Alkyne hydrogenation using Pd-Ag hybrid nanocatalysts in surface-immobilized dendrimers

Article abstract of DOI:10.1002/aoc.3367

A series of Pd-Ag mixed-metal nanocatalysts were prepared by reduction of Pd-Ag salts in the presence of poly(propylene imine) dendrimers, which were covalently bound to the surface of a silica polyamine composite, BP-1 (polyallylamine covalently bound to a silanized amorphous silica gel). Three different Pd-to-Ag ratios were evaluated (50:50, catalyst 1; 40:60, catalyst 2; 60:40, catalyst 3) with the goal of determining how the amount of Ag effects selectivity, rate and conversion in the selective reduction of alkynes, such as phenylacetylene and 1- or 4-octyne, to the corresponding alkenes. Conditions for the catalysis are reported where there is improved selectivity without a serious reduction in rate when compared with the analogous Pd-only catalysts. Catalyst 2 worked best for phenylacetylene and catalyst 3 worked best for the octynes. The catalysts could be reused seven times without loss of activity.

Full text of DOI:10.1002/aoc.3367

Full paper
Received: 1 July 2015
Accepted: 21 July 2015
Published online in Wiley Online Library: 11 September 2015
(wileyonlinelibrary.com) DOI 10.1002/aoc.3367
Alkyne hydrogenation using Pd–Ag hybrid
nanocatalysts in surface-immobilized dendrimers
Edward A. Karakhanova*, Anton L. Maximova,b, Anna V. Zolotukhinaa,
Nadezhda Yatmanovaa and Edward Rosenbergc*
A series of Pd–Ag mixed-metal nanocatalysts were prepared by reduction of Pd–Ag salts in the presence of poly(propylene imine)
dendrimers, which were covalently bound to the surface of a silica polyamine composite, BP-1 (polyallylamine covalently bound to
a silanized amorphous silica gel). Three different Pd-to-Ag ratios were evaluated (50:50, catalyst 1; 40:60, catalyst 2; 60:40, catalyst
3) with the goal of determining how the amount of Ag effects selectivity, rate and conversion in the selective reduction of alkynes,
such as phenylacetylene and 1- or 4-octyne, to the corresponding alkenes. Conditions for the catalysis are reported where there is
improved selectivity without a serious reduction in rate when compared with the analogous Pd-only catalysts. Catalyst 2 worked
best for phenylacetylene and catalyst 3 worked best for the octynes. The catalysts could be reused seven times without loss of
activity. Copyright © 2015 John Wiley & Sons, Ltd.
Additional supporting information may be found in the online version of this article at the publisher’s web site.
Keywords: hydrogenation; palladium; silver; dendrimer; polyamine composite
We have recently reported hydrogenations, catalyzed by Pd
nanoparticles, which are stabilized by poly(propylene imine)
Introduction
Catalytic hydrogenation is one of the most important processes in
the modern petrochemical industry. Selective hydrogenation of
the acetylenic and dienic fragments in C2–C5 hydrocarbons to their
corresponding olefins avoids further polymerization and oxidation
of these petroleum components.[1,2] The selective hydrogenation
of phenylacetylene, which poisons polymerization catalysts, is more
efficient than its separation from the feedstocks.[1,3,4]
The traditional heterogeneous catalysts used in industry fre-
quently require vigorous reaction conditions and do not always
have the ability to selectively reduce alkynes to alkenes, resulting
in the necessity for additional separation processing. Homoge-
neous catalysts, which have high activity and selectivity, are not
advantageous due to their high cost, low stability and the necessity
to separate them from the reaction products for reuse.
Therefore, the synthesis of hybrid hydrogenation catalysts, which
combine the merits of both homogeneous (high activity and
selectivity) and heterogeneous (stability, recyclability) catalysts, is
of current interest in the petrochemical industry. One approach to
this type of system is the immobilization of metal complexes
and/or metal nanoparticles in soluble polymers, bound to an
inorganic carrier.[2,5–12]
Dendrimers, globular macromolecules with regular hyperbranched
structures, are promising ligands for this goal.[13,14] Dendrimers possess
the following advantages for use in catalysis:[13,15]
(PPI) dendrimers immobilized on silica polyamine composites
(SPCs).[16] This catalytic system proved very active for the selec-
tive hydrogenation of phenylacetylene to styrene and the reduc-
tion of dienes to monoenes. Very high turnover rates were
realized and the catalyst could be reused multiple times without
loss of activity.
The use of bimetallic nanoparticles in catalysis can offer higher
selectivity and, sometimes, activity.[7,17–19] The addition of silver to
a heterogeneous catalyst has been shown to be an effective addi-
tive for the semi-hydrogenation of phenylacetylene and other al-
kynes in the presence of palladium catalysts.[20,21]
In this paper we report the synthesis of catalysts based on
palladium–silver nanoparticles, stabilized by PPI dendrimers, co-
valently grafted to a SPC in order to understand how the addi-
tion of silver would impact the selectivity and activity toward
reduction of alkynes for this already efficient catalyst system. As
in a previous study we used BP-1, a SPC based on polyallylamine
(Fig. 1).
*
Correspondence to: Edward A. Karakhanov, Department of Petroleum Chemistry
and Organic Catalysis, Moscow State University, 119991 Moscow, Russia.
E-mail: kar@petrol.chem.msu.ru Edward Rosenberg, Department of Chemistry
and Biochemistry, University of Montana, Missoula, MT 59812, USA.
E-mail: edward.rosenberg@mso.umt.edu
(1) good control over the quantity of metal sorption;
(2) provide a more even dispersion of particles on the carrier;
(3) better control of metal particle size, providing a material with
a narrow size distribution;
a Department of Petroleum Chemistry and Organic Catalysis, Moscow State
University, 119991 Moscow, Russia
b Institute of Petrochemical Synthesis RAS, 119991 Moscow, Russia
(4) improved resistance to metal leaching; and
(5) allow the tuning of catalyst stability and substrate selectivity
by means of dendrimer terminal group modification.
c
Department of Chemistry and Biochemistry, University of Montana, Missoula MT
59812 USA
Appl. Organometal. Chem. 2015, 29, 777–784
Copyright © 2015 John Wiley & Sons, Ltd.

