] G. Sastre, C.R.A. Catlow, A. Corma, J. Phys. Chem. B 103 (1999)
5187.
[14] (a) R. Breslow, A. Graff, J. Am. Chem. Soc. 115 (1993) 10988;
(b) R. Breslow, Acc. Chem. Res. 29 (1991) 317.
[15] (a) B. Pugin, H. Landert, F. Spindler, H.U. Blaser, Adv. Synth.
Catal. 344 (2002) 974;
(b) R. Dorta, D. Broggini, R. Stoop, H. Rüegger, F. Spindler, A. Togni,
Chem. Eur. J. 10 (2004) 267.
[16] T. Ohkuma, H. Takeno, Y. Honda, R. Noyori, Adv. Synth. Catal. 4
(2001) 43.
[7] (a) C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck,
Nature 359 (1992) 710;
(b) J.S. Beck, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt,
C.T.-W. Chu, K.H. Olson, E. Sheppard, S.B. McCullen, J.B. Higgins,
J.L. Schlenk, J. Am. Chem. Soc. 114 (1992) 10834.
[8] A. Corma, V. Fornés, J.M. Guil, S. Pergher, T.L.M. Maesen, J.G. Bu-
glass, Micropor. Mesopor. Mater. 38 (2000) 301.
[17] (a) Ch.A. Brown, V.K. Ahuja, J. Org. Chem. 38 (1973) 2226;
(b) D.R. Anton, R. Crabtree, Organometallics 2 (1983) 855.