Catalytically active nano-porous cobalt-palladium alloys

Article abstract of DOI:10.1016/j.jallcom.2021.161936

The potential advantages of nano-alloys and particularly, nano-porous alloys, place them at the “spotlight” of heterogeneous catalysis. Nevertheless, controlling the precise compositions of these materials is still a synthetic challenge. Previous research introduced the fabrication of metals and alloys with a high nano-scale porosity and controllable compositions, via a hydrogen-mediated chemical reduction process of metal complex salts. We have used this procedure to obtain two magnetic nano-porous Co-Pd alloys, pure porous palladium and pure porous cobalt. Single crystal X-ray diffraction studies enabled structural determination of the two Co-Pd bi-complex salts that were used as precursors for these alloys. Powder X-ray diffraction studies determined the crystalline phases of the alloys and indicated the nanometric size of their crystallites. High-resolution scanning electron microscopy indicated that these alloys assemble as highly porous clusters of interconnected nano-crystallites. It also indicated that each alloy cluster preserves the micrometric morphologies of its salt precursor. Energy dispersive X-ray spectroscopy showed that the alloys exhibit uniform composition down to the micro-level, which preserved the Co/Pd ratio within the salts. Focused ion beam tomography enabled 3D structural representation of the alloys and metals. Geometrical analysis of the 3D reconstructed data determined 90% porosity and a specific surface area of ~100 m²/g for the alloys. In addition, the alloys showed improved catalytic activity in the semi-hydrogenation of phenylacetylene, compared to the pure metals and commercial Pd/C. Moreover, their magnetic properties enabled facile recovery at the end of the reaction. The yield for styrene in this reaction was increased using “design of experiments” (DOE), a method for optimization of reaction conditions. Furthermore, our experiments implied that a highly porous structure significantly improves the selectivity of styrene in the reaction. These results demonstrated the advantage of fabricating nano-porous alloys with uniform compositions that may exhibit special properties and serve as new and efficient catalysts.

Full text of DOI:10.1016/j.jallcom.2021.161936

Journal of Alloys and Compounds 891 (2021) 161936
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Catalytically active nano-porous cobalt-palladium alloys
Shay Avisara, Yahel Shnera, Raed Abu-Reziqa, Inna Popovb, Avi Binoa,⁎
a Institute of Chemistry, The Hebrew University of Jerusalem, The Edmond J. Safra Campus, Givat Ram, 9190401 Jerusalem, Israel
b The Center for Nanoscience and Nanotechnology, The Hebrew University of Jerusalem, The Edmond J. Safra Campus, Givat Ram, 9190401 Jerusalem, Israel
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 23 May 2021
Received in revised form 12 August 2021
Accepted 7 September 2021
Available online 12 September 2021
The potential advantages of nano-alloys and particularly, nano-porous alloys, place them at the "spotlight"
of heterogeneous catalysis. Nevertheless, controlling the precise compositions of these materials is still a
synthetic challenge. Previous research introduced the fabrication of metals and alloys with a high nano-
scale porosity and controllable compositions, via a hydrogen-mediated chemical reduction process of metal
complex salts. We have used this procedure to obtain two magnetic nano-porous Co-Pd alloys, pure porous
palladium and pure porous cobalt. Single crystal X-ray diffraction studies enabled structural determination
of the two Co-Pd bi-complex salts that were used as precursors for these alloys. Powder X-ray diffraction
studies determined the crystalline phases of the alloys and indicated the nanometric size of their crystal-
lites. High-resolution scanning electron microscopy indicated that these alloys assemble as highly porous
clusters of interconnected nano-crystallites. It also indicated that each alloy cluster preserves the micro-
metric morphologies of its salt precursor. Energy dispersive X-ray spectroscopy showed that the alloys
exhibit uniform composition down to the micro-level, which preserved the Co/Pd ratio within the salts.
Focused ion beam tomography enabled 3D structural representation of the alloys and metals. Geometrical
analysis of the 3D reconstructed data determined 90% porosity and a specific surface area of ~100 m2/g for
the alloys. In addition, the alloys showed improved catalytic activity in the semi-hydrogenation of phe-
nylacetylene, compared to the pure metals and commercial Pd/C. Moreover, their magnetic properties
enabled facile recovery at the end of the reaction. The yield for styrene in this reaction was increased using
"design of experiments" (DOE), a method for optimization of reaction conditions. Furthermore, our ex-
periments implied that a highly porous structure significantly improves the selectivity of styrene in the
reaction. These results demonstrated the advantage of fabricating nano-porous alloys with uniform com-
positions that may exhibit special properties and serve as new and efficient catalysts.
Keywords:
Transition metal alloys and compounds
Gas-solid reactions
Catalysis
Vacancy formation
Scanning electron microscopy (SEM)
X-ray diffraction
© 2021 Elsevier B.V. All rights reserved.
1. Introduction
examples of alloys with nanoscale porosity and controllable com-
position are still rare.
Multi-element nano-dimensional and nano-porous metallic
materials are capable of combining synergistic properties of alloys
and nanoscale matter such as high surface area. Therefore, these
materials attract considerable attention in several fields including
heterogeneous catalysis. Many techniques have been developed for
the synthesis of nano-alloys in the gas-phase, in solution, as a sup-
ported phase on a substrate, or in a matrix [1,2].
