Preparation of Organophilic Pd-Montmorillonite, An Efficient Catalyst in Alkyne Semihydrogenation

Article abstract of DOI:10.1006/jcat.2000.2929

Palladium particles incorporated into organophilic montmorillonite (Pd-M) were prepared via a novel synthetic method, mediated by a cationic surfactant stabilizer. The materials were characterized by UV-Vis, ICP-AES, and TEM. Two representative samples (Pd-M1 and Pd-M2), with metal contents of 0.1% and 0.46%, respectively, were investigated in detail. TEM measurements indicated that the diameters of the Pd particles observed were in the range 1.5-6 nm. It is suggested that most of the Pd particles are situated on the external surface of the clay lamellae. Both Pd-M samples exhibited marked catalytic activities and stereoselectivities in the liquid-phase hydrogenation of 1-phenyl-1-butyne. For the production of the cis-alkene stereoisomer, Pd-M2 proved to be less active but more stereoselective than Pd-M1. The stereoselectivities obtained for Pd-M2 in n-hexane (86-88%) were nearly as high as those experienced for the Lindlar catalyst.

Full text of DOI:10.1006/jcat.2000.2929

Journal of Catalysis 194, 146–152 (2000)
doi:10.1006/jcat.2000.2929, available online at http://www.idealibrary.com on
Preparation of Organophilic Pd–Montmorillonite, An Efficient
Catalyst in Alkyne Semihydrogenation
�
,1
�
�
´
˝
A. Mastalir, Z. Kir a´ ly,† Gy. Sz o¨ llosi, and M. Bart o´ k
�
Department of Organic Chemistry and Organic Catalysis Research Group of the Hungarian Academy of Sciences, University of Szeged, D o´ m t e´ r 8,
H-6720 Szeged, Hungary; and †Department of Colloid Chemistry, University of Szeged, Aradi vt. 1, H-6720 Szeged, Hungary
Received February 24, 2000; revised May 9, 2000; accepted May 9, 2000
Pd complexes in the interlayer space of montmorillonite,
Palladium particles incorporated into organophilic montmoril- subsequently applied as catalysts in the hydrogenations of
lonite(Pd–M)wereprepared via a novelsyntheticmethod, mediated unsaturated substrates (14, 15). More recently, the forma-
by a cationic surfactant stabilizer. The materials were characterized
by UV–Vis, ICP-AES, and TEM. Two representative samples (Pd–
M1 and Pd–M2), with metal contents of 0.1% and 0.46%, respec-
tively, were investigated in detail. TEM measurements indicated
tion and stabilization of highly dispersed Pd particles have
been investigated. Ultrafine Pd nanoparticles may be pre-
pared with alcohols (12, 16, 17), with hydro-organoborates
(
18, 19), and also by sonochemical methods (20, 21). An ef-
that the diameters of the Pd particles observed were in the range
.5–6 nm. It is suggested that most of the Pd particles are situated
ficient means of protecting the metal nanoparticles from ag-
gregation is the application of cationic surfactants for stabi-
lization (22–24). The catalytic activity of montmorillonite–
intercalated Pd crystallites has been confirmed in alkene
1
on the external surface of the clay lamellae. Both Pd–M samples
exhibited marked catalytic activities and stereoselectivities in the
liquid-phase hydrogenation of 1-phenyl-1-butyne. For the produc-
tion of the cis-alkene stereoisomer, Pd–M2 proved to be less active and alkyne hydrogenations (25).
but more stereoselective than Pd–M1. The stereoselectivities ob-
Previous investigations revealed that, of all the metals
tained for Pd–M2 in n-hexane (86–88%) were nearly as high as studied for alkyne semihydrogenation, Pd is the most se-
those experienced for the Lindlar catalyst.
�c 2000 Academic Press
lective catalyst (26, 27). The aim of the present work was
Key Words: montmorillonite; palladium; nanoparticles; surfac- the preparation of finely dispersed Pd nanoparticles inter-
tant; 1-phenyl-1-butyne; hydrogenation; stereoselectivity.
calated in montmorillonite by applying a cationic surfac-
tant that renders the clay surface hydrophobic and prevents
metal aggregation. The catalytic activity of the Pd–mont-
morillonite (Pd–M) formed was tested in the liquid-phase
INTRODUCTION
hydrogenation of 1-phenyl-1-butyne.
