Structure and Properties of Bimetallic Colloids Formed in Polystyrene-block-Poly-4-vinylpyridine Micelles: Catalytic Behavior in Selective Hydrogenation of Dehydrolinalool

Article abstract of DOI:10.1006/jcat.2000.3039

Catalytic properties of palladium and bimetallic (PdAu, PdPt, and PdZn) nanoparticles formed in block copolymer micelles derived from polystyrene-block-poly-4-vinylpyridine (PS-b-P4VP) were studied in dehydrolinalool (DHL) hydrogenation. FTIR spectroscopy on CO adsorption and XPS show that the second metal (Au, Pt, or Zn) acts as a modifier toward Pd, changing both its electronic structure and its surface geometry. In turn, this change provides higher catalytic activity of bimetallic particles formed in PS-b-P4VP micelles compared to Pd micelles, which can be ascribed mainly to an increase in the number of active centers on the particle surface. High selectivity of DHL hydrogenation (99.8% at 100% conversion) was achieved for all the Pd and bimetallic micellar catalysts, by chemical modification of the nanoparticle surface with pyridine units. Kinetic study of DHL hydrogenation, along with computational kinetic models, allowed us to describe a hydrogenation mechanism with these catalysts.

Full text of DOI:10.1006/jcat.2000.3039

Journal of Catalysis 196, 302–314 (2000)
doi:10.1006/jcat.2000.3039, available online at http://www.idealibrary.com on
Structure and Properties of Bimetallic Colloids Formed in
Polystyrene-block-Poly-4-vinylpyridine Micelles: Catalytic
Behavior in Selective Hydrogenation of Dehydrolinalool
Lyudmila M. Bronstein, ,1 Dmitrii M. Chernyshov,† Ilya O. Volkov,† Marina G. Ezernitskaya,† Peter M.
Valetsky,† Valentina G. Matveeva,‡ and Esther M. Sulman‡
Chemistry Department, Indiana University, Bloomington, Indiana 47405; †Nesmeyanov Institute of Organoelement Compounds, 28 Vavilov Street,
Moscow 117813, Russia; and ‡Tver Technical University, Tver 170000, Russia
Received April 17, 2000; revised August 7, 2000; accepted August 7, 2000
synthesis of vitamins (A, E, and K) and fragrant substances
(linalool) (9). Synthesis of linalool (3,7- dimethyloctadiene-
1,6-ol-3, LN) can be carried out by selective hydrogena-
tion of dehydrolinalool (3,7-dimethyloctaen-6-yne-1-ol-3,
DHL). If hydrogenation goes on non selectively, dihydroli-
nalool (3,7-dimethyloctaen-6-ol-3, DiHL) forms as well
(Scheme 1).
Catalytic properties of palladium and bimetallic (PdAu, PdPt,
and PdZn) nanoparticles formed in block copolymer micelles
derived from polystyrene-block-poly-4-vinylpyridine (PS-b-P4VP)
were studied in dehydrolinalool (DHL) hydrogenation. FTIR spec-
troscopy on CO adsorption and XPS show that the second metal
(Au, Pt, or Zn) acts as a modifier toward Pd, changing both its
electronic structure and its surface geometry. In turn, this change
provides higher catalytic activity of bimetallic particles formed in
PS-b-P4VP micelles compared to Pd micelles, which can be ascribed
mainly to an increase in the number of active centers on the parti-
cle surface. High selectivity of DHL hydrogenation (99.8% at 100%
conversion) was achieved for all the Pd and bimetallic micellar cata-
lysts, by chemical modification of the nanoparticle surface with
pyridine units. Kinetic study of DHL hydrogenation, along with
computational kinetic models, allowed us to describe a hydrogena-
c
tion mechanism with these catalysts.
2000 Academic Press
Key Words: catalysis; hydrogenation; dehydrolinalool; poly-
styrene-block-poly-4-vinylpyridine; block copolymers; micelles;
bimetallic colloids.
SCHEME 1
Traditionally, DHL hydrogenation is carried out using a
Lindlar catalyst (Pd/CaCO3 modified with Pb acetate and
quinoline), which provides selectivity of 95% at 100% con-
version (10). However, the use of these modifiers leads to
the pollution of end products and has an unfavorable im-
pact on the environment. Another trend in triple-bond hy-
drogenation was the use of polymer Pd-containing cata-
lysts, which gave higher selectivity (up to 98% in the
DHL hydrogenation) and stability without any modifica-
tion (11, 12); however, the activity of such catalysts was
not sufficiently high. Recently, we developed new col-
loidal catalysts formed in block copolymer micelles de-
rived from the polystyrene-block-poly-4-vinylpyridine (PS-
b-P4VP) block copolymer (13, 14). In selective solvents
(toluene, and THF) these block copolymers form micelles
with the P4VP cores, which serve as nanoreactors for metal
nanoparticle formation. In our preceding paper (5) we de-
scribed the hydrogenation of DHL with Pd colloids stabi-
lized in PS-b-P4VP micelles. A crucial feature of colloidal
catalysts formed in block copolymer micelles is a polymeric
INTRODUCTION
Nowadays, acceleration of chemical reactions, along with
development of highly selective and robust catalysts, re-
main one of the most vigorously discussed topics in scien-
tific literature. This strongly stimulates the development of
new catalytic systems, which might combine the advantages
of both heterogeneous and homogeneous catalysts. Among
such systems, polymeric catalysts, i.e., catalysts where metal
complexes or metal particles are located in polymeric ma-
trices, were expected to be promising (1, 2). They were ac-
tively studied in hydrogenation (3–6) and oxidation (7, 8) of
various substrates. Hydrogenation of long-chain acetylene
alcohols and unsaturated ketones is of particular interest
because hydrogenation products are intermediates in the
1 To whom correspondence should be addressed. Fax: (812) 855-8300.
E-mail: lybronst@indiana.edu.
302
0021-9517/00 $35.00
c
Copyright
2000 by Academic Press
All rights of reproduction in any form reserved.

