Adsorption-Induced Segregation as a Method for the Target-Oriented Modification of the Surface of a Bimetallic Pd–Ag Catalyst

Article abstract of DOI:10.1134/S0023158418050154

The effect of palladium segregation was studied which resulted from the effect of CO and O₂ on the surface structure and catalytic characteristics of the Pd–Ag₂/Al₂O₃ catalyst. The IR-spectroscopic study of adsorbed CO showed that Pd₁ centers isolated from each other by silver atoms predominated on the surface of reduced Pd–Ag₂/Al₂O₃, as evidenced by the almost complete absence of absorption bands typical for the multicentred CO adsorption. In the course of catalyst treatment with CO and O₂, the intensity of absorption bands characteristic of the multicenter CO adsorption considerably increased due to the transformation of a portion of monatomic Pd₁ centers into multiatomic Pd_n ones as a result of the surface segregation of Pd. In this case, a substantial increase in the catalyst activity in the liquid-phase hydrogenation of diphenylacetylene was observed. It was established that, after treatment with CO, the catalyst selectivity for the formation of a target olefin (stilbene) remained almost constant, whereas the treatment with O₂ led to a decrease in the selectivity because of more considerable surface modification.

Full text of DOI:10.1134/S0023158418050154

ISSN 0023-1584, Kinetics and Catalysis, 2018, Vol. 59, No. 5, pp. 610–617. © Pleiades Publishing, Ltd., 2018.
Original Russian Text © A.Yu. Stakheev, N.S. Smirnova, P.V. Markov, G.N. Baeva, G.O. Bragina, A.V. Rassolov, I.S. Mashkovsky, 2018, published in Kinetika i Kataliz, 2018,
Vol. 59, No. 5, pp. 601–609.
Adsorption-Induced Segregation as a Method for the Target-
Oriented Modification of the Surface of a Bimetallic Pd–Ag Catalyst
A. Yu. Stakheeva, *, N. S. Smirnovaa, P. V. Markova, G. N. Baevaa, G. O. Braginaa,
A. V. Rassolova, and I. S. Mashkovskya
aZelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Moscow, 119991 Russia
*e-mail: st@ioc.ac.ru
Received March 15, 2018
Abstract—The effect of palladium segregation was studied which resulted from the effect of CO and O2 on the
surface structure and catalytic characteristics of the Pd–Ag2/Al2O3 catalyst. The IR-spectroscopic study of
adsorbed CO showed that Pd1 centers isolated from each other by silver atoms predominated on the surface
of reduced Pd–Ag2/Al2O3, as evidenced by the almost complete absence of absorption bands typical for the
multicentred CO adsorption. In the course of catalyst treatment with CO and O2, the intensity of absorption
bands characteristic of the multicenter CO adsorption considerably increased due to the transformation of a
portion of monatomic Pd1 centers into multiatomic Pdn ones as a result of the surface segregation of Pd. In
this case, a substantial increase in the catalyst activity in the liquid-phase hydrogenation of diphenylacetylene
was observed. It was established that, after treatment with CO, the catalyst selectivity for the formation of a
target olefin (stilbene) remained almost constant, whereas the treatment with O2 led to a decrease in the
selectivity because of more considerable surface modification.
Keywords: liquid-phase hydrogenation of diphenylacetylene, segregation, Pd–Ag alloy, IR spectroscopy of
adsorbed CO
DOI: 10.1134/S0023158418050154
INTRODUCTION
prevent coalescence. Considerable disadvantages of
this procedure consist in the complexity of the per-
formed operations and in the use of various chemical
compounds as stabilizers, which lead to the contami-
nation of nanoparticles. The catalytic systems thus
prepared require further thorough removal of the
impurities of organic ligands [9, 10].
Directions in the science of catalysis related to the
development and characterization of new supported
catalytic systems with bi- or polymetallic particles as
the active centres have been intensively developed in
the recent decades [1–3]. The advantages of these cat-
alysts include increased activity and selectivity for the
desired product, stability and resistance to deactiva-
tion. An improvement in the catalytic characteristics
was explained by the interaction of the polymetallic
catalyst particles with each other to form combined
phases (solid solutions, alloys, or intermetallic com-
pounds) [4, 5]. The basic factors responsible for the
properties of these catalytic systems are the composi-
tion and surface structure of nanoparticles [6]. A pre-
requisite to the production of a highly selective catalyst
is the uniformity of active centers on its surface. This
requirement should be obeyed for the target-oriented
synthesis of complex organic compounds with ultra-
high selectivity [7], which requires effective methods
for the control of the surface structure of catalyst.