The preparation and catalytic application of clay–
intercalated metal complexes and nanoparticles, initiated
by Pinnavaia et al. (1, 2), have attracted considerable atten-
EXPERIMENTAL
tion (3–9). One of the most widely used support materials,
montmorillonite, belongs in the group of smectite clay min-
Materials
erals. Montmorillonite, a 2 : 1-type layered aluminosilicate
The calculated amount of a 0.5–3% solution of Pd(acac)2
and a good cation exchanger, permits the incorporation of (Aldrich) in CHCl was added to an aqueous solution of
3
various metal cations or positively charged complexes in- a cationic surfactant, tetradecyltrimethylammonium bro-
to the interlamellar space (1, 10). Metal insertion may be mide (C TAB, a Sigma product) under stirring. The con-
14
enhanced by utilizing the swelling property of montmoril- centration of the C TAB solution was 10–30 times the crit-
14
lonite (1, 11); i.e., the application of an appropriate disper- ical micelle concentration (cmc = 3.9 mM at 298 K in H O).
2
sion medium gives rise to expansion of the clay lamellae Upon reduction with an excess of 55% hydrazine dissolved
and hence results in an increased basal spacing available in water of the bright-yellow colloidal solution obtained, a
for guest species (12, 13).
dark-brown Pd sol was formed. Pd–montmorillonite was
Palladium incorporation into montmorillonite has been prepared by the addition of 50–200 ml of the Pd sol to
the subject of several studies. An important preparation 100–500 ml of a 0.1–2% Na–montmorillonite suspension in
method is the formation of heterogenized homogeneous 0.01% NaCl under vigorous stirring, which proceeded for
2
4 h. The reaction mixture was subsequently centrifuged
1
with a Sorwall RC centrifuge for 30 min (3500 rpm).
Any residual surfactant was removed by washing the
To whom correspondence should be addressed. Fax: 36 62 544200.
E-mail: mastalir@chem.u-szeged.hu.
146
0021-9517/00 $35.00
Copyright �c 2000 by Academic Press
All rights of reproduction in any form reserved.

Pd� MONTMORILLONITE IN ALKYNE HYDROGENATION
147
product with ethanol and then with toluene, in several
centrifugation–redispersion cycles. The products were air- GC and an HP 5970 MSD apparatus, a DB-1 capillary col-
dried and stored in a vacuum desiccator until use (28).
The reaction products were identified with an HP 5890
umn (i.d. = 0.25 mm, l = 60 m) being used for separation.
For comparative experiments, the Lindlar catalyst (5% Quantitative analysis was performed with an SRI 8610 A
Pd/CaCO3 poisoned with lead, an Aldrich product) was gas chromatograph equipped with a DB-Wax capillary col-
applied.
umn (i.d. = 0.53 mm, l = 30 m) and FID. The experimental
1
-Phenyl-1-butyne, the substrate for catalytic hydrogena- error in the product analysis was �0.01–0.15% .
tion, was an Aldrich product of 99% purity and was used
as received.
RESULTS AND DISCUSSION
The proposed mechanism for the successive formation
of the Pd hydrosol and the Pd–organoclay is illustrated in
Fig. 1. While the precursor Pd(acac)2 is readily soluble in
CHCl3, it is almost insoluble in water. To accomplish the
dissolution of Pd(acac)2 in an aqueous system, C14TAB is
added, which tends to increase the apparent solubility of
CHCl3 in water. If the concentration of the surfactant ex-
ceeds the cmc (the concentration at which association of
the surfactant molecules, i.e., micelle formation, occurs),
an aqueous colloidal solution of Pd(acac)2 previously dis-
solved in CHCl3 may be formed via the solubilization ef-
fect. The latter means the formation of a thermodynami-
cally stable colloidal solution in the presence of surfactant
molecules.