E. A. Karakhanov et al.
Synthesis of dendrimer-based hybrid materials
The synthesis of dendrimer-modified support DAB(NH2)16@BP-1
was carried out similar to a previously described procedure.[16] In
a 10 ml single-neck round-bottom flask, equipped with a magnetic
stirrer and a reflux condenser, 995 mg of the BP-1 support
(1.6mmolN gꢀ1, particle size range 250–550 μm), 300 μl of 37%
aqueous formaldehyde solution (fivefold excess relative to the
mmol N gꢀ1 of BP-1) and 5 ml of methanol were combined. The
reaction was carried out for 2 h with vigorous stirring, the product
swelling to form a gel. The imine derivative of BP-1 formed was
combined with the third-generation PPI dendrimer (600 mg,
0.356 mmol) and the reaction was conducted for 24 h at 70°C with
vigorous stirring (the yellow-brown dendrimer solution gradually
discolored). The mixture was evaporated using a rotary evaporator
at 50°C. The product was isolated as a milky-white powder
weighing 1.4 g (88% yield based on the dendrimer used).
XPS (eV): 102.2 (Si 2p, 6.0%); 285.2 (C 1s, 61.1%); 399.0 (N 1s,
16.0%); 531.2 (O 1s, 16.9%). 13C NMR (δ, ppm): 71.0 (HNCH2OH,
HNCH2NH); 53.5 (br s, NCH2CH2CH2N, (CH2)2CHCH2NHCH2,
NCH2CH2CH2CH2N); 41.6 (br s, NCH2CH2CH2NH2, (CH2)2CHCH2NH2);
28.0 (br s, NCH2CH2CH2NH2, NCH2CH2CH2N); 18 (br s,
NCH2CH2CH2Si); 13 (br s, NCH2CH2CH2Si). 29Si NMR (δ, ppm):
ꢀ65.2 (br, Si(OSi)3(CH2)3); ꢀ97.0 (br, Si(OSi)3OH); ꢀ111 (br s, Si
(OSi)4).
Figure 1. Synthesis of the bimetallic Pd–Ag catalyst.
Experimental
Materials
Reagents: phenylacetylene C6H5C≡CH (98%, Aldrich); styrene
C6H5CH¼CH2 (≥99%, Aldrich); ethylbenzene C6H5CH2CH3 (Reachim,
Purum); 1-octyne (97%, Aldrich); 4-octyne (98%, Aldrich); 1-octene
(Reachim, Purum); and n-octane (Reachim, Purum). Their purity
was monitored by means of gas–liquid chromatography. Solvents
methanol CH3OH (>99%, Acros Organics), ethanol C2H5OH (Irea
2000, Purum p.a.) and n-hexane С6H14 (Chimmed, Purum p.a.) were
used without additional purification.
For the synthesis of the hybrid supports and catalysts the follow-
ing substances were used: silver(I) sulfate Ag2SO4 (99%, Aldrich);
palladium(II) sulfate PdSO4 (98%, Aldrich); and sodium borohydride
NaBH4 (98%, Aldrich). PPI dendrimers of the third generation with a
diaminobutane (DAB) core and 16 terminal amino groups (DAB
(NH2)16)) were synthesized according to a literature procedure.[22]
The SPC material BP-1 was synthesized according to published liter-
ature procedures.[23,24]
Catalyst 1: DAB(NH2)16*Pd50Ag50@BP-1
In a single-neck round-bottom flask (capacity of 50 ml), equipped
with a magnetic stirrer and a reflux condenser, 500 mg of the com-
posite material DAB(NH2)16@BP-1 and 20 ml of water were placed.
Then 95 mg (0.4675 mmol) of PdSO4 and 73 mg (0.234 mmol) of
Ag2SO4 (Pd:Ag mole ratio of 50:50) were added to the resulting sus-
pension with stirring.
Analyses
Chromatographic analysis of the reaction mixtures was done with a
ChromPack CP9001 gas chromatograph equipped with a flame ion-
ization detector and a 30 m × 0.2 mm column containing a grafted
SE-30 phase. The chromatograms were recorded and analyzed with
a computer using the Maestro 1.4 program. Conversion was deter-
mined from the change in the relative area (%) of substrate and
product peaks.
Transmission electron microscopy (TEM) analysis was performed
using a LEO912 AB Omega microscope with an electron tube volt-
age of 100 kV. The particle size distribution was measured by pro-
cessing the obtained microscopy images using the ImageJ
program. High-resolution TEM (HRTEM) analysis was performed
using a JEM-2100 microscope with an electron tube voltage of
200 kV (LaB6 cathode). The d-spacing of the metal nanoparticles
was made by means of fast Fourier transforms of the micrographs
obtained, using the ImageJ program.
X-ray photoelectron spectroscopy (XPS) studies were performed
using a LAS-3000 instrument, equipped with a photoelectronic
analyzer with retarding potential OPX-150. To excite photoelectrons,
aluminum anode X-ray radiation was used (Al Kα = 1486.6 eV) with
a tube voltage of 12 kV and emission current of 20 mA. The calibra-
tion of photoelectron peaks was performed along the C 1s line with
binding energy of 285.0 eV. Deconvolution of palladium and silver
peaks was processed with the PeakFit 4.11 program using the
Gauss–Lorentz–Ampere approximation.
Quantitative determination of palladium and silver in the
samples was performed using atomic emission spectroscopy with
inductively coupled plasma (ICP-AES) using an IRIS Interpid II XPL
instrument (Thermo Electron Corp., USA).
The reaction was carried out for 12 h at 80°C and with vigorous
stirring. The color of the reaction mixture was brown. After reaction,
the suspension was evaporated using a rotary evaporator at 60°C.