A promising alternative can be the application of
methods based on the surface segregation of nanopar-
ticle components, which makes it possible to regulate
the composition and surface properties of the
nanoparticle over wide limits. The driving force of sur-
face segregation is the tendency to decrease the surface
energy of a bimetallic system [11]. In this case, segre-
gation can be controlled by the adsorption of different
adsorbate molecules on the surface to enrich the cata-
lyst with a component whose binding energy with the
adsorbate molecule is higher. Because the composi-
tion and the structure of surface layers strongly depend
on sample treatment conditions (temperature, reac-
tion atmosphere, and reaction time), surface segrega-
One of these methods is based on the preparation tion can be used as a method of controlling the struc-
of a core–shell structure [8]. The formation of parti- ture of catalyst surface. The advantage of this method
cles is accomplished by a colloid method, in which is the absence of surface contamination which elimi-
nanoparticles are stabilized in solution by the adsorp- nates the need for additional purification. Molecules
tion of organic compounds on their surface in order to with high adsorption energies, such as NO, CO, and
610

ADSORPTION-INDUCED SEGREGATION AS A METHOD
611
O2, can be used as the adsorbates [12]. Co-induced were studied in the liquid-phase hydrogenation of
diphenylacetylene (DPA).
EXPERIMENTAL
surface segregation in the bimetallic systems based on
noble metals changes the configuration of active cen-
ters; in turn, this can substantially influence the activ-
ity and selectivity of multicomponent catalysts [13].
Catalyst Preparation
Despite the fact that the application of adsorption-
induced segregation has not been adequately investi-
gated up to now, it was found that the catalytic proper-
ties of systems such as Pt–Cu [13], Pd–Au [14], and
Cu–Ni [15] could be effectively affected with the aid
of this method. Very interesting results were obtained
for the Pd–Cu catalysts with different Pd : Cu ratios
[6]. It was found that, after the treatment in CO, a sur-
face rearrangement occurred in the Pd–Cu system,
which was retained under reaction conditions and
changed catalytic characteristics in the gas-phase reac-
tion of selective acetylene hydrogenation.
In this work, we attempted to use the adsorption-
induced segregation for controlling the catalytic char-
acteristics of the Pd–Ag system. This was related to the
fact that the Pd–Ag catalysts are widely used in the
selective hydrogenation of hydrocarbons with multiple
bonds. An enormous number of publications were
devoted to the research and development of Pd–Ag
alloys for the industrial processes of the removal of
acetylene impurities from ethylene [16–19]. Further-
more, these systems are highly selective catalysts in the
liquid-phase hydrogenation of terminal and internal
alkynes [20].
An increase in the selectivity of the Pd–Ag alloys is
related to a number of factors: thus, upon the modifi-
cation with silver, the electronic state of palladium is
changed and the formation of palladium hydrides,
which are responsible for nonselective hydrogenation,
is suppressed. The formation of a surface structure in
which palladium atoms are isolated from each other by
the atoms of silver (a so-called single-atom structure)
[21, 22] is of importance; because of this, the active
centers possess a high degree of homogeneity.
The effect of adsorbates on the surface characteris-
tics of Pd–Ag catalysts was examined in a number of
experimental and theoretical studies [23]; however,
there are almost no publications dedicated to the tar-
get-oriented change of the catalytic properties of Pd–
Ag systems. McCue and Anderson [6] demonstrated
the possibility of controlling the surface structure of
bimetallic Pd–Cu catalysts by treating the samples in
an CO medium; however, studies of this kind were not
performed for the Pd–Ag systems.
The sample containing 1 wt % Pd and 2 wt % Ag
(the molar ratio Ag : Pd = 2) was prepared by the
incipient wetness co-impregnation of aluminum oxide
(Sasol, SBET = 56 m2/g) with an aqueous solutions of
palladium and silver nitrates. After the impregnation,
the catalyst was dried at room temperature for a day
and then reduced in a flow of 5% H2/Ar at 550°C for
2 h. Monometallic 0.5 wt % Pd/Al2O3, which was pre-
pared in accordance with this procedure, served as a
reference sample. Before measurements, the prepared
samples were ground in a mortar to obtain uniform
fine powder.
Transmission Electron Microscopy
The microstructure of the samples was studied by
transmission electron microscopy (TEM) on a
HT7700 electron microscope (Hitachi, Japan). The
freshly prepared Pd/Al2O3 and Pd–Ag2/Al2O3 cata-
lysts were investigated. All manipulations with the
samples were performed in air. The images were
obtained in the bright-field regime at an accelerating
voltage of 100 kV. Before imaging, a powder sample
was ultrasonically dispersed from a suspension in iso-
propanol to copper gauze with a diameter of 3 mm
covered with a carbon film. The analytical measure-
ments were optimized within the framework of an
approach described earlier [24].
IR Spectroscopy of Adsorbed CO
The diffuse reflectance IR spectra of adsorbed CO
were measured using a Tensor 27 IR spectrometer
(Bruker, Germany) equipped with a Harrick Diffuse
Reflectance Kit for in situ measurements (Harrick
Scientific Products, the United Kingdom). Since the
changes in the surface structure of the catalyst after
segregation in CO and synthetic air were studied in the
present work, the treatment with adsorbates and the
characterization of the sample surface were performed
in situ in a thermostatically controlled cell.
For the measurement of spectra, a 20-mg test sam-
ple was placed in a thermostatically controlled cell
with CaF2 windows; then, it was heated to 500°C in a
flow of argon (30 mL/min). Thereafter, it was reduced
at 500°C for 1 h in a flow of 5% H2/Ar (30 mL/min)
and cooled initially to 300°C in a flow of 5% H2/Ar
and then to 50°C in a flow of argon, and the back-
ground spectrum was recorded. The spectra of
adsorbed CO were measured at 50°C in a flow of
The aim of this work was to study the target-ori-
ented regulation of the catalytic characteristics of pal-
ladium–silver catalysts with the use of the adsorption-
induced surface segregation. CO and O2 were chosen
as adsorbates. The diffuse reflectance IR spectroscopy
of adsorbed CO was used for the monitoring of the sur-
face structure of Pd–Ag nanoparticles, and the cata-
lytic properties of the freshly reduced Pd–Ag catalyst
and the samples after adsorption-induced segregation 0.5 vol % CO/He (500 scans; resolution, 4 cm–1).
KINETICS AND CATALYSIS Vol. 59 No. 5 2018

612
STAKHEEV et al.
At the following stage, the catalyst was processed at rate of alkene into alkane conversion (r2) for the Pd–
200°C either in an atmosphere of 20% O2/N2 (syn-
Ag catalysts is much lower than that for a monometal-
thetic air, 30 mL/min) for 30 min or in a flow of 30% lic analog, this parameter was determined as follows:
CO/N2 (30 mL/min) for 30 min; thereafter, the tem- in a range of 1.3–1.8 equiv H2 for Pd/Al2O3, in a range
perature was decreased to 50°C in an inert atmo- of 1.1–1.4 equiv H2 for Pd–Ag2/Al2O3 treated in a flow
sphere, and the spectra of adsorbed CO were mea-
sured.
of CO and O2, and in a range of 1.05–1.15 equiv H2 for
freshly reduced (Pd–Ag2/Al2O3 + H2). The kinetic
selectivity of reaction was evaluated from a ratio
between the rates of DPA hydrogenation at the first
and second stages (r1/r2). The specific activity (TOF)
of the samples was calculated as a ratio of the number
of converted substrate molecules to the total number
of palladium atoms in the catalyst per second.
To reveal a correlation between spectral and cata-
lytic data and to minimize sample contact with the
environment, an accurately weighed catalyst portion
was placed in a cell of the spectrometer and required
treatments (reduction in H2/Ar; treatment in a mix-
ture of 20% O2/N2 or 30% CO/N2) were performed in
accordance with the above procedure but without the
measurement of the IR spectra because of an insuffi-
cient quantity of the sample. Thereafter, the sample
was removed from the cell of the spectrometer in a
flow of argon and quantitatively transferred to a cata-
lytic reactor.
RESULTS AND DISCUSSION
Transmission Electron Microscopy
The samples of Pd/Al2O3 and Pd–Ag2/Al2O3 were
studied by transmission electron microscopy. Small
spherical contrasting particles in the micrographs
(Fig. 1) correspond to the supported metals, and large
bright ones correspond to the carrier. An analysis of
Diphenylacetylene Hydrogenation
The liquid-phase hydrogenation of diphenylacety- several micrographs allowed us to conclude that the
lene (99%, Aldrich) was carried out at room tempera- uniform distribution of metallic particles in the struc-
ture (25°C) in an autoclave-type reactor mounted on ture of the carrier was characteristic of both of the
a magnetic stirrer. A catalyst sample, 6 mL of an n- samples. For the monometallic catalyst, the average
hexane solvent (98%, Merck), 170 mg of DPA, and a size of nanoparticles was ~3–4 nm; bimetallic parti-
magnetic stirring bar were placed in the reactor; there- cles were somewhat larger, and their mean diameter
after, the system was connected to a gas supply unit was ~5–7 nm.
and an electronic pressure sensor for the control of the
degree of hydrogen absorption. The initial pressure of
H2 was 5 bar, and the rate of stirring of the reaction
IR Spectroscopy of Adsorbed CO
mixture was 1000 rpm. The sample weight was selected
individually (1.5–2.8 mg) for each experiment in order
to ensure the kinetic regime of the process. Before the
experiment, a flow of argon (~30 mL/min) was passed
through the solvent for 3 h for the removal of oxygen
impurities dissolved in n-hexane. Selectivity for olefin
(Sen) was calculated based on the data of the gas-chro-
matographic analysis of a reaction mixture on a Crys-
tall 5000 chromatograph (Chromatek, Russia) with a
flame-ionization detector. The calculation procedure
was described elsewhere [25].
The infrared spectroscopy of adsorbed CO was
used for studying the surface structure; the advantages
of this method include high information content and
relative simplicity of the experiments. The state of Pd–
Ag2/Al2O3 was studied after its in situ reduction in the
cell of the spectrometer. For studying the effect of CO-
and O2-induced segregation on the surface structure,
a freshly reduced catalyst was treated in a flow of 30%
CO/N2 or 20% O2/N2; thereafter, the cell was cooled
in a flow of argon, and the spectra were recorded. The
reduced Pd/Al2O3 catalyst (H2/Ar, 500 °C, 1 h) was
used as a reference sample.
Because the hydrogenation of DPA consecutively
occurs first to diphenylethylene (DPE) and then to
diphenylethane, the amount of the complete hydroge-
nation product formed at the first stage of hydrogena-
tion is insignificant. The hydrogenation of the residual
DPA to DPE at the second stage of the process can
also be disregarded. Taking into account these
assumptions, we determined the reaction rates r
−1
Two wide absorption bands were observed in the IR
spectra of CO adsorbed on the Pd/Al2O3 catalyst
(Fig. 2, spectrum 1) in a range of 2200–1800 cm–1. An
absorption band in a region of 2100–2050 cm–1 with a
maximum at 2090 cm–1 corresponds to vibrations of
CO molecules adsorbed on the sample surface in a lin-
ear form, and it is characteristic of palladium catalysts
supported onto aluminum oxide [26]. A band in a
range of 2000–1800 cm–1 with a maximum at
1980 cm–1 belongs to the two- and three-coordinated
forms of CO, and it is indicative of the presence of the
multicenter adsorption of CO on the adjacent atoms
(ensembles) of palladium [27].
(
min–1) for each stage of hydrogenation
molH
g
Cat
2
from a graph of the dependence of the quantity of
absorbed hydrogen on the reaction time. The rates of
alkyne to alkene hydrogenation (r1) for all samples
were calculated from the slopes of straight portions in
an absorption range of 0.1–0.6 equiv H2. Because the
KINETICS AND CATALYSIS
Vol. 59
No. 5
2018

ADSORPTION-INDUCED SEGREGATION AS A METHOD
613
50 nm
50 nm
(а)
(b)
Fig. 1. Typical TEM micrographs of the catalysts (a) Pd/Al O and (b) freshly reduced Pd–Ag /Al O .
2
3
2
2 3
The spectrum of the Pd–Ag2/Al2O3 catalyst absorption bands were almost completely absent from
a region of <2000 cm–1; this fact is also indicative of
the formation of an ordered alloy structure (a so-
called single-atom structure), in which each palla-
dium atom is isolated by the atoms of a diluent element
(silver) to prevent the appearance of a bridged-
adsorbed CO [28].
(Fig. 2, spectrum 2) exhibited an intense absorption
band with a maximum at 2049 cm–1. A shift of the
band maximum by ~40 cm–1 toward lower wavenum-
bers, in comparison with the absorption band position
of the linear adsorbed CO in the spectrum of mono-
metallic Pd/Al2O3, can be explained by a decrease in
the dipole–dipole interaction between CO molecules
adsorbed on the adjacent atoms of palladium due to
the dilution of Pd with Ag atoms. This can also be
related to an increase in the electron density on the Pd
atoms due to the formation of a Pd–Ag alloy; because
of this, donation to the π-antibonding orbital of the
adsorbed CO molecule becomes more pronounced
and leads to C–O bond weakening [28]. Note that
The treatment of Pd–Ag2/Al2O3 in a flow of
CO/N2 at 200°C (Fig. 2, spectrum 3) led to the
appearance of an absorption band at 2000–1900 cm–1,
which is characteristic of the two- and three-point
adsorption of CO. Its intensity is only slightly lower
than the absorption band intensity of the linear
adsorbed CO. Furthermore, a small shift of the
absorption band of linearly adsorbed CO to the region
of higher wavenumbers (2054 cm–1) was observed in
the spectrum of Pd–Ag2/Al2O3 after treatment with
CO. These changes indicate the transformation of iso-
lated monatomic centers (Pd1) to polyatomic centers
(Pdn), which occurs during the surface segregation of
palladium. The surface is enriched in palladium due to
the formation of strong Pd–CO bond; because of this,
the surface segregation of Pd becomes a thermody-
namically favorable process. It should be noted that
there were no absorption bands with a frequency lower
than 1925 cm–1 in the spectrum of Pd–Ag2/Al2O3 after
treatment in CO; this fact indicates the predominance
of the two-coordinated (bridge) form of CO adsorp-
tion and the absence of the more highly coordinated
forms. Thus, it is possible to conclude that the active
centers of two adjacent atoms (dimers) of palladium
are formed on the catalyst surface.
Absorbance
1985
2071
4
2054
2049
1979
3
2
1
1959
1978
2090
The exposure of the Pd–Ag2/Al2O3 catalyst to a
flow of O2/N2 at 200°C led to the more essential sur-
face modification than treatment in CO. Thus, the IR
spectrum of adsorbed CO also exhibited a peak with a
maximum at 1985 cm–1, which corresponds to multi-
center CO adsorption on Pd; however, its intensity was
noticeably higher than the one of the linear adsorbed
CO (Fig. 2, spectrum 4). Upon the oxidation of Pd–
Ag2/Al2O3, the maximum of linearly adsorbed CO
2200 2150 2100 2050 2000 1950 1900 1850 1800
Wavenumber, cm–1
Fig. 2. Diffuse reflection IR spectra of CO adsorbed on the
catalysts: (1) Pd/Al O (reference sample); (2) freshly
2
3
reduced Pd–Ag /Al O ; (3) the Pd–Ag2/Al O catalyst
2
2
3
2 3
after treatment in a flow of 30% CO/N ; and (4) Pd–
2
Ag /Al O after treatment in a flow of 20% O /N .
(2071 cm–1) considerably shifted to the greater wave-
2
2
3
2
2
KINETICS AND CATALYSIS Vol. 59 No. 5 2018

614
A ,
STAKHEEV et al.
Diphenylacetylene Hydrogenation
mol H2/mol DPA
H2
Catalyst activity. Figure 3 shows the kinetic curves
of hydrogen absorption for the initial freshly reduced
Pd–Ag2/Al2O3, the samples in situ treated in atmo-
spheres of CO and O2 and data for the reference
monometallic Pd/Al2O3. Because of a higher activity,
the sample weight of the monometallic catalyst was
reduced for retaining the kinetic regime of the process.
Characteristic bend in the curves at ~1 equiv H2 in
Fig. 3 are related to a decrease in the reaction rate on
going from the first stage of hydrogenation (DPA into
DPE) to the second. The process of complete hydro-
genation most rapidly proceeds on the monometallic
palladium catalyst and completes in ~140 min. The
modification with silver substantially slows down the
rate of the second stage of hydrogenation, as evi-
denced by a pronounced bend of the kinetic curves in
the region of >1 equiv H2 (Fig. 3).
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
Pd–Ag2/Al2O3 + H2
Pd–Ag2/Al2O3 + CO
Pd–Ag2/Al2O3 + O2
Pd/Al2O3
0
20
40
60
80 100 120 140 160
τ, min
The activity of Pd–Ag2/Al2O3 is substantially lower
than that of monometallic Pd/Al2O3, which is confirmed
by a decrease in the rate of hydrogenation of the initial
Fig. 3. Kinetic curves of hydrogen absorption in the hydro-
genation of DPA on the Pd/Al O and Pd–Ag/Al O cat-
2
3
2 3
alysts after different treatments.
−1
DPA (r ) from 15.1 to 3.06
min–1 (Table 1).
molH
g
1
Cat
2
Treatment with CO leads to a noticeable increase in
the activity. This is evident from an increase in the rate
numbers, which are characteristic of the adsorption of
CO on the surface of monometallic Pd nanoparticles,
and the shape of this absorption band is complex and
includes several components. The experimental
results are consistent with published data, according to
which oxygen molecules are predominantly localized
on palladium atoms upon adsorption on the
PdAg(111) faces [29]; as a result, palladium segregates
to the catalyst surface in the course of oxidation in air
at elevated temperatures [30].
In spite of the more considerable changes in the
surface structure observed after the treatment of Pd–
Ag2/Al2O3 in oxygen, an absorption band at 2090 cm–1,
which is characteristic of the adsorption of CO on
monometallic Pd, does not appear. A shoulder in the
region of smaller wave numbers also indicates that a
part of the bimetallic alloy occurs on the surface in an
unchanged state.
Thus, according to the results obtained by the IR
spectroscopy of adsorbed CO, the treatment of the
bimetallic Pd–Ag catalyst in O2 and CO medium leads
to the partial disappearance of the isolated centers Pd1
and their transformation into the polyatomic Pdn as a
result of the adsorption-induced surface segregation of
Pd. After treatment in CO, the absorption band maxi-
mum of linear CO changed insignificantly. This fact
suggests that the geometry of active centers in the sample
somewhat changed after segregation in CO with the
retention of the electronic state of palladium. Note that
the observed changes in the surface structure of the Pd–
Ag nanoparticles are similar to the transformations
observed earlier by McCue and Anderson [4, 6] in the
Pd–Cu/SiO2 catalysts.
−1
of the first stage r from 3.06 to 6.72
min–1
molH
g
1
Cat
2
(Table 1). The surface segregation of palladium in the
Pd–Ag sample as a result of treatment in O2 exerts an
even larger effect on the rate change: r1 increased from
−1
Cat
3.06 to ~8.4
min–1.
molH
g
2
The values of TOF1 also demonstrate that the addi-
tion of silver to the catalyst strongly (by a factor of ~10)
decreased the specific activity of the bimetallic cata-
lyst relative to that of Pd/Al2O3; however, the above
parameter increased by a factor of ~2 after segregation,
as compared with that of freshly reduced Pd–
Ag2/Al2O3. Based on the calculated values of TOF for
the second stage of reaction (the undesirable process
of stilbene hydrogenation), it is possible to conclude
that, in the case of bimetallic systems, the essential (by
almost two orders of magnitude) suppression of the
formation of diphenylethane is observed, as compared
with the monometallic analog.
The rate ratio r1/r2, which characterizes the kinetic
process efficiency, is an important parameter of the
process of hydrogenation (Table 1). For the monome-
tallic palladium catalyst, the ratio r1/r2 is 4.10, whereas
it increases to 40.7 for Pd–Ag2/Al2O3. After the treat-
ments of Pd–Ag2/Al2O3 in CO and O2, the ratio r1/r2
increased to ~52–53 because of a predominant
increase in the rate of the first stage.
The decrease in the catalytic activity observed for
Pd–Ag2/Al2O3 relative to that of Pd/Al2O3 is related to
a number of factors:
(1) In the Pd–Ag2/Al2O3 samples the number of
the active centers of hydrogenation on the surface of
KINETICS AND CATALYSIS
Vol. 59
No. 5
2018

ADSORPTION-INDUCED SEGREGATION AS A METHOD
615
Table 1. Kinetic characteristics of the hydrogenation of diphenylacetylene on Pd and Pd–Ag catalysts
r1 × 103,
r2 × 103,
TOF1, s–1
TOF2, s–1
mCat, mg
r1/r2
Sample
−1
min–1
molH
g
Cat
2
Pd–Ag2/Al2O3 + H2
Pd–Ag2/Al2O3 + CO
Pd–Ag2/Al2O3 + O2
Pd/Al2O3
2.7
2.6
2.8
1.5
3.06
6.72
8.37
15.1
0.08
0.13
0.16
3.70
0.54
1.19
1.49
5.38
0.013
0.023
0.028
1.31
40.7
52.2
53.4
4.10
metallic nanoparticles decreased as a result of the sub- tion of active centers from several palladium atoms
stitution of silver atoms with Pd ones.
(Pdn ensembles) on the catalyst surface. The oxidation
of Pd–Ag2/Al2O3 led to more substantial changes in
the structure than the treatment with CO, providing
higher catalytic activity. These observations are con-
sistent with data published by McCue and Anderson
[6], who found that the presence of palladium dimers
on the surface considerably increased catalyst activity
in the hydrogenation of acetylene.
Selectivity for the formation of stilbene. Selectivity
for olefin is the most important characteristic of cata-
lysts for the hydrogenation of alkyne compounds;
therefore, one of the tasks of this work was to compar-
atively evaluate selectivity for the formation of stilbene
(Sen) on Pd/Al2O3 and Pd–Ag/Al2O3 before and after
in situ treatments (Fig. 4). According to the experi-
mental data, the highest values of Sen (97–95% at DPA
conversions (XDPA) of 53.9 and 96.7%, respectively)
were obtained on the freshly reduced Pd–Ag2/Al2O3
catalyst. The selectivity of the sample treated in CO
was almost the same: 96.8% at XDPA = 44.6% and
94.6% at XDPA = 98.6%. The selectivity of the catalyst
(2) Upon the introduction of silver, electronic state
of palladium changed to cause a decrease in the
adsorption energy of reactant molecules on the cata-
lyst surface [31].
(3) In the bimetallic catalyst, the number of poly-
atomic Pdn centers, which are in charge for high
hydrogenation rate, decreased.
It is likely that the observed increase in the Pd–
Ag2/Al2O3 catalyst activity after the treatment in CO
and O2 occurred because of a quantitative increase in
the active centers as a result of the segregation of Pd
onto the surface and due to the transformation of
monatomic centers into the multiatomic Pdn, which
can exhibit higher activity. Because only the isolated
centers of palladium occurred on the surface of the
freshly reduced Pd–Ag catalyst, this sample exhibited
the lowest activity in the hydrogenation of DPA. The
spectroscopic data obtained after treatment in an
atmosphere of adsorbates are indicative of the forma-
treated with O2 was much lower (94.0% at XDPA
=
90.5%). In the case of the monometallic palladium
Sen, %
catalyst, this characteristic had the lowest values.
100
It is well known that the reduction of Pd–Ag cata-
lysts with hydrogen at 500°C significantly increases
the selectivity of ethylene formation in the course of
acetylene hydrogenation [32, 33]. Zea et al. [33]
related this phenomenon to the dilution of palladium
atoms with silver upon the formation of alloy (a geo-
metric effect). The dilution of the polyatomic centers
Pdn with another metal prevents the strong adsorption
of a substrate on the metal thus preventing its further
hydrogenation [34]. Furthermore, in contrast to
metallic palladium, the Pd–Ag alloy does not dissolve
hydrogen, which negatively influences selectivity for
the desired alkene [35]. It is likely that, in our case, the
high selectivity of the stilbene formation on the freshly
reduced Pd–Ag2/Al2O3 can also be explained by the
presence of isolated palladium centers.
98
96
94
Pd–Ag2/Al2O3 + H2
Pd–Ag2/Al2O3 + CO
Pd–Ag2/Al2O3 + O2
Pd/Al2O3
92
90
0
20
40
60
80
100
DPA conversion, %
A small decrease in Sen for the catalyst treated in a
flow of CO is related to the formation of active centers
consisting of two adjacent atoms of palladium (dimers
[6]). Oxidation in a flow of oxygen is accompanied not
Fig. 4. Dependence of the selectivity of alkene formation
on the conversion of DPA for the Pd/Al O and Pd–
2
3
Ag/Al O catalysts after different treatments.
2
3
KINETICS AND CATALYSIS Vol. 59 No. 5 2018

616
STAKHEEV et al.
only by the appearance of a larger quantity of dimers grateful to the Department of Structural Studies of the
but also by a significant change in the electronic state Zelinsky Institute of Organic Chemistry, Russian
of Pd atoms, as evidenced by a shift of the absorption Academy of Sciences for studying the samples by elec-
band maximum of the IR spectrum of adsorbed CO. tron microscopy.
As found by Venezia at al. [30], treatment in O2 at an
elevated temperature (350°C) causes the segregation
REFERENCES
1. Furukawa, S. and Komatsu, T., ACS Catal., 2016,
of palladium onto the surface of Pd–Ag nanoparticles
with the subsequent formation of PdO nanoparticles.
In our study, the treatment temperature was lower
(200°C); however, according to the results obtained by
IR spectroscopy, it was sufficient for changing the
electronic state of palladium. Thus, the effect of O2 on
the Pd–Ag catalyst leads to the more significant sur-
face modification accompanied by the segregation of
palladium and, probably, by the formation of Pd metal
nanoclusters, resulting in a marked decrease in stil-
bene selectivity.
vol. 6, p. 2121.
2. Armbruster, M., Behrens, M., Cinquini, F., Fottinger, K.,
Grin, Yu., Haghofer, A., Klotzer, B., Knop-Gericke, A.,
Lorenz, H., Ota, A., Penner, S., Prinz, J., Rameshan, C.,
Revay, Z., Rosenthal, D., Rupprechter, G., Sautet, Ph.,
Schlogl, R., Shao, L., Szentmiklosi, L., Teschner, D.,
Torres, D., Wagner, R., Widmer, R., and Wowsnick, G.,
ChemCatChem, 2012, vol. 4, p. 1048.
3. Ananikov, V.P., Eremin, D.B., Yakukhnov, S.A., Dil-
man, A.D., Levin, V.V., Egorov, M.P., Karlov, S.S.,
Kustov, L.M., Tarasov, A.L., Greish, A.A., Shester-
kina, A.A., Sakharov, A.M., Nysenko, ZN., Shereme-
tev, A. B., Stakheev, A.Yu., Mashkovsky, I.S., Sukho-
rukov, A.Yu., Ioffe, S.L., Terent’ev, A.O., Vil’, V.A.,
Tomilov, Yu.V., Novikov, R.A., Zlotin, S.G., Kuche-
renko, A.S., Ustyuzhanina, N.E., Krylov, V.B., Tsvet-
kov, Yu.E., Gening, M.L., and Nifantiev, N.E., Men-
deleev Commun., 2017, vol. 27, p. 425.
CONCLUSIONS
Thus, the study showed that the treatment of the
Pd–Ag2/Al2O3 catalyst in the atmospheres of different
adsorbates (CO, O2) at an elevated temperature (200°C)
facilitated the appearance of structures that, appar-
ently, consisted of two neighboring palladium atoms
(dimers) on the surface. The IR spectrum of CO
adsorbed on the catalyst treated in O2 is indicative of a
more significant surface modification than that upon
the treatment in CO accompanied by a change in the
electronic state of supported Pd. The results of study-
ing the properties of the catalysts in a model reaction
of the liquid-phase hydrogenation of diphenylacety-
lene made it possible to conclude that the treatment in
an atmosphere of CO facilitated an increase in the cat-
alytic activity of diphenylacetylene hydrogenation; in
this case, selectivity for the formation of stilbene did
almost not decrease. After the oxidation of Pd–
Ag2/Al2O3 by oxygen, its activity further increased;
however, selectivity in the region of high DPA conver-
sions noticeably decreased (from 95.4 to ~94.0%).
Hypothetically, this was related to the transformation
of the isolated palladium active centers Pd1 into the
polyatomic Pdn.
4. McCue, A.J., and Anderson, J.A., Front. Chem. Sci.
Eng., 2015, vol. 9, p. 142.
5. Furukawa, S., and Komatsu, T., ACS Catal., 2017,
vol. 7, p. 735.
6. McCue, A.J. and Anderson, J.A., J. Catal., 2015,
vol. 329, p. 538.
7. Ananikov, V.P., Khemchyan, L.L., Ivanova, Y.V., Dil-
man, A.D., Levin, V.V., Stakheev, A.Yu., Turova, O.V.,
Mashkovsky, I.S., Terent’ev, A.O., Krylov, I.B., Bukh-
tiyarov, V.I., Sorokin, A.M., Prosvirin, I.P., Roma-
nenko, A.V., Simonov, P.A., Bukhtiyarova, G.A., Kop-
tyug, I.V., Kovtunov, K.V., Zhivonitko, V.V., Vatsadze, S.Z.,
et al., Rus. Chem. Rev., 2014, vol. 83, no. 10, p. 885.
8. Semagina, N. and Kiwi-Minsker, L., Catal. Rev., 2009,
vol. 51, p. 147.
9. Lopez-Sanchez, J.A., Dimitratos, N., Hammond, C.,
Brett, G.L., Kesavan, L., White, S., Miedziak, P., Tiru-
valam, R., Jenkins, R.L., Carley, A.F., Knight, D.,
Kiely, C.J., and Hutchings, G.J., Nature Chem., 2011,
vol. 3, p. 551.
10. Crespo-Quesada, M., Andanson, J.-M., Yarulin, A.,
Lim, B., Xia, Y., and Kiwi-Minsker, L., Langmuir,
2011, vol. 27, p. 7909.
The data obtained in this work indicate that the
adsorption-induced segregation can be a convenient
method for the target-oriented modification of the
surface structure and, as a consequence, the catalytic
properties of bimetallic Pd–Ag catalysts.
11. Zafeiratos, S., Piccinin, S., and Teschner, D., Catal.
Sci. Technol., 2012, vol. 2, p. 1787.
12. Karski, S., Witonska, I., Rogowski, J., and
Goluchowska, J., J. Mol. Catal. A: Chem., 2005,
vol. 240, p. 155.
13. Andersson, K.J., Calle-Vallejo, F., Rossmeisl, J., and
Chorkendorff, I., J. Am. Chem. Soc., 2009, vol. 131,
p. 2404.
ACKNOWLEDGMENTS
The IR-spectroscopic study of adsorbed CO and
the catalytic tests of the synthesized catalyst samples
were supported by the Russian Foundation for Basic
Research (project no. 16-29-10788). The develop-
ment of a procedure for the synthesis of the bimetallic
Pd–Ag2/Al2O3 catalyst was supported by the Russian
Science Foundation (grant no. 17-13-01526). We are
14. Gao, F., Wang, Y., and Goodman, D.W., J. Phys.
Chem. C, 2009, vol. 113, p. 14993.
15. Yao, Y. and Goodman, W., Phys. Chem. Chem. Phys.,
2014, vol. 16, p. 3823.
KINETICS AND CATALYSIS
Vol. 59
No. 5
2018

ADSORPTION-INDUCED SEGREGATION AS A METHOD
617
16. Pachulski, A., Schödel, R., and Claus, P., Appl. Catal.,
A, 2011, vol. 400, p. 14.
G., Freund, H.-J., and Lennon, D., J. Chem. Phys.,
2005, vol. 123, p. 174706.
17. Kuhn, M., Lucas, M., and Claus, P., Ind. Eng. Chem. 27. Cabilla, G.C., Bonivardi, A.L., and Baltanas, M.A.,
Res., 2015, vol. 54, p. 6683.
Catal. Lett., 1998, vol. 55, p. 147.
28. Rassolov, A.V., Markov, P.V., Bragina, G.O., Baeva,
G.N., Krivoruchenko, D.S., Mashkovskii, I.S., Yaku-
shev, I.A., Vargaftik, M.N., and Stakheev, A.Yu.,
Kinet. Catal., 2016, vol. 57, no. 6, p. 859.
18. Ravanchi, M.T. and Sahebdelfar, S., Appl. Catal., A,
2016, vol. 525, p. 197.
19. Ravanchi, M.T., Sahebdelfar, S., Fard, M.R., Fadaeer-
ayeni, S., and Bigdeli, P., Chem. Eng. Technol., 2016,
vol. 39, p. 301.
20. Rassolov, A.V., Markov, P.V., Bragina, G.O., Baeva, G.N.,
Mashkovskii, I.S., Stakheev, A.Y., Yakushev, I.A., and
Vargaftik, M.N., Kinet. Catal., 2016, vol. 57, no. 6,
p. 853.
29. Wouda, P.T., Schmid, M., Nieuwenhuys, B.E., and
Varga, P., Surf. Sci., 1999, vol. 423, p. 229.
30. Venezia, A.M., Liotta, L.F., Deganello, G., Schay, Z.,
Horvath, D., and Guczi, L., Appl. Catal., A, 2001,
vol. 211, p. 167.
31. Nikolaev, S.A., Smirnov, V.V., Zanaveskin, L.N.,
Zanaveskin, K.L., and Averyanov, V.A., Rus. Chem.
Rev., 2009, vol. 78, no. 3, p. 231.
21. Pei, G., Liu, Xi.Y., Yang, Xi., Zhang, L., Wang, A., Li, L.,
Wang, H., Wang, Xi., and Zhang, T., ACS Catal., 2017,
vol. 7, p. 1491.
32. Jin, Y., Datye, A.K., Rightor, E., Gulotty, R., Water-
man, W., Smith, M., Holbrook, M., Maj, J., and
Blackson, J., J. Catal., 2001, vol. 203, p. 292.
22. Pei, G., Liu, Xi.Y., Wang, A., Lee, A.F., Isaacs, M.A.,
Li, L., Pan, Xi., Yang, Xi., Wang, Xi., Tai, Z., Wilson, K.,
and Zhang, T., ACS Catal., 2015, vol. 5, p. 3717.
33. Zea, H., Lester, K., Datye, A.K., Rightor, E., Gulotty, R.,
Waterman, W., and Smith, M., Appl. Catal., A, 2005,
vol. 282, p. 237.
23. Jaatinen, S., Salo, P., Alatalo, M., Kulmala, V., and
Kokko, K., Surf. Sci., 2003, vol. 529, p. 403.
24. Kachala, V.V., Khemchyan, L.L., Kashin, A.S., Orlov, N.V.,
Grachev, A.A., Zalesskiy, S.S., and Ananikov, V.P.,
Russ. Chem. Rev., 2013, vol. 82, no. 7, p. 648.
34. Borodzinski, A. and Bond, G.C., Catal. Rev., 2008,
vol. 50, p. 379.
35. Witonska, I., Karski, S., Frajtak, M., Krawczyk, N.,
and Krolak, A., React. Kinet. Catal. Lett., 2008, vol. 93,
no. 2, p. 241.
25. Bragina, G.O., Smirnova, N.S., Krivoruchenko, D.S.,
Markov, P.V., Baeva, G.N., and Stakheev, A.Yu.,
Kinet. Katal., 2017, vol. 58, no. 6, p. 763.
26. Lear, T., Marshall, R., Lopez-Sanchez, J.A., Jack-
Translated by V. Makhlyarchuk
son, S.D., Klapotke, T.M., Baumer, M., Rupprechter,
KINETICS AND CATALYSIS Vol. 59 No. 5 2018