Methods
The metal contents of the Pd–M samples were obtained
indirectly by monitoring the turbidity of the Pd sol before
and after adsorption on montmorillonite. The measure-
ments were carried out with a UVIKON 930 spectropho-
tometer operated at a fixed wavelength of 600 nm. A cal-
ibration curve indicated a linear relationship between the
turbidity and the Pd concentration (29). The amount of Pd
particles adsorbed on the montmorillonite was calculated
from the turbidity change in the supernatant, due to ad-
sorption.
Inductively coupled plasma-atomic emission spectro-
scopy (ICP-AES) was employed to determine the metal
contents of the Pd–M samples. The measurements were
carried out with a Jobin Yvon 24 sequential ICP-AES spec-
trometer at 229.7 nm. Before investigation, the samples
were dissolved in a 1 : 2 mixture of aqua regia and 0.5 M
In our case, according to Fig. 1, the apolar system
Pd(acac)2 dissolved in CHCl3 is situated in the internal
apolar part of the micelles formed in the aqueous so-
lution. The UV–Vis spectrum of Pd(acac)2 indicated a
shift in the absorbance maximum of the micellar solution
(
NH4)HF2. The Pd contents were obtained from the emis-
sion intensities by means of a calibration curve.
TEM investigations were performed with a Philips C-10
transmission electron microscope at 100 kV of accelerating
(
340 nm) as compared with that in pure CHCl3 (327 nm).
This may be attributed to a change in the microenviron-
ment of Pd(acac)2, induced by solubilization or, more likely,
by the formation of an adduct between the surfactant and
the palladium salt (21, 31). The addition of an aqueous
hydrazine solution to the micellar system resulted in the
5
voltage, with a magnification of 10 . The samples were dis-
persed in ethanol and subsequently mounted and air-dried
on a plastic film supported by a Formvar grid. The mean
particle diameters d were determined as number-average
P
P
values ( ni di / ni ) from the magnified TEM images, for
2
00 particles on average. As a consequence of the spherical
symmetry of the Pd particles observed, the active surface
area of the Pd (DS) was calculated as DS = 0.885/d (30).
The test reaction was the hydrogenation of 1-phenyl-1-
butyne, performed in an automated liquid-phase reactor
5
at 10 Pa. For each measurement, the mass of catalyst used
was 10 mg and the amount of reactant related to the Pd con-
tent of the catalyst was S : C = 2500 (mol of substrate/mol of
5
Pd). Before reaction, the sample was pretreated in 10 Pa
3
of H2 at 298 K for 1 h, 1 cm of solvent was then added,
and further treatment in H2 was applied under stirring
(
1400 rpm) at the same temperature for 45 min. Stirring
was maintained after introduction of the reactant and the
reaction was monitored with a PC. After the reaction had
been completed, the catalyst was removed by gravity filtra-
tion and the filtrate was analyzed to establish the product
distribution.
FIG. 1. Schematic illustration of the mechanism suggested for the suc-
cessive formation of Pd–hydrosol and Pd–montmorillonite in a micellar
system.

1
48
MASTALIR ET AL.
TABLE 1
Characterization of Organophilic Pd–Montmorillonite Catalysts
Pd [% ]
Pd [% ]
d [nm]
Sample
UV–Vis
ICP-AES
TEM
DS
Pd–M1
Pd–M2
0.09
0.51
0.11
0.42
2.42
3.68
0.36
0.24
0
formation of Pd nanoparticles, stabilized both sterically
and electrostatically by the cationic surfactant molecules
adsorbed on the surface of the particles (Fig. 1). The stabi-
lizing effect of the C14TAB molecules prevented the aggre-
gation of nanosized Pd clusters.
Pd–M was obtained by mixing the Pd hydrosol with
FIG. 2. Particle size distribution obtained for Pd–M1.
+
a dilute Na –montmorillonite suspension. The reaction
+
of Na –montmorillonite with the stabilizing surfactant
molecules resulted in the formation of organophilic alky-
Montmorillonite is well known to be a hydrophilic clay
lammonium montmorillonite, with the simultaneous re- mineral that is usually employed in aqueous systems (1, 10).
lease and the subsequent deposition of the Pd nanopar- However, organophilic Pd–M samples may be readily dis-
ticles on the surface of the clay lamellae (Fig. 1). It should persed in a variety of organic solvents generally used in
be stressed that the surfactant molecules had a key role in liquid-phase catalytic hydrogenation reactions. The trans-
the preparation of Pd–M since they ensured both the sta- formation of 1-phenyl-1-butyne on Pd–M1 at 298 K is il-
bilization of the Pd particles and the cationic exhange of lustrated in Fig. 4. The reaction was carried out in tetrahy-
the clay host by rendering it hydrophobic and thus readily drofuran (THF), which led to a considerable swelling of
applicable in organic media.
