Lattice distortion induced electronic coupling results in exceptional enhancement in the activity of bimetallic PtMn nanocatalysts

Article abstract of DOI:10.1016/j.apcata.2017.01.021

Lattice strain plays a critical role in structural heterogeneity and surface electronic properties of bimetallic nanocatalysts. However, understanding of how to engineer optimal electron transfer in anisotropic bimetallic crystals remains a grand challenge to achieve enhanced catalytic performances. We investigate beyond conventional polymer based core-shell and alloy structures, and present unique lattice distorted PtMn catalysts fabricated via a cooperative self-assembly method. The strong internal strain between Pt and Mn lattices is found to induce the structural distortion of anisotropic PtMn crystals and formation of asymmetric flower shapes, leading to stretched Pt and contracted Mn lattices. Such distorted bimetallic crystals exhibit unusual electronic coupling and an eight-fold synergistic enhancement in catalytic oxidation of renewable biomass feedstocks compared with monometallic Pt catalysts. The novel synthesis technique and revealed electronic coupling mechanism described herein opens the door for the rational discovery of other bimetallic nanocatalysts with positive synergy.

Full text of DOI:10.1016/j.apcata.2017.01.021

Applied Catalysis A: General 534 (2017) 46–57
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Lattice distortion induced electronic coupling results in exceptional
enhancement in the activity of bimetallic PtMn nanocatalysts
Xin Jina,b, Chun Zenga, Wenjuan Yana,b, Meng Zhaoa, Pallavi Bobbaa, Honghong Shia,
Prem S. Thapac, Bala Subramaniama, Raghunath V. Chaudharia,∗
a Center for Environmentally Beneficial Catalysis, Department of Chemical and Petroleum Engineering, 1501 Wakarusa Drive, University of Kansas,
Lawrence, KS 66047, USA
b State Key Laboratory for Heavy Oil Processing, College of Chemical Engineering, China University of Petroleum, Qingdao 266580, China
c Microscopy and Analytical Imaging Laboratory, Haworth Hall, 1200 Sunnyside Ave, University of Kansas, Lawrence, KS 66045, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Lattice strain plays a critical role in structural heterogeneity and surface electronic properties of bimetal-
lic nanocatalysts. However, understanding of how to engineer optimal electron transfer in anisotropic
bimetallic crystals remains a grand challenge to achieve enhanced catalytic performances. We investigate
beyond conventional polymer based core-shell and alloy structures, and present unique lattice distorted
PtMn catalysts fabricated via a cooperative self-assembly method. The strong internal strain between Pt
and Mn lattices is found to induce the structural distortion of anisotropic PtMn crystals and formation of
asymmetric flower shapes, leading to stretched Pt and contracted Mn lattices. Such distorted bimetallic
crystals exhibit unusual electronic coupling and an eight-fold synergistic enhancement in catalytic oxi-
dation of renewable biomass feedstocks compared with monometallic Pt catalysts. The novel synthesis
technique and revealed electronic coupling mechanism described herein opens the door for the rational
discovery of other bimetallic nanocatalysts with positive synergy.
Received 23 November 2016
Received in revised form 23 January 2017
Accepted 26 January 2017
Available online 28 January 2017
Keywords:
Lattice mismatch
Distortion
Anisotropic growth
Bimetallic cluster
Oxidation reaction
© 2017 Published by Elsevier B.V.
1. Introduction
chemical upgrading, but also in emerging areas such as biomass
conversion and solar cells [10].
The development of novel catalytic materials is
a
key
In the past decade, major research efforts have been made
on fabricating bimetallic Pt-3d metal nanomaterials due to their
tunable synergy in catalysis [6,11–15]. It is established in classic
catalysis theories that atoms on highly indexed sites display supe-
rior activities than others. Therefore, conventional strategies on
tuning the surface geometries of bimetallic crystals, and thereby to
enhance catalytic activity, is by altering the morphologies of Pt-3d
bimetallic structures through adding ligands [16]. Manipulation of
crystal growth kinetics using ligand-metal interaction as the exter-
nal driving force results in enlargement of surface roughness. This
well-known methodology has been established by several research
groups and summarized in several reviews [4–10,16,17]. Numerous
results have confirmed that PtCo, PtCu, PtNi and PtFe particles with
cubic, spherical, truncated octahedral and wired structures can
be fabricated by mixing ligands [e.g. polyvinylpyrrolidone (PVP)]
with metal precursors, in solvents containing polyols or amines
[18–21]. For example, a two-step method involving PVP mixed with
Cu seeds led to hollow nanocaged or nanoframe structured PtCu
particles [22]. Branched PtNi nanostructures are reported using Pt
seeds as templates [23,24], while didodecyldimethylammonium
bromide stabilized Ni seeds generate PtNi nanodendrites [25].
for resource-efficient technologies for platform chemicals with
reduced environmental impact [1–3]. Extensive research has
already shown that bimetallic and multimetallic catalysts
often exhibit enhanced catalytic performances compared with
monometallic ones due to tunable morphology and composition
[4,5]. Although these materials can be synthesized with precise
control and particle size and tailored structures such as cubic,
rhombic and octahedral shapes, they often consist of low indexed
facets with limited surface catalytic activities [6–8]. Therefore, the
discovery of novel nanomaterials with enhanced anisotropy and
tunable electron configuration, which display high surface cat-
alytic activities, has received much attention in recent years [9]. In
the area of bimetallic materials, Pt-based nanostructured catalysts
have been one of the most extensively studied systems because
they not only show exceptional performances in fuel cell and petro-
∗
Corresponding author.
E-mail address: rvc1948@ku.edu (R.V. Chaudhari).
http://dx.doi.org/10.1016/j.apcata.2017.01.021
0926-860X/© 2017 Published by Elsevier B.V.

X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
47
Similar techniques have also been employed to synthesize other
3d metal based nanomaterials [26–32], as well as other bimetallic
combination such as PtAu, PtPd, etc. [10]. It is clear that two-step
ligand directed seed/template growth is the main methodology to
create anisotropic high index nanoparticles. Although the ligand-
metal force is effective in facilitating anisotropic growth, a major
drawback is that such bimetallic crystals often lack strong internal
interaction and structural coherence, easily decomposing or deac-
tivating [9,33] when exposed to complex reaction environment
[34].
of these catalysts obtained from biomass oxidation results pro-
vides guidance for rational design of synergistic bimetallic catalysts
for selective activation of C H and C O bonds encountered in the
processing of renewable fuels and chemicals.
2. Experimental section
2.1. Chemicals
Glycerol, lactic acid, glycolic acid, formic acid and dimethyl-
formide (DMF) were purchased from Sigma Aldrich. Glyceric and
tartronic acids were obtained from Fisher Scientific. Metal precur-
sors such as Pt(acac)2 and Mn(NO3)2 as well CeO2 powders were
also purchased from Sigma Aldrich.
Classic bottom up synthetic methodologies are primarily
focused on ligand-mediated surface index for morphologically
influenced catalysis (facet dependent catalysis), which can be
attributed to activity enhancements of bimetallic nanocatalysts
[35]. But recent research in nanomaterials has discovered that elec-
tron transfer and metal–metal interfacial strain can intrinsically
enhance structural coherence and surface properties of bimetallic
crystals [36–38], where extensive studies have shown that rather
than forcing two metal species to form bimetallic nanoparticles
using polymer stabilizers, lattice strain induced by difference in
lattice parameters among the metal species can be used as a driv-
ing force to generate unique nanoparticles. Interfacial strain often
induce unusual lattice stretch or contraction in bimetallic crystals.
Such lattice distortion undoubtedly leads to tunable surface elec-
tronic properties and thus affecting catalytic activities of bimetals.
This is because that the strong cohesive interaction at metal–metal
interface (“ligand” effect) involves charge transfer or rehybridiza-
tion of electron orbitals of two metals. The perturbation of a metal’s
orbitals intrinsically shift d band towards/away from Fermi level
[36]. Therefore, in the area of catalysis, this finding offers a new
methodology and provides immense potential for innovative cat-
alyst design [34,35]. Engineering optimal electronic coupling by
altering metal–metal interfacial lattice strain and understanding
its impact on catalytic activity of bimetallic materials have been
attracting extensive attention in recent years [9,36,39].
There are three fundamental aspects which need to be
addressed in present work: (a) proposing a ligand-free method
for cooperative assembly of exchange coupled bimetallic nanocrys-
tals, (b) investigating the plausible mechanism of lattice distorted
and coupled growth of bimetallic crystals and (c) understanding
how lattice distortion of crystals influences electronic coupling
and surface catalytic properties. Using bimetallic PtMn crystal as a
model system, we studied in this paper a polymer free one pot self-
assembly method to synthesize exchange coupled nanoclusters.
The anisotropic growth is induced by large lattice constant mis-
match between the two metals (Pt: fcc, 0.39 nm; Mn: bcc, 0.89 nm).
In this approach, well-controlled bud- and cauliflower-shaped PtMn
clusters were grown in the absence of polymer stabilizers and
immobilized on heterogeneous support (i.e. CeO2) in one pot pro-
cess. The influence of Pt/Mn ratio, solvent and chemical promoters
on the lattice strain of PtMn clusters were systematically studied.
Surface morphologies of PtMn catalyst samples were characterized
using transmission electron microscopy (TEM), while information
on Pt and Mn interaction, surface electronic properties and lattice
distortion were obtained from X-ray photoelectron spectroscopy
(XPS). A plausible mechanism for lattice distorted anisotropic
growth of bimetallic PtMn is discussed. While reported studies on
PtMn nanocatalysts deal with reactions involving simple substrates
such as methanol [40–43], we investigated the performances of
PtMn clusters for catalytic oxidation of renewable biomass-derived
feedstocks (C2-3 polyols) in aqueous phase under relatively mild
conditions (70 ◦C, 1 atm O2).
2.2. Catalyst preparation
In a typical catalyst synthesis process, 1.2 g of CeO2 powders
and approximately 10 mL DMF solvent were first mixed at room
temperature. Known amounts of Pt and Mn precursors were then
charged and further mixed with the slurry till dissolved completely.
Another 10 mL DMF was also added to the mixture. After transfer-
ring into a 100 mL autoclave, the whole slurry was flushed thrice
with N2 at room temperature. The whole mixture was heated at
200 ◦C under N2 atmosphere (∼2.2 MPa) for about 12 h before cool-
ing down to room temperature. Catalyst powders from each batch
of preparation were washed with ethanol/water (2/1 v/v) and cen-
trifuged several times before drying under vacuum overnight. The
catalyst samples were tested for oxidation without any further pre-
treatment. The details of catalyst preparation method are described
as follows:
Catalyst preparation was conducted via a simple solvother-
mal method. A schematic description of the preparation step is
presented in Fig. 1. In particular, cerium oxide (CeO2) supported
Pt and PtMn catalysts were synthesized using DMF as a solvent.
Unlike two-pot or seeded growth methods, reduction of metal
precursors, Pt(acac)2 and Mn(NO3)2·4H2O, was performed in the
presence of CeO2 powder. In other words, both growth of nanocrys-
tals and simultaneous immobilization on the catalyst support were
achieved in one pot. The Pt content on CeO2 support is 1 wt% for
all the catalyst samples (identical Pt precursor concentration in all
solution), while the atomic ratios of Pt/Mn (1/x) were varied from
1/0.5 to 1/2 (denoted as PtMnx). In a typical catalyst preparation
experiment, approximately 1.2 g catalysts were obtained in 20 mL
DMF solution. While the CeO2 powder and metal precursor solu-
tion in DMF are yellowish-white and brownish-yellow respectively,
the solid products resulting from the one-pot synthesis are grey,
suggesting reduction of metal precursors and immobilization of
nanoparticles on the CeO2 support. In addition, the gas atmosphere
in the autoclave is also believed to be important in the formation
and continuous growth of bimetallic nanocyrstals. In most cases,
N2 was used as the inert blanket, while the reduction of metal
precursors was induced by DMF (a weak reducing agent). In some
control experiments, H2 (a strong reducing agent) was added, while
in another case benzoic acid (BA, a template agent) was used as a
promoter in the DMF medium (see Table S1 for notation).
2.3. Oxidation tests
The experimental procedure for testing catalyst activity for glyc-
erol oxidation was similar to that described previously [44]. Briefly,
about 0.05 g of solid catalyst was added to 25 mL aqueous solution
containing glycerol (1.0 g) and NaOH (1.7 g), which was transferred
to a 100 mL of three neck flask. The slurry was heated in an oil bath
with precise temperature control before heating up to a predeter-
mined reaction temperature. Once the liquid slurry attained the
This work shows that anisotropic growth induced lattice dis-
tortion and electronic coupling leads to exceptional (8 fold)
enhancement in the activity of bimetallic PtMn nanocatalysts for
oxidation of biomass feedstocks. The structure-activity correlation

48
X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
Fig. 1. Schematic description of one pot solvothermal synthesis of PtMn catalyst.
reaction temperature, stirring rate was set at 1000 RPM, and O2
was introduced into the reactor at a constant rate (60 std mL/min),
which signified the start of an experimental run. For recycle stud-
ies, catalyst samples were centrifuged to separate from reaction
slurries and used for next run without any pretreatment.
regularly-shaped structures. As in the case of lattice mismatched
PtFe heteroclusters, both anisotropic growth and galvanic displace-
ment between Pt2+ and Fe as well as Pt and Fe3+ play critical roles in
the formation of nanoclusters [46]. However, in the case of PtMn, it
is strongly believed that lattice mismatch is the main driving force
to influence anisotropic growth of PtMn clusters.
It is already established that kinetic rate of reduction of metal
cations (e.g. Pt2+, Mn2+) can be controlled to tailor the morphology
of the final particles [6,23]. Previous literature using NaBH4 (strong
reducing reagent) and electron deposition (reduction) have con-
firmed that fast reduction of metal cations results in predominantly
small seeds/clusters with low surface indexed cubic and spherical
shapes [25,42,47]. The DMF used in this study as a mild reduc-
ing agent can potentially generate different crystal structures. This
hypothesis is confirmed when adding more Mn species in Pt system.
As shown in Fig. 3(a–c), novel nano-bud shaped bimetallic clusters
were formed in PtMn1 sample. Each bud is approximately 5 nm
wide and 23.8 4.3 nm in diameter. Further inspection of selected
PtMn1 clusters show that most of them have four to eight buds
[Fig. 3(d, e)]. The multiplicity of buds in each structure indicates
anisotropic growth possibly orienting from the surface plane of
ordered Pt octahedral or cubic structures, which has already been
shown in the PtMn0.5 sample. In other words, it is plausible that
when Mn is present in the Pt-based system, newly formed Pt species
tend to deposit on mismatched Mn template and further grow in an
epitaxial and anisotropic fashion [41]. This hypothesis is confirmed
by the fact that between each bud within one particular cluster,
clear disordered boundaries are present [Fig. 3(f)]. This information
suggests that the lattice mismatch between Pt and Mn facilitates the
formation of anisotropic and disordered layers wherein the buds
can grow further. Thus, the heterogeneity of such materials may be
further enhanced.
2.4. Product analysis
Small amounts of liquid samples were withdrawn during batch
studies and acidified with H2SO4 solution before injecting in HPLC
(SHIMADZU with SH1011 column, see Fig. S3 for an example HPLC
result). After each batch experiment, the volume of liquid mixture
was measured to ensure negligible loss of solvent (<0.4 mL). From
the concentration-time profiles, conversion (X), selectivity (S), car-
bon balance (C%) and turnover frequency (TOF based on surface Pt
atoms, in h−1) were calculated as defined below. Due to the com-
plex structures of PtMn bimetallic nanocatalysts, it is difficult to
distinguish surface Pt metal atoms from Mn. But TOF based on Pt
dispersion was still used in this manuscript as an approximation
to evaluate the performances of Pt and PtMn catalysts in oxidation
reactions (Table S1) [45].
3. Results and discussion
3.1. Particle morphologies of Pt and PtMn nanoclusters
The TEM images of Pt, PtMn0.5, PtMn1, PtMn1.5 and PtMn2
catalysts were first collected. Particle size and lattice spacing mea-
surements were carried out in detail. It is seen from Figs. 2–4 that
Pt and PtMn with increasing Mn content show different morpholo-
gies. As shown in Fig. 2(a, b), in the case of monometallic Pt/CeO2
catalyst, 11.8 3.8 nm sized Pt nanoparticles exhibit polyhedron
shapes on the surface of CeO2 support. The monometallic Pt cat-
alyst without Mn doping displays ordered structures with (111)
and (100) exposed as the dominant surface planes. When a small
amount of Mn species was present, we observed that PtMn0.5 sam-
ple has slightly bigger particle size distribution in the range of
14.2 3.5 nm with irregular shapes, as shown in Fig. 2(c). This dif-
ference suggests that the addition of Mn species interrupts the
regular growth of Pt crystals during catalyst preparation and cre-
ates structures with defects. In particular, unlike monometallic Pt
sample, it is found that PtMn0.5 nanoparticles are more irregularly
shaped with clear formation of several heteroclusters. For exam-
ple, most of the bimetallic PtMn0.5 particles in Fig. 2(c) inset display
mixed shapes including branched and cubic particle features. There
is evidence that some particles in this sample begin to form het-
eroclusters with twin structures within one particle. A detailed
inspection of PtMn0.5 in Fig. 2(d) indicates that twined structures
with clearly different lattice spacing is present in several observed
regions. This is possibly because the large lattice mismatch between
Pt (0.39 nm) and Mn (0.89 nm) facilitates the anisotropic growth
of bimetallic PtMn clusters forming heterogeneous rather than
Clearly, the formation of bud-shaped PtMn1 clusters results
from internal lattice strain between Pt and Mn metals rather
than being induced by template containing polymer ligands. Thus,
the proposed novel methodology involving lattice-strain induced
growth shows advantageous features over conventional method-
ologies, where ligands are often used to induce anisotropic growth
of Pt-3d metals [4,5,17].
3.2. Lattice distortion and asymmetric structures
A close examination of selected domains in PtMn0.5 and PtMn1
reveals the lattice distortion at the interface. Specifically, we mea-
sured lattice spacing in Fig. 2(f) and found that lattice spacing of Pt
(100) facet has been stretched from approximately 0.19 nm (pure
Pt (100)) to 0.23 nm. Such lattice stretching for Pt species obvi-
ously exists close to the boundaries of Pt and Mn lattices. Such
unusual lattice distortion for PtMn crystals is possibly caused by
DMF solvent annealing effect. Since it is known that DMF also act
as ligand to mediate Pt (111) growth [16], it would be interesting to
compare the significance of ligand and annealing effects during the

X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
49
Fig. 2. TEM images of (a, b) monometallic Pt and (c–f) bimetallic PtMn0.5 catalysts.
assembly of PtMn crystals. Therefore, detailed inspection of Fig. 2(f)
was compared with that for Fig. 3(f). When more Mn species is
present, Pt (100) surface has been minimized while Pt (111) is more
predominant. Considering the morphological changes for PtMn1
from PtMn0.5, it is plausible that annealing and mediating influence
of DMF solvent occur on Pt (111) surface.
The composition dependent morphology changes observed in
this work motivated us to further understand how the presence of
Mn species affects the size, shape and heterogeneity of PtMn clus-
ters. Fig. 4 presents TEM images of PtMn samples with increased Mn
contents (PtMn1.5, PtMn2). As shown in Fig. 4(a), when Mn content
increases from PtMn1 to PtMn1.5, some of the nano-buds appear to
have merged into a flower shape. High resolution TEM (HR-TEM)
images in Fig. 4(b, c) confirm that the length of each bud signifi-
cantly increases from 5 nm to 10 nm for the PtMn1.5 sample. While
the lattice spacing of bud is consistent in PtMn1 and PtMn1.5 sam-
ples, HR-TEM images of the PtMn1.5 catalyst sample shows that the
particle size (flower size) is enhanced significantly to 35.9 5.8 nm.
Enhanced bud length does not show isotropic features but displays
several mismatched layers within the structure of one bud. This

50
X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
Fig. 3. TEM images of bimetallic PtMn1 catalysts.
observation implies that Pt species, rather than forming additional
nuclei on the support, tend to deposit on existing “bud templates”
and facilitate size growth when more Mn species are present (via
the Oswald mechanism). Thereby, the bud cluster evolves into a
flower-shape cluster with increased dimensions. In the PtMn2 sam-
ple [Fig. 4(d)], the average particle size of bimetallic clusters was
enhanced to approximately 49 nm. This suggests that the increased
Mn content in the Pt system facilitates the epitaxial growth of PtMn
clusters rather than forming new Pt seeds, resulting in the forma-
tion of cauliflower-shaped structures with >15 buds in each cluster
[Fig. 4(e)]. It is further found that as the length of the bud grows,
the gap between each bud shrinks significantly. Flower structures
thus become very compact, which could reduce the overall surface
area of nanoparticles exposed for catalytic reactions.
The TEM images for PtMn1, PtMn1.5 and PtMn2 samples also
reveal the existence of several asymmetric structures [see Figs.
3(c–e) and 4(b, c, e, f)]. As mentioned previously, conventional
methods usually involve the formation of nanoparticles or nan-
ocluster from nuclei emanating in a homogeneous solution (oil
or water phase). Symmetric structures typically result from these
conventional approaches. In addition, the as prepared nanoclus-
ters were washed and impregnated on solid supports (e.g. carbon)

X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
51
Fig. 4. TEM images of bimetallic (a–c) PtMn1.5 and (d–f) PtMn2 catalysts [boundaries in (f) was marked in black lines].
[6,48–50]. Therefore, the support does not play any role during clus-
3.3. Factors influencing morphology of PtMn clusters
ter formation. In sharp contrast, in this study, we observed that the
flower-shaped clusters actually grow and orient from solid support
(i.e. CeO2). In other words, the flower-shaped PtMn clusters tend to
form at liquid-solid interface and further grow towards the liquid
phase (source of metal precursors). While the influence of hetero-
geneous support on final particle morphologies could be a topic of
another report, it is clear from this work that the presence of CeO2
plays an important role in the formation of the observed bimetallic
PtMn nanoclusters with novel features.
Based on the TEM characterization shown above, it is clear
that the morphology of PtMn displays a strong dependence on
Mn content. We also investigated the dependence of the growth
of PtMn nano-buds on other experimental variables, such as the
gas deployed (in the presence of H2 instead of N2) and chemi-
cal promoters/inhibitors (e.g., BA) to gain mechanistic insights into
cluster growth and also understand how chemical reagents affect
the cluster morphologies.

52
X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
Fig. 5. TEM images of bimetallic (a and b) PtMn1(H), (c and d) PtMn1(BA) catalysts and (e) plausible reduction mechanism of metal ion reduction by (i) DMF, (ii) DMF + H2
and (iii) DMF + BA.
3.3.1. H2 atmosphere
found that a large number of particles tend to form string-shaped
structures on the support surface rather than nano-bud shapes. In
addition, isolated particles were also found within the PtMn1(BA)
sample. The differences among PtMn1, PtMn1(H) and PtMn1(BA)
suggest that carboxylic groups in benzoic acid molecules also play
a role of weak template during cluster formation. However, the
relatively large benzoic ring in BA structure possibly tends to ster-
ically hinder the closed packing of nano-bud in the sample more
significantly compared with amine group in DMF. The PtMn1(BA)
sample therefore does not display clear three dimensional cluster
structures [see (iii) in Fig. 5(e)].
We also performed EDX mapping for selected bimetallic PtMn
catalyst samples to understand how lattice mismatch and rela-
tive Pt/Mn content influence Pt and Mn dispersion. Fig. 6(a)–(d)
present elemental mapping for PtMn0.5, (b) PtMn1, (c) PtMn2 and
(d) PtMn1(H) catalyst samples, respectively. It is observed from
Fig. 6(a) that Pt and Mn elements are not evenly distributed
within selected nanoclusters. In other words, it is plausible that
Pt and Mn species tend to deposit on different sites, which enables
anisotropic growth of bimetallic PtMn clusters. Therefore, it is plau-
sible that lattice mismatch force (strain) may prevent Pt and Mn
from forming uniform alloy structures during reduction of metal
precursors. When more Mn is present in PtMn system, the het-
erogeneity of clusters is further enhanced. From the TEM images
(Figs. 3 and 4) and the element mapping, it appears that the
strain generated from the lattice mismatch creates more inter-
facial boundaries that might be important for enhanced surface
properties. This finding further confirms our hypothesis regarding
anisotropic growth of PtMn clusters, resulting in defect induced
structures. For increased Mn content in the PtMn system, it seems
plausible that the particle growth and the agglomeration of Mn
(via phase segregation) become more significant in comparison to
lattice mismatch induced phase boundaries.
PtMn1(H) was prepared in a H2 (rather than inert N2) atmo-
sphere under otherwise similar synthesis conditions used for the
PtMn1 sample. As shown in Fig. 5(a, b), rather than forming
nano-bud structures, PtMn1(H) sample tends to form much larger
flowers, although the size of each bud is still small (approximately
6 nm). Clearly, the enhanced crystal size reduces the overall sur-
face area of the PtMn1(H) sample. The difference in morphology
between PtMn1 and PtMn1(H) suggests that (a) the amine groups
in DMF slowly reduce metal precursors in the liquid/solid slurry
medium, (b) the growth kinetics are controlled by C O groups in
DMF [see (i) in Fig. 5(e)], and (c) H2 tends to competitively interact
with both metal precursors in the liquid phase and existing metallic
species in the solid phase against the C O groups. Thus enhanced
reduction of metal cations to metallic clusters is facilitated, leading
to facile anisotropic growth of the bimetallic PtMn clusters [51]. In
other words, while C O group interacts mildly with metal precur-
sors, H2 is a strong reducing agent that enhances the deposition rate
of the metals, resulting in larger cluster sizes relative to the PtMn1
sample prepared in N2 atmosphere. This hypothesis appears to be
inconsistent with previously reported findings because the bonding
between C O and metal is known to be very stable at low temper-
ature [9,10,16]. Considering the slow reduction kinetics associated
with DMF (weak reducing agent) compared to H2 and solvother-
mal conditions (200 ◦C), it is concluded that the influence of C
groups on the morphologies of PtMn1(H) is negligible compared to
the PtMn1 sample. Such slow reduction kinetics in the presence of
O
C
O is favorable for the formation of nano-bud structures [see (ii)
in Fig. 5(e)].
3.3.2. Chemical promoter/inhibitor
The interesting observation with H2 atmosphere led us to fur-
ther explore how the nature of solvents, chemical promoters or
inhibitors affects the growth of PtMn clusters. The TEM images in
Fig. 5(c, d) are shown for the PtMn sample prepared by adding ben-
zoic acid (BA) in DMF solvent [PtMn1(BA)]. We find that the addition
of benzoic acid (BA) corrodes the surface of CeO2 support leading to
formation of needle like structures (Fig. S2). More importantly, it is
3.4. Plausible growth mechanism for PtMn clusters
Based on the parametric studies discussed in the previous sec-
tion, a plausible mechanism for PtMn cluster formation and growth

X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
53
Fig. 6. EDX Mapping of (a) PtMn0.5, (b) PtMn1, (c) PtMn2 and (d) PtMn1(H) catalyst samples (white and black bars indicate 10 nm range).
is proposed. The nuclei for cluster formation is believed to occur as
follows. Once Mn species is added to Pt system, the formation of Mn
atoms on existing Pt crystals creates a strong surface tension due
to large lattice mismatch between the two metals. A trace amount
of Mn on Pt surface results in anisotropic growth of the Pt nuclei.
In other words, the stacking faults break the surface orientation
of Pt template and enables the formation of anisotropic structures
[52]. Hence, it is plausible that PtMn twin structures, as opposed
to Pt clusters, act as a template for the formation of nano-buds.
While the formation of PtMn template occurs at a relatively fast
rate, further growth of each bud might require relatively longer
time even though they become thermodynamically favorable with
the formation of new seeds in the liquid/solid medium [9,10,16].
The further growth of PtMn nano-bud into cauliflower shape is
believed to occur as follows. If the isotropic growth of nano-bud
is dominant, straight and uniform long bud structures should be
expected when Mn loading is higher in PtMn system (e.g. PtMn1,
PtMn1.5). However, from Figs. 4 and S1, it is found that most of
the long buds (branches) consist of lattice mismatched layers. This
observation suggests that the surface deposition of both Pt and Mn
species during the reduction of metal precursor forms new atomic
metal layers with a large quantity of defects. These flawed surface
geometries may enhance the overall active sites for catalytic reac-
tions. As the Mn content is further increased, the surface deposition
becomes dominant while nucleation of seeds is almost negligible
under solvothermal conditions. In addition, it is worth mentioning
that Pt/Mn ratios also have direct impact on cluster sizes. It is clear
in Fig. 3 that when Mn content is low, the average size of bimetallic
PtMn particles is also small. But when more Mn is present in the
system, we find that the size of bimetallic PtMn is enhanced signif-
icantly, which is also much larger than monometallic Pt particles.
Based on existing theories of metallic atom deposition, we propose
that the presence of Mn itself could act as a chemical promoter for
crystal growth.
and substrate) mass transfer limitations are insignificant compared
to the observed reaction rates. The activity for each investigated
catalyst was calculated based on the intrinsic kinetic rate at low
conversion levels (∼23%) and the measured Pt metal dispersion,
the details of which are presented in Table S1.
The influence of cluster size on the activity of the PtMn nanocat-
alysts for GLY oxidation is shown in Fig. 7(a). We observed a
peak activity value of 17,836.9 h−1 on PtMn1 (Pt/Mn: 1/1 molar
ratio) clusters, dropping significantly for PtMn1.5 (9911 h−1) and
PtMn2 (3231 h−1) catalysts. Concentration-time profiles of GLY
oxidation over PtMn1 catalyst [Fig. 7(b)] show that GLY concen-
tration dropped from 0.43 kmol/m3 to zero within 6 h with glyceric
acid (GLYA), the primary GLY oxidation product, being the domi-
nant one. Interestingly, the GLYA concentration decreases gradually
with increasing concentration of tartronic acid (TAR), glycolic acid
and oxalic acid, implying that secondary oxidation (GLYA to TAR)
and C C cleavage occur after GLY was consumed. The PtMn1
catalyst outperforms most investigated nanocatalysts reported in
the literature (see Table S2 in supporting information). Although
the activity of PtMn1 catalyst is slightly lower than the recently
reported activity for bimetallic PtFe catalysts (TOF: 20,978 h−1), the
selectivity towards TAR (53% on PtMn1), a valuable dicarboxylic
acid product, is higher than observed on PtFe catalysts (42%).
As already shown in Fig. 5, the addition of H2 and BA has
been found to influence the morphology of the bimetallic PtMn
clusters. While the observed activity of PtMn1(H) and PtMn1(BA)
catalysts is less than the activity of PtMn1 catalyst, they also exhibit
enhanced TOFs [see Fig. 7(c) and Table S1 for details] compared to
the monometallic Pt catalyst. The observed oxidation activity for
other bio-derived alcohols and polyols such as EG and EtOH reveal
that as the carbon number decreases from 3 to 2, the reactivity of
polyols decreases. Furthermore, we also observed that reactivity of
GLY and EG is higher than that of PDO and EtOH. This difference
implies that substrates with more electron withdrawing groups
(higher oxygen content) exhibit relatively higher reactivity during
aqueous phase oxidation reactions.
3.5. Catalytic oxidation of biomass-derived substrates to
carboxylic acids
Recycle tests were also carried out to study the durability of
PtMn1 catalyst during oxidation of GLY. Results in Table 1 show
good stability of PtMn1 catalyst in terms of activity and selectivity.
It is also found that the total metal leaching for Pt and Mn after
two recycles is 0.50% and 3.3%, indicating good stability of PtMn1
catalyst under reaction conditions.
We studied how the novel PtMn cluster structures influence the
catalytic oxidation of biomass-derived feedstocks. Specifically, oxi-
dation of glycerol (GLY, a C3 polyol), 1,2-propanediol (PDO, a C3
polyol), ethanol (EtOH, a C2 alcohol) and ethylene glycol (EG, a C2
polyol) were investigated on selected PtMn catalysts to understand
the oxidation activity trends for polyols and alcohols. Catalytic oxi-
dation reactions were carried out under very mild conditions (70 ◦C,
1 atm O2) with carboxylic acids being the target products.
3.6. Lattice distortion induced reconfiguration of electrons
The results obtained from the experimental conversion studies,
in conjunction with TEM characterization on bimetallic nanocrys-
tals, provide fundamental insights into how lattice distortion
Using the methodology described elsewhere [46], we estimated
that gas-liquid (for O2 molecule), liquid-solid and intraparticle (O2

54
X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
Fig. 7. Catalytic performances of PtMn catalysts for GLY oxidation. (a) composition-dependent activity, (b) a concentration-time profile in the presence of PtMn1 catalyst
at 70 ◦C and 0.1 MPa O2, (c) catalytic performance comparison of mono Pt and bimetallic PtMn1(H), PtMn1(BA) and PtMn1 (TAR yield obtained after 24 h reaction time), (d)
oxidation of other bio-derived alcohols in the presence of PtMn1 catalyst.
Table 1
Recycle tests of PtMn1 catalyst for GLY oxidation at 70 ◦C.
PtMn1 catalyst
TOF (h−1
)
Selectivity after 24 h reaction (%)
Metal leaching (%)
GLYA
TAR
Others
Fresh
1st recycle
2nd recycle
17836.9
18100.1
16966.7
24.9
22.1
26.4
52.7
55.8
51.2
20.1
18.1
19.9
Pt: 0.5%
Mn: 3.3%
Refer “Experimental Section” for other experimental conditions. 100% conversion were obtained for all tests.
influences surface morphologies and thereby catalytic performance
during aqueous phase oxidation over supported bimetallic nan-
oclusters. The mechanism of how lattice strain leads to electron
reconstruction and synergistic enhancement in catalytic perfor-
mance is yet to be fully understood. Therefore, several bimetallic
PtMn catalyst samples were characterized using XPS to reveal the
surface metal valence state, metal–metal coupling effect and possi-
ble reconfiguration of surface electrons. It is important to mention
that each sample was repeatedly scanned to ensure reproducibil-
ity of XPS spectra. In Fig. 8, XPS spectra of Pt, PtMn0.5, PtMn1 and
PtMn1.5 catalyst samples are presented in (a, b), (c–e), (f–h) and
(i–k), respectively. In particular, spectra in Fig. 8(a) and (b) were
obtained for monometallic Pt catalysts and used as reference for
bimetallic PtMn catalysts. Fig. 8(a) shows two characteristic peaks
for Pt4f5/2 and Pt4f7/2 at 71–75 eV of binding energy, while Ce3d5/2
spectra at 883 eV and 892 eV are shown in Fig. 8(b). In addition, it is
found from all bimetallic catalyst samples that Pt element exists in
metallic form while Mn in oxide (MnO) form [53,54]. In bimetallic
PtMn samples, as more Mn species were doped into Pt system, the
binding energy for Pt4f5/2 and Pt4f7/2 began to shift. Specifically,
the separated spin-orbits Pt4f5/2 and Pt4f7/2 exhibit shift towards
lower binding energy in PtMn0.5 sample [see Fig. 8(c)]. This shift

X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
55
Fig. 8. XPS spectra of (a, b) Pt, (c–e) PtMn0.5, (f–h) PtMn1, (i–k) PtMn1.5 catalyst samples (XPS spectra for Pt, Mn and Ce in blue, green and purple respectively). (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
becomes gradually more significant for PtMn1 [Fig. 8(f–h)] and
PtMn1.5 [Fig. 8(i–k)] samples. In contrast, we observed that Mn2p3/2
exhibits a slight shift to higher binding energy. Obviously the lat-
tice mismatch creates internal strain, leading to changes in binding
energies for Pt and Mn species.
Changed local environment on metal surface atoms indicates
that there might display an electronic coupling effect for Pt and
Mn atoms. The lattice distance for Pt cells was stretched while
that for Mn was contracted [39]. The changed lattice parameters
modify Pt-Mn bond length and induce formation of strained lay-
ers within bimetallic crystals (as schematically described in Fig. 1).
Such changes obviously alter the electronic structures of the two
metals by orbital overlap and electronic coupling [36]. Although,
a detailed mechanism is still unclear, the interaction between the
Fig. 9. Schematic description of possible surface reaction mechanism.
two types of species obviously results in higher electron density
on the Pt surface. This provides a plausible explanation for the
enhanced oxidation activity observed with bimetallic catalysts.
catalyst, Pt sites tend to attack side C atoms in polyol molecules
Quantitative XPS data showing that surface Pt/Mn atomic ratios are
[55], while adjacent O activated Pt sites tend to interact with
also dependent on bulk compositions, have been added to Table S3.
the C H bond [56]. On the other hand, on bimetallic PtMn
Clearly, such changes in the surface properties may influence sur-
catalyst surface, MnO is more easy to be activated due to rel-
face catalytic activity and selectivity. Based on XPS characterization,
atively higher oxidation potential than Pt [57]. This is because
it appears that the interaction between Pt and Mn might be the key
compressed Mn lattices leads to a broadened d-band along the
for synergistic catalytic activity in aqueous phase oxidation.
direction of strain (distortion) [39]. Theoretical predictions find
Based on the foregoing discussion, we propose the following
that adsorption of atomic oxygen on MnO surface could be a
surface reaction mechanism, a schematic description of which is
key step for catalytically breaking O H bond [58] by form-
shown in Fig. 9:
ing an intermediate ring on catalyst surface with adjacent Pt
atoms [59]. Higher surface electron density on Pt metal surface
is believed to facilitate activation of C O bond [60] to form C
bond.
O
(1) Monometallic Pt or bimetallic PtMn nanocrystals can be acti-
vated by molecular O2 on the surface. In monometallic Pt

56
X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
(2) Compared with monometallic Pt catalyst, bimetallic PtMn
nanocrystals may have slightly lower binding energy of
References
C
O
[1] X. Jin, L.N. Dang, J. Lohrman, B. Subramaniam, S.Q. Ren, R.V. Chaudhari, ACS
Nano 7 (2) (2013) 1309.
[2] W.J. Yan, A. Ramanathan, M. Ghanta, B. Subramaniam, Catal. Sci. Technol. 4
(12) (2014) 4433.
[3] W.J. Yan, A. Ramanathan, P.D. Patel, S.K. Maiti, B.B. Laird, W.H. Thompson, B.
Subramaniam, J. Catal. 336 (2016) 75.
[4] X.M. Lu, M. Rycenga, S.E. Skrabalak, B. Wiley, Y.N. Xia, Annu. Rev. Phys. Chem.
60 (2009) 167.
groups [35], thus secondary reactions such as aldehydes to
monocarboxylicacids as well as monocarboxylic to dicarboxylic
acids can occur more easily on bimetallic surface. The pres-
ence of Mn species has clearly improved activity and selectivity
of Pt catalysts. When no Mn is present, the strong interaction
between
C O group and Pt inhibit secondary oxidation reac-
[5] R.G. Chaudhuri, S. Paria, Chem. Rev. 112 (4) (2012) 2373.
[6] H.H. Li, C.H. Cui, S. Zhao, H.B. Yao, M.R. Gao, F.J. Fan, S.H. Yu, Adv. Energy
Mater. 2 (10) (2012) 1182.
[7] Y.J. Xiong, J.Y. Chen, B. Wiley, Y.N. Xia, S. Aloni, Y.D. Yin, J. Am. Chem. Soc. 127
(20) (2005) 7332.
[8] G.Q. Ren, Y.C. Xing, Nanotechnology 20 (46) (2009).
[9] Z.W. Quan, Y.X. Wang, J.Y. Fang, Acc. Chem. Res. 46 (2) (2013) 191.
[10] N.S. Porter, H. Wu, Z.W. Quan, J.Y. Fang, Acc. Chem. Res. 46 (8) (2013) 1867.
[11] B. Geboes, I. Mintsouli, B. Wouters, J. Georgieva, A. Kakaroglou, S.
Sotiropoulos, E. Valova, S. Armyanov, A. Hubin, T. Breugelmans, Appl. Catal.
B—Environ. 150 (2014) 249.
[12] I. Mintsouli, J. Georgieva, S. Armyanov, E. Valova, G. Avdeev, A. Hubin, O.
Steenhaut, J. Dille, D. Tsiplakides, S. Balomenou, et al., Appl. Catal. B—Environ.
136 (2013) 160.
[13] B.I. Podlovchenko, V.A. Krivchenko, Y.M. Maksimov, T.D. Gladysheva, L.V.
Yashina, S.A. Evlashin, A.A. Pilevsky, Electrochim. Acta 76 (2012) 137.
[14] A. Sarkar, A. Manthiram, J. Phys. Chem. C 114 (10) (2010) 4725.
[15] B.I. Podlovchenko, T.D. Gladysheva, A.Y. Filatov, L.V. Yashina, Russ. J.
Electrochem. 48 (2) (2012) 173.
[16] J.P. Lai, W.X. Niu, R. Luque, G.B. Xu, Nano Today 10 (2) (2015) 240.
[17] J. Gu, Y.W. Zhang, F. Tao, Chem. Soc. Rev. 41 (24) (2012) 8050.
[18] M. Wang, W.M. Zhang, J.Z. Wang, A. Minett, V. Lo, H.K. Liu, J. Chen, J. Mater.
Chem. A 1 (7) (2013) 2391.
[19] D.H. Chen, Y.C. Zhao, X.L. Peng, X. Wang, W.J. Hu, C. Jing, S.S. Tian, J.N. Tian,
Electrochim. Acta 177 (2015) 86.
[20] C. Gumeci, D.U. Cearnaigh, D.J. Casadonte, C. Korzeniewski, J. Mater. Chem. A 1
(6) (2013) 2322.
tions but promoting side reactions [low selectivity to tartronic
acid on monometallic Pt catalyst, Fig. 7(c)]. However, the inter-
action is weakened due to the addition of Mn content, which is
evident by enhanced secondary oxidation reaction and higher
tartronic acid selectivity [Fig. 7(b)]. Therefore, we conclude that
lower binding energy is a dominant factor for enhanced selec-
tivity towards tartronic acid, while stretched Pt lattice main
contribute to enhanced catalytic activity (higher d-band center
energy) [21,40,43]. Oxidation reactions consume surface oxy-
gen species thus catalytic sites can be available for subsequent
cycles.
(3) Interestingly, the overall reactivity of PDO and EtOH is lower
than GLY and EG. This suggests that the R groups in these
substrate molecules also affect their overall reactivity. Methyl
(CH3-) is an electron donating group and the existence of such a
group in the substrate molecules increases the electron density
of reactive C, therefore unfavorably preventing attack by Pt sites
during catalytic reactions. This mechanism is validated by XPS
characterization of Pt and PtMn catalysts, their catalytic perfor-
mances, and reactivity trends of polyols. Therefore, it is clear
that the addition of Mn in Pt system alters the lattice parame-
ters, resulting in electronic coupling and surface electronic thus
synergistically enhancing the catalytic properties.
[21] J.P. Lai, R. Luque, G.B. Xu, ChemCatChem 7 (20) (2015) 3206.
[22] F. Nosheen, Z.C. Zhang, J. Zhuang, X. Wang, Nanoscale 5 (9) (2013) 3660.
[23] R. Sakamoto, K. Omichi, T. Furuta, M. Ichikawa, J. Power Sources 269 (2014)
117.
[24] K.A. Kuttiyiel, K. Sasaki, Y.M. Choi, D. Su, P. Liu, R.R. Adzic, Nano Lett. 12 (12)
(2012) 6266.
[25] W.Y. Wang, D.S. Wang, X.W. Liu, Q. Peng, Y.D. Li, Chem. Commun. 49 (28)
(2013) 2903.
4. Conclusions
[26] H. Zhang, D.R. Lin, G.T. Xu, J.B. Zheng, N.W. Zhang, Y.H. Li, B.H. Chen, Int. J.
Hydrogen Energy 40 (4) (2015) 1742.
[27] Z.R. Dai, S.H. Sun, Z.L. Wang, Nano Lett. 1 (8) (2001) 443.
[28] J.Y. Li, G.X. Wang, J. Wang, S. Miao, M.M. Wei, F. Yang, L. Yu, X.H. Bao, Nano
Res. 7 (10) (2014) 1519.
[29] J. Yao, Y.F. Yao, Renew. Energy 81 (2015) 182.
[30] L. Chen, C. Bock, P.H.J. Mercier, B.R. MacDougall, Electrochim. Acta 77 (2012)
212.
[31] V.R. Stamenkovic, B.S. Mun, M. Arenz, K.J.J. Mayrhofer, C.A. Lucas, G.F. Wang,
P.N. Ross, N.M. Markovic, Nat. Mater. 6 (3) (2007) 241.
[32] N. Wakabayashi, M. Takeichi, H. Uchida, M. Watanabe, J. Phys. Chem. B 109
(12) (2005) 5836.
[33] J.E. Mondloch, E. Bayram, R.G. Finke, J. Mol. Catal. A: Chem. 355 (2012) 1.
[34] H. Liao, A. Fisher, Z.J. Xu, Small 11 (27) (2015) 3221.
[35] A. Holewinski, H. Xin, E. Nikolla, S. Linic, Curr. Opin. Chem. Eng. 2 (3) (2013)
312.
[36] E. Gross, G.A. Somorjai, Top. Catal. 56 (12) (2013) 1049.
[37] J.B. Wu, P.P. Li, Y.T. Pan, S. Warren, X. Yin, H. Yang, Chem. Soc. Rev. 41 (24)
(2012) 8066.
[38] A. Kirkeminde, S. Ren, Nano Lett. 14 (8) (2014) 4493.
[39] J. Wu, P. Li, Y.-T.F. Pan, S. Warren, X. Yin, H. Yang, Chem. Soc. Rev. 41 (24)
(2012) 8066.
[40] C.W. Xu, Y.Z. Su, L.L. Tan, Z.L. Liu, J.H. Zhang, S.A. Chen, S.P. Jiang, Electrochim.
Acta 54 (26) (2009) 6322.
In this paper, we report a unique approach for the synthe-
sis of novel structured PtMn nanocatalysts using lattice strain
technique, which have exceptionally high catalytic activity in oxi-
dation of polyols and alcohols. The formation of nuclei, growth of
nanoclusters and immobilization on heterogeneous support were
achieved in a one pot wet chemistry method. The influences of
Pt/Mn ratios, synthesis conditions and chemical promoters on sur-
face morphologies and catalytic properties of bimetallic PtMn were
studied systematically. It is found that the large lattice mismatch
between Pt and Mn metals leads to anisotropic growth while lat-
tice distortion of bimetallic crystals is believed to be the key for
synergistic enhancement of catalytic activity and selectivity. Aque-
ous phase oxidation of biomass-derived polyols was used as the
model reaction to reveal the correlation of crystal morphologies
with catalytic properties of bimetallic catalysts. The methodology
for catalyst synthesis proposed in this work provides a new tech-
nique for designing active and selective nanomaterials for biomass
as well as other renewable energy applications.
[41] Y.J. Kang, C.B. Murray, J. Am. Chem. Soc. 132 (22) (2010) 7568.
[42] M. Ammam, E.B. Easton, J. Power Sources 236 (2013) 311.
[43] M. Ammam, E.B. Easton, J. Electrochem. Soc. 160 (2) (2013) F212.
[44] X. Jin, M. Zhao, J. Shen, W.J. Yan, L.M. He, P.S. Thapa, S.Q. Ren, B. Subramaniam,
R.V. Chaudhari, J. Catal. 330 (2015) 323.
Acknowledgement
[45] S.T. Oyama, X.M. Zhang, J.Q. Lu, Y.F. Gu, T. Fujitani, J. Catal. 257 (1) (2008) 1.
[46] X. Jin, M. Zhao, W. Yan, C. Zeng, P. Bobba, P.S. Thapa, B. Subramaniam, R.V.
Chaudhari, J. Catal. 337 (2016) 272.
This work was partially supported by NSF/EPA grant CHE-
1339661.
[47] K. Ono, R. Okuda, Y. Ishii, S. Kamimura, M. Oshima, J. Phys. Chem. B 107 (9)
(2003) 1941.
[48] S.H. Sun, C.B. Murray, D. Weller, L. Folks, A. Moser, Science 287 (5460) (2000)
1989.
Appendix A. Supplementary data
[49] H.M. Song, W.S. Kim, Y.B. Lee, J.H. Hong, H.G. Lee, N.H. Hur, J. Mater. Chem. 19
(22) (2009) 3677.
[50] L. Gan, C.H. Cui, M. Heggen, F. Dionigi, S. Rudi, P. Strasser, Science 346 (6216)
(2014) 1502.
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.apcata.2017.01.
021.

