Article

pubs.acs.org/cm

Oxidative Etching of Pd Decahedral Nanocrystals with a Penta-

twinned Structure and Its Impact on Their Growth Behavior
Aleksey Ruditskiy,† Madeline Vara,† Hongwen Huang,‡ and Younan Xia*,†,‡
†
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332, United States
‡
The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta,
Georgia 30332, United States
*
S Supporting Information

ABSTRACT: We report a systematic study of the oxidative

etching of penta-twinned Pd decahedral nanocrystals by O2/I−
under different conditions and its impact on their subsequent
growth behavior. Analysis by transmission electron microscopy
shows significant rounding of the decahedral structure. More
specifically, the etching is found to begin at the equatorial vertices,
due to their high surface free energy, and proceed along the
adjacent, equatorial edges through the dissolution of low-
coordination atoms. Comparison of the etching behaviors under
different conditions reveals the critical role of a reductive
environment for the initiation of oxidative etching, possibly due
to the presence of a protective oxide layer on the surface of Pd
decahedra. Overgrowth on the seeds with a rounded profile
generates penta-twinned Pd nanorods with an asymmetric, tapered structure as a result of simultaneous axial and radial growth.
In comparison, the original decahedral seeds only show axial growth, leading to the formation of penta-twinned nanorods with a
uniform size along the axial direction. A good understanding of the etching and growth behaviors of Pd decahedral nanocrystals
will be useful for the successful adoption of these nanomaterials in real-world applications, including their use as catalysts and as a
platform for the development of more complex nanostructures.

■ INTRODUCTION
Oxidative etching is a commonly occurring electrochemical
and co-workers used in situ transmission electron microscopy
(TEM) to investigate the oxidative etching of various Au
process, which is exemplified by the formation of iron rust in nanocrystals inside a graphene pocket.16 Notably, the authors
the presence of water and air. When applied to metallic observed a nonequilibrium intermediate during the etching of a
nanocrystals, the power of this process can be harnessed to Au nanocube, before the intermediate was converted to the
manipulate the properties of nanocrystals, including their size, expected spherical shape. In principle, one would expect to
shape, internal structure, and composition.1,2 This is typically glean more information when moving from single-crystal
achieved by employing an etchant pair, such as O2/halide, to nanocubes to nanocrystals with more complex internal twin
systematically remove atoms from a nanocrystal. Over the past structures. To this end, Pd decahedral nanocrystals would
decade or so, this approach has resulted in metal nanocrystals provide a rich platform for the study of oxidative etching due to
with a diverse array of morphologies, including octapods,3,4 the presence of a diversified set of surface features.8,17 These
nanocages,5−8 and nanoframes.9−11 The preferential etching of include the presence of twin boundaries, different types of
multitwinned products has facilitated the preparation of single- facets, and multiple subtypes of edges and vertices. Such a wide
crystal and single-twinned Ag nanocrystals with high purity.12,13 variety of surface features potentially allows for the
Moreover, the coupling of oxidative etching with reduction- identification of sites most prone to oxidative etching, as well
based regrowth has allowed for the creation of a variety of as a systematic study of etching in terms of initiation and
structures from a single type of Pd nanocrystal.14 A related propagation. Furthermore, Pd decahedra have shown potential
approach has also been used to create Ag−Au alloyed as electrocatalysts,18 as well as a platform material for the
nanostructures with hollow interiors, which would be otherwise creation of more complex structures, such as core−shell
difficult to fabricate.15 nanocrystals7 and pentagonal nanorods,19−21 with notable
Because metal nanocrystals are typically synthesized in an catalytic and optical properties. Taken together, it can be
oxygenated environment, and often from halide-based
precursors, oxidative etching is inevitable. As such, it would Received: May 1, 2017
be sensible to systematically explore its impacts on the physical Revised: May 27, 2017
properties of metal nanocrystals. In a recent study, Alivisatos Published: May 30, 2017