L
the clay host (the basal spacings d obtained from previous
Two Pd–M samples with different levels of metal load- XRD data for Pd–M1 in THF and in the dry state were 3.74
ing were selected for further investigations. Table 1 lists and 1.72 nm, respectively) (13).
the mean particle diameters of the samples, obtained from
Pd–M1 exhibited an appreciable catalytic activity in the
TEM, and the Pd contents, determined by UV–Vis spec- hydrogenation of 1-phenyl-1-butyne. The conversion in-
troscopy and ICP-AES. The Pd contents obtained by the creased rapidly with the reaction time, and at 90 min,
two methods are in good agreement, and the actual Pd load- transformation was complete. After this time, there was
ings of Pd–M1 and Pd–M2 are therefore taken as 0.1% and no appreciable change in the product distribution. With re-
0
.46% , respectively.
gard to the product selectivities, the main reaction pathway
TEM measurements revealed the presence of spherical, was the semihydrogenation of 1-phenyl-1-butyne, result-
monodispersed Pd particles in the range 1.5–6 nm, situ- ing in the formation of cis- and trans-alkenes, with a
ated mostly on the external surface sites of the montmoril-
lonite. The particle size distributions are displayed in Figs. 2
and 3.
Figures 2 and 3 indicate a fairly narrow range of particle
diameters for both Pd–M samples. For Pd–M1, the major-
ity of the Pd nanoparticles fell in the range 1–3 nm, and the
largest particle diameter observed was 5 nm. For Pd–M2,
a different size distribution was obtained, which suggested
the appearance of a small amount of Pd particles larger
than 6 nm. This may account for an increased mean parti-
cle diameter as compared with Pd–M1, which resulted in a
lower active surface area of Pd, as can be seen in Table 1.
The occurrence of a large amount of ultrafine (1–2 nm)
Pd particles for Pd–M2 may be attributed to the stabilizing
effect of the surfactant, which prevents aggregation if the
metal loading is increased. In fact, only a few aggregates ap-
peared, whereas the frequency of the 1- to 2-nm particles
increased considerably.
FIG. 3. Particle size distribution obtained for Pd–M2.

Pd� MONTMORILLONITE IN ALKYNE HYDROGENATION
149
It is clear that the introduction of n-hexane decreased the
conversion of 1-phenyl-1-butyne to some extent. This may
be explained by the evidence that in n-hexane no swelling
of Pd–M1 occurred (dL = 1.92 nm), which implies that the
clay lamellae are more aggregated as compared with those
in THF. It follows that the exposed Pd surface area avail-
able for reactant molecules is lower in n-hexane than in
THF. However, the application of n-hexane had a positive
effect on the stereoselectivity of 1-phenyl-cis-1-butene for-
mation, which displayed little variation with the reaction
temperature. Nevertheless, the highest value of 82% was
observed at the subambient temperature of 283 K. Sim-
ilarly, the low stereoselectivity of trans-alkene formation
(
3–6% ) underwent no significant change upon an increase
FIG. 4. Transformation of 1-phenyl-1-butyne on Pd–M1 as a function of the temperature. There was no significant change of
5
of the reaction time. Mass of catalyst, 10 mg; T = 298 K; p = 10 Pa; solvent,
conversion below 298 K, but a marked increase was ex-
perienced at this temperature. It appears that reduction
of the reaction temperature neither eliminates nor appre-
ciably decreases the formation of 1-phenylbutane, which
suggests that the formation of stereoisomeric alkenes and
3
1
cm of THF; S : C = 2500.
predominance of the cis isomer. The highest cis stereoselec-
tivities were observed at the beginning of the reaction. The
selectivity of the formation of the overhydrogenation prod-
uct 1-phenylbutane increased at the expense of 1-phenyl-
cis-1-butene and reached its highest value at 60 min. The
selectivities of the formation of both 1-phenylbutane and
the trans stereoisomer exhibited no appreciable change af-
ter 90 min, with the amount of 1-phenyl-trans-1-butene be-
ing substantially lower in the reaction. It is apparent that a
prolonged reaction time is not favoured in terms of cis se-
lectivity since it tends to reinforce the overhydrogenation
of 1-phenyl-cis-1-butene.