Destructive techniques such as de-alloying [3,4] or galvanic dis-
placement [5,6] produce metallic materials with nano-scale porosity.
However, these techniques restrict composition control and often
produce structures with diverse compositions [7–12]. Consequently,
A different preparative approach has demonstrated the pre-
paration of numerous alloys with nano-scale porosity through the
chemical reduction of metal complex salts [13–18]. These metallic
materials assemble as highly porous (~90% porosity) nanoparticle
clusters that preserve the morphologies and volumes of their mi-
crocrystalline salt precursors. In addition, they preserve the precise
stoichiometric ratio between the two different metal atoms that
exist in the bi-complex salt precursors and exhibit a uniform spatial
composition down to the sub-micro-level. Some studies have also
shown examples of the catalytic reactivity of these alloys [19–22].
Yet, there is only little knowledge about the correlation between the
structure of these types of alloys and their catalytic reactivity.
Correlating between structure and reactivity of catalysts in a
reaction requires information about the conditions of the reaction in
which each catalyst operates such as the temperature or the reaction
⁎
Corresponding author.
E-mail address: bino@mail.huji.ac.il (A. Bino).
https://doi.org/10.1016/j.jallcom.2021.161936
0925-8388/© 2021 Elsevier B.V. All rights reserved.

S. Avisar, Y. Shner, R. Abu-Reziq et al.
Journal of Alloys and Compounds 891 (2021) 161936
period. A better comparison also requires the optimized conditions
for each catalyst. OVAT ("one variable at a time") is the traditional
approach for this kind of optimization. Nevertheless, this approach
often does not lead to the real optimum and is only valid if there is
no interaction between the variables. Moreover, OVAT approach re-
quires many experiments for gaining a small amount of information
regarding the investigated system. An alternative statistical method
– DOE ("design of experiments"), also noted as "experimental de-
sign", enables collection of precise information from fewer experi-
ments. Particularly, DOE enables to consider the effect of the
interaction between the variables upon the reaction results [23].
Consequently, this approach has recently gained some popularity as
an optimization method, including in heterogeneous cata-
lysis [24–29].
Herein we utilize chemical reduction processes, mentioned
above, in order to produce two cobalt-palladium alloys, pure cobalt,
and pure palladium with nano-scale porosity. The catalytic reactivity
of these alloys was examined using the semi-hydrogenation of
phenyl-acetylene as a model reaction. A specific focus was directed
towards conversion, selectivity, and yield of styrene. DOE method
was applied to optimize the conditions of the reaction for each alloy
catalyst, including hydrogen pressure, temperature, and reaction
periods. Next, the reactivity of these alloys were compared with each
other, with the pure metals and with commercial Pd/C. Finally, a
specific focus was directed at the correlation between these mate-
rials structures and compositions and their catalytic reactivity.
solved and refined by the SHELXTL software package [32]. These data
can be obtained free of charge from The Cambridge Crystallographic
Data Centre at https://www.ccdc.cam.ac.uk/structures/. Further de-
tails of the crystal structures may be obtained from the Fachinfor-
mationszentrum Karlsruhe, 76344 Eggenstein-Leopoldshafen
(Germany), on quoting the depository numbers CSD-2026589,
2026590.
d. Powder X-ray diffraction (PXRD) analysis
Powder X-ray diffraction patterns were measured using a Philips
PW1820 diffractometer with a Philips PW1710 proportional detector.
The measurements were carried out using Cu Kα (Kα1 λ = 1.54060 Å,
Kα2 λ = 1.54439 Å) X-ray source. The Cu tube was operated at a
voltage of 40 kV and a current of 30 mA. XRD measurements were
recorded in Bragg–Brentano reflection geometry. The scan was per-
formed in the 2θ ranges of 10–60° for BCS and 30–90° for metals and
alloys, with a step width of 0.02° and a step time of 1 s
e. Electron microscopy
High-resolution scanning electron microscopy (HR-SEM) and
energy dispersive x-ray spectroscopy (EDS) measurements were
conducted using an FEI Sirion HR-SEM equipped with EDAX EDS and
FEI Magellan 400 L XHR-SEM equipped with an Oxford Instruments
XMAX SDD EDS working on an INCA450 platform.
2. Materials and methods
f. FIB tomography
a. Bi-complex salts (BCS) preparation
3D structure of nano-porous alloys and metals prepared in this
work was studied with Focused Ion Beam (FIB) tomography. Milling
and data collection was done at FEI Dual beam FIB Helios 460 F Lite
equipped with Auto Slice &View software. Reconstruction of ex-
perimental data and geometrical analysis of 3D reconstructed vo-
lume was performed with FEI AVIZO 9.2 software.