CATALYTIC PROPERTIES OF BIMETALLIC COLLOIDS
303
environment in the micelle cores (15, 16). The P4VP micelle at room temperature (molar ratio N/Pd = 3/1, Pd/Au = 4/1,
cores not only control the nanoparticle size and size distri- Pd/Pt = 4/1, and Pd/Zn = 4/1). For bimetallic systems,
bution (13, 14, 17, 18) but also provide a modification of the addition of both salts was carried out simultaneously.
the nanoparticle surface due to pyridine units. Moreover, After 24 h of stirring, the resulting metal-containing poly-
this kind of modification is fairly permanent, unlike the mer solution was filtered from the unreacted metal com-
low molecular modifiers mentioned above. This stability of pounds or insoluble by-products of the reaction (KCl in the
modification, which governs the stability of catalytic prop- case of Zeise salt) through a Millipore 0.45- m PTEF filter.
erties and high selectivity, was discussed in our preceding Reduction of Pd or bimetallic samples was performed in a
papers (5, 19). Pd colloids formed in PS-b-P4VP micelles two-neck flask equipped with a stopcock, a rubber septum,
were found to be promising catalysts for hydrogenation of and a Teflon stir bar. The reducing agent (SH) was used
long-chain acetylene alcohols combining high selectivity, in fivefold molar excess. In all the cases reduction was car-
stability, and satisfying activity. Nevertheless, an increase ried out under argon after a standard degassing procedure
in catalytic activity is always desirable. Incorporation of a (14). Airless transfer of solutions was achieved with the
second metal into metal colloids often modifies their prop- Hamilton syringe technique. To isolate metal-colloid-
erties, increasing activity of the catalysts (20–22). This can containing block copolymers for further examination,
be caused by a change in the electronic structure of the main block copolymer solutions were precipitated with degassed
metal (for one, Pd or Pt), by a change in the morphology of petroleum ether and isolated under Ar. Resulting solids
the particle, or by some other changes (23–25). Recently, we were dried for 24 h in vacuo (0.5 mbar). Colloid-containing
described the higher activity of bimetallic PdAu colloids in block copolymers were easily redispersable in all the sol-
hydrogenation of cyclohexene and cyclohexadiene (14). In vents that are good for polystyrene (THF and toluene). To
the present paper we report the synthesis and properties of avoid oxidation, metal-containing block copolymer sam-
PdAu, PdPt, and PdZn bimetallic colloids synthesized at a ples were stored under Ar. Metal contents in the end poly-
molar ratio of 4/1 in PS-b-P4VP micelles and their catalytic mers were determined by X-ray fluorescence analysis as
behavior in DHL hydrogenation. Because the amount of described elsewhere (28) (see Table 1).
Pd in all the bimetallic nanoparticles is much higher than
that of the second metal, Pd mainly determines the catalytic
properties.
Measurements
FTIR spectra of CO adsorbed on colloidal catalysts were
recorded with a Nicolet FTIR spectrometer in the spectral
region of 2500–1500 cm 1 with a resolution of 1 cm 1. Typ-
ically, a preparation of samples for the FTIR examination
was performed as follows. A freshly synthesized solution
of metal colloids in block copolymer micelles was concen-
trated to 15–20 wt% solid content using an Amicon ultra-
filtration device. The given solution was degassed under
vacuum of 1 mbar and exposed to a slow carbon monoxide
bubbling. Saturation of colloidal solutions with CO contin-
ued for 2 h at room temperature and then FTIR spectra
were recorded.
METHODS
Materials
Diblock copolymer polystyrene-block-poly-4-vinylpy-
ridine (PS-b-P4VP), Mn = 19,400, Mw = 22,500, and X4-vp
(relative 4-VP content) = 0.340, was prepared via living an-
ionic polymerization (Max Plank Institute of Colloids and
Interfaces, Potsdam/Golm, Germany) according to a syn-
thetic procedure described elsewhere (26, 27). HAuCl4
3H2O, Pd(CH3COO)2, Zn(CH3COO)2, K[Pt(C2H4)Cl3]
H2O (Zeise salt), LiB(C2H5)3H (SH, 1 M solution of
LiB(C2H5)3H in THF)) were obtained from Aldrich and
used as received. Toluene (Aldrich) was twice distilled un-
der KOH. Petroleum either (Fluka) was distilled under
metallic sodium in the Ar counterflow. Other solvents were
used without additional purification.
TABLE 1
Elemental Analysis Data for Mono- and Bimetallic-Colloid-
Containing Block Copolymers
Metal
compound
used
Metal
content
wt%
Molar ratio
obtained
Pd : Mt
Polymer
sample
Preparation of Catalysts
PS-b-P4VP/Pd
Pd(CH3COO)2
Pd: 6.73
1
Synthesis of block copolymer micelles containing mono-
and bimetallic nanoparticles was carried out by the method
described elsewhere (13, 14). For this, PS-b-P4VP samples
were dissolved in toluene at a concentration of 10 g/L un-
der stirring (for 1 h). The metal compounds were added to
the polymer solutions under air to avoid partial reduction
of Pd2+ by 4-vinylpyridine units (14) and stirred for 24 h
PS-b-P4VP/PdAu
Pd(CH3COO)2
HAuCl4 3H2O
Pd: 5.89
Au: 3.01
3.62/1
PS-b-P4VP/PdPt
PS-b-P4VP/PdZn
Pd(CH3COO)2
K[Pt(C2H4)Cl3] H2O
Pd: 5.61
Pd: 2.68
3.86/1
3.67/1
Pd(CH3COO)2
Zn(CH3COO)2
Pd: 6.16
Zn: 1.03