Previous literature results indicated that the application
of subambient temperature and an apolar solvent tends to
increase the stereoselectivity of cis-alkene formation (32).
We therefore changed the dispersion medium to n-hexane
and studied the temperature dependence by applying a re-
action time of 60 min. The results obtained are illustrated
in Fig. 5.
1
-phenylbutane is more likely to occur via simultaneous
processes than in consecutive reactions (27, 33). Although
an augmentation of alkane selectivity in response to a tem-
perature increase could have been expected (27, 34), there
was no evidence of this in our case.
Figure 6 demonstrates that the catalytic activity of Pd–
M2 in THF was lower than that of Pd–M1. This may be
ascribed to the increase in the mean particle size of the
Pd particles (Figs. 2 and 3), which implies a lower active
metal surface area available for reactants, despite the ap-
pearance of ultrafine Pd particles (Table 1). Nevertheless,
a full transformation of the reactant can be achieved over
a period of 180 min. With regard to the product selectivi-
ties, the stereoselective formation of 1-phenyl-cis-1-butene
is clearly established; it was more pronounced than that
for Pd–M1. The variations of the product selectivities with
FIG. 5. Transformation of 1-phenyl-1-butyne on Pd–M1 as a function
FIG. 6. Transformation of 1-phenyl-1-butyne on Pd–M2 as a function
5
5
of the reaction temperature. Mass of catalyst, 10 mg; p = 10 Pa; reaction
of the reaction time. Mass of catalyst, 10 mg; T = 298 K; p = 10 Pa; sol-
3
3
time, 60 min; solvent, 1 cm of n-hexane; S : C = 2500.
vent, 1 cm of THF; S : C = 2500.

1
50
MASTALIR ET AL.
solvent effect is suggested by the differences in the initial
rates determined in THF and n-hexane. As indicated by the
basal spacings mentioned above, swelling of the clay host
was much more important in THF than in n-hexane. The
increased swelling observed in THF is accompanied by a
considerable disaggregation of the clay lamella packages.
Hence, the amount of active Pd particles exposed to the
reactant molecules is also increased, despite the evidence
that the Pd particles formed were mainly situated on the ex-
ternal sites of the montmorillonite. For a given catalyst, the
stereoselectivities of cis alkene formation in different sol-
vents were very close (in fact, they were exactly the same for
Pd–M2), and thus they were clearly unaffected by the dis-
persion medium. The stereoselectivities obtained for the
formation of 1-phenyl-cis-1-butene were in the same or-
der as those reported for the semihydrogenation of pheny-
lacetylene on hydrophilic Pd–montmorillonite (25). How-
ever, the initial rates determined for the present Pd–Ms
were considerably higher, especially for Pd–M1. Thus, it is
concluded that hydrophobic Pd–M samples are more effi-
cient catalysts than hydrophilic ones in alkyne semihydro-
genation.
FIG. 7. Transformation of 1-phenyl-1-butyne on Pd–M2 as a function
5
of the reaction temperature. Mass of catalyst, 10 mg; p = 10 Pa; reaction
3
time, 60 min; solvent, 1 cm of n-hexane; S : C = 2500.
reaction time were found to be similar to those obtained
for Pd–M1. However, the selectivity of 1-phenylbutane for-
mation was significantly lower for Pd–M2 than for Pd–M1
and was very close to that of the trans isomer. It follows
that overhydrogenation is less of a problem for Pd–M2,
which was found to be more stereoselective than Pd–M1.
As revealed by Fig. 7, the stereoselectivity of cis-alkene for-
mation may be further increased by applying n-hexane as
a dispersion medium.