Microcrystalline
bi-complex
metal
salts
(BCS)
[Co
(NH3)6]2[PdCl4]3·8H2O (1) and [Co(NH3)5Cl][PdCl4] (2) were syn-
thesized in a similar manner. Mixing concentrated cold (cooled in an
ice bath) aqueous solutions (about 20 mg/mL) of [Co(NH3)6]Cl3 and
(NH4)2[PdCl4] in a molar ratio of 2:3 respectively, yielded a pre-
cipitation of the brown-orange salt 1. The precipitate was separated,
washed three times with deionized cold water, and left to dry in the
open air. Mixing concentrated aqueous solutions (about 10 mg/mL)
of [Co(NH3)5Cl]Cl2 and K2[PdCl4] in a molar ratio of 1:1 led to the
formation of a reddish powder of 2. The precipitate was separated,
washed three times with deionized water, one time with acetone,
and left to dry in the open air. Single crystals of 1 and 2 were also
prepared using membrane diffusion technique.
g. Catalytic studies
Catalytic reactions were performed in a 50 mL stainless steel
high-pressure reactor (Parr Instruments). In a typical reaction,
2.5 mg of the catalyst powder (Co-Pd alloy, Co metal, Pd metal or
Sigma Aldrich Pd/C 10% wt.) and 500:1 mol/mol ratio of phenyla-
cetylene/catalyst were dissolved in 1–2 mL of chloroform-D in a
4 mL glass vessel. The vessel was inserted to a larger 50 mL glass
vessel that was placed in the reactor. The reactor was purged three
times with 10 bar of H2 (99.999%) prior to the reaction. Hydrogen
pressure and temperature of the reactor were adjusted (as described
in Table 2 and Table 3 in section 3.2 below), and the reaction vessel
was kept with constant stirring under these conditions (pressure &
temperature) for 1–4 h. Following the reaction, fractions of the re-
action mixture were extracted, diluted with chloroform-D and se-
parately analyzed by GC-MS (Fisher-Scientific) and 1H NMR (Bruker).
b. Preparation of metals and alloys
All metallic materials were prepared according to a previously
published procedure [16]. Co-Pd alloys were prepared by chemical
reduction of salt 1 and salt 2 in a tube furnace under a flow of
1 atm H2-Ar mixture (5%/95%) at 200 °C for 24 h. Porous palladium
and cobalt were prepared by reduction of K2[PdCl4] and [Co
(NH3)5Cl]Cl2, respectively, in
a tube furnace under a flow of
1 atm H2-Ar mixture (5%/95%) at 350 °C for 24 h. The resulting black
powders of metals or alloys were cooled to room temperature and
kept under a stream of Ar for additional 10 min
3. Results and discussion
a. Preparation and characterization of bi-complex salts (BCS), nano-
porous alloys and porous metals
c. Single-crystal X-ray diffraction (SC-XRD) analysis
Two Co-Pd BCS, their corresponding alloys, and pure cobalt and
palladium metals were prepared through modifications of a pre-
viously published procedure [16]. The two Co-Pd bi-complex salts
[Co(NH3)6]2[PdCl4]3∙8H2O (1) and [Co(NH3)5Cl][PdCl4] (2) were pre-
pared by mixing concentrated aqueous solutions of counter-charged
cobalt and palladium complexes as described in Eqs. (1a) and (1b),
respectively:
Single crystals of BCS 1 and 2 were transferred to a Bruker SMART
APEX CCD X-ray diffractometer equipped with a graphite mono-
chromator. The system was controlled by a pentium-based PC run-
ning the SMART software package [30]. Data were collected at room
temperature using Mo Kα radiation (λ = 0.71073 Å). The raw data
frames were transferred to a second PC computer for integration and
reduction by the SAINT program package [31]. The structures were
2

S. Avisar, Y. Shner, R. Abu-Reziq et al.
Journal of Alloys and Compounds 891 (2021) 161936
Table 1
resulted from the various electrostatic interactions between the two
counter-charged cobalt and palladium complexes.
PXRD analysis of the BCS (Fig. S1) demonstrates high corre-
spondence between the experimental and simulated patterns, thus
indicating the purity and the ordered structure of the micro-
crystalline BCS samples.
Thermal treatment of the BCS 1 and 2 under a reducing en-
vironment (1 atm flow of 5% H2 in 95% Ar, 200 °C) for 24 h resulted in
formation of the corresponding highly porous alloys Co2Pd3 and
CoPd, respectively, as described in Eq. (2):
Values of porosity and specific surface area of Co-Pd alloys, pure palladium, and pure
cobalt.
Metallic material
Porosity (%)
Specific surface area (m2/g)
Co2Pd3
CoPd
Pd
90
90
90
22
109
99
50
5
Co
Table 2
The levels of the variables for the DOE.
5%H
2(g)
[CoLm]x [PdLn]y
CoxPdy + Volatile products
(2)
200 °C, 24 h
Variable H2 Pressure (psi) Temperature (°C) Reaction time (h)
Application of a similar reduction process to [Pd(NH3)4]Cl2 and
[Co(NH3)5Cl]Cl2 provided pure Pd and Co porous metals respectively
(Fig. S2).