The resulting material was a grey-brown powder weighing 667
mg (97% yield, based on the metal compounds used).
XPS (eV): 102.2 (Si 2p, 18.8%); 285.1 (C 1s, 45.9%); 336.2 (PdOx 3d5/2
,
0.5%); 337.1 (PdO 3d5/2, 0.3%); 338.3 (Pd2+ 3d5/2, 0.2%); 340.9
(PdOx 3d3/2, 0.2%); 341.7 (PdO 3d3/2, 0.1%); 342.8 (Pd2+ 3d3/2
,
0.1%); 366.8 (AgIAgIIO2 3d5/2, 0.7%); 368.3 (Ag0 3d5/2, 0.5%); 373.0
(AgIAgIIO2 3d3/2, 0.6%); 368.3 (Ag0 3d3/2, 0.3%); 398.6 (N 1s, 5.2%);
400.4 (Ag+, PdOx←N 1s, 5.1%); 402.8 (Pd2+←N 1s, 4.1%); 531.4
(O 1s, 17.4%). ICP-AES (%): Pd, 4.16; Ag, 0.25.
Reduction of this precursor was performed using 300 mg
suspended in 10 ml of methanol and 5 ml of deionized water in a
single-neck round-bottom flask (capacity of 50 ml), equipped with
a magnetic stirrer and a reflux condenser. Then 160 mg (9.935 mmol)
of sodium borohydride and 5 ml of water were added portion-wise
to the resulting suspension with stirring. The color of the reaction
mixture turned black and vigorous gas evolution was observed.
The reaction was carried out for 12 h at 80°C with stirring. After
the reaction, the resulting solid mixture was washed with water
and methanol to remove sodium borate salts, and the residue
was isolated by centrifugation and dried. The resulting material
was a black powder weighing 268 mg (97% yield, based on
reduction of the metal in the precursor).
XPS: 102.9 (Si 2p, 26.1%); 285.1 (C 1s, 24.1%); 334.7 (Pd0 3d5/2
0.5%); 336.1 (PdOx 3d5/2, 0.4%); 337.8 (Pd2+ 3d5/2, 0.3%); 339.5
,
(Pd0 3d3/2, 0.4%); 340.5 (PdOx 3d3/2, 0.3%); 342.4 (Pd2+ 3d3/2
,
0.1%); 368.1 (Ag0 3d5/2, 0.6%); 369.8 (AgIAgIIO2, sat 3d5/2, 0.5%);
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 777–784

Alkyne hydrogenation with Pd–Ag hybrid nanocatalysts
371.2 (Ag2O, sat 3d5/2, 0.3%); 373.9 (Ag0 3d3/2, 0.3%); 375.1
(AgIAgIIO2, sat 3d3/2, 0.2%); 376.4 (Ag2O, sat 3d3/2, 0.2%); 399.6 (N
1s, 4.8%); 401.1 (Ag+, PdOx←N 1s, 6.6%); 402.5 (Pd2+←N 1s, 7.7%);
532.9 (O 1s, 26.6%). ICP-AES (%): Pd, 6.21; Ag, 0.49.
temperature and depressurized. The reaction products were sepa-
rated by decantation and analyzed using gas–liquid chromatogra-
phy. The catalyst located in vitro was used without additional
loading, drying or purification in subsequent reaction cycles.
The turnover frequency in each catalytic reaction is reported as
specific activity (SA) and was calculated as the ratio of the reacted
substrate to the metal loading, expressed in moles per unit time:
Catalyst 2: DAB(NH2)16*Pd40Ag60@BP-1
The synthesis was carried out according to the procedures de-
scribed above. Amounts of 200 mg of DAB(NH2)16@BP-1 support,
30 mg (0.1496 mmol) of PdSO4 and 35 mg (0.1122 mmol) of Ag2SO4
were combined in 10 ml of deionized water providing a nominal
Pd:Ag ratio of 40:60. The resulting catalyst precursor was grey-
brown powder weighing 263 mg (99% yield based on the metal
compounds used).
Conversionꢁvsubstr
SA ¼
(1)
vPdꢁt
Results and discussion
The reduction of this precursor was carried out using 263 mg
with 142 mg (3.74 mmol) of sodium borohydride in a mixture of 5
ml of deionized water and 5 ml of methanol. The resulting species
was a black powder weighing 202 mg (84% yield based on reduc-
tion of the metal in the precursor).
Synthesis and characterization of hybrid catalysts
Immobilization of the dendrimers on the SPC BP-1 surface was car-
ried out using the Mannich reaction, with formaldehyde as the
linking agent.[16] The latter was added in fivefold excess relative
to the available amino groups on BP-1 based on N analysis. The
dendrimer was added to the reaction mixture 2 h later to minimize
crosslinking between the dendrimer molecules.
The composition and structure of the synthesized hybrid mate-
rials were elucidated by means of XPS, TEM and solid-state 13C
NMR and 29Si NMR spectroscopy.
XPS (eV): 102.6 (Si 2p, 31.9%); 285.5 (C 1s, 45.9%); 367.6 (Ag2O
3d5/2, 0.3%); 369.0 (AgIAgIIO2, sat 3d5/2, 0.8%); 372.4 (Ag2O 3d3/2
,
0.2%); 373.9 (AgIAgIIO2, sat 3d3/2, 0.4%); 398.0 (N 1s, 1.8%); 400.2
(Ag+, PdOx←N 1s, 4.6%); 402.2 (Pd2+←N 1s, 1.9%); 532.8 (O 1s,
30.2%). ICP-AES (%): Pd, 4.34; Ag, 0.44.
Catalyst 3: DAB(NH2)16*Pd60Ag40@BP-1
The XPS spectrum of the modified support shows the presence of
(CH2)3N fragments, which are typical for dendritic and poly-
allylamine moieties, and the presence of Si–O and Si–C bonds.[25,26]
In comparison with the original BP-1 material[23,24] there is a de-
crease in surface oxygen and silicon content by a factor of two (6.0
and 16.9%, respectively, versus 14.7 and 34.6%), while the atomic
concentrations of carbon and nitrogen increase by approximately
50% (61.1 and 16.0%, respectively, versus 40.2 and 10.5%), indicating
the dendrimers are successfully bound to the BP-1 surface.