304
BRONSTEIN ET AL.
TABLE 2
X-ray photoelectron spectroscopy (XPS) was carried
out with a two-chamber XSAM-800 spectrometer (Kratos,
UK). For photoelectron excitation, the characteristic
MgK radiation was used; the power (h = 1253.6 eV) of
the X-ray gun did not exceed 75 W (15 kV, 5 mA). The
spectra were recorded at a pressure of 10 9 Torr.
Wide-angle X-ray scattering (WAXS) measurements
were performed on a Rigaku D/max-RC X-ray diffrac-
tometer operating at 12-kV CuK radiation. Because of
weak diffraction signals, spectra were recorded with an an-
gle resolution of 0.1 . Estimation of the average crystallite
size (or zones of the coherent scattering) was performed
using the Debye–Scherer formula.
Electronic Properties of Metals Studied
Electro-
Ionization
Standard
Metal
negativity (30)
potential (eV) (31)
potential (V) (32)
Pd
2.20
2.54
8.33
9.22
+0.915a
Au
+1.002a
+1.348b
Pt
Zn
2.28
1.65
9.0
9.391
+0.749a
0.77b
a Aqueous medium.
b Nonaqueous medium.
Samples for transmission electron microscopy (TEM)
were prepared by evaporation of 10 mol/L THF solu-
tions under air. Electron micrographs of the samples were
obtained with a Zeiss 912 Omega electron microscope.
UV spectra were recorded with a Beckman DU-530 spec-
trophotometer.
4
pends on the ability of the metal to be reduced in the partic-
ular environment. Traditionally, this ability is characterized
by the ionization potential (Table 2) (14, 23). Comparing
Pd, Pt, and Au, one can assume that reduction of Pt and
Au compounds proceeds faster than reduction of Pd com-
pounds, providing core–shell structures for PdAu and PdPt
bimetallics (23–25). On the other hand, Zn, which cannot
be reduced easily, has a high ionization potential (Table 2).
This means that the ionization potential value should not be
used in the estimation of the ability of metal ions to be re-
duced. Among known parameters, the ability of metal ions
to accept electrons and to be reduced can be characterized
most closely by standard (or formal) potentials (Table 2),
which, however, strongly depend on the reaction medium
(aqueous or nonaqueous, presence of ligands) and condi-
tions of examination (29–31). From the data presented in
Table 2 one can see that AuCl4 (in any media) can be re-
duced easier than Pd2+ (from Pd acetate), which can govern
the possibility of core(Au)–shell(Pd) structure. In contrast,
the values of standard potentials for Pd and Pt do not as-
sume this kind of structure. The negative value for Zn2+
(from Zn acetate) shows rather the inability of zinc ions to
be reduced by common reducing agents (for one, superhy-
dride). So from these facts one can propose that the core–
shell structure is possible solely for the PdAu pair, though
the polymeric environment can change the ability of ions
to be reduced.
DHL Hydrogenation
Reactor and analysis. The catalytic reactions were car-
ried out in a glass batch isothermal reactor installed in a
shaker (maximum 960 shakings min) and connected to a
gasometric burette. The reactor was equipped with two in-
lets: for catalyst, solvent, and substrate loading and for hy-
drogen feeding. Before the substrate was charged into the
reactor, the catalyst was pretreated with H2 for 60 min at
90 C. DHL concentration (Co), catalyst amount (Cc), and
hydrogenation temperature (th) have been varied. The ex-
periments were carried out at atmospheric pressure.
Analysis of the reaction mixture was carried out by gas
chromatography (GC) using a CHROM-5 chromatograph
with FID and glass column 3 m/3 mm. The column has been
filled with solid-phase Chromaton N (0.16–0.20 mm) satu-
rated with 20 M carbowax (10% of liquid phase to support
weight).
RESULTS AND DISCUSSION
1. Physicochemical Characterization of Mono-
and Bimetallic Colloids Formed in Block
Copolymer Micelles
As known, FTIR study of carbon monoxide adsorbed on
colloidal particles can give useful information about struc-
tural peculiarities and active sites of metallic systems (2).
The choice of second metal or metal modifier for Pd was Some investigations in this area have shown that there are
based on its electronic properties. Au and Pt have higher correlations between the size of colloidal particles and their
electronegativity than Pd, so they can be acceptors of elec- crystallinity and the relative occupancies of CO molecules
trons from Pd (Table 2). In contrast, Zn, whose electroneg- in terminal (2040–2060 cm 1) or in bridging (1920–
ativity is very low, is rather a donor to Pd, so its influence 1940 cm 1) positions (32–34). According to Ref. (32), small
should be the opposite.
and probably amorphous metal particles exhibit presum-
Another feature of the synthesized bimetallics derives ably linear (terminal) CO adsorption. In contrast, an in-
from the method of preparation. As mentioned in Meth- crease of metal particle size gives rise to the fraction of
ods, two metal compounds were incorporated into block bridging CO molecules (34). In the case of bimetallic parti-
copolymer micelles and their reduction occurred simulta- cles, adsorbed carbon monoxide is also regarded as a useful
neously. The structure of the particle formed strongly de- infrared probe of surface composition (35). Relying on this

CATALYTIC PROPERTIES OF BIMETALLIC COLLOIDS
305
FIG. 1. FTIR spectra of CO adsorption on Pd (a) and bimetallic PdAu (b), PdPt (c), and PdZn (d) colloids formed in the PS-b-P4VP micelles.
information, we studied CO adsorption on monometallic solutions (23–25), where PdPt colloids prepared by simul-
Pd and bimetallic PdAu, PdPt, and PdZn colloids formed in taneous reduction of Pd and Pt compounds had core–shell
the PS-b-P4VP micellar solutions (Fig. 1). One can see that structures.
all the samples except PdAu show both terminal and bridg-
As mentioned above, the presence of bridging CO ad-
ing CO adsorption. Moreover, PdPt and PdZn colloids dis- sorption might be an indication of larger and crystalline par-
play two bands in the region of terminal absorption, while ticles. According to our earlier publication (14), the mean
Pd and PdAu samples exhibit only one band in this region. crystallite size of Pd colloids formed in the PS-b-P4VP mi-
The latter should indicate linear CO molecules adsorbed on celles, estimated from WAXS, was 1.8 0.2 nm. This value
the Pd surface (32–34). The band at 2058 cm 1 (Fig. 1c) can was in agreement with TEM data (14). However, along with
be ascribed to the terminal CO adsorption on Pt (36), while this kind of particles, the overall structure contained a large
a band at 1994 cm 1 is due to linear CO adsorption on Zn. number of isotropic scatterers, which should be small amor-
The lack of the second terminal band for CO adsorption for phous clusters. A TEM micrograph of PdPt colloids is pre-
PdAu (at 2112 cm 1 (37)) shows that there are no Au atoms sented in Fig. 2. One can see that the micelle cores contain
on the particle surface. Then, we can surmise that, for PdAu small nanoparticles, whose mean diameter does not exceed
bimetallics, the core–shell structure does form, while PdPt 2 nm.
and PdZn colloids form cluster-in-cluster structures. This
Typical /2 diffractograms of PdAu (a) and PdPt(b)
demonstrates the difference in the formation of bimetal- bimetallic colloids stabilized in the PS-b-P4VP micelles
lic particles in block copolymer micelle cores, which are (recorded after subtraction of the background) are pre-
closer to a solid state, and in poly(N-vinyl-2-pyrrolidone) sented in Fig. 3. One can see very broadened Bragg

306
BRONSTEIN ET AL.
FIG. 2. TEM image of PdPt colloids formed in the PS-b-P4VP micelles.
reflections at positions typical for Pd crystallites. The aver-
age size of the crystallites derived from Bragg peaks does
not exceed 1.5–2 nm for both samples. No Bragg peaks re-
lated to Au, Pt, or PdO were detected. Similar results were
obtained for PdZn colloids. Thus, the presence or absence
of bridging CO groups in the FTIR spectra of bimetallic
particles formed in the PS-b-P4VP micelles cannot indicate
the difference in particle size or particle crystallinity. We
assign the difference in the CO adsorption to different sur-
face properties of bimetallic colloids: for PdAu colloids only
one kind of active center exists (only linear adsorption),
while for Pd, PdPt, and PdZn colloids, at least two types
of active center are located on the particle surfaces. Prob-
ably this difference is caused by the different (core–shell)
structures of PdAu colloids.
To probe the influence of the modifying metals (Au, Pt,
and Zn) on the Pd particle surface, XPS was employed. The
XPS data are presented in Figs. 4 and 5 and Table 3. One can
see that monometallic Pd colloids formed in the PS-b-P4VP
FIG. 3. WAXS diffractograms of the PdAu (a) and PdPt (b) samples.

CATALYTIC PROPERTIES OF BIMETALLIC COLLOIDS
307
FIG. 4. The MgK induced X-ray photoelectron spectra of monometallic Pd (run 1) and bimetallic PdZn (run 2), PdAu (run 3), and PdPt (run 4)
colloids formed in the PS-b-P4VP micelles.
micelles core have a binding energy of 335.1 eV, which is ation fits the acceptor ability of the metal: in the PdAu sam-
characteristic of zerovalent Pd, so the experimental con- ples the Pd(II) species strongly prevail, while in the PdPt
ditions during superhydride reduction do not allow even species (Pt is a weaker acceptor compared to Au), the frac-
surface oxidation of nanoparticles (38). On the other hand, tion of Pd(0) is rather noticeable. On the other hand, the
XPS spectra of bimetallic colloids show two components in presence of a high fraction of the Pd(II) species in the PdZn
the Pd3d5/2 region. One might speculate that the presence colloids (Fig. 4) does not fit the suggested idea of easier oxi-
of a second metal having electron acceptor properties (Pt dation in the presence of an acceptor because Zn is a donor
and Au) might provide easier oxidation of the particle sur- toward Pd. To clarify the phenomenon, a series of XPS spec-
face, even in conditions (see Methods) where reduction is tra were recorded immediately after preparation and after
carried out in an inert atmosphere. Moreover, this consider- 12 h of standing in the camera of the spectrometer. Figure 5
shows the spectra recorded for each bimetallic system. As
can be seen, the spectrum of PdAu bimetallics remains un-
TABLE 3
changed (Fig. 5a), for PdPt, the fraction of Pd(0) notice-
ably increases (Fig. 5b), and for PdZn, nearly all Pd(II) is
converted to Pd(0) (Fig. 5c). Obviously, the changes that
occurred in the camera of the spectrometer (in a vacuum of
10 9 Torr) should not be discussed in terms of “oxidation–
reduction.” The X-ray diffraction data for monometallic Pd
nanoparticles (Fig. 3) characteristic of Pd(0) show that high
binding energy values can be attributed to charging of the
nanoparticles in a nonconductive polymer environment,
which shows an oxidation state higher than the one present
in a real sample (38). After treatment of PdAu colloids with
H2 at 90 C for 1 h (typical pretreatment of the catalyst be-
fore DHL hydrogenation), the X-ray diffractogram does
not change, while in the XPS spectrum of this sample the
Pd(0) species prevail. This can be attributed to a change
in the nanoparticle charging after hydrogen pretreatment:
when the environment changes (hydrogen is adsorbed onto
Binding Energy Values in the Pd, Au, Pt, and Zn Compounds
Compound
E Pd 3d5/2
E Au 4f7/2
E Zn 2p3/2
E Pt 4f7/2
Pd0
335.7
339
335.1
—
—
—
338.7
—
—
335.5
—
—
—
—
—
83.8
84.5
87.9
84.5
—
—
—
—
—
—
—
—
—
—
—
—
1021.6
1022.5
1022.1
—
—
—
—
—
—
—
—
—
—
Pd(CH3COO)2
PS-b-P4VP/Pd
Au0
AuCl
KAuCl4
PS-b-P4VP/PdAu
Zn0
Zn(CH3COO)2
PS-b-P4VP/PdZn
Pt0
K[Pt(C2H4)Cl3]
K2PtCl6
PS-b-P4VP/PdPt
—
71.3
73.6
75.6
73.4
—
—
—
—
338.2
—
—

308
BRONSTEIN ET AL.
2. Study of Catalytic Properties of Mono- and Bimetallic
Colloidal Micellar Catalysts
2.1. Dependence of selectivity of catalysts on reaction con-
ditions. To find the conditions providing the highest selec-
tivity of colloidal catalysts in DHL hydrogenation, several
parameters were varied: temperature, stirring rate, ratio of
substrate, and catalyst amounts. These parameters influ-
ence both the selectivity and activity of catalysts. Table 4
shows the variation of reaction conditions and selectivity
value for the PdAu catalyst as an example. For all the cata-
lysts studied the highest selectivity (99.8% at 100% conver-
sion) was obtained at 90 C, at 480 shakings per minute (with
diffusion limitations (19)), at an initial DHL concentration
Co of 0.44 mol/L, and with a catalyst amount Cc of 2.3
5
10 mol of Pd/L). Toluene was chosen as the reaction
medium to provide the same environment of colloids in the
micelle cores, which existed during synthesis of metal col-
loids (13, 14) and the most favorable conditions for micelle
core availability for the substrate.
2.2. Kinetics of DHL hydrogenation. To find regular-
ities of selective DHL hydrogenation with mono- and
bimetallic colloids formed in block copolymer micelles,
we studied the kinetics of the reaction. All experiments
were carried out in toluene with a stirring rate of 960
shakings/min (without diffusion limitations) at 90 C and
varyingconcentrationsofsubstrate and catalyst in the range
5
of 0.22–0.88 mol/L (Co) and (1.01 10 5)–(4.08 10
mol of Pd/L (Cc), respectively.
)
As a criterion allowing evaluation of the dependence of
DHL conversion on time at various Co and Cc values, the
ratio q = Co/Cc (mol/mol) was used. The analysis of the
primary experimental data, presented in Fig. 6, shows that
the higher the Co/Cc ratio, the higher the hydrogenation
time, which is typical for most catalytic reactions. As a result
of experimental data processing, the 50% DHL conversion
time ( 0.5) has been determined. The value of 0.5 was found
to depend on the ratio (Co/Cc)n,
(Co/Cc)n,
[1]
0.5
FIG. 5. The MgK induced X-ray photoelectron spectra of bimetallic
where n is the formal parameter characterizing the slope of
the ln 0.5 dependence on ln Co/Cc. In the case of the PdAu
catalyst, n = 2, and for other catalysts, n = 1. These differ-
ent values will be interpreted in terms of different mecha-
nisms (see below).
PdAu (a), PdPt (b), and PdZn (c) colloids recorded immediately after
sample preparation (run 1) and after 12 h standing in the vacuum camera
of spectrometer (run 2).
GC analysis of all the samples obtained after DHL hy-
drogenation with mono- and bimetallic micellar catalysts
showed mainly LN formation, though further saturation of
the double bond to the single one yielding DiHL is also pos-
sible (Scheme 1). Other products have not been detected
in the reaction mixture. Thus, the following reaction route
can be suggested,
the Pd nanoparticle surface), the charge on the nanoparti-
cles can change. Thus, for this particular polymeric envi-
ronment, XPS data on bimetallic nanoparticles could not
be considered as a reliable source of structural information
because of artifact charging. On the other hand, because
this charging does not occur for Pd nanoparticles, but is
rather pronounced for bimetallic ones, this confirms the in-
fluence of a modifying metal on the electronic structure of
bimetallic colloids.
k1
k2
DHL → LN → DiHL,
+ H 2
+ H 2

CATALYTIC PROPERTIES OF BIMETALLIC COLLOIDS
309
TABLE 4
Selectivity of DHL Hydrogenation with PS-b-P4VP/PdAu Depending on Reaction Conditionsa
Catalyst amount
Cc 105 (mol of Pd/L)
DHL concentration
Co (mol/L)
Temperature
Shaking per
minute
Hydrogenation
time (s)
Selectivity
(% )
Run
th ( C)
1
2
3
4
5
6
7
8
9
2.30
1.15
4.45
2.30
2.30
2.30
2.30
2.30
2.30
2.30
0.44
0.44
0.44
0.66
0.33
0.44
0.44
0.44
0.44
0.44
90
90
90
90
90
70
80
95
90
90
480
480
480
480
480
480
480
480
360
600
1140
1320
780
1260
840
1680
1500
900
1380
840
99.8
98.7
98.5
98.4
98.6
98.2
99.0
98.7
99.0
97.6
10
a In toluene.
where k1 and k2 are the hydrogenation rate constants. Be-
To discuss the experimental data obtained at various val-
cause DHL hydrogenation here is characterized by high ues of the substrate/catalyst ratio Co/Cc, we used relative
selectivity and the DiHL amount in the reaction mixture is concentrations of substrate (DHL) and product (LN),
negligible, k1 k2, the second stage can be neglected in the
computation.
The linear dependence of ln
xi = Ci /Co,
[3]
on ln(Co/Cc) allows us
0.5
to introduce an independent variable, relative time 2,
where i = 1, 2 for DHL and LN, respectively, and Ci is
the current substrate or product concentration. In so do-
ing, the experimental data, presented in Fig. 6, have been
brought together into a family of curves (Fig. 7) in the
x2–2 coordinates. Thus, a mathematical description of the
2 = /(Co/Cc)n,
[2]
where is the current reaction time.
FIG. 6. Dependence of DHL conversion on reaction time with Pd (a) and bimetallic PdAu (b), PdZn (c), and PdPt (d) catalysts for different
substrate/catalysts ratios (q = Co/Cc mol of DHL/mol of Pd) at th 90 and 960 shakings/min.

310
BRONSTEIN ET AL.
experimental data can be represented as a system of differ- form assignment, the linear model has been assumed,
ential equations,
W1 = kx1,
[5]
dx1
=
W1
d2
dx2
d2
where k is the kinetic parameter.
[4]
In this model the absence of adsorption (or coordination)
interactions in the catalytic system has been presumed. To
check this hypothesis as well as other models of DHL hy-
= W2,
where Wi is the reaction rate at initial concentration of the drogenation, the inverse problem has been solved using an
substrate Co = 1 mol/L and concentration of the catalyst explicit integral method (45).
Cc = 1 mol/L.
According to the results of the inverse problem solv-
Different mathematical methods are known (39, 40) for ing, for all the catalysts, the kinetic models have been cho-
the description of the reaction kinetics. In the present sen (Table 5) that describe well the hydrogenation kinetics
paper the integral method (41) has been used to esti- (Fig. 7).
mate kinetic model parameters. The cubic spline serves
The models obtained are a formal description of the
as interpolating function in this method. The integral effi- DHL hydrogenation kinetics with mono- and bimetallic
ciency function formulated has been minimized by the com- colloidal catalysts. One can see that for the Pd catalyst the
bined gradient method of Levenberg–Marquardt (42, 43). model equation is the same kind as that for PdPt and PdZn
The kinetic equations obtained describing the kinetics of colloidal catalysts, but the coordination (adsorption) pa-
DHL hydrogenation correspond to the form of Langmuir– rameter Q is very small, so the second component in the
Hinshelwood equations (44).
denominator can be omitted. Thus, for the Pd catalyst,
For the computation of the kinetic model parameters, the rate of hydrogenation does not depend on the current
the following initial conditions have been defined: 2 = 0, substrate concentration, while for bimetallic catalysts com-
x1 = 1, and x2 = 0. In a first approximation, for W1 explicit plex dependence of the hydrogenation rate on substrate
FIG. 7. Dependence of DHL conversion on relative time with Pd (a) and bimetallic PdAu (b), PdZn (c), and PdPt (d) catalysts. For conditions see
Fig. 6.

CATALYTIC PROPERTIES OF BIMETALLIC COLLOIDS
311
TABLE 5
Kinetics Models for DHL Hydrogenation with Mono- and Bimetallic Colloids Formed in Block Copolymer Micelles
1
Catalytic system
Model
W = k
k (mol/mol)n s
Qa
102 b
1.65
2
2
PS-b-P4VP/Pd
(2.08 0.02) 10
0.14
—
kx
1
PS-b-P4VP/PdAu
PS-b-P4VP/PdZn
PS-b-P4VP/PdPt
W =
W =
W =
0.140 0.009
2.07
2.72
2.57
2
)
(x + Qx
1
2
kx
2
2
1
(3.75 0.08) 10
(5.8 0.1) 10
(1.3 0.1) 10
(5.2 0.2) 10
x
+ Qx
1
1
2
kx
2
1
x
+ Qx
2
a Q is the coordination parameter, taking into account the coordination of the product and the substrate with catalyst.
b
is the mean square deviation of computation curves from the experimental data.
concentration is demonstrated. Moreover, for bimetallic ders of magnitude. At the same time, the activation energy
catalysts, in the equation of the model the denominator- E increases approximately by a factor of 2 compared to that
containing coordination (adsorption) parameter is intro- of other catalysts. It can be speculated that this difference
duced, which shows a more pronounced impact of the co- is determined by the peculiar structure of PdAu colloids.
ordination (adsorption) processes on DHL hydrogenation
An increase of the activation energy for PdAu is in agree-
compared to the monometallic one. Among all the cata- ment with two facts: Au being a strong acceptor toward Pd
lysts, PdAu appears different: its model equations contain and being located in the core of the particle (which is con-
a squared term in the denominator.
firmed by CO adsorption) pulls electron density from Pd.
To calculate the parameters in the Arrhenius equation, The low electron density of these bimetallic colloids pro-
the temperature of DHL hydrogenation was varied from vides the favorable conditions for hydrogenation of dou-
50 to 95 C. Statistical analysis performed by the method ble bonds as described in (23, 24), but at the same time for
described elsewhere (46) showed that within experimen- triple-bond hydrogenation, the high electron density on the
tal error the coordination parameter Q can be considered catalyst surface is preferable (48). Then, low electron den-
constant and not dependent on the temperature. From the sity increases the energy barrier of reaction. On the other
experimental results the Arrhenius dependence was drawn hand, at the same nanoparticle size for bimetallic core–shell
in the coordinates ln k 1/T and frequency factor k0 of the particles, nearly all the Pd atoms can be located on the sur-
Arrhenius equation (which characterizes the amount of ef- face of bimetallic particles (23, 24), while for Pd colloids, a
fective collisions or the amount of active centers (47)) and fraction of Pd atoms is in the center of the particle. Thus, an
apparent activation energy E were calculated:
increase of the number of active centers for PdAu colloids
(compared to Pd nanoparticles) can be explained both by
defects on the particle surface generated by the formation
of core–shell structures and by the fact that all Pd atoms
are on the surface and available for the substrate.
E/RT
k = k0e
.
[6]
The data obtained are presented in Table 6. For all the
bimetallic catalysts one can see the significant increase of
active center amount. For PdPt and PdZn colloidal catalysts
this can be explained by an increase of defects in bimetal-
lic particles (24). However, particular attention should be
drawn to the PdAu catalyst, which k0 increasesbyseveralor-
As discussed above, for the PdPt catalyst, CO adsorption
was observed both for Pd and Pt species, confirming the
cluster-in-cluster structure. The activation energy for this
catalyst is similar to the Pd one, so an increase of cata-
lytic activity can be ascribed to an increase of defects in
bimetallic colloids (23–25). For the PdZn system, according
to CO adsorption, we surmise the presence on the surface of
both Pd atoms and Zn ions; the latter should not be reduced
in the reaction conditions, which is confirmed by XPS data.
The turnover frequency (TOF) values (Table 6) for
mono- and bimetallic colloidal catalysts show that, for
bimetallics, the catalytic activity is higher than that for the
Pd ones, which can be explained by modifying the influence
of gold, platinum, and zinc. It is known that the addition of
a second metal can change the electronic properties of cata-
lyst (ligand effect) (49–53). As discussed in (54), this change
can influence the energy of metal–hydrogen and metal–
substrate bonds and the amount of hydrogen adsorbed. The
addition of a modifying metal to a basic component can
TABLE 6
Parameters of Arrhenius Equation and TOF for Mono-
and Bimetallic Colloidal Catalysts
Activation energy Frequency Turnover frequencya
1
Catalyst
PS-b-P4VP/Pd
PS-b-P4VP/PdAu
PS-b-P4VP/PdZn
PS-b-P4VP/PdPt
(kJ/mol)
factor
(s
)
23
55
26
26
66
107
356
659
18.5
36.9
34.4
49.2
a
Experimental conditions: th 90 C, toluene, volume of reaction mix-
3
5
ture V = 30 10 L, C0 = 0.44 mol/L, Cc = 2.3 10 mol of Pd/L, 960
shakings/min.

312
BRONSTEIN ET AL.
also change the surface geometry (ensemble effect) (51,
Because DHL hydrogenation is practically irreversible
55–60). Thus, a combination of all these factors determines in the conditions used, the surface reaction between the
the change of catalytic activity of bimetallic catalysts, which coordinated substrate and hydrogen with the formation of
is reflected in TOF values. Moreover, according to the ki- LN (P) should be irreversible as well,
netic results, DHL hydrogenation with mono- and bimetal-
kS
SZ0 + 2H Z → P Z0 + 2Z,
[10]
lic catalysts occurs by different mechanisms. This should be
caused by the modifying influence of the second metal as
well.
where P is the product (LN) and ks is the rate constant of
the surface reaction.
Let us assume that stages of adsorption and desorption
of the product are fast as well and in dynamic equilibrium,
3. Hypothesis on Mechanism of the Selective
DHL Hydrogenation
From the above results, we can propose a hypothesis
on the mechanism for DHL hydrogenation with Pd and
bimetallic catalysts and compare it with the computed ki-
netic models presented in Table 5.
K2
P Z0 ↔ P + Z0,
[11]
where K2 is the equilibrium constant.
Then, the equation of the chemical reaction rate tak-
ing into account sorption–desorption of DHL and LN on
organometallic centers and of hydrogen on metallic centers
is given by
3.1. Catalysts PS-b-P4VP/Pd, PS-b-P4VP/PdZn, and PS-
b-P4VP/PdPt. Presumably, hydrogen is activated on one
kind of reactive center, which we call “metallic” (Z), and
then diffuses along the surface to substrate molecules (S)
adsorbed on other centers called “organometallic” (Z0).
Organometallic centers might be formed due to coordina-
tion of metal with polymer ligands (in the P4VP micelle
core). For traditional heterogeneous catalysts, organo-
metallic centers are formed due to coordination of metal
with a substrate, solvent, or support or can be synthesized
purposely (61). These two kinds of active centers should
exist on each single colloidal particle. Their presence is
confirmed by FTIR data on CO adsorption: existence of
terminal and bridging CO.
W = ks[SZ0][H Z]2.
[12]
According to the assumption of fast equilibrium for
stages [7]–[9], we get
[SZ0] = K1C1 Z0 C1 = [S],
[H Z]2 = Kn[H2 Z2],
[13]
[14]
[15]
[16]
[17]
[H2 Z2] = Km[H2]Z2,
[H Z]2 = Kn Km[H2]Z2,
[P Z0] = K2C2 Z0 C2[P].
Let us assume that in the reaction system there are in-
termediates SZ0 formed due to associative adsorption of
DHL on organometallic active centers. The coordination
of a substrate can be envisioned as
Then, we use a mass balance equation for surface com-
pounds,
Z0 + [SZ0] + [P Z0] = Z00 ,
Z2 + [H2 Z2] = Z0,
[18]
where Z0 and Z are the occupied organometallic and metal-
lic active centers and Z00 and Z0 are the total numbers of ac-
tive centers of both types; the latter values are proportional
to the catalyst amount.
As this takes place, we propose that adsorption and des-
orption are fast and are in equilibrium,
K1
S + Z0 ↔ SZ0,
[7]
Assuming that 1 (K1C1 + K2C2), then
where K1 is the equilibrium constant.
Z00 = Z0 + K1C1 Z0 + K2C2 Z0 = Z0(1 + K1C1 + K2C2)
Hydrogen is dissociatively adsorbed and activated with
formation of atomic hydrogen on metallic active centers
Z, which is in agreement with published data (40). Again,
we assume that sorption–desorption occurs fast and these
stages are in equilibrium,
= Z0(K1C1 + K2C2).
[19]
Asfollowsfrom above, similarlyto the metalsurface (40),
Z00
Z0 =
,
[20]
Kn
(K1C1 + K2C2)
H2 + 2Z ↔ H2 Z2,
[8]
[9]
Z0 = Z2 + [H2 Z2] = Z2(1 + Kn Km[H2]),
[21]
[22]
Km
H2 Z2 ↔ 2H Z,
Z0
Z2 =
.
(1 + Kn Km[H2])
where Kn and Km are the equilibrium constants.

CATALYTIC PROPERTIES OF BIMETALLIC COLLOIDS
313
Taking into consideration the equations [13]–[19] and
[20]–[22], the rate of chemical reaction can be figured as
Using all approaches discussed above, we obtain the
following equation for the rate of the chemical reaction,
W = ks[SZ][H2 Z] = ks K1[S]Z2 Kn[H2],
Z0 = Z + K1C1 Z + K2C2 Z + Z Kn[H2]
W = ks K1C1 Z0 Kn Km[H2]Z2,
[23]
ks K1 Kn KmC1 Z00 [H2]Z0
W =
[32]
= Z(1 + K1C1 + K2C2 + Kn[H2])
= Z(K1C1 + K2C2),
(K1C1 + K2C2)(1 + Km Kn[H2])
kC1Ck
=
.
[24]
(K1C1 + K2C2)
because K1C1 and K2C2
Kn[H2] and 1,
One can see that the equation obtained is a Langmuir–
Hinshelwood equation (44) and it coincides with the equa-
tion for the model (Table 5) when x1 = C1 and x2 = C2:
Z0
Z =
,
[33]
[34]
(K1C1 + K2C2)
ks K1C1 Kn[H2]Z02
(K1C1 + K2C2)2
W =
,
kS K1 Kn Km[H2]Z0
k =
,
[25]
[26]
ks K1 KnC1(Z0)2[H2]
(K1C1 + K2C2)2
(1 + Kn Km[H2])
K2
W = ks[SZ0][H2 Z0] =
Q =
.
kC1Ck2
(K1C1 + K2C2)2
K1
=
.
[35]
The coincidence of the kinetic equation obtained with the
equation of the model allows us to assume the correctness
of our hypothesis and shows the physical meaning of the
parameters of the model, where k is the kinetic parameter
takinginto account hydrogen adsorption on the catalyst and
Q is the adsorption parameter equal to the ratio of adsorp-
tion constants of the product and substrate, respectively.
The equation above fits the equation of the model
(Table 5) at
x1 = C1; x2 = C2,
k = ks K1 Kn[H2],
[36]
[37]
K2
Q =
.
[38]
K1
3.2. Catalyst PS-b-P4VP/PdAu. As discussed above,
the PdAu catalyst differs from others and has presumably a
core–shell structure. Similar to Pd colloids, PdAu bimetallic
nanoparticles have only Pd atoms on the surface; however,
judging by FTIR adsorption, on CO, the surface structure
for a bimetallic catalyst is quite different. Here, we can as-
sume the existence of active centers of only one type and
probably competitive coordination of the substrate and hy-
drogen on these centers. The surface reactions of sorption–
desorption for the substrate and hydrogen can be figured as
In the kinetic equation [35], similarly to the model equation
presented in Table 5, there is an inverse square dependence
of the reaction rate on the concentration of adsorbed sub-
strate and product. This coincidence confirms the correct-
ness of our hypothesis. We surmise that the surface of the
PdAu colloid is poor in electron density due to the Au core
influence and that it attracts pyridine units more strongly
from the P4VP micelle core and thus provides a denser
packing of the modifying ligands. This packing yields a more
uniform particle surface with only one kind of active center.
K1
S + Z ↔ SZ,
[27]
[28]
[29]
Kn
SUMMARY
H2 + 2Z ↔ H2 Z2,
Km
H2 Z2 ↔ 2H Z,
Monometallic (Pd) and bimetallic (PdAu, PdPt, and
PdZn) colloids formed in the PS-b-P4VP block copolymer
micelles were studied in DHL hydrogenation. FTIR spectra
of CO adsorbed onto PdPt and PdZn nanoparticles demon-
strate two bandsfor CO terminaladsorption and also a band
for the bridging one, while the spectrum of the PdAu sample
shows only one band for terminal adsorption. This testifies
to the presence of only Pd atoms on the PdAu nanopar-
ticle surface and the existence of only one type of active
center. The PdPt and PdZn bimetallics display both Pd and
Pt(or Zn) atoms on the particle surface and the presence
of two types of active centers on Pd. Due to the influence
of the modifying metal, the catalytic activities of bimetal-
where K1, Kn, and Km are equilibrium constants.
For irreversible DHL hydrogenation, the surface reac-
tion between the coordinated substrate and hydrogen with
a formation of LN (P) is irreversible as well,
kS
SZ + 2H Z → P Z + Z,
[30]
where ks is the rate constant of the surface reaction, fol-
lowed by the desorption of the product:
K2
P Z ↔ P + Z0.
[31] lic catalysts are higher than that for the Pd one. For all

314
BRONSTEIN ET AL.
the catalysts, the optimal conditions were found, providing 23. Toshima, N., Harada, M., Yamazaki, Y., and Asakura, K., J. Phys.
Chem. 92, 9927 (1992).
24. Harada, M., Asakura, K., Ueki, Y., and Toshima, N., J. Phys. Chem.
high selectivity of hydrogenation (up to 99.8% ), which is
mainly determined by the modifying influence of pyridine
groups in the P4VP cores. The study of the kinetics of DHL
hydrogenation with these new catalysts allowed us to sug-
96, 9730 (1992).
25. Harada, M., Asakura, K., and Toshima, N., J. Phys. Chem. 98, 2653
(1994).
gest a hypothesis for a hydrogenation mechanism, which 26. Antonietti, M., Heinz, S., Rosenauer, C., and Schmidt, M., Macro-
molecules 27, 3276 (1994).
27. Antonietti, M., Fo¨rster, S., Oestreich, S., Hartmann, J., and Wenz, E.,
fits computational kinetic models well.
Nachr. Chem. Lab. Tech. 44, 579 (1996).
28. Bronstein, L. M., Mirzoeva, E. Sh., Valetsky, P. M., Solodovnikov,
S. A., and Register, R. A., J. Mater. Chem. 5, 1197 (1995).
29. Bowser, J. R., “Inorganic Chemistry.” Brooks Cole, Pacific Grove, CA,
1997.
30. Kleinberg, J., Argersinger, W. J., Griswold, E., Jr., “Inroganic Chem-
istry.” Heath, Boston, 1960.
ACKNOWLEDGMENTS
The authors acknowledge the financial support provided by the Rus-
sian Foundation for Basic Research (Grants 98-03-33372) and NATO SfP-
974173 Colloidal Catalysts. We also thank Prof. Dr. M. Antonietti for the
PS-b-P4VP sample and Dr. S. N. Polyakov for help in the WAXS exami-
nations.
31. “Encyclopedia of Electrochemistry of Elements” (A. J. Bard, Ed.).
Vol. 4–6. Dekker, New York /Basel. 1976.
32. Shieu, L. L., Karpinski, Z., and Sachtler, W. M. H., J. Phys. Chem. 93,
4890 (1989).
REFERENCES
33. Bradley, J. S., Millar, J. M., Hill, E. W., Behal, S., Chaudret, B., and
Duteil, A., Faraday Disc. 92, 255 (1991).
34. Bradley, J. S., Hill, E. W., Behal, S., Klein, C., Chaudret, B., and
Duteil, A., Chem. Mater. 4, 1234 (1992).
1. Ciardelli, F., Tsuchida, E., and Wo¨hrle, D., “Macromolecule–Metal
Complexes.” Springer-Verlag, Berlin, 1996.
2. Schmid, G., “Clusters and Colloids.” VCH, Weinheim, 1994.
3. Jacobs, P. W., Wind, S. J., Ribeiro, F. H., and Somorjai, G. A., Surf. Sci. 35. Bradley, J. S., Hill, S., Klein, C., Chaudret, B., and Duteil, A., Chem.
372, 249 (1997). Mater. 5, 254 (1993).
4. Michalska, Z. M., Ostaszewski, B., Zientarska, J., and Sobczak, J. W., 36. Duteil, A., Que´au, R., Chaudret, B., Mazel, R., Roucau, C., and Bradly,
J. Mol. Catal. A 129, 207 (1998). J. S., Chem. Mater. 5, 345 (1993).
5. Sulman, E., Matveeva, V., Usanov, A., Kosivtsov, Yu., Demidenko, G., 37. Bollinger, M., and Vannice, M. A., Appl. Catal. B 8, 417 (1996).
Bronstein, L., Chernyshov, D., and Valetsky, P., J. Mol. Catal. A 146, 38. Wagner, C., Riggs, W., Davis, L., and Mullenberg, G., “Handbook
265 (1999).
of X-Ray Photoelectron Spectroscopy.” Perkin–Elmer, Minneapolis,
MN 1978.
6. Bo¨nnemann, H., Brijoux, W., Brinkmann, R., Fretzen, R., Joussen, T.,
Ko¨ppler, R., Korall, B., Neiteler, P., and Richter, B., J. Mol. Catal. 86, 39. Schbib, N. S., Garcia, M. A., Gigola, C. E., and Errazu, A. F., Ind. Eng.
129 (1994).
Chem. Res. 35, 1496 (1996).
7. Bo¨nnemann, H., Brijoux, W., Brinkmann, R., Schulze, A., Tilling., and
Schilling, T., et al., Inorg. Chim. Acta 270, 95 (1998).
8. Pollmann, J., Franke, R., Hormes, J., Bo¨nnemann, H., Brijoux, W.,
Schulze, A., and Tilling, A., J. Electron Spectrosc. Related Phenom. 94,
219 (1998).
9. Sulman, E. M., Russ. Chem. Rev. 63, 936 (1994).
10. Lindlar, H., Helv. Chim. Acta 35, 44 (1952).
11. Sulman, E. M., Matveeva, V. G., Bronstein, L. M., Ankudinova, T. V.,
Mirzoeva, E. Sh., Sidorov, A. I., and Valetsky, P. M., Russ. Chem. Farm.
J. 5, 149 (1995).
40. Rajashekharam, M. N., Nikalje, D. D., Jaganathan, R., and Chaudhari,
R. V., Ind. Eng. Chem. Res. 36, 592 (1997).
41. Himmelblau, D. M., Jones, C. R., and Bischoff, K. B., Ind. Eng. Chem.
Fundam. 6, 539 (1967).
42. Bard, Y., SIAM J. Numer. Anal. 7, 157 (1970).
43. Marquardt, D. W., J. Soc. Ind. Appl. Math. 11, 431 (1963).
44. Brundage, M. A., Balakos, M. W., and Chuang, S. S. C., J. Catal. 173,
122 (1998).
45. Yermakova, A., Bibin, V. S., and Brin, E. F., Kinet. Catal. (Russ.) 26,
711 (1985).
12. Mirzoeva, E. Sh., Bronstein, L. M., Valetsky, P. M., and Sulman, E. M.,
React. Polym. 24, 243 (1995).
13. Antonietti, M., Wenz, E., Bronstein, L. M., and Seregina, M. S., Adv.
Mater. 7, 1000 (1995).
46. Vajda, S., Valko, P., and Turani, T., Int. J. Chem. Kinet. 17, 55 (1985).
47. Sakakini, B. N., and Verbrugge, A. S., J. Chem. Soc. Faraday Trans. 93,
1637 (1997).
48. Rajaram, J., Narula, A. P. S., Chawla, H. P. S., and Sukh, Dev., Tetra-
hedron 39, 2315 (1983).
14. Seregina, M. S., Bronstein, L. M., Platonova, O. A., Chernyshov, D. M.,
Valetsky, P. M., Hartmann, J., Wenz, E., and Antonietti, M., Chem. 49. Recchia, S., Dossi, C., Poli, N., Fusi, A., Sordelli, L., and Rzaro, R.,
Mater. 9, 923 (1997).
J. Catal. 11 (1999).
15. Fendler, J. H., “Nanoparticles and Nanostructured Films.” Wiley–
VCH, Weinheim /New York, 1998.
16. Fo¨rster, S., and Antonietti, M., Adv. Mater. 10, 195 (1998).
50. Adams, R. D., and Barnard, T. S., Organometallics 17, 2885 (1998).
51. Candy, J. P., Santini, C. C., and Basset, J. M., J. Top. Catal. 4, 211 (1997).
52. Bertolini, J. C., J. Surf. Rev. Lett. 3, 1857 (1996).
17. Platonova, O. A., Bronstein, L. M., Solodovnikov, S. P., Yanovskaya, 53. Lopezgaona, A., Martin, N., and Viniegra, M., J. Mol. Catal. A 96, 155
I. M., Obolonkova, E. S., Valetsky, P. M., Wenz, E., and Antonietti, M.,
Colloid Polym. Sci. 275, 426 (1997).
18. Bronstein, L., Chernyshov, D., Valetsky, P., Tkachenko, N.,
Lemmetyinen, H., Hartmann, J., and Fo¨rster, S., Langmuir 83, 15
(1999).
19. Sulman, E., Bodrova, Yu., Matveeva, V., Semagina, N., Cerveny, L.,
Kurtc, L., Bronstein, L., Platonova, O., and Valetsky, P., Appl. Catal.
A 176, 75 (1999).
20. Farris, T. M., and Armor, J. N., Appl. Catal. A 96, 25 (1993).
21. Schmidt, T. G., Noeske, M., Gasteiger, H. A., Behm, R. J., Britz, P.,
Brijoux, W., and Bo¨nnemann, H., Langmuir 13, 2591 (1997).
22. Hable, C. T., and Wrighton, M. S., Langmuir 9, 3284 (1993).
(1995).
54. Krishnamurthy, K. R., Stud. Surf. Sci. Catal. 113, 139 (1998).
55. Aranda, D. A. G., Noronha, F. B., Ordine, A. P., and Schmal, M., Phys.
Status Solid. A Appl. Res. 173, 109 (1999).
56. Naito, S., Hasebe, T., and Miyao, T., J. Chem. Lett. 11, 1119 (1998).
57. Vieira, A., Tovar, M. A., Pfaff, C., Mendez, B., Lopez, C. M., Machado,
F. J., Goldwasser, J., and de Agudelo, M. M. R., J. Catal. 177, 60 (1998).
58. Brandford, M. C. J., and Vannice, M. A., J. Catal. Lett. 48, 31 (1997).
59. Mendes, F. M. T., and Schmal, M., J. Appl. Catal. A 163, 153 (1997).
60. Aranda, D. A. G., and Schmal, M., J. Catal. 171, 398 (1997).
61. Agnelli, M., Swaan, H. M., Marquezalvarez, C., Martin, G. A., and
Mirodatos, C., J. Catal. 175, 117 (1998).