The most important observation provided by Fig. 7 is
the remarkably high stereoselectivity of 1-phenyl-cis-1-
butene formation, which remains practically unchanged
86–88% ) throughout the entire temperature range. For
Pd–M2, the conversion of 1-phenyl-1-butyne at 298 K for
0 min decreased from 71% to 41% when the disper-
As reported in the literature, trans-alkenes are always
formed in the hydrogenation of alkynes, either as initial
products or as a result of isomerization of the cis stereoiso-
mer (27, 35). It is generally accepted that direct trans-alkene
formation may occur via the addition of molecular hydro-
gen to the adsorbed alkyne molecule (36). In our case, the
experimental evidence indicated that a small amount of
trans-alkene was always formed in the reaction mixture be-
fore the alkyne had been consumed. On the other hand,
a drop in the cis stereoselectivity parallel to an increase in
the transstereoselectivity wasobserved for both Pd–M sam-
ples at reaction times between 30 and 60 min (Figs. 4 and 6),
which suggested that cis → trans isomerization took place.
Hence, the current study supported the evidence that 1-
phenyl-trans-1-butyne may be formed via both routes (35).
Nevertheless, it should be stressed that, for Pd–M2, the iso-
merization pathway was far less important. This may be
related to the large number of ultrafine Pd particles (see
Fig. 3). The activation energiesobtained from the Arrhenius
(
6
sion medium was changed from THF to n-hexane (see
also Fig. 6). This loss of activity was far more pronounced
than that observed for Pd–M1 under the same conditions
97% ⇒ 79% ). Moreover, the application of subambient
(
temperatures decreased the catalytic activity of Pd–M2
even further. On the other hand, no noteworthy changes in
the selectivities of the minor products as a function of tem-
perature could be detected. Whereas the stereoselectivi-
ties obtained for 1-phenyl-trans-1-butene were very similar
for the two Pd–M samples, the selectivity of the overhy-
drogenation product for Pd–M2 was decreased consider-
ably (6–9% ). It follows that Pd–M2 may be regarded as an
extremely efficient catalyst in the stereoselective semihy-
drogenation of 1-phenyl-1-butyne. The initial rates and the
product selectivities obtained at the same conversion level
for Pd–M1 and Pd–M2 in THF and n-hexane are listed in
Table 2.
�
1
plots were 91.5 and 88.5 kJ mol for Pd–M1 and Pd–M2,
respectively. Such high values may reflect the adsorption
TABLE 2
Initial Rates and Selectivities of Pd–M1 and Pd–M2 Determined
at 298 K in THF and n-Hexane
3
Conv. ri [cm /(min mmol Stotal alkene
Scis/
Sample
Solvent
[% ]
of Pd)]
[% ]
S(cis+trans)
For both THF and n-hexane, the initial rates obtained for Pd–M1 THF
78
78
40
40
1789
842
1224
369
84
81
94
94
0.929
0.938
0.936
0.936
Pd–M1 n-Hexane
Pd–M1 were higher than those observed for Pd–M2, which
confirms that the increase in the mean particle diameter of
Pd decreases the catalytic activity. Moreover, a pronounced
Pd–M2 THF
Pd–M2 n-Hexane

Pd� MONTMORILLONITE IN ALKYNE HYDROGENATION
151
TABLE 3
Transformation of 1-Phenyl-1-butyne on Lindlar Pd and Pd–M2
Conversion and selectivities [% ] Lindlar Pda
Conversion and selectivities [% ] Pd–M2b
T [K]
Conv.
Scis
Strans
Scis/S(cis+trans)
Conv.
Scis
Strans
Scis/S(cis+trans)
2
2
2
78
83
98
10
17
42
93
93
92
3
3
5
0.969
0.969
0.948
13
15
41
87
86
88
6
5
6
0.925
0.945
0.936
a
3
m = 10 mg; solvent, 1 cm of n-hexane; S : C = 250; reaction time, 60 min.
b
3
m = 10 mg; solvent, 1 cm of n-hexane; S : C = 2500; reaction time, 60 min.