The synthesis was carried out according to the procedures described
above. Amounts of 1000 mg of DAB(NH2)16@BP-1 support, 227 mg
(1.122 mmol) of PdSO4 and 117 mg (0.374 mmol) of Ag2SO4 in 25
ml of deionized water provided a nominal Pd:Ag ratio of 60:40. The
resulting catalyst precursor was a black-brown powder weighing
1273 mg (95% yield based on the metal compounds used).
The reduction of this precursor was carried out using 1273 mg
with 711 mg (18.7 mmol) of sodium borohydride in a mixture of
10 ml of deionized water and 15 ml of methanol. The resulting spe-
cies was a black powder weighing 955 mg (74% yield based on re-
duction of the metal in the precursor).
The 13C NMR spectrum of the dendrimer-modified DAB(NH2)
16@BP-1 carrier (Fig. S1) contains the following signals: 13–20
ppm, corresponding to the alkyl chains that are directly bound to
silicon; 28 ppm, corresponding to the β-CH2 groups in the
polyalkylamine fragments; 41.6 ppm, typical for α-CH2 groups at
primary amino groups both in dendrimers and in polyalkylamine
fragments of the original BP-1 support; 53.5 ppm, assigned to the
α-CH2 groups at secondary and tertiary amino groups of the
polyalkylamine fragments.[27–29] The signals at 71 ppm can be assigned
to the methylene groups of the polymer attached to the N=CH2,
groups fragments, which are formed due to the interaction of
formaldehyde with the surface amino groups of the BP-1 carrier.
TEM revealed areas with two different types of structure in DAB
(NH2)16@BP-1. The first type is homogeneous and contains cells
and channels of 3 nm in diameter, which are characteristic of micro-
porous materials (Fig. S2). This corresponds to the silica. The second
type of structure consists of dark sphere-like formations of 15–20
nm in diameter. This corresponds to the dendrimer-modified areas
of the BP-1 carrier. The small spherical ‘particles’ of 1.5–1.7 nm in
diameter can be assigned to the dendrimers themselves.
To investigate the effect of the addition of silver on the activity
and selectivity of the synthesized hybrid catalysts, we prepared
samples with nominal Pd:Ag ratios of 50:50 (1), 40:60 (2) and 60:40
(3). The synthesis of the bimetallic Pd–Ag nanoparticles was carried
out using the co-complexation method, with aqueous solutions of
PdSO4 and Ag2SO4 followed by reduction with NaBH4 (Fig. 1).[30]
All the materials obtained were characterized using ICP-AES, XPS
and TEM. For the catalyst with Pd-to-Ag ratio of 50:50, the
unreduced precursor was also characterized. Physicochemical
XPS (eV): 103.2 (Si 2p, 2.8%); 285.1 (C 1s, 24.8%); 335.9 (PdOx 3d5/2
0.4%); 337.1 (PdO 3d5/2, 0.2%); 338.3 (Pd2+ 3d5/2, 0.1%); 340.5
(PdOx 3d3/2, 0.4%); 341.7 (PdO 3d3/2, 0.2%); 342.9 (Pd2+ 3d3/2
,
,
0.1%); 367.1 (AgIAgIIO2 3d5/2, 0.6%); 368.4 (AgI/Ag2O2 3d5/2, 0.3%);
369.6 (AgIAgIIO2, sat 3d5/2, 0.2%); 372.9 (AgIAgIIO2 3d3/2, 0.4%);
374.1 (AgI/Ag2O2 3d3/2, 0.3%); 375.8 (AgIAgIIO2, sat 3d5/2, 0.2%);
399.6 (N 1s, 7.5%); 401.4 (Ag+, PdOx←N 1s, 4.6%); 402.8 (Pd2+←N
1s, 2.7%); 532.7 (O 1s, 35.3%). ICP-AES (%): Pd, 6.96; Ag, 0.22.
Protocol for catalytic experiments
The desired amounts of catalyst and substrate were placed into a
thermostatically controlled steel autoclave, equipped with a glass
test-tube insert and a magnetic stirrer. The autoclave was sealed,
pressurized with hydrogen up 1 or 3 MPa and connected to the
thermostat. The reaction was carried out at 80°C for 1 h or 15
min. The reactor was then cooled below room temperature and
depressurized. The reaction products were analyzed using gas–
liquid chromatography.
Reuse of catalyst 3 was carried out by hydrogenation of 1-octyne.
For this purpose, 1 mg of catalyst 3 and 250 μl of 1-octyne were
placed in a thermostatically controlled stainless steel autoclave,
equipped with a glass test tube-insert and a magnetic stirrer. The
autoclave was sealed, pressurized with hydrogen up to 1 MPa
and connected to the thermostat. The reaction was conducted at
80°C for 15 min. Then, the reactor was cooled below room
Appl. Organometal. Chem. 2015, 29, 777–784
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

E. A. Karakhanov et al.
characteristics of the catalysts synthesized are summarized in the
Tables 1 and S1.
According to deconvolution of the XPS data (Table 1; Fig. 2), silver
prevails on the surface of all samples. The surface palladium
content in catalyst 2 is apparently below the instrument detection
limit, i.e. below 0.5%. Therefore, in spite of the co-complexation
method used, we obtained so-called ‘core–shell’ catalysts, in which
one of the metals (palladium) is predominantly located in the core
of each particle, and the other metal (silver) is located on its surface,
in agreement with previous reports.[21]
In the unreduced catalyst, both palladium and silver are present
in both zero-valent and oxidized states (Table S2; Fig. S3). The pres-
ence of zero-valent palladium in the spectrum of the unreduced
catalyst can be explained in terms of electron transfer from the
dendritic amino groups to the bimetallic palladium–silver clusters,
resulting in enamine formation.[31]
It should be noted that the interpretation of the silver spectra is
complicated due to overlapping of the binding energy signals that
are typical of each of three common valence states of silver: Ag0
(367.9–368.4 eV), Ag2O (367.6–368.5 eV) and AgIAgIIIO2 (367.3–
368.1 eV).[32–35] The silver oxide signals are negatively shifted due
to decreasing binding energy relative to the Ag0 state.[36,37]
Reduction with sodium borohydride results in an increase of the
zero-valent forms of silver and, especially, palladium. Thus, the Pd0/
(PdO + Pd2+) ratio in catalyst 1 is approximately 3, whereas in the
unreduced precursor this ratio does not exceed 1 (Table S1).
The prevalence of the oxidized forms of palladium in catalyst 3
(Table S1) can be explained in terms of the sample’s facile oxidation
on storage under ambient conditions.[38] The predominance of
surface silver oxides in catalysts 2 and 3 (Table S2) can be explained
in the same way. All the silver in catalyst 2 is apparently located on
the nanoparticle surfaces, and results in its occurrence only as
oxides (Table S1). This explains the absence of palladium in the
spectrum of catalyst 2.
TEM reveals that both large particles of 5–15 nm in diameter and
small particles of 2–4 nm in diameter are present in the sample of
the unreduced catalyst (Fig. 3(a)), but the smaller particles predom-
inate, accounting for up to 99% of the total particles. Reduction by
sodium borohydride of catalyst 1 results in a slight decrease in the
number of small particles (down to 96%) and a shift of the large par-
ticles to a maximum of 4–6 nm (Fig. 3(b)). On the evidence of the
data presented, we suggest that small particles are formed
predominantly inside the dendrimer branches, while large particles
grow in the areas with low grafted dendrimer density.
Figure 2. Deconvoluted XPS spectra: (a) catalyst 1 (50:50 Pd/Ag); (b)
catalyst 2 (40:60 Pd/Ag); (c) catalyst 3 (60:40 Pd/Ag). The 3d silver spectra
are left and the 3d palladium spectra are right.
to the fast Fourier transform data, obtained after micrograph
processing, the sample contains palladium, silver and palladium–silver
nanoparticles, where the silver is dominant and both metals are
predominantly in the oxidized states, in agreement with the XPS data.
Hydrogenation of unsaturated compounds in the presence of
hybrid bimetallic catalysts
Phenylacetylene hydrogenation
The activity of the catalysts synthesized was investigated for
phenylacetylene hydrogenation and calculated in terms of the ratio
of the substrate reacted to the metal loading, expressed in moles
per unit time, according to equation (1).
The results for catalyst 1 (Table 2) show that the presence of silver
significantly increases the selectivity for styrene, while reducing
activity in comparison with analogous palladium-only hybrid
catalysts.[16] Thus, the selectivity for styrene is 97% at 15 min and
a substrate-to-Pd ratio of 7780, but the conversion is only 40%
(Table 2, entry 3). This indicates that the decreased activity of the
Changing the Pd-to-Ag ratio does not affect the mean diameter
and distribution curve of the particles (Figs. 3(c) and (d)). However,
increasing the silver content in catalyst 2 increases the average
diameter of the larger particles to 4–8 nm (Fig. 3(c)).
Catalyst 3, with a nominal Pd-to-Ag ratio of 60:40, was also charac-
terized by means of HRTEM, which reveals both large (6–10 nm in
diameter) and small (2–4 nm in diameter) particles (Fig. 4). According
Table 1. Physicochemical characteristics of synthesized catalysts
SSample
Catalyst
Pd
(mass
%)
Ag
(mass
%)
d (nm)
XPS (at.%)
N
Si
С
O
Pd
Ag
Not reduced DAB(NH2)16*Pd50Ag50@BP-1
4.16
6.21
4.34
6.96
0.25
0.49
0.44
0.22
2.5 (99%); 10.8 (1%)
2.5 (96.5%); 5.1 (3.5%)
2.4 (96%); 6.9 (4%)
18.8
26.1
31.9
2.8
45.9
24.1
27.9
24.8
14.4
19.1
8.3
17.4
26.6
30.2
35.3
1.4
2.0
-
2.1
2.1
1.7
2.0
1
2
3
DAB(NH2)16*Pd50Ag50@BP-1
DAB(NH2)16*Pd40Ag60@BP-1
DAB(NH2)16*Pd60Ag40@BP-1
2.5 (99.5%); 9.2 (0.5%)
14.8
1.4
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 777–784

Alkyne hydrogenation with Pd–Ag hybrid nanocatalysts
Figure 4. HRTEM image of catalyst 3.
Table 2. Phenylacetylene hydrogenation (at 80°C) in the presence of
catalyst 1
Substrate/Pd Conversion SA (hꢀ1
)
Selectivity
a
Entry
t
P
(min) (MPa) (mol/mol)
(%)
to styrene (%)
1
2
3
4
5
60
15
15
15
15
1
1
1
3
3
3 890
3 890
7 780
7 780
15 565
100
100
40
3 890
15 560
12 450
30 505
26 145
71
86
97
91
98
98
42
aSpecific activity calculated from equation (1).
5565 the conversion is close to quantitative after 90 min with a se-
lectivity of 96% (Table 3, entry 2). Further decreasing the reaction
time and increasing the substrate-to-Pd ratio lead to predictable
decreases in the reaction yield, but the selectivity is still 96–98%.
Catalyst 3 appears to be more active than catalyst 2, but less
active than catalyst 1. Also, it is significantly less selective than all
the other catalysts (Table 4). We suggest that this behavior can be
explained in terms of both the lower surface palladium content in
comparison with silver and the palladium oxidized states being pre-
dominant. The latter results in a change in the adsorption proper-
ties of the catalyst; that is, the electron-deficient palladium oxide
areas of the catalyst nanoparticles will more easily adsorb styrene
as it is formed than areas containing palladium in reduced states
(catalyst 1) or of lower palladium content (catalyst 2).
Figure 3. TEM images and particle size distributions of the synthesized
hybrid Pd–Ag catalysts.
synthesized Pd–Ag catalyst is due to the location of the catalytically
active palladium atoms predominantly in the core of the particles,
while the inactive silver atoms dominate on the surface.[20,21] In-
creasing the hydrogen pressure to 3 MPa results in an increase of
the conversion (from 40 to 98%) while only slightly decreasing
the selectivity (from 98 to 91%; Table 2, entry 4). We consider
these reaction conditions to be the best for phenylacetylene
semi-hydrogenation using catalyst 1. Further increases in the
substrate-to-Pd ratio lead to a decrease in conversion and increase
in selectivity again. When starting with the unreduced precursor of
1, the resulting catalyst is significantly less active, but more
selective, than reduced catalyst 2 (Table S2).
Table 3. Phenylacetylene hydrogenation (at 80°C) in the presence of
catalyst 2
a
Entry
t
P
Substrate/Pd Conversion SA (hꢀ1
)
Selectivity
(min) (MPa) (mol/mol)
(%)
to styrene (%)
1
2
3
4
5
6
120
90
60
15
15
60
1
1
1
1
1
3
5 565
5 565
100
97
83
34
17
35
2 780
3 600
4 620
7 570
7 570
15 590
75
96
97
97
98
96
5 565
5 565
11 135
11 135
Catalyst 2 does not contain Pd on its surface, as the XPS data
reveal, and this is manifested in its much lower activity for
phenylacetylene hydrogenation. Thus, at a substrate-to-Pd ratio of
aSpecific activity calculated from equation (1).
Appl. Organometal. Chem. 2015, 29, 777–784
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

E. A. Karakhanov et al.
Table 4. Phenylacetylene hydrogenation (at 80°C) in the presence of
catalyst 3
of 54% for 15 min, with the selectivity to 1-octene decreasing from
93 to 88% (Table 5, entry 6).
a
The optimal conditions for hydrogenation of 1-octyne to 1-octene
with catalyst 3 are found to be a substrate-to-Pd ratio of 6940, a
hydrogen pressure of 1 MPa and a reaction time of 15 min. This
results in a conversion of 96% (SA = 26 660 hꢀ1), a total selectivity
for alkenes of 86% and selectivity for 1-octene of 77% (Table 5,
entry 3). These values for selectivity are slightly inferior to those
obtained at a substrate-to-Pd ratio of 13 885 and a hydrogen pressure
of 3 MPa (SA = 29 995 hꢀ1, total selectivity to alkenes of 91% and
selectivity to 1-octene of 88%, but a lower conversion of only 54%).
The higher activity of catalyst 3 in the hydrogenation of 1-octyne
as compared with phenylacetylene can probably be explained in
terms of electron transfer from palladium to silver, as XPS results
show (Table S2). On the other hand, dendrimer-based catalysts,
containing only palladium, exhibit, as a rule, higher activity in the
hydrogenation of conjugated double bonds.[16,39,40] We suggest
this phenomenon can have two possible causes. The first one is that
due to the silver prevailing on the surface of catalyst nanoparticles,
there are left only a few sites favoring the plane-parallel adsorption
of phenylacetylene and its subsequent hydrogenation. The second
is that the C≡C triple bond in the molecule of phenylacetylene is
more electron-deficient due to the weak ꢀM-effect of the aromatic
ring. In contrast, the alkyl substituent in the 1-octyne molecule pro-
vides a weak +I-effect. Therefore, its adsorption on the electron-
deficient palladium surface is more favorable in comparison with
phenylacetylene.
Hydrogenation of 4-octyne can proceed under both kinetic and
thermodynamic control (Scheme 2; Fig. 5). Due to the facile isomer-
ization of С¼С bond in the presence of palladium catalysts, the ste-
rically less hindered 2- and 3-octenes are formed at first as well as a
small proportion of 1-octene. The order of their formation agrees
with the remoteness of the newly formed С¼С bond in the alkene
from the С≡С triple bond in the initial 4-octyne. The higher concen-
tration of the isomeric octenes is observed at a reaction time of 30
min, i.e. in the middle of the reaction process (Fig. 5). Subsequently,
the reversal of isomerization takes place, and the thermodynami-
cally more stable 4-octene is reformed. Hence, at a conversion of
100% for 60 min, the selectivity for 4-octene reaches 98%.
It should also be noted that due to the internal position of
the С≡С bond in 4-octyne, its hydrogenation proceeds more
slowly than that of 1-octyne. The specific activity for 4-octyne is
6940 hꢀ1 at a substrate-to-Pd ratio of 3470. The further hydrogenation
of 4-octyne at larger substrate-to-Pd ratios was not carried out,
because the conversion for 15 min was only 49%.
Entry
t
P
Substrate/Pd Conversion SA (hꢀ1
)
Selectivity
to styrene (%)
(min) (MPa) (mol/mol)
(%)
1
2
3
4
60
15
15
15
1
1
1
1
3470
3470
6940
6940
99
89
40
77
3 435
12 360
11 110
21 385
50
90
91
92
aSpecific activity calculated from equation (1).
It should be noted that catalyst 3 appears to be much more
resistant to air oxidation on storage. Therefore, it was used in further
experiments for the hydrogenation of 1-octyne and 4-octyne.
Octyne hydrogenation
The hydrogenation of 1-octyne proceeds with the formation of
1-octene, octane and isomeric octenes, the latter being formed by
palladium-catalyzed proton migration along the carbon chain
(Scheme 1). Also, due to the sterically hindered dendritic matrix,
grafted to the SPC surface, octane formation is possible only from
1-octene with its terminal double bond.
Conversion of 1-octyne to mainly 1-octene (68%) is observed at a
reaction time of 15 min and a substrate-to-Pd ratio of 3470 (Table 5,
entry 2). Increasing the reaction time to 60 min results in further
hydrogenation of 1-octene to octane, whose yield reaches 60%,
and an increased proportion of isomerized alkenes. Increasing the
substrate-to-Pd ratio leads to a gradual decrease in conversion and
a simultaneous increase in the selectivity for 1-octene. Thus, conver-
sion to 1-octene is 31% and selectivity is 93% (SA = 17 220 hꢀ1) for
15 min at substrate-to-Pd ratio of 13 885. Increasing the reaction
time to 60 min at this substrate-to-Pd ratio yields 100% conversion,
but the proportion of 1-octene among the reaction products de-
creases to 52%. Conducting the experiment at the same substrate-
to-Pd ratio with a hydrogen pressure of 3 MPa results in a conversion
Catalyst recycling
Recycling of catalyst 3 was carried out for the hydrogenation of 1-
octyne. The protocol for the experimental procedure was similar
to that for the standard hydrogenation procedure with the follow-
ing difference. At the end of the each reaction cycle the products
were separated by decantation, whereas the catalyst was kept in
the glass test-tube and reused without additional loading and re-
generation. Hexane was used as a solvent for the better resettling
of the dendrimer-based catalysts.[6,41]
The results of the recycling test are shown in Fig. 6. The total turn-
over number for seven reaction cycles is 24 300. A slight coloring of
the reaction mixture is observed after the first cycle, possibly
indicating the formation of 1-octyne complexes with leached
palladium in solution.[42–44] However, this decreases from cycle to
cycle, and a decrease in the conversion is not observed suggesting
Scheme 1. Hydrogenation of 1-octyne in the presence of Pd-containing
catalysts.
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 777–784

Alkyne hydrogenation with Pd–Ag hybrid nanocatalysts
Table 5. Hydrogenation of 1-octyne (at 80°C) in the presence of catalyst 3
SA (hꢀ1
)
Products
a
Entry
1
t (min)
P (MPa)
Substrate/Pd (mol/mol)
3 470
Conversion (%)
100
60
1
3 470
13 885
26 660
13 885
1-Octene, 2%; 4-octene, 12%;
other octenes, 26%; octane, 60%
1-Octene, 68%; 4-octene 6%;
other octenes, 8%; octane, 18%
1-Octene, 77%; 4-octene, 3%;
other octenes, 6%; octane, 14%
1-Octene, 52%; 4-octene, 7%;
other octenes, 12%; octane, 29%
1-Octene, 93%; octane, 7%
2
3
4
15
15
60
1
1
1
3 470
6 940
100
96
13 885
100
5
6
15
15
1
3
13 885
13 885
31
54
17 220
29 995
1-Octene, 88%; 4-octene, 1%;
other octenes, 2%; octane, 9%
aSpecific activity calculated from equation (1).
Figure 6. Recycling of catalyst 3 in the hydrogenation of 1-octyne.
(Reaction conditions: 80°С, 1 MPa of H2, 15 min, substrate/Pd ≈ 3470.)
that active metal leaching decreases after the first cycle.[45] Hence,
the hybrid materials, based on the dendrimer-encapsulated
Pd–Ag nanoparticles, can be reused without any loss of activity.
Scheme 2. Hydrogenation of 4-octyne in the presence of Pd-containing
catalysts.
Conclusions
Two important conclusions can be drawn from this brief investiga-
tion. First, the XPS studies clearly demonstrate that the addition of
silver to the hybrid palladium catalysts results in the formation of
core–shell nanoparticles where silver dominates the outer shell.
This leads to the second conclusion that the addition of silver miti-
gates catalyst activity and, as a result, improves selectivity for reduc-
tion of alkynes to alkenes due to the decreased palladium surface
sites. However, by careful adjustment of hydrogenation conditions,
good conversion of alkyne to alkene can be achieved with a reason-
able selectivity. The arylalkyne phenylacetylene showed distinctly
different behavior from that of 1-octyne and 4-octyne due to their
different electronic properties (ꢀM versus +I, respectively). Finally,
the Pd–Ag hybrid catalyst can be reused without loss of activity.
Acknowledgements
This work was supported by RFBR grant no. 14-03-31531. We also
acknowledge the National Science Foundation (CHE-1049569) for
Figure 5. Kinetics of 4-octyne hydrogenation in the presence of catalyst 3.
(Reaction conditions: 80°С, 1 MPa of H2, substrate/Pd ≈ 3470.)
Appl. Organometal. Chem. 2015, 29, 777–784
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

E. A. Karakhanov et al.
[23] M. A. Hughes, D. Nielsen, E. Rosenberg, R. Gobetto, A. Viale, S. D. Burton,
J. Ferel, Ind. Eng. Chem. Res. 2006, 45, 6538–6547.
[24] M. A. Hughes, E. Rosenberg, Sep. Sci. Technol. 2007, 42, 261–283.
[25] A. Tóth, I. Bertóti, V. S. Khotinsky, Surf. Interface Anal. 1994, 22, 551–555.
[26] P. Laoharojanaphand, T. J. Lin, J. O. Stoffer, J. Appl. Polym. Sci. 1990, 40,
369–384.
support of this research. Also, we thank Natalia Tabachkova
(National University of Science and Technology ‘MISIS’, Moscow,
Russia) for HRTEM facilities.
[27] G. R. Krishnan, K. Sreekumar, Soft Matter 2010, 8, 114–129.
[28] B. González, M. Colilla, C. López de Laorden, M. A. Vallet-Regí, J. Mater.
Chem. 2009, 19, 9012–9024.
[29] A. Papp, K. Miklós, P. Forgo, Á. Molnár, J. Mol. Catal. A 2005, 229, 107–116.
[30] E. A. Karakhanov, A. L. Maximov, A. V. Zolotukhina, T. Y. Filippova,
S. V. Kardashev, Pet. Chem. 2012, 52, 323–332.
References
[1] T.-B. Lin, T.-C. Chou, Appl. Catal. A 1994, 108, 7–19.
[2] J. Allen, E. Rosenberg, E. Karakhanov, S. V. Kardashev, A. Maximov,
A. Zolotukhina, Appl. Organomet. Chem. 2011, 25, 245–254.
[3] S. Huh, R. Touroude, J. Catal. 1988, 114, 411–421.
[4] I. M. Zhvanetskii, F. D. Klebanova, E. A. Katsman, A. S. Berenblyum, Kinet.
Catal. 1990, 31, 777–782.
[5] H. Brunner, P. Panstel, S. Wieland, in Applied Homogeneous Catalysis
with Organometallic Compounds, (Eds.: B. Cornils, W. Hoffmann),
Wiley-VCH, Weinheim, 2002, pp. 646–663.
[6] R. W. J. Scott, O. M. Wilson, R. M. Crooks, J. Phys. Chem. B 2005, 109,
692–704.
[7] Y.-M. Chung, H.-K. Rhee, Korean J. Chem. Eng. 2004, 21, 81–97.
[8] R. Andrés, E. de Jesús, J. C. Flores, New J. Chem. 2007, 31, 1161–1191.
[9] J. M. Thomas, B. F. G. Johnson, R. Raja, G. Sankar, P. A. Midgley, Acc.
Chem. Res. 2003, 36, 20–30.
[10] S. Boujday, J. Blanchard, R. Villanneau, J.-M. Krafft, C. Geantet, C. Louis,
M. Breysse, A. Proust, ChemPhysChem 2007, 8, 2636–2642.
[11] J. Shen, X. Yin, D. Karpuzov, N. Semagina, Catal. Sci. Technol. 2013, 3,
208–221.
[12] S. Miao, Z. Liu, B. Han, J. Huang, Z. Sun, J. Zhang, T. Jiang, Angew. Chem.
Int. Ed. 2006, 45, 266–269.
[13] Y. Niu, R. M. Crooks, C. R. Chim. 2003, 6, 1049–1059.
[14] G. R. Newkome, C. D. Shreiner, Polymer 2008, 49, 1–173.
[15] G. Jayamurugan, C. P. Umesh, N. Jayaraman, J. Mol. Catal. A 2009, 307,
142–148.
[16] E. Karakhanov, A. Maximov, Y. Kardasheva, V. Semernina,
A. Zolotukhina, A. Ivanov, G. Abbott, E. Rosenberg, V. Vinokurov, ACS
Appl. Mater. Interfaces 2014, 6, 8807–8816.
[31] O. P. Parenago, V. M. Frolov, Zh. Vses. Khim. O-va im. D. I. Mendeleeva
1989, 34, 659–665(In Russian.).
[32] C. D. Wagner, A. V. Naumkin, A. Kraut-Vass, J. W. Allison, C. J. Powell,
J. R. Rumble, NIST X-ray Photoelectron Spectroscopy Database, NIST
Standard Reference Database 20, version 3.3, 2003. Web version:
http://www.srdata.nist.gov/xps/ (accessed April 2012).
[33] B. V. R. Chowdari, K. F. Mok, J. M. Xie, R. Gopalakrishnan, J. Non-Cryst.
Solids 1993, 160, 73–81.
[34] P. Ptacek, Z. Bastl, Appl. Surf. Sci. 1990, 45, 319–323.
[35] V. K. Kaushik, J. Electron Spectrosc. Relat. Phenom. 1991, 56, 273–277.
[36] G. Schoen, Acta Chem. Scand. 1973, 27, 2623–2633.
[37] S. W. Gaarenstroom, N. Winograd, J. Chem. Phys. 1977, 67, 3500–3506.
[38] Z. Xiao, Y. Zhao, A. Wang, J. Perumal, D.-P. Kim, Lab Chip 2011, 11, 57–62.
[39] E. Karakhanov, A. Maximov, S. Kardashev, Y. Kardasheva,
A. Zolotukhina, E. Rosenberg, J. Allen, Macromol. Symp. 2011, 304,
55–64.
[40] E. A. Karakhanov, A. L. Maximov, V. A. Skorkin, A. V. Zolotukhina,
A. S. Smerdov, A. Y. Tereshchenko, Pure Appl. Chem. 2009, 81, 2013–2023.
[41] M. Murata, Y. Tanaka, M. Tomoo, K. Ebitani, K. Kaneda, Chem. Lett. 2005,
34, 272–273.
[42] M. Crespo-Quesada, R. R. Dykeman, G. Laurenczy, P. J. Dyson,
L. Kiwi-Minsker, J. Catal. 2011, 279, 66–74.
[43] G. He, Y. Song, X. Kang, S. Chen, Electrochim. Acta 2013, 94, 98–103.
[44] M. Ahlquist, G. Fabrizi, S. Cacchi, P.-O. Norrby, Chem. Commun. 2005,
4196–4198.
[45] W. Long, N. A. Brunelli, S. A. Didas, E. W. Ping, C. W. Jones, ACS Catal.
2013, 3, 1700–1708.
[17] A. R. Vilchis-Nestor, M. Avalos-Borja, S. A. Gómez, A. Hernández José,
A. Olivas, T. A. Zepeda, Appl. Catal., B 2009, 90, 64–73.
[18] R. M. Crooks, M. Zhao, L. Sun, V. Chechik, L. K. Yeung, Acc. Chem. Res.
2001, 34, 181–190.
[19] H. K. Rhee, Y. M. Chung, Catal. Lett. 2003, 85, 159–164.
[20] G. C. Bond, A. F. Rawle, J. Mol. Catal. A 1996, 109.
[21] N. A. Khan, A. Uhl, S. Shaikhutdinov, H. J. Freund, Surf. Sci. 2006, 600,
1849–1853.
[22] E. M. M. de Brabander-van den Berg, E. W. Meijer, Angew. Chem. Int. Ed.
1993, 32, 1308–1312.
Supporting information
Additional supporting information may be found in the online ver-
sion of this article at the publisher’s web site.
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 777–784