A.V. Rassolov et al.
Molecular Catalysis 506 (2021) 111550
% was observed within the whole conversion range, which was almost
10-fold higher as compared to the monometallic counterpart with the
same metal loading. Later formation of SAA structures for Pd-Cu/Al2O3
catalysts was confirmed by Cao et al. using HAADF-STEM and
H2-chemsorption [24]. Due to their SAA structure Pd-Cu catalysts
exhibit excellent (>73 %) selectivity to ethylene at acetylene conversion
over 99 %.
obtained using the same preparation procedure (designated as Pd). To
obtain similar metal particle sizes for both Pd1Ag3/Al2O3 and Pd/Al2O3
catalysts the latter one was reduced at 700 ◦C for 3 h.
Catalyst characterization
To analyze the particle size distribution and to determine the average
particle size transmission electron microscopy (TEM) was performed.
The TEM images were collected with a Hitachi HT 7700 electron mi-
croscope (Japan) in the bright-field regime at an accelerating voltage of
100 kV. The catalysts were powdered and ultrasonically dispersed from
a suspension in isopropanol to copper gauze with a diameter of 3 mm
covered with a carbon film. The size distribution was estimated by
measuring 150–200 particles located in different parts of the samples.
The average nanoparticle size was calculated as follows:
High selectivity of SAA catalysts in acetylene hydrogenation was also
revealed by Pei et al. for PdCu, PdAu and PdAg catalysts [19–22]. By
diluting Pd with a second metal, the authors found that an excess of the
latter is a prerequisite for the successful preparation of SAAC. Excellent
selectivity to ethylene was explained by a higher electron density on
single Pd atoms within the alloy decreasing the strength of the π-bonded
ethylene adsorption over such electron-rich atoms [20]. This explana-
tion was supported by a microcalorimetric study of ethylene adsorption
on SAA Pd-Ag catalyst, which demonstrated a lower ethylene adsorption
enthalpy compared to monometallic Pd analogue [21,22].
dav = Σnidi/n,
As can be seen from the studies cited above, the performance of SAA
catalysts in the gas-phase selective hydrogenation of acetylene has been
intensively investigated. For example catalysts based on the Pd-Ag
composition are efficient in gas-phase hydrogenation of acetylene to
ethylene [25–29]. However, their catalytic characteristics in the
liquid-phase hydrogenation of substituted alkyne compounds remains
insufficiently studied, despite the fact that this is a key process for
important transformations of internal alkynes into cis- and trans-alkene
moieties serving as building blocks in pharmaceutical and fine chemical
industries. Therefore, in this study the investigation was focused on the
properties of PdAg single-atom alloy catalyst in this process.
where dav is the average diameter of nanoparticles, nm; ni is the number
of nanoparticles with a diameter di; n is the total number of nano-
particles. The methodology of analytical measurements was described
elsewhere [37].
The diffuse reflectance IR spectra of adsorbed CO (DRIFTSꢀ 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, UK) under flow conditions. The reduced
catalyst (20 mg) was placed in a thermostatically controlled cell with
CaF2 windows and heated to 500 ◦C for 1 h in Ar flow (30 mL/min).
Thereafter, it was reduced at 550 ◦C for 1 h in a 5 % H2/Ar (30 mL/min)
flow and cooled first in 5% H2/Ar (30 mL/min) flow to 300 ◦C and then
in Ar (30 mL/min) flow to 50 ◦C. The background spectrum was recor-
ded under Ar flow (30 mL/min) at 50 ◦C. The spectra of adsorbed CO
were collected at 50 ◦C in a 0.5 vol. % CO/He flow for 10 min (30 mL/
min; 250 scans; resolution, 4 cm–1). Fig. 2 shows FTIRꢀ CO spectra ob-
tained after 10 min of CO exposure.
The main idea of this investigation was to study the kinetics of se-
lective alkyne hydrogenation on monometallic Pd and PdAg SAA cata-
lysts and to develop kinetic models, adequately describing experimental
data. Selective liquid-phase hydrogenation of diphenylacetylene (DPA)
previously studied in detail [30–32], was used as a model reaction. A
significant advantage of this reaction is a possibility to carry it out at
room temperature, therefore minimizing the effect of the adsorbate
induced surface segregation on the structure of surface active sites.
Otherwise an increase in the reaction temperature enhances mobility of
Pd atoms, which, in turn, facilitates the formation of multiatomic Pdn
sites (n ≥ 2) upon surface segregation of Pd. The latter, induced by
adsorption of CO or alkyne, decreases the desirable selectivity to the
alkene [33,34].
Catalytic tests
Commercially available diphenylacetylene (98 %, Sigma-Aldrich)
was used as a substrate, being additionally purified by distillation
under Ar atmosphere. The solvent was n-hexane (98 %, Merck,
Germany).
For preparation of PdAg SAA catalyst we use the methodology
developed in our recent study of the formation of the structure of Pd1
sites isolated by Ag atoms on the surface of PdAg bimetallic nano-
particles [35]. It was found that a stable surface structure of single-atom
Pd1 sites can be attained at Ag/Pd ratio of 2/1 or above. These data are
in agreement with the theoretical calculations, demonstrating that a
significant excess of Ag is a prerequisite for the formation of Pd1 isolated
sites [36]. In order to ensure formation of SAA structure in this study the
Ag/Pd ratio was increased to ca. 3/1.
The liquid-phase hydrogenation was performed as described in de-
tails elsewhere [31]. Typical catalytic tests were carried out at 25 ◦C,
initial hydrogen pressure of 5 bar, and 1000 rpm stirring.
External mass-transfer limitations were avoided on the basis of the
experiments with different stirring rates [38]. It was shown that at
stirring rates exceeding 600 rpm identical activity and selectivity were
obtained. To minimize the internal mass-transfer limitations the catalyst
was finely grinded to obtain a powder with a particle size below 10
as suggested in [39].
μm
Calculations of the Weisz-Prater criterion [40] for the highest reac-
tion rate considering hydrogen concentration in the solvent and effec-
tive diffusion coefficient of hydrogen, a porosity to tortuosity ratio of
0.1, gave the values of the Weisz-Prater criterion below 0.02, pointing
out that the internal mass transfer limitations can be ruled out. Experi-
ments with different catalyst amounts demonstrated that the gas-liquid
mass transfer was not the limiting factor.
Experimental
Catalyst preparation
First, alumina powder (98.9 %, PURALOX 200/55, Sasol, Germany;
SBET = 56 m2/g) was calcined in dry air flow at 500 ◦C for 4 h and then
was impregnated by the incipient-wetness method with solutions of Pd
(NO3)2 (99.9 %, Sigma-Aldrich) and Ag(NO3)2 (≥ 99.0 %, Sigma-
Aldrich). After impregnation sample was dried at room temperature
overnight and calcined in dry air flow (300 mL/min) at 550 ◦C for 4 h.
To obtain the Pd1Ag3/Al2O3 catalyst the parent material was reduced in
5%H2/Ar flow at 550 ◦C for 3 h, cooled down in 5 % H2/Ar to 200 ◦C,
and then in N2 (99.999 %) flow to room temperature. According to ICP-
AES analysis, the metal content was 2.0 wt% Pd and 5.97 wt %Ag
(designated as Pd1Ag3). The reference 0.5 wt. % Pd/Al2O3 catalyst was
The liquid probes were taken at regular intervals and analyzed by gas
chromatography (GC) using a Crystal 5000 instrument (Chromatek,
Russia) equipped with a flame-ionization detector and an HP5-MS col-
umn (5% phenyldimethylsiloxane; 30 m, 0.25 mm I.D., 0.25
thickness).
μm film
2