Minimum level (−1) 40
Maximum level (+1) 300
25
40
1
4
PXRD analysis of the alloys indicated that they were crystalline
and composed of two phases. Fig. 2 shows that the major crystalline
phase in both samples is isostructural to the fcc Pd (JCPDF #
04–015–0492). The second crystalline phase presented in the sam-
2[Co(NH3)6]Cl3
[Co(NH3)6]2 [PdCl4]3(s) + 6(NH4)+(aq) + 6Cl(aq)
+ 3(NH4)2[PdCl4](aq)
(aq)
(1a)
(1b)
ples appeared as an isostructural to the hcp Co (JCPDF
#
[Co(NH3)5Cl]Cl2
[Co(NH3)5Cl][PdCl4](s) + 2K+(aq) + 2Cl(aq)
+ K2[PdCl4](aq)
(aq)
01–089–7373). A systematic shift of the diffraction peaks indicates
formation of solid solutions in both structures. The estimated size of
the crystallites calculated with Scherrer formula was a few nan-
ometers in the hcp Co-like phase and in the range of tens of nan-
ometers in the fcc Pd-like phase (Table S1). The phase content of
each alloy was only qualitatively estimated, due to the large FHWM
(full width of half-maximum) and low intensity of the diffraction
peaks of Co-based phase. This estimation revealed that the amount
of Pd-like phase was higher in Pd-rich Co2Pd3 sample, while the
amount of Co-like phase was higher in Co-rich CoPd sample.
Single crystals of these salts were also prepared and structurally
characterized using X-Ray crystallography. Earlier studies already
reported the characterization of salt 2 using powder diffraction [13],
yet we have crystallized and determined its structure using single-
crystal X-ray crystallography for the first time. The unit cell of each
of the mentioned crystals comprises cobalt and palladium atoms in a
specific ratio – 2:3 for salt 1 and 1:1 for salt 2 (Fig. 1). These ratios
Table 3
The set of eight DOE experiments, and their responses for: (a) Co2Pd3 catalyst, and (b) CoPd catalyst.
Experiment number
H2 pressure (psi)
Temperature (°C)
Reaction time (h)
Conversion of phenylacetylene (%)
Styrene's selectivity (%)
Styrene's yield (%)
3a. Co2Pd3 catalyst
1
2
3
4
5
6
7
8
300
40
300
40
300
40
300
40
40
40
25
25
40
40
25
25
4
4
4
4
1
1
1
1
100
88
100
49
100
31
61
1
1
92
13
93
70
94
95
93
81
13
46
70
29
58
20
21
3b. CoPd catalyst
1
2
3
4
5
6
7
8
300
40
300
40
300
40
300
40
40
40
25
25
40
40
25
25
4
4
4
4
1
1
1
1
100
97
100
65
100
39
100
12
0
88
0
91
75
92
20
92
0
85
0
59
75
36
20
11
Fig. 1. The unit cells of the single crystals: (a) [Co(NH3)6]2[PdCl4]3·8H2O, and (b) [Co(NH3)5Cl][PdCl4]. The hydrogen atoms of the ammonia ligands were omitted for clarity.
3

S. Avisar, Y. Shner, R. Abu-Reziq et al.
Journal of Alloys and Compounds 891 (2021) 161936
the relatively low reduction temperature. This is since at elevated
temperatures, the metal atoms have rather high diffusion rates that
prevent typical metallurgical processes from manufacturing nano-
dimensional metals and alloys [34,35].
Fig. 4 shows that the morphologies and the volumes of the BCS
micro-crystallites were almost perfectly preserved in the alloys
during the reduction process, yet the initially smooth faces of micro-
particles (4a, 4c) became porous (4b, 4d). HR-SEM images of porous
palladium and cobalt metals, prepared using the same process, have
shown similar results (Fig. S3, Fig. S4). Nevertheless, while the
structure of the porous Co-Pd alloys and palladium were more alike,
cobalt acquired a coarser and a less porous structure (Fig. S3).
HR-SEM imaging shown in Fig. 5 reveals that the reduced alloys
acquired a highly porous structure of interconnected nano-dimen-
sional metal ligaments separated by the pores of the similar size, as
already known for this process [16]. Besides, the dimension of these
ligaments was very similar to the crystal size estimated from the
PXRD data (Table S1), i.e., the final metal network was nano-di-
mensional, nanocrystalline and nano-porous. This specific structure
was certainly obtained during the reduction process at which all
ligands and solvent molecules evaporated leaving only metallic
material behind, as already shown earlier [16].
Fig. 2. Comparisons of powder diffractions of Co-Pd nano-porous alloys to the ICDD
database.
3D reconstruction of the reduced alloys volume using FIB to-
mography shown in Fig. 6 had further confirmed their nano-porous
structure. 3D reconstruction of pure palladium and pure cobalt
metals revealed that the former had acquired a highly nano-porous
structure, while the latter has acquired a coarser and a less porous
structure (Fig. S6).
Geometrical analysis of the 3D reconstructed data of the alloys
and metals enabled to determine their porosity and specific surface
area (Table 1, Equations S1, S2 and Table S2). Both alloys and pure
palladium acquired 90% porosity while pure cobalt acquired only
22% porosity. The specific surface area of the alloys (~100 m2/g) was
twice higher than that of pure palladium (50 m2/g), and 20 times
higher than that of pure cobalt (5 m2/g). These results agreed with
the SEM and XRD observations and indicated the catalytic potential
of the alloys, due to their relatively high specific surface areas.
The equilibrium phase diagram for Co-Pd system predicts a
single fcc – based solid solution phase for the binary compositions
prepared in this work (60/40 and 50/50 at. ratio of Pd/Co) below
200 °C [33]. Therefore, we assumed that both Co-Pd alloys obtained
in this work were metastable two-phase bimetallic alloys composed
of nano-crystalline solid-solution phases. This assumption was fur-
ther probed via direct identification of the elemental composition in
the BCS and the corresponding alloys.
Fig. 3 demonstrates a very good agreement between the average
Pd/Co ratio in BCS 1 and 2 and their corresponding alloys, con-
sidering the statistical errors of the EDS results. It also shows a si-
milar agreement between the Pd/Co ratios directly measured by EDS
and calculated from the data of SC-XRD. These results indicated that
the initial Pd/Co ratio of the crystalline BCS was preserved almost
perfectly at the reduction. Therefore, we concluded that the reduced
alloys are indeed composed of two nano-crystalline solid solution
phases. We assumed that the preservation of Pd/Co ratio during the
reduction was provided by relatively low diffusion rates induced by
b. Catalytic activity of nano-porous alloys and porous metals
The catalytic activity of the Co-Pd alloys was examined in the
semi-hydrogenation of phenyl-acetylene as a model reaction for
hydrogenation of alkynes. A specific focus was directed at max-
imizing conversion and the selectivity towards styrene, as a model of
selectivity toward alkenes (see Scheme 1).
GC-MS was used to qualitatively identify the products of the
reaction, while 1D 1H NMR was used to quantitatively determine the
conversion and the selectivity of the reactions. The yield of the re-
action was calculated by multiplying the conversion and the se-
lectivity for each experiment. The alloys were easily separated at the
end of the reactions using a simple magnet, due to their magnetic
properties arising from the presence of cobalt atoms. HR-SEM ima-
ging of the alloys after the reaction did not reveal any obvious
change in their porous structure (Fig. S7).
Design of experiments (DOE, noted also as experimental design
or factorial design) method was used to optimize the conditions for
the reactions – including reaction time, temperature, and hydrogen
pressure. This technique allowed us to consider the influence of any
variable and more important, the influence of the interaction be-
tween these variables, upon the conversion, selectivity, and thus
yield of the reaction. Each variable had a minimum level and a
maximum level, as shown in Table 2, coded as − 1 and + 1, respec-
tively.
Since there were three variables, eight experiments were per-
formed for each alloy catalyst according to 2k DOE (where k is the
number of variables) model matrix (table S3). Finally, the results of
Fig. 3. A comparison between: the Pd:Co ratios in the BCS calculated using SC-XRD
data, the Pd:Co average ratios in the BCS determined using EDS, and the Pd:Co average
ratios in the Co-Pd alloys determined using EDS.
4

S. Avisar, Y. Shner, R. Abu-Reziq et al.
Journal of Alloys and Compounds 891 (2021) 161936
Fig. 4. HR-SEM images of BCS microcrystals and their corresponding nano-porous alloys: (a) salt 1, (b) Co2Pd3 alloy, (c) salt 2, (d) CoPd alloy.
Fig. 5. HR-SEM images acquired under increased magnifications (scale marker bars 5 µm, 1 µm, and 200 nm) at the nano-porous alloys (a)–(c) Co2Pd3 and (d)–(f) CoPd.
5

S. Avisar, Y. Shner, R. Abu-Reziq et al.
Journal of Alloys and Compounds 891 (2021) 161936
Fig. 6. 3D representations of the nano-porous alloys (a) Co2Pd3 and (b) CoPd reconstructed from FIB tomography data.
Table 4
Probing experiments for Co-Pd catalysts.
Catalyst
H2 pressure (psi)
Temperature (°C)
Time (h)
Conversion of phenylacetylene (%)
Styrene's selectivity (%)
Styrene's yield (%)
Co2Pd3
CoPd
60
60
35
35
4
4
93
99
97
89
90
88
these experiments (noted as responses), shown in Table 3, were used
to create a "response models" that describes the effects of the
mentioned variables on the conversion, selectivity, and yield of the
reaction for each catalyst.
According to the mathematical models for Co2Pd3 and CoPd
catalysts (equations S2-S7), the reaction yield was mostly negatively
influenced by the interactions between the hydrogen pressure and
time. Namely, if the pressure was increased the time should be de-
creased (and vice versa) to retain high reaction yield. In addition,
both models showed a negative influence of the interaction between
the three variables (time, pressure and temperature) on the yield.
Therefore, they suggested that setting one of the variables in an
opposite direction to the others could help increasing the yield. The
models for both catalysts also indicated an increase in the yield by
decreasing the pressure and by increasing the temperature. The
models did not agree regarding the influences (both their direction
and their extent) of the reaction time, the interaction pressure-
temperature, and the interaction time-temperature on the yield.
Despite that, identical probing experiments were conducted for both
catalysts, based on the agreement between the models of the opti-
mized conditions for maximizing the yield. In these experiments, the
reaction time was maximized to 4 h, the temperature was set to a
high value of 35 °C (but not maximized) and the H2 pressure was set
to a relatively low value of 60 PSI although not minimized. The yields
achieved in these experiments, as shown in Table 4, mostly agreed
with the predications of the models for both catalysts. In fact, this
correspondence emphasized the advantage of the DOE method that
well predict the influence of variables, and more important, the in-
fluence of their interaction, on the examined parameter. Further-
more, the high yields achieved in the probing experiments
Fig. 7. Comparison between the catalytic activities of Co2Pd3, CoPd, Co, Pd and
commerical Pd/C. The seletivity and yield of the reaction with Pd/C were 0% and thus
do not appear on the bar chart.
emphasized the importance of DOE method for optimization of
catalytic reactions.
The catalytic activity of Co-Pd nano-porous alloy was compared
to the activity of cobalt and palladium porous pure metals and
commercial Pd/C (10% Pd, sigma) in the optimized conditions
mentioned (60 psi H2, 35 °C, and 4 h). The results, shown in Fig. 7
clearly indicate improvement in the activity of the alloys compared
to the pure metals, owing to higher conversions. These results could
be attributed to the relatively higher surface area of the alloys
compared to the pure metals or to synergistic effects. The results
also show that Co-Pd alloys and porous palladium remarkably im-
proved the selectivity of the reaction, hence increasing the yield of
styrene, compared to porous cobalt and commercial Pd/C.a This may
imply that the highly porous structure of the alloys and palladium
a
Scheme 1. Hydrogenation of phenylacetylene reaction.
The seletivity and yield of the reaction with Pd/C were 0%
6

S. Avisar, Y. Shner, R. Abu-Reziq et al.
Journal of Alloys and Compounds 891 (2021) 161936
had a positive effect on the selectivity of the reaction, due to the
coarser structure of porous cobalt and the non-porous structure of
Pd/C (Figs. S3-S5). Moreover, comparison between the two alloys
revealed that CoPd was more active but less selective than Co2Pd3,
meaning that they "obeyed" the reactivity-selectivity principle
myth [36].
References
[1] M. Sankar, N. Dimitratos, P.J. Miedziak, P.P. Wells, C.J. Kiely, G.J. Hutchings,
Designing bimetallic catalysts for a green and sustainable future, Chem. Soc. Rev.
41 (2012) 8099–8100, https://doi.org/10.1039/c2cs35296f
[2] R. Ferrando, J. Jellinek, R.L. Johnston, Nanoalloys: from theory to applications of
alloy clusters and nanoparticles, Chem. Rev. 108 (2008) 845–910, https://doi.org/
10.1021/cr040090g
[3] A.J. Forty, Corrosion micromorphology of noble metal alloys and depletion
gilding, Nature 282 (1979) 597–598, https://doi.org/10.1038/282597a0
[4] J. Erlebacher, M.J. Aziz, A. Karma, N. Dimitrov, K. Sieradzki, Evolution of nano-
porosity in dealloying, Nature 410 (2001) 450–453, https://doi.org/10.1038/
35068529
[5] Y. Sun, Y. Xia, Mechanistic study on the replacement reaction between silver
nanostructures and chloroauric acid in aqueous medium, Am. Chem. Soc. 126
(2004) 3892–3901, https://doi.org/10.1021/ja039734c
[6] S.E. Skrabalak, J. Chen, Y. Sun, X. Lu, L. Au, C.M. Cobley, Y. Xia, Gold nanocages:
synthesis, properties, and applications, Acc. Chem. Res. 41 (2008) 1587–1595,
https://doi.org/10.1021/ar800018v
[7] J.C. Thorp, K. Sieradzki, L. Tang, P.A. Crozier, A. Misra, M. Nastasi, D. Mitlin,
S.T. Picraux, Formation of nanoporous noble metal thin films by electrochemical
dealloying of PtxSi1−x, Appl. Phys. Lett. 88 (2006) 033110, https://doi.org/10.
1063/1.2161939
[8] X. Wang, Z. Qi, C. Zhao, W. Wang, Z. Zhang, Influence of alloy composition and
dealloying solution on the formation and microstructure of monolithic nano-
porous silver through chemical dealloying of Al−Ag alloys, J. Phys. Chem. C. 113
(2009) 13139–13150, https://doi.org/10.1021/jp902490u
[9] J. Weissmüller, R.C. Newman, H.-J. Jin, A.M. Hodge, J.W. Kysar, Nanoporous
metals by alloy corrosion: formation and mechanical properties, MRS Bull. 34
(2009) 577–586, https://doi.org/10.1557/mrs2009.157
[10] J. Yu, Y. Ding, C. Xu, A. Inoue, T. Sakurai, M. Chen, Nanoporous metals by deal-
loying multicomponent metallic glasses, Chem. Mater. 20 (2008) 4548–4550,
https://doi.org/10.1021/cm8009644
[11] H. Rösner, S. Parida, D. Kramer, C.A. Volkert, J. Weissmüller, Reconstructing a
nanoporous metal in three dimensions: an electron tomography study of deal-
loyed gold leaf, Adv. Eng. Mater. 9 (2007) 535–541, https://doi.org/10.1002/adem.
200700063
4. Conclusions
We introduced two Co-Pd bi-complex salts, namely [Co
(NH3)6]2[PdCl4]3∙8 H2O (1) and [Co(NH3)5Cl][PdCl4] (2) as precursors
for Co-Pd alloys. Applying hydrogen-mediated chemical reduction
process on these BCS induced the formation of Co2Pd3 and CoPd
nano-porous alloys with relatively high surface areas. Under the
same process, cobalt complex salt and palladium complex salt in-
duced the formation of porous cobalt and nano-porous palladium,
respectively. These metallic materials preserved the morphologies of
their micro-crystallites salt precursors. The alloys also preserved the
stoichiometric Co/Pd ratio in their BCS precursors. The catalytic re-
activity of these alloys was examined in hydrogenation of phenyl-
acetylene, using DOE to optimize the conditions of the reaction. The
reactivity of the alloys in the optimized conditions was compared
with pure cobalt porous metal, pure palladium nano-porous metal
and Pd/C reactivities. These experiments illustrated the utility of
using DOE instead of "one variable at a time" method to reach the
optimized conditions using only a few experiments. In addition, the
presence of cobalt atoms within the alloys enabled their facile se-
paration at the end of the reaction using a simple magnet. Moreover,
the alloys showed improved catalytic properties compared to the
pure metals and high yields towards styrene. Furthermore, based on
our experiments, we suggest that nano-porous structure may im-
prove the selectivity achieved by bimetallic alloy catalysts.
Subsequently, further investigation of Co-Pd and other bimetallic
nano-porous alloys catalytic and recycling properties may extend
their applicability in other catalytic reactions or for energy storage.
[12] H. Li, A. Misra, J.K. Baldwin, S.T. Picraux, Synthesis and characterization of na-
noporous Pt–Ni alloys, Appl. Phys. Lett. 95 (2009) 201902, https://doi.org/10.
1063/1.3265744
[13] Y.V. Shubin, S.V. Korenev, K.V. Yusenko, T.M. Korda, A.B. Venediktov, Powder X-
ray diffraction study of the double complexes [M(NH 3)5Cl][M′Cl4] as precursors
of metal powders (M = Ir, Rh, Co; M′ = Pt, Pd), Russ. Chem. Bull. 51 (2002) 41–45,
https://doi.org/10.1023/A:1015045310216
[14] S.V. Korenev, A.B. Venediktov, Y.V. Shubin, S.A. Gromilov, K.V. Yusenko, Synthesis
and structure of binary complexes of platinum group metals - Precursors of
metallic materials, J. Struct. Chem. 44 (2003) 46–59, https://doi.org/10.1023/
A:1024980930337
[15] A. Zadesenets, E. Filatov, P. Plyusnin, I. Baidina, V. Dalezky, Y. Shubin, S. Korenev,
A. Bogomyakov, Bimetallic single-source precursors [M(NH3)4] [Co(C2O4)
2(H2O)2]·2H2O (M = Pd, Pt) for the one run synthesis of CoPd and CoPt magnetic
nanoalloys, Polyhedron 30 (2011) 1305–1312, https://doi.org/10.1016/j.poly.2011.
02.012
[16] M. Avisar-Levy, O. Levy, O. Ascarelli, I. Popov, A. Bino, Fractal structures of highly-
porous metals and alloys at the nanoscale, J. Alloy. Compd. 635 (2015) 48–54,
https://doi.org/10.1016/j.jallcom.2015.02.073
[17] Y.V. Shubin, P.E. Plyusnin, S.V. Korenev, Determination of the equilibrium mis-
cibility gap in the Pd – Rh alloy system using metal nanopowders obtained by
decomposition of coordination compounds, J. Alloy. Compd. 622 (2015)
1055–1060, https://doi.org/10.1016/j.jallcom.2014.10.187
CRediT authorship contribution statement
Shay Avisar: Synthesis, data analysis, Writing manuscript. Yahel
Shner: Synthesis. Raed Abu-Reziq: Supervising catalytic processes.
Inna Popov: Analytical procedures, Data interpretation. Avi Bino:
Project coordinator.
Declaration of Competing Interest
[18] Y. Shubin, P. Plyusnin, M. Sharafutdinov, E. Makotchenko, S. Korenev, Successful
synthesis and thermal stability of immiscible metal Au-Rh, Au-Ir and Au-Ir-Rh
nanoalloys, Nanotechnology 28 (2017) 205302, https://doi.org/10.1088/1361-
6528/aa6bc9
[19] P.V. Snytnikov, K.V. Yusenko, S.V. Korenev, Y.V. Shubin, V.A. Sobyanin, Co-Pt bi-
metallic catalysts for the selective oxidation of carbon monoxide in hydrogen-
containing mixtures, Kinet. Catal. 48 (2007) 276–281, https://doi.org/10.1134/
S0023158407020127
[20] D.I. Potemkin, E.Y. Semitut, Y.V. Shubin, P.E. Plyusnin, P.V. Snytnikov,
E.V. Makotchenko, D.Y. Osadchii, D.A. Svintsitskiy, S.A. Venyaminov,
S.V. Korenev, V.A. Sobyanin, Silica, alumina and ceria supported Au-Cu nano-
particles prepared via the decomposition of [Au(en)2]2[Cu(C2O4)2]3·8H2O single-
source precursor: synthesis, characterization and catalytic performance in CO
PROX, Catal. Today 235 (2014) 103–111, https://doi.org/10.1016/j.cattod.2014.04.
026
[21] S. Nijem, S. Dery, M. Carmiel, G. Horesh, J. Garrevoet, K. Spiers, G. Falkenberg,
C. Marini, E. Gross, Bimetallic Pt-Re nanoporous networks: synthesis, char-
acterization, and catalytic reactivity, J. Phys. Chem. C. 122 (2018) 24801–24808,
https://doi.org/10.1021/acs.jpcc.8b07863
The authors declare that they have no known competing fi-
nancial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgments
We are grateful to Dr. Benny Bogoslavsky and Dr. Shmuel Cohen
for assistance with SC-XRD and PXRD analysis. We wish to thank Dr.
Benny Bogoslavsky and Anat Shvets with proof reading of this ar-
ticle. We thank Atzmon Vakahi for his assistance with FIB tomo-
graphy analysis. We also thank Dr. Vladimir Uvarov for structural
characterization and theoretical support, and Suheir Omar for her
assistance with preliminary catalytic experiments.
[22] M. Carmiel-Kostan, S. Nijem, S. Dery, G. Horesh, E. Gross, Composition-reactivity
correlations in platinum-cobalt nanoporous network as catalyst for hydro-
deoxygenation of 5-hydroxymethylfurfural, J. Phys. Chem. C. 123 (2019)
30274–30282, https://doi.org/10.1021/acs.jpcc.9b07453
Appendix A. Supporting information
[23] R. Leardi, Experimental design in chemistry: a tutorial, Anal. Chim. Acta 652
(2009) 161–172, https://doi.org/10.1016/j.aca.2009.06.015
Supplementary data associated with this article can be found in
the online version at doi:10.1016/j.jallcom.2021.161936.
7

S. Avisar, Y. Shner, R. Abu-Reziq et al.
Journal of Alloys and Compounds 891 (2021) 161936
[24] P.M. Murray, S.N.G. Tyler, J.D. Moseley, Beyond the numbers: charting chemical
reaction space, Org. Process Res. Dev. 17 (2013) 40–46, https://doi.org/10.1021/
op300275p
[25] G.D. Bowden, B.J. Pichler, A. Maurer, A. Design, of experiments (DoE) approach
accelerates the optimization of copper-mediated 18F-fluorination reactions of
[29] A.F. Silva, P. Neves, S.M. Rocha, C.M. Silva, A.A. Valente, Optimization of con-
tinuous-flow heterogeneous catalytic oligomerization of 1-butene by design of
experiments and response surface methodology, Fuel 259 (2020) 116256,
https://doi.org/10.1016/j.fuel.2019.116256
[30] SMART-NT V5.6, Bruker AXS GmbH, D-76181 Karlsruhe, Germany, 2002.
[31] SAINT-NT V5.0, Bruker AXS GmbH, D-76181 Karlsruhe, Germany, 2002.
[32] SHELXTL-NT V6.1, Bruker AXS GmbH, D-76181 Karlsruhe, Germany, 2002.
[33] T.B. Massalski, H. Okamoto, P.R. Subramanian, L. Kacprzak (Eds.), Binary Alloy
Phase Diagrams, Second ed.,, ASM International, 1990.
arylstannanes, Sci. Rep.
47846-6
9 (2019) 1–10, https://doi.org/10.1038/s41598-019-
[26] T.A. Shear, F. Lin, L.N. Zakharov, D.W. Johnson, “Design of experiments” as a
method to optimize dynamic disulfide assemblies: cages and functionalizable,
in: Macrocycles, Chemie- Angew (Eds.), Int, 59 2020, pp. 1496–1500, https://doi.
org/10.1002/anie.201912169
[34] J. Erlebacher, R. Seshadri, Hard materials with tunable porosity, MRS Bull. 34
(2009) 561–568, https://doi.org/10.1557/mrs2009.155
[27] H. Tye, M. Whittaker, Use of a design of experiments approach for the optimi-
[35] M. Ames, J. Markmann, R. Karos, A. Michels, A. Tschöpe, R. Birringer, Unraveling
the nature of room temperature grain growth in nanocrystalline materials, Acta
Mater. 56 (2008) 4255–4266, https://doi.org/10.1016/j.actamat.2008.04.051
[36] H. Mayr, A.R. Ofial, The reactivity-selectivity principle: an imperishable myth in
organic chemistry, Angew. Chem. - Int. Ed. 45 (2006) 1844–1854, https://doi.org/
10.1002/anie.200503273
sation of
a microwave assisted Ugi reaction, Org. Biomol. Chem. 2 (2004)
813–815, https://doi.org/10.1039/b400298a
[28] H. Tye, Application of statistical “design of experiments” methods in drug dis-
covery, Drug Discov. Today 9 (2004) 485–491, https://doi.org/10.1016/S1359-
6446(04)03086-7
8