© 2017 American Chemical Society 5394 DOI: 10.1021/acs.chemmater.7b01776

Chem. Mater. 2017, 29, 5394−5400
Chemistry of Materials Article

concluded that understanding the effects of oxidative etching collected precipitate was redispersed in DI water and centrifuged three
on decahedral nanocrystals is critical to their effective use in times at 16 500 rpm for 25 min.
real-world applications. Instrumentation. TEM images were obtained on a Hitachi HT-
Here we report a systematic study of oxidative etching of Pd 7700 microscope operated at 120 kV. HRTEM and HAADF-STEM
images were collected with a Hitachi HD-2700 operated at 200 kV. An
decahedra by the O2/I− etchant pair. This pair was chosen Eppendorf centrifuge (5430) was used for the centrifugation and
owing to its ability to effectively etch Pd nanocrystals at a washing of all samples. X-ray photoelectron spectroscopy (XPS)
controllable rate,14 allowing us to analyze the etching process measurements were carried out using a spectrometer (Thermo
over a reasonable time frame. The etched samples were Scientific, KAlpha) with Al Kα X-ray (1486.6 eV) as the light source.
characterized using TEM, high-resolution TEM (HRTEM),
and high-angle annular dark-field scanning transmission
electron microscopy (HAADF-STEM) to ascertain the
■ RESULTS AND DISCUSSION
Figure 1 shows the effects of O2/I−-based oxidative etching on
initiation and propagation of etching over time. The effects of Pd decahedral nanocrystals. Figure 1A shows a TEM image of
the etching environment were also explored by conducting
experiments in diethylene glycol (DEG) and deionized (DI)
water, in conjunction with the introduction of various types of
additives. Finally, we demonstrated the use of oxidative etching
to modulate the growth behavior of Pd decahedral seeds in
generating nanocrystals with more complex structures.

■ EXPERIMENTAL SECTION
Chemicals and Materials. DEG (≥99.0%, lot no. BCBP6902V),
poly(vinylpyrrolidone) (PVP, MW ≈ 55 000), sodium sulfate
(Na2SO4, ≥99.0%), sodium iodide (NaI, ≥99.5%), sodium chloride
(NaCl, ≥99.0%), iron(III) chloride (FeCl3, 97%), and ascorbic acid
(AA, ≥99.0%) were all procured from Sigma-Aldrich. Sodium
tetrachloropalladate (Na2PdCl4, ≥99.0%) was obtained from Acros
Organics. All chemicals were used as received. DI water with a
resistivity of 18.2 MΩ·cm was used throughout the experiment.
Synthesis of Pd Decahedra. In a typical synthesis, 174 mg of
Na2SO4 was added into a 50 mL three-neck flask. Next, 8 mL of DEG-
based PVP solution (a colloidal stabilizer, 20 mg/mL) was introduced,
along with a 6 × 25 mm polygonal, Teflon-coated stirring bar. The
flask was then placed in an oil bath set to 117.5 °C under magnetic
stirring (450 rpm). Note that, at this oil bath temperature, the
temperature of the reaction solution was only measured to be 103.3
°C. Immediately after the stirring began in the flask, 4 mL of a DEG- Figure 1. TEM and HAADF-STEM images of Pd decahedral
based Na2PdCl4 solution (a precursor to elemental Pd, 15.5 mg/mL) nanocrystals (A, B) before and (C, D) after oxidative etching with
was introduced and the timer was started. The reaction was terminated O2/I− in DEG at 90 °C for 30 min. The etched nanocrystals show a
after 3 h by immersing the flask in an ice bath. The product was significantly rounded profile while retaining the penta-twinned
collected by adding acetone in a 3:1 ratio to the reaction solution and structure. The red arrows denote examples of initial decahedra (A)
centrifuging the mixture at 6 000 rpm for 10 min. The precipitate was and rounded particles (C) imaged from the side.
redispersed in DI water, centrifuged three times at 16 000 rpm for 25
min, and finally stored in DI water.
Etching of Pd Decahedra. In a standard procedure, 52.5 mg of the initial Pd decahedra used for the etching studies, which had
PVP and 10 mg of NaI were dissolved in 3.5 mL of DEG. The an average size of ∼17 nm, defined as the distance (d) between
resultant solution was placed in a 20 mL dram vial with a 10 × 15 mm two vertices of a decahedron (see Figure S1). For clarity, we
polygonal stirring bar. The vial was loosely closed and then heated in will henceforth refer to the five in-plane vertices as “equatorial”
an oil bath set to 90 °C for 10 min under stirring (450 rpm). After the and the two remaining vertices as “axial”. The red arrows in
preheating was complete, 0.5 mL of the Pd decahedra suspension Figure 1A denote an example of a decahedral nanocrystal
(1.35 mg of Pd/mL) was injected with a pipet and the vial was capped imaged from the side. Figure 1B shows the HAADF-STEM
again. The etching was allowed to proceed for 30 min and was
terminated by immersion in an ice bath. The product was collected by
image of an initial decahedron, highlighting the flat edges
adding acetone in a 3:1 ratio to the reaction solution and centrifuging running along the equator of the structure, as well as the slight
at 6 000 rpm for 10 min. The precipitate was redispersed in DI water, truncation at the equatorial vertices. Exposing the decahedra to
centrifuged three times at 16 000 rpm for 25 min, and then stored in O2/I− in DEG for 30 min at 90 °C yielded particles with a
DI water. notably rounded profile, compared to the initially well-defined
Synthesis of Tapered, Penta-Twinned Pd Nanorods. In a decahedra, as shown in Figure 1C and D. Moreover, the
typical process, 7 mL of DEG containing 105 mg of PVP, 100 mg of rounded particles that were imaged from the side (for example,
AA, and 100 mg of NaI was added into a 20 mL vial equipped with a denoted by the red arrows in Figure 1C) showed evidence of
10 × 15 mm polygonal, Teflon-coated stirring bar. The vial was then noticeable rounding at the axial vertices as well. Crucially, the
capped and placed in an oil bath held at 160 °C under magnetic
penta-twinned structure of the starting nanocrystals was clearly
stirring (450 rpm). After 10 min, 1 mL of rounded particles in DEG
was added, immediately followed by the introduction of 3 mL of a preserved. After the etching, the average size of the rounded
DEG solution of Na2PdCl4 (10 mg/mL). The reaction was allowed to particles decreased by 2−3 nm, although the size polydispersity
proceed for 1 h and terminated by immersing the vial in an ice-water of the initial decahedra made an accurate comparison difficult.
bath. The product was collected by adding acetone in a 3:1 ratio to the As a control experiment, Ar gas was bubbled through the DEG
reaction solution and centrifuging at 6 000 rpm for 10 min. The solutions prior to heating. The resultant particles maintained a
5395 DOI: 10.1021/acs.chemmater.7b01776
Chem. Mater. 2017, 29, 5394−5400
Chemistry of Materials Article

decahedral shape, as shown in Figure S2, thereby confirming

the need for the presence of O2 gas to initiate and drive the
oxidative etching process.
Interestingly, increasing the reaction time and/or the
temperature did not produce a noticeable difference in particle
size or structure. As shown in parts A and B of Figure 2,

Figure 3. (A) Higher-resolution TEM and (B) HAADF-STEM images

of rounded Pd decahedral nanocrystals. (C, D) Magnified HRTEM
and HAADF-STEM images, respectively, taken from the equatorial
edge regions of the rounded Pd decahedron shown in (B). The red
lines denote the outlines and twin boundaries of the decahedron. The
extent of truncation in the outline was estimated by benchmarking to
the image shown in Figure 1B. The white lines in (D) highlight the
formation of a new vertex. A version of this figure without the outlines
Figure 2. TEM images showing the effects of increasing temperature can be found in Figure S2.
and reaction time on the etching of Pd decahedral nanocrystals. (A) 90
°C, 60 min, (B) 90 °C, 90 min, (C) 120 °C, 60 min, and (D) 120 °C,
90 min. All the etching experiments were conducted in DEG. The oxidative etching process utilized in this study can be
summarized as an electrochemical reaction between the Pd0
atoms on the particle surface and the O2 gas molecules
dissolved in the DEG, with the I− anions serving as
maintaining the temperature at 90 °C and increasing the coordination ligands and charge carriers.1 More specifically,
reaction time to 60 and 90 min, respectively, resulted in the O2 gas is reduced to oxide species, while the Pd0 is
rounded particles with sizes of 15 nm. Similar products were consequently oxidized to Pd2+. The metal cations are then
obtained after increasing the reaction temperature to 120 °C rapidly complexed with I− to form the highly stable PdI42−.22
and subjecting the decahedra to etching for 60 and 90 min, as This species is difficult to reduce back into Pd0, thus reducing
shown in parts C and D of Figure 2, respectively. These results the concentration of free Pd2+ and driving the etching reaction
suggest that the rate of etching was slowed once the initial forward. In general, the susceptibility of specific nanocrystal
rounding of the decahedra was complete. surface sites to oxidative etching depends on their respective
To gain a better understanding of the morphology of the surface free energies, with higher energy sites being more prone
rounded particles, further characterization was performed using to dissolution. In turn, the free energy of an atom is determined
TEM, HRTEM, and HAADF-STEM, as shown in Figure 3. by the number of bonds connecting that atom with its nearest
TEM imaging (Figure 3A) revealed significant rounding at the neighbors, as well as the extent of lattice strain those bonds
equatorial vertices of the initial decahedra. Furthermore, new experience. The former is known as the coordination number
vertices could be clearly observed midway between the twinned (CN), with the free energy of an atom increasing as the CN
boundaries, although not all of the particles showed this feature. decreases. Lattice strain can be described as the expansive or
The magnitude of the rounding can be clearly seen in the compressive distortion in bond length, relative to the ideal bulk
HAADF-STEM image (Figure 3B). The red lines represent the value, which results in higher free energies for the atoms
outlines of a regular decahedron with truncated equatorial involved.23 This phenomenon is typically caused by tensions at
vertices, aligned to the twin boundaries of the imaged particle. the interface between two different phases within a crystal, such
The extent of truncation in the outline was estimated using as those found at the boundaries between different facets or
measurements taken from Figure 1B. These images indicate twin defect planes.
that the etching process preferentially removed Pd atoms from By combining these basic principles with the TEM
the equatorial vertices and the adjacent, equatorial edges. A observations, we can posit a mechanism to account for the
closer look at the equator of the rounded particles (Figure 3C etching of Pd decahedra. A schematic illustration is shown in
and D) showed a smooth atomic edge with hexagonal packing, Figure 4. For a truncated regular decahedron, the atoms with
as well as evidence of a new vertex denoted by white lines in the lowest CN are located at the equatorial (CN = 4) and axial
Figure 3D. Electron microscopy images without the overlays (CN = 5) vertices, followed by those found at the {100} edges
can be found in Figure S3. (CN = 6).17 The {211}-enclosed twinned ridges separating the
5396 DOI: 10.1021/acs.chemmater.7b01776
Chem. Mater. 2017, 29, 5394−5400
Chemistry of Materials Article

Figure 4. Schematic illustration of the etching process, which begins

from the equatorial vertices of a decahedron due to their highest
energy. The I− anions rapidly complex with the dissolved Pd2+ to
generate highly stable PdI42−, thus driving the etching reaction
forward. The dissolution then proceeds by removing the low-
coordination atoms along the equatorial edges of the particle until a
rounded particle is formed.

{111} facets experience increasing expansive strain with

increasing radial distance from the decahedral center.8 Addi-
tionally, the I− ions have been previously shown to selectively
adsorb to the {100} surfaces of Pd decahedra, located at the
equatorial edges.21 The combination of these factors suggests
that the etching process should begin at the equatorial vertices.
Moreover, the rounding of the axial vertices seen in Figure 1C
suggests that these sites are likewise susceptible to dissolution
by the O2/I− etchant. The Pd atoms present in the twinned
ridges likely resist dissolution, despite their strain-induced high
free energy, due to comparatively higher CN values (CN = 7−
8), thus preserving the penta-twinned structure. The removal of
Pd atoms during the etching of the equatorial vertices reduces
the CN values for the adjacent edge atoms, making them highly
susceptible to dissolution. As such, the etching proceeds
symmetrically away from the equatorial vertices along the
equatorial edges until the new vertices form midway between Figure 5. TEM images of Pd decahedral nanocrystals obtained by
the twinned ridges. oxidative etching with O2/I− in DI water. (A) Replacement of DEG
Because the DEG possesses inherent reducing capability, it with DI water in the standard procedure. (B, C) Addition of 2 mg/mL
would be prudent to study the etching process in a more inert of Cl− and Fe3+ to the water-based etching solution, respectively. (D−
medium. To that end, the standard procedure was modified to F) Addition of 60, 100, and 300 mg of AA to the water-based etching
use DI water as the solvent, instead of DEG. These experiments solution, respectively. All experiments were conducted at 90 °C for 30
were performed with the same batch and concentration of the min, using the same batch of Pd decahedra at the same concentration.
initial Pd decahedral nanocrystals, as well as the same batches
of all chemicals involved. Both reactions were conducted at 90 °C, the Ostwald coefficient of O2 in water (0.0347) is found to
°C for 30 min. While the etching of decahedra performed in be ∼14.5% higher than that of EG (0.0303).24 Because gas
DEG produced the expected rounded particles (Figure 1C), solubility decreases with increasing temperature, the O2
etching in an aqueous environment resulted in particles with solubility is expected to decrease substantially in both solvents
well-preserved decahedral features (Figure 5A). This result is at 90 °C. As such, the higher O2 solubility in water, compared
counterintuitive, because the reducing capability of DEG is to EG, suggests that the difference in O2 gas concentration
expected to counteract the etching process. Thus, the products cannot explain the unusual etching behavior. A second possible
collected from the water-based reaction were expected to show explanation lies in the other half of the etchant pair: the ion.
evidence of more extensive dissolution, when compared to the DEG is known to contain trace impurities, present as
decahedra etched in DEG. There exist several potential reasons byproducts of the chemical manufacturing and storage
that may explain the disparate etching behaviors in DEG and processes. According to the manufacturer’s assay of the specific
water. First, there exists a difference in O2 gas solubility DEG batch used for these experiments, there was a measurable
between the two solvents. Because the rate of the etching concentration of Cl− (∼2 mg/kg) and Fe3+ (∼2 mg/kg)
reaction depends on the concentration of the dissolved O2 gas, dissolved in the solvent. Both of these ions are known as strong
a low solubility of O2 in a solvent may result in depressed rates etchants and would be absent in the filtered DI water (with a
of etching, and vice versa. While there is a lack of data on the resistivity of 18.2 MΩ·cm).1 To test the impact of these
solubility of O2 in DEG, we can attempt to approximate it by impurities, equivalent concentrations of the ions in question
comparing the Ostwald coefficients for water and ethylene were added to the DI water, in addition to the standard amount
glycol (EG), a similar molecule also used as a reducing solvent. of I−, by using NaCl and FeCl3, respectively. Parts B and C of
The Ostwald coefficient is defined as the ratio between the Figure 5 show the results of the corresponding etching studies.
volume of dissolved gas and the volume of the solvent. At 25 In both cases, the pentagonal shape of the initial decahedra was
5397 DOI: 10.1021/acs.chemmater.7b01776
Chem. Mater. 2017, 29, 5394−5400
Chemistry of Materials Article

still preserved, indicating that the impurities were not

responsible for the observed differences in the etching products
between DEG and water.
Because the reducing capability of DEG is one of the primary
differences between the two solvents, it would be sensible to
explore its impact on the etching process. To that end, we
introduced varying amounts of ascorbic acid (AA), a reducing
agent, into the water-based etching solution. The experiments
were conducted at 90 °C for 30 min. It should be noted that
the reducing strength of a glycol increases with continued
application of heat while in air, due to the oxidation of the
glycol molecules into aldehyde derivatives.25 For 60 mg of AA
added, the collected products largely consisted of decahedra, as
shown in Figure 5D. Addition of 100 mg of AA (Figure 5E)
resulted in a mixture of decahedra and rounded particles.
However, when the amount of AA was increased to 300 mg
(Figure 5F), the products were largely comprised of rounded
particles with distinct penta-twinned structures. These particles
appear identical to those produced with DEG as the solvent,
such as those shown in Figure 1C. Notably, the average size of
the particles shown in Figure 5F was 2−3 nm smaller than the
initial decahedra, which was consistent with the etching results
obtained in DEG.
These results suggest that a sufficiently reducing environ- Figure 6. TEM images of Pd (A) standard and (B) tapered nanorods
ment is crucial to the oxidative etching process. An explanation obtained through seeded growth from the original and rounded Pd
decahedra, respectively. Insets show TEM images of the corresponding
for this counterintuitive phenomenon may lie in the methods seeds (scale bars: 40 nm). Red arrows in (B) highlight tapered rods
used for the cleaning and storage of the initial Pd decahedral that stopped growing radially. (C, D) HAADF-STEM images of the
nanocrystals. The decahedra were prepared, cleaned, and stored end and side of a tapered nanorod, respectively.
under exposure to air.26 Furthermore, DI water was used to
both purify and store the final nanocrystals. Exposure to such
oxidative environments likely caused an oxide PdO layer to
form on the particle surfaces.27 As with bulk-scale oxidative shown in Figure 6B. The lengths of tapered nanorods were
etching, an oxide layer on the surface can inhibit further measured to be 22 nm on average, compared with 30 nm for
oxidation of a structure. On the basis of this principle, we the standard nanorods. HAADF-STEM characterization of the
hypothesize that a sufficiently strong reducing environment is ends (Figure 6C) and the sides (Figure 6D) of the rods
necessary to remove the oxide layer, thereby making the revealed that the penta-twinned structure was preserved
nanocrystal susceptible to oxidative etching. We attempted to throughout the nanorod after the Pd outgrowth. The sloped
validate this hypothesis by characterizing the initial Pd sides of the tapered rod were likely enclosed by high index
decahedra, as well as the products obtained from the DEG facets, rather than the {100} facets found on the sides of the
and aqueous etching systems by XPS, as shown in Figure S4. standard nanorods. Importantly, Figure 6C shows that the
The decahedra were etched for 30, 60, and 90 min in both larger end of the tapered rod contained well-defined decahedral
systems. The results confirmed the presence of PdO in all of features, which were much sharper than those of the initial Pd
the samples tested. Therefore, it was impossible to ascertain for decahedra used for the etching studies (Figure 1B). This
certain that the oxide layer was removed during the etching information suggests that overgrowth occurred not only in the
process without switching to an in situ technique. This result axial direction but also radially along the etched equatorial
has implications for general metal nanocrystal synthesis and edges of the rounded particles. Previous seeded growth
storage, because it is commonly accepted that a reducing syntheses of Pd−Ag and Pd−Cu nanorods,19,20 utilizing Pd
environment is favorable to preventing nanocrystal degradation decahedra as seeds, showed that the axial growth proceeds from
over time. Instead, it may be more sensible to expose the newly one side of a decahedron, i.e., one set of five {111} surfaces.
prepared nanocrystals to a mildly oxidative environment in Therefore, the initial seed remains at one of the ends of the
order to form an oxide shell and protect against further metal final nanorod. In the case of the tapered rods, the axial growth
dissolution. initially occurs on one of the available {111} surface sets of the
Finally, it would be interesting to explore the utility of the rounded particle. The radial growth along the dissolved edge
rounded particles as platforms for the seeded growth of more occurs simultaneously, steadily adding new atomic sites for
complex structures. To achieve this, a recently developed seed- further axial growth. However, because these new sites begin
mediated synthesis of Pd pentagonal nanorods was adapted to axial growth after the initially available {111} surfaces, the result
use the rounded particles as growth surfaces.21 Standard is the formation of a radial gradient along the length of the rod
nanorods, produced through the seeded growth of the initial as it elongates, i.e., tapering. This gradient disappeared when
decahedra, are shown in Figure 6A. These structures are the radial growth terminated upon the reformation of a sharp
symmetric along their length and terminated by rounded ends. decahedral configuration (Figure 6C), resulting in constant rod
Employing the rounded particles as seeds resulted in the diameters at the wider end of the tapered structure. This can be
formation of asymmetric, tapered nanorods, with notably seen in the longer tapered rods, marked with red arrows in
smaller diameters at one end of the structure than the other, as Figure 6B.
5398 DOI: 10.1021/acs.chemmater.7b01776
Chem. Mater. 2017, 29, 5394−5400
Chemistry of Materials Article

■ CONCLUSION
In summary, we have carried out a systematic study to
(3) Mulvihill, M. J.; Ling, X. Y.; Henzie, J.; Yang, P. Anisