Optimizing the activity of Pd based catalysts towards roomerature formic acid decomposition by Au alloying

Article abstract of DOI:10.1039/c8cy02402b

Herein, a series of PdAu/C alloyed catalysts were synthesized via a modified coprecipitation-reduction method by using carbon powder as a support, and their activities towards formic acid decomposition (FAD) at room temperature (30 °C) were evaluated. It was identified that the content of surface active species, PdO, in the alloyed catalysts is well related to the feed ratio of Au used in the syntheses, attributed to both the lattice strain and ligand effects induced by Au alloying. As a result, the alloyed catalyst with the optimal Pd and Au atomic ratio, Pd_0.69Au_0.31/C, exhibited a turnover frequency (TOF) as high as 6634 h^-1 towards room temperature FAD, 4.3 times that of the pure Pd/C catalyst (1539 h^-1).

Full text of DOI:10.1039/c8cy02402b

Catalysis
Science &
Technology
View Article Online
View Journal
COMMUNICATION
Optimizing the activity of Pd based catalysts
towards room-temperature formic acid
Cite this: DOI: 10.1039/c8cy02402b
decomposition by Au alloying†
Received 25th November 2018,
Accepted 7th January 2019
Zihao Xing, Zilong Guo, Xiangyu Chen, Peng Zhang and Wensheng Yang
*
DOI: 10.1039/c8cy02402b
rsc.li/catalysis
Herein, a series of PdAu/C alloyed catalysts were synthesized via a
modified coprecipitation–reduction method by using carbon powder
as a support, and their activities towards formic acid decomposition
Inspired by this speculation, in this work, a series of PdAu
alloyed catalysts with different atomic ratios of Pd and Au
were synthesized via coprecipitation–reduction of the Pd
(H PdCl ) and Au (HAuCl ) precursors modified from a litera-
(FAD) at room temperature (30 °C) were evaluated. It was identified
2
4
4
1
3
that the content of surface active species, PdO, in the alloyed
catalysts is well related to the feed ratio of Au used in the syntheses,
attributed to both the lattice strain and ligand effects induced by Au
alloying. As a result, the alloyed catalyst with the optimal Pd and Au
atomic ratio, Pd0.69Au0.31/C, exhibited a turnover frequency (TOF) as
ture method,
and their performance towards room-
temperature FAD was evaluated. It is identified that the con-
tent of PdO on the surface of the catalysts and thus their ac-
tivities were readily tunable by changing the feed ratio of Au
used in the syntheses. The alloyed catalyst with the optimal
−1
high as 6634 h towards room temperature FAD, 4.3 times that of
Pd and Au ratio, i.e., Pd0.69Au , presented a turnover fre-
0.31
−1
−1
the pure Pd/C catalyst (1539 h ).
quency (TOF) as high as 6634 h for room-temperature FAD,
attributed to the optimized balance of lattice strain and li-
gand effects.
Transmission electron microscopy (TEM) imaging was
employed to characterize the size and morphology of the as-
prepared alloyed catalysts supported on carbon powders
Formic acid decomposition (FAD) into hydrogen and carbon
dioxide offers one of the most ideal approaches towards
hydrogen production due to its renewable resource, safe
storage in the liquid form, and high energy density (0.057 kg
−
1
1
H L , 4.4 wt%). Since the pioneering work of Zhou et al., a
2
(
Fig. 1a–f). All the catalysts are nanoparticles (NPs) nearly
variety of noble metal (Pd, Pt, Ru, and Au) based
heterogeneous catalysts have been explored to promote the
on-demand release of hydrogen produced from FAD due to
spherical in shape, which were well-dispersed on the carbon
support. The average sizes of the NPs are in the range of 3.5–
2
–5
the feasibility of separating them from the reactant. Among
the catalysts investigated, Pd based ones have attracted inten-
sive attention since they are able to initiate FAD at room-tem-
perature, which is of vital importance for its applications in
6
,7
fuel cells. Recent work showed that alloying Pd with other
noble metals, such as Au or Ag, is effective to improve the ac-
tivity of Pd based catalysts towards room-temperature
8–11
FAD.
It is documented that introduction of the second no-
ble metal may affect the electronic density of Pd in the cata-
lysts by both lattice strain and ligand effects, which provide
us the opportunity to tune the content of surface active spe-
1
2
cies, PdO, and thus the room-temperature FAD performance
of the Pd based catalysts.
State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, College of
Chemistry, Jilin University, Changchun 130022, China.
Fig. 1 TEM images of the carbon supported PdAu/C alloyed catalysts
prepared with different feed ratios of Au. (a) Pd/C, (b) Pd0.90Au0.10/C,
(c) Pd0.82Au0.18/C, (d) Pd0.75Au0.25/C, (e) Pd0.69Au0.31/C, and (f)
Pd0.64Au0.36/C. The image of the pure Au/C sample prepared under the
same experimental conditions is given in Fig. S2† for comparison.
E-mail: wsyang@jlu.edu.cn; Fax: +86 0431 85168868; Tel: +86 0431 85168185
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8cy02402b
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Communication
Catalysis Science & Technology
Table 1 Interplanar spacing, lattice parameter and lattice strain of Pd, Au
and PdAu NPs derived from the XRD patterns
5
.5 nm when the feed ratios of Au atoms were changed in the
range of 0–0.36 (Fig. 2a, see Fig. S1† for the size distribution
curves). The X-ray diffraction (XRD) patterns of the samples
Interplanar
spacing/nm
Lattice
parameter/nm
Lattice
strain
(Fig. 2b) prepared by using the Pd and Au precursors well
Sample
2θ/degrees
matched the standard files (JCPDS, file No. 46-1043 for Pd
and file No. 04-0784 for Au), indicating the formation of pure
Pd and Au NPs. The catalysts prepared by coprecipitation–re-
duction of the Pd and Au precursors presented diffraction
peaks located between those of Pd and Au, suggesting the
formation of alloyed PdAu NPs rather than a mixture of
monometallic Pd and Au NPs. Take the diffraction peaks cor-
responding to the (111) planes of face cantered cubic (fcc) Pd
and Au for example, as shown in Table 1, the value of the 2θ
angles shifted from 40.04° to 39.60°, indicating the increased
interplanar spacing (from 0.2250 to 0.2274 nm) when the
feeding ratio of Au increased from 0 to 0.36. Based on the
patterns, the lattice parameters of the alloyed NPs are deriv-
able according to Vegard's law (see the ESI† for details),
which increased from 0.6365 to 0.6431 nm when the feed ra-
tio of Au increased from 0 to 0.36. Such lattice expansion
leads to an increase in lattice strain, which increased from
Pd
40.04
39.80
39.73
39.69
39.64
39.60
38.18
0.2250
0.2263
0.2267
0.2269
0.2272
0.2274
0.2355
0.6365
0.6404
0.6412
0.6417
0.6425
0.6431
0.6669
—
Pd0.90Au0.10
Pd0.82Au0.18
Pd0.75Au0.25
Pd0.69Au0.31
Pd0.64Au0.36
Au
0.0061
0.0073
0.0081
0.0093
0.0103
—
by the simple coprecipitation–reduction method, the lattice
strain increased almost linearly with the feed ratio of Au used
in the syntheses.
The volume of the gas generated from room-temperature
FAD was subsequently tested to evaluate the performance of
the alloyed catalysts. Within 10 min of the reaction, the vol-
ume of the gas was 192 mL for the Pd/C catalyst, which in-
creased to 307, 324, 370 and 442 mL when the ratio of Au in
the PdAu alloyed NPs increased to 0.10, 0.18, 0.25 and 0.31
and then decreased to 364 mL when the feed ratio of Au fur-
ther increased to 0.36 (Fig. 3a). In comparison, no gas was
detectable when the Au/C sample was employed as the cata-
lyst under the same experimental conditions (Fig. S3†).
Electrochemical measurements were performed to further
evaluate the intrinsic activity of the PdAu catalysts. Based on
the results of the carbon monoxide stripping experiments
(Fig. S4†), the electrochemically active surface area (ECSA) of
0
.0061 to 0.0103 (see the ESI† for the detailed arithmetic
method to calculate the lattice strain) when the ratio of Au
increased from 0 to 0.36, nearly in a linear relationship with
the increase in content of Au in the alloys (Fig. 2c). The above
lattice expansion was further illustrated by high resolution
transmission electron microscopy (HRTEM) imaging. As
shown in Fig. 2d and e, an interplanar spacing of 0.230 nm
was observed for the Pd/C sample, which is attributed to the
1
4
2
−1
(
111) facet of Pd. For the Pd0.90Au0.10/C and Pd0.69Au0.31/C
the Pd/C catalyst was calculated to be 323 m g and those
samples, the spacings were observed to be 0.245 and 0.253
nm, respectively. Since the relative difference between the
atomic radii of Pd (1.38 Å) and Au (1.44 Å) is only about 5%
of the PdAu/C catalysts fluctuated in the range of 253–308
2
−1
m
g
(Fig. S5†). The Pd/C sample exhibited the highest
ESCA, attributed to its relatively small average size (Fig. 2a).
The fact that the Pd/C sample has the highest ECSA but the
lowest FAD activity (Fig. 3a) precludes the possibility of Au
acting as a site diluter. Fig. 3b shows the turnover frequency
(TOF, see the ESI† for the detailed arithmetic method to cal-
1
5
and both of them have the same fcc crystal structure, the
Au atoms readily substitute Pd to form the Pd/Au alloy to in-
duce the lattice expansion according to the binary alloy phase
1
6
diagram. As a result, in the PdAu alloyed catalysts prepared
5
,17
culate the TOF)
of the PdAu catalysts towards room-
temperature FAD. The monometallic Pd/C catalyst presents a
−
1
TOF value of 1539 h , which increased to 2819, 3446, 5411
Fig. 2 (a) Variation in average sizes of the alloyed NPs in the catalysts
with the atomic fraction of Au. (b) XRD patterns of the catalysts with
different atomic fractions of Au. The pattern of pure Au is given as a
dotted curve for comparison. (c) Variation in lattice strain in the
alloyed NPs with the Au atomic fraction. HRTEM images of the NPs in
Fig. 3 (a) Variations in volume of the gas generated from FAD over
the Pd/C and PdAu/C catalysts prepared with the feed ratio of Au in
the range of 0–0.36 along with the reaction time at room-temperature
(30 °C). (b) Variation in TOF for H production with the atomic fraction
2
of Au in the alloyed catalysts. The variation in lattice strain of the cata-
lysts is given as a dotted line for comparison.
(
d) Pd/C, (e) Pd0.90Au0.10/C, and (f) Pd0.69Au0.31/C catalysts.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Communication
−
1
and 6634 h when the feed ratio of Au increased to 0.10,
0
ever, the content of PdIJII) decreased to 59% when the ratio of
Au further increased to 0.36. The increase in the content of
PdIJII) in the range of Au ratios of 0.1–0.31 is well consistent
with the lattice strain effect of Au as shown in Fig. 2c, in
which the increase in content of Au in the alloyed catalysts
promoted the lattice expansion and thus the capability of Pd
towards the dissociative adsorption of O2.
In contrast to the lattice strain effect, charge transfer from
Au to Pd atoms, which is known as the ligand effect, may
weaken the capability of Pd towards the dissociative adsorp-
−
1
.18, 0.25 and 0.31. However, the TOF decreased to 5473 h
when the feed ratio of Au further increased to 0.36. It is docu-
mented that the introduction of the second metal with a
larger atomic radius, such as Au, should result in lattice
strain and thus upward movement of the d-band centre of
Pd. According to the increased lattice strain with the in-
creased feed ratio of Au (Fig. 3b), it is deduced that the disso-
1
8
2
ciative adsorption of Pd to O and thus the formation of cata-
lytically active surface species, PdO, should be promoted with
the increased fraction ratio of Au in the alloyed catalysts. As
a result, the TOF increases with the feed ratio of Au. The per-
formance of the catalyst with the optimized Au ratio,
Pd0.69Au0.31/C, is greatly improved compared with the Pd
based catalysts reported in previous studies (Table S1†). The
conflict between the increased lattice strain and decreased
activity for the catalyst with the Au ratio of 0.36 means there
may exist other effects of Au that contribute to the decreased
activity of the catalyst.
tion of O . XPS analyses showed that the binging energy of
2
Pd 3d3/2 decreased from 342.24 to 342.20 and 341.90 eV, and
that of Au 4f5/2 increased from 87.61 to 87.74 and 87.81 eV
when the Au ratio increased from 0.10 to 0.25, and both of
them remained less changed when the ratio further increased
to 0.31 and 0.36 (Fig. 4b, see Fig. S7† for the variations in
binding energies of Pd 3d5/2 and Au 4f7/2), suggesting the ob-
19
vious depletion of Au d electrons and thus weakened capa-
bility of Pd towards the dissociative adsorption of O in the
2
The contents of Pd (0) and PdIJII) on the surface of the cat-
alysts were investigated by X-ray photoelectron spectroscopy
alloyed catalysts with the high Au ratios (≥0.25). XPS results
showed that the content of lattice oxygen increased from 0.51
to 0.62 when the ratio of Au increased from 0.10 to 0.31, and
then decreased to 0.55 at the Au ratio of 0.36 (Fig. S6c and
S8†). The existence of lattice oxygen, combined with the ab-
sence of the peaks corresponding to PdO in the XRD patterns
(Fig. 2b), suggests the formation of PdO species on the sur-
face of the alloyed catalysts. These results suggested that the
(
XPS) analyses (Fig. S6a†) to illustrate the effect of Au alloying
on the capability of the catalysts towards dissociative adsorp-
tion of O . As illustrated in Fig. 4a, for the PdAu/C sample
2
with the Au ratio of 0.10, ca. 53% of the surface Pd were
found to be PdIJII) ions, which increased to 55%, 58% and
6
0
1% with increasing ratio of Au in the alloyed catalysts to
.18, 0.25 and 0.31, accompanied by a decrease in the con-
capability of Pd towards dissociative adsorption of O in the
2
tent of Pd (0) atoms. Meanwhile, the spectra of Au 4f show a
couple of distinct peaks at 84.02 and 87.61 eV, corresponding
to Au 4f7/2 and 4f5/2 of Au (0), respectively (Fig. S6b†). These
results indicate the high capability of Pd and the very low ca-
alloyed catalysts with the appropriate Au ratios (0.10–0.31)
was mainly dominated by the lattice strain effect of Au. The
formation of surface-active species, PdO, and thus the activity
of the catalysts were promoted by Au alloying. For the catalyst
with the highest Au ratio (0.36), the capability of Pd towards
1
8
pability of Au towards dissociative adsorption of O
2
.
How-
dissociative adsorption of O (ref. 20) was weakened by the li-
2
gand effect of Au, which resulted in the decreased content of
the surface-active species, PdO, and thus the decreased activ-
ity of the catalysts towards room-temperature FAD.
Variations in surface compositions of the catalysts were
further summarized based on the XPS results (Fig. 4c). When
the feed ratio of Au increased from 0.10, 0.18, 0.25, 0.31 and
0.36, the fraction of Pd on the catalyst surface decreased
from 0.89, 0.77, 0.72, 0.68, and 0.66, while the fraction of Au
on the surface increased from 0.11, 0.23, 0.28, 0.32 and 0.34,
suggesting the complete reduction of the Au and Pd precur-
sors and the formation of homogeneous (equilibrated) alloys.
Interestingly, the variation of the PdIJII)/Pd (0) ratio, which in-
creased from 1.13 to 1.22, 1.38, 1.56 while the Au ratio in-
creased from 0.10 to 0.18, 0.25 and 0.31, and then decreased
to 1.43 while the Au ratio further increased to 0.36, which is
well consistent with the change in the TOF of the catalysts to-
wards room-temperature FAD (see Fig. S9† for comparison in
variations in content of PdIJII) and TOF of the catalysts). This
is to say that the catalyst with the highest PdO : Pd ratio leads
Fig. 4 (a) Variations in relative contents (RC) of PdIJII) and Pd (0), (b)
variations in binding energies of Pd 3d3/2 and Au 4f5/2, (c) variations in
the relative atomic fraction of Pd and Au on the surface, and (d)
variation in the ratio of PdIJII)/Pd (0) on the catalyst surface with the
atomic fraction of Au in the catalysts. Variation in TOF of the catalysts
is given as a dotted line in (d) for comparison.
−
1
to the highest TOF (6634 h ). According to the above results
and discussion, the optimized capability of Pd towards disso-
ciative adsorption of O
2
and thus the formation of active
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Communication
Catalysis Science & Technology
is primarily related to the PdO : Pd ratio on the catalyst sur-
face, which is optimizable by changing the ratio of Au in the
PdAu alloys. The PdAu catalyst with the optimized Pd : Au
−
1
ratio, Pd0.69Au0.31/C, exhibited a TOF as high as 6634 h ,
among the highest ones reported so far in the literature. It is
expected that the optimization of the capability of Pd towards
dissociative adsorption of O in the Pd based catalysts should
2
be profitable to further boost their performance towards
room-temperature FAD.
Conflicts of interest
Fig. 5 Schematic representation of lattice strain and ligand effects of
Au on the capability of Pd towards dissociative adsorption of O and
2
There are no conflicts to declare.
lattice oxygen content (LOC) at the low and high Au ratios in the
alloyed catalysts.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (No. 21773089, 51372097).
species (PdO) on the surface of the PdAu catalysts by chang-
ing the ratio of Au are attributed to the balanced lattice strain
and ligand effects induced by Au alloying. As summarized in
Fig. 5, at low Au ratios (≤0.31), the strain effect of Au plays a
dominate role in optimizing the structure and performance
of the resulting alloyed catalysts. Au alloying induces the lat-
tice expansion of Pd and upshift of its d band centre, which
Notes and references
1 X. Zhou, Y. Huang, W. Xing, C. Liu, J. Liao and T. Lu, Chem.
Commun., 2008, 3540–3542.
2 C. Fellay, P. J. Dyson and G. Laurenczy, Angew. Chem., Int.
Ed., 2008, 47, 3966–3968.
promote dissociative adsorption of O on Pd. As a result, the
3 R. Amos, F. Heinroth, B. Chan, A. Ward, S. Zheng, B.
Haynes, C. Easton, A. Masters, T. Maschmeyer and L.
Radom, ACS Catal., 2015, 5, 4353–4362.
4 K. Jiang, K. Xu, S. Zou and W. Cai, J. Am. Chem. Soc.,
2014, 136, 4861–4864.
5 K. Tedsree, T. Li, S. Jones, C. Chan, K. Yu, P. Bagot, E. Marquis,
G. Smith and S. Tsang, Nat. Nanotechnol., 2011, 6, 302–307.
6 O. Metin, X. Sun and S. Sun, Nanoscale, 2013, 5, 910–912.
7 R. Williams, R. S. Crandall and A. Bloom, Appl. Phys. Lett.,
1978, 33, 381–383.
2
lattice oxygen content (LOC) and thus the relative amount of
the surface active species increase with the increased ratio of
Au in the alloyed catalysts. At high Au ratios (>0.31), the li-
gand effect (electron transfer from Au to Pd) becomes domi-
nant, leading to a decrease in binding energies of Pd 3d and
decreased capability of Pd towards dissociative adsorption of
2
O . As a result, both the LOC and the relative amount of the
surface active species decrease with further increasing Au ra-
tio in the alloyed catalysts. The optimized activity of the
alloyed catalysts comes from the appropriate balance of lat-
tice strain and ligand effects of Au.
8 N. Wang, Q. Sun, R. Bai, X. Li, G. Guo and J. Yu, J. Am.
Chem. Soc., 2016, 138, 7484–7487.
Cycling experiments were carried out to evaluate the recycla-
bility of the catalyst. The catalyst with the optimized Au ratio,
Pd0.69Au0.31/C, showed no obvious change in the catalytic activ-
ity after five cycles (Fig. S10†), suggesting the good recyclability
of the catalyst. GC analysis showed that the gas generated was
9 X. Gu, Z. H. Lu, H. L. Jiang, T. Akita and Q. Xu, J. Am. Chem.
Soc., 2011, 133, 11822–11825.
10 Z. Wang, X. Hao, D. Hu, L. Li, X. Song, W. Zhang and M. Jia,
Catal. Sci. Technol., 2017, 7, 2213–2220.
11 S. Akbayrak, Y. Tonbul and S. Ozkar, Appl. Catal., B,
2017, 206, 384–392.
2 2 2
composed of by 46.91% H , 1.45% O , 6.82% N and 44.82%
CO , and no CO was detectable in our experiments (Table S2†).
12 Q. Lv, Q. Meng, W. Liu, N. Sun, K. Jiang, L. Ma, Z. Peng, W.
Cai, C. Liu, J. Ge, L. Liu and W. Xing, J. Phys. Chem. C,
2018, 122, 2081–2088.
13 Z. Yan, Z. Xu, J. Yu and M. Jaroniec, Environ. Sci. Technol.,
2015, 49, 6637–6644.
2
TEM and XPS analyses indicated that there are no obvious
changes in size and binding energy of Pd 3d of the catalyst af-
ter the reaction, excluding the aggregation of the nanoparticles
and the existence of surface-bound Pd–CO after the reaction
(
Fig. S11†). These results suggested the good recyclability of the
14 V. Mazumder and S. Sun, J. Am. Chem. Soc., 2009, 131,
4588–4589.
PdAu alloyed catalysts, attributed to the high selectivity of the
catalysts towards the production of H
2
and CO
2
.
15 H. Yao and R. Kobayashi, J. Colloid Interface Sci., 2014, 419, 1–8.
1
6 Y. Ding, F. Fan, Z. Tian and Z. Wang, J. Am. Chem. Soc.,
010, 132, 12480–12486.
2
Conclusions
1
7 A. P. Umpierre, E. D. Jesus and J. Dupont, ChemCatChem,
2011, 3, 1413–1418.
In this work, PdAu/C alloyed catalysts were synthesized and
their activities towards room-temperature FAD were evalu-
ated. It was revealed that the activity of the alloyed catalysts
18 F. Gao and D. W. Goodman, Chem. Soc. Rev., 2012, 41,
8009–8020.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Communication
1
9 F. P. Leisenberger, G. Koller, M. Sock, S. Surnev, M. G.
20 Y. S. Lee, Y. Jeon, Y. D. Chung, K. Y. Lim, C. N.
Whang and S. J. Oh, J. Korean Phys. Soc., 2000, 37,
451–455.
Ramsey, F. P. Netzer, B. Klotzer and K. Hayek, Surf. Sci.,
2
000, 445, 380–393.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.