X. Jin et al. / Applied Catalysis A: General 534 (2017) 46–57
57
[51] C.H. Cui, L. Gan, H.H. Li, S.H. Yu, M. Heggen, P. Strasser, Nano Lett. 12 (11)
[57] G.F. Nordberg, B.A. Fowler, M. Nordberg, Handbook on the Toxicology of
Metals, Academic Press, 2014.
(2012) 5885.
[52] H. You, S. Yang, B. Ding, H. Yang, Chem. Soc. Rev. 42 (7) (2013) 2880.
[53] M.S. El-Deab, T. Ohsaka, Angew. Chem. Int. Ed. 45 (36) (2006) 5963.
[54] A.E. Vilian, M. Rajkumar, S.-M. Chen, C.-C. Hu, K.M. Boopathi, C.-W. Chu, RSC
Adv. 4 (78) (2014) 41387.
[55] B. Liu, J. Greeley, J. Phys. Chem. C 115 (40) (2011) 19702.
[56] M. Salciccioli, W.T. Yu, M.A. Barteau, J.G.G. Chen, D.G. Vlachos, J. Am. Chem.
Soc. 133 (20) (2011) 7996.
[58] H.-Y. Su, Y. Gorlin, I.C. Man, F. Calle-Vallejo, J.K. Nørskov, T.F. Jaramillo, J.
Rossmeisl, Phys. Chem. Chem. Phys. 14 (40) (2012) 14010.
[59] J. Trawczyn´ ski, B. Bielak, W. Mis´ta, Appl. Catal. B: Environ. 55 (4) (2005) 277.
[60] M.U. Khan, L. Wang, Z. Liu, Z. Gao, S. Wang, H. Li, W. Zhang, M. Wang, Z. Wang,
C. Ma, Angew. Chem. (2016).