heat of the substrate at saturation coverage, as reported cationic exchange. The catalytic activities of both Pd–M
for the liquid-phase hydrogenation of 2-hexyne, where an samples studied were confirmed in the liquid-phase semi-
�
1
equally high activation energy of 87.9 kJ mol was deter- hydrogenation of 1-phenyl-1-butyne. Nevertheless, their
mined (37). catalytic performances were different. Pd–M2, which was
For Pd catalysts, the Pd–H system readily undergoes con- less active but more stereoselective than Pd–M1, is re-
version into the �-hydride, which may affect the catalytic garded as more efficient for the above reaction. At the same
performance (38). In our case, the Pd contents of both Pd– conversion level, the application of n-hexane as a disper-
M samples appear as small clusters, and according to the sion medium afforded stereoselectivities similar to those
mean particle diameters displayed in Table 1, they are very obtained in THF, and the role of subambient temperatures
unlikely to form �-hydrides, as reported by Nandi et al. was rather insignificant. As a consequence of the high
(
39). Hence, our experimental results are in accordance dispersion of the Pd particles, the formation of �-hydride
with those of previous studies on the semihydrogenation is unlikely. The amount of Pd–M2 required for similar
of phenylacetylene, where it was suggested that the best catalytic performance was considerably lower than that of
stereoselectivities were obtained for highly dispersed Pd the Lindlar catalyst.
particles, which displayed no �-hydride formation at 298 K
(
40).
The experimental data in Table 3 indicate that the stere-
ACKNOWLEDGMENT
Financial support through OTKA Grants T026430 and T025002 is
gratefully acknowledged.
oselectivity of Pd–M2 is comparable with that of the Lindlar
catalyst, which is considered to be the most stereoselective
of all Pd catalysts investigated in alkyne semihydrogenation
REFERENCES
(
27).
There was no appreciable difference between the con-
1
2
. Pinnavaia, T. J., Science 220, 4595 (1983).
. Pinnavaia, T. J., Raythatha, R., Lee, J. G., Halloran, L. J., and Hoffman,
J. F., J. Am. Chem. Soc. 101, 6891 (1979).
3. Taqui Khan, M. M., Samad, S. A., and Siddiqui, M. R. H., J. Mol. Catal.
0, 97 (1989).
. Adams, J. M., Bylina, A., and Graham, S. H., J. Catal. 75, 190
1982).
. Raythatha, R., and Pinnavaia, T. J., J. Catal. 80, 47 (1983).
6. Bart o´ k, M., Sz o¨ ll o˝ si, Gy., Mastalir, A´ ., and D e´ k a´ ny, I., J. Mol. Catal.
A: Chem. 139, 227 (1999).
. Ballantine, J. A., Purnell, H., Rayanakorn, M., Thomas, J. M., and
Williams, K. J., Chem. Commun. 9 (1981).
. Shimazu, S., Ro, K., Sento, T., Ichikuni, N., and Uematsu, T., J. Mol.
Catal. A: Chem. 107, 297 (1996).
. Halligudi, S. B., Bhatt, K. N., and Venkatasubramanian, K., React.
Kinet. Catal. Lett. 48, 547 (1992).
versions obtained for the two samples, and the stereoselec-
tivities of 1-phenyl-cis-1-butene formation on Pd–M2 ap-
proached those observed for Lindlar Pd reasonably well,
especially at 298 K. The efficiency of Pd–M2 is further sup-
ported by the evidence that the reaction on the Lindlar
catalyst was performed with S : C = 250, which was 1 mag-
nitude lower than that for Pd–M2. That is, to achieve a sim-
ilar transformation for a given amount of substrate, Pd–M2
may be applied in a considerably smaller amount than the
Lindlar catalyst.
5
4
5
(
7
8
9
CONCLUSIONS
1
0. Laszlo, P., Science 235, 1473 (1987).
´
Pd–montmorillonite was prepared via a novel synthetic 11. Patzk o´ , A., and D e´ k a´ ny, I., Colloids Surf. 71, 299 (1993).
´
1
1
1
2. Kir a´ ly, Z., D e´ k a´ ny, I., Mastalir, A., and Bart o´ k, M., J. Catal. 161, 401
method that provided finely distributed Pd particles
situated mostly on the external surface of the clay lamellae.
The surfactant C14TAB had a key role in preparation since
it ensured both the stabilization of nanosized Pd clusters
and the formation of organophilic montmorillonite by
(
1996).
´
3. Mastalir, A., Notheisz, F., Kir a´ ly, Z., Bart o´ k, M., and D e´ k a´ ny, I., Stud.
Surf. Sci. Catal. 108, 477 (1997).
4. Ravi Kumar, K., Choudary, B. M., Jamil, Z., and Thyagarajan, G.,
Chem. Commun. 130 (1986).

1
52
MASTALIR ET AL.
1
1
1
1
1
2
5. Choudary, B. M., Sharma, G. V. M., and Bharathi, P., Angew. Chem. 27. Bart o´ k, M., “Stereochemistry of Heterogeneous Metal Catalysis.”
Int. Ed. Engl. 28, 465 (1989). Wiley, Chichester, 1985.
6. Bradley, J. S., Hill, E. W., Klein, C., Chaudret, B., and Duteil, A., Chem. 28. Kir a´ ly, Z., Veisz, B., Mastalir, A, R a´ zga, Z., and D e´ k a´ ny, I., Chem.
Mater. 5, 254 (1993). Commun. 1925 (1999).
7. Esumi, K., Itakura, T., and Torigoe, K., Colloids Surf. 82, 111 29. Furlong, D. N., Launikonis, A., Sasse, H. F., and Sanders, J. V., J. Chem.
1994). Soc. Faraday Trans. 80, 571 (1984).
8. B o¨ nnemann, H., Brijoux, W., and Joussen, T., Angew. Chem. Int. Ed. 30. Aben, P. C., J. Catal. 10, 224 (1968).
´
(
Engl. 29(3), 273 (1990).
31. Bouquillon, S., du Moulinet d’Hardemare, A., Averbuch-Pouchot,
9. B o¨ nnemann, H., Brijoux, W., Brinkmann, R., Dinjus, E., Joussen, T.,
and Korall, B., Angew. Chem. Int. Ed. Engl. 30(10), 1312 (1991).
0. Okitsu, K., Bandow, H., and Maeda, Y., Chem. Mater. 8, 315
M-Th., H e´ nin, F., and Muzart, J., Polyhedron 18, 3511 (1999).
32. Gutman, H., and Lindlar, H., in “Chemistry of Acetylenes” (H. G.
Viehe, Ed.), Chap. 6. Dekker, New York, 1969.
(
1996).
33. Bond, G. C., Webb, G., Wells, P. B., and Winterbottom, J. M., J. Catal.
1, 74 (1962).
2
2
1. Arul Dhas, N., and Gedanken, A., J. Mater Chem. 8, 445 (1998).
2. Drognat Landr e´ , P., Richard, D., Draye, M., Gallezot, P., and Lemaire, 34. Mann, R. S., and Khulbe, K. C., Can. J. Chem. 45, 2755 (1967).
M., J. Catal. 147, 214 (1994). 35. Burwell, R. L., Jr., Chem. Rev. 57, 895 (1957).
3. Yonezawa, T., Tominaga, T., and Toshima, N., Langmuir 11, 4601 36. Bond, G. C., and Wells, P. B., J. Catal. 5, 65 (1966).
2
2
2
2
(
1995).
37. Ulan, J. G., and Maier, W. F., J. Mol. Catal. 54, 243 (1989).
38. Palczewska, W., in “Hydrogen Effects in Catalysis” (Z. Pa a´ l and P. G.
Menon, Eds.), Chap. 14. Dekker, New York, 1988.
4. Reetz, M. T., Helbig, W., Quaiser, S. A., Stimming, U., Breuer, N., and
Vogel, R., Science 267, 367 (1995).
5. Crocker, M., Buglass, J. G., and Herold, R. H. M., Chem. Mater. 5, 105 39. Nandi, R. K., Georgopoulos, P., Cohen, J. B., Butt, J. B., Burwell, R. L.,
(
1993).
and Bilderbeck, D. H., J. Catal. 77, 421 (1982).
40. Carturan, G., Facchin, G., Cocco, G., Enzo, S., and Navazio, G.,
J. Catal. 76, 405 (1982).
6. Dobson, N. A., Exglinton, G., Krishnamurti, M., Raphael, R. A., and
Willis, R. G., Tetrahedron 16, 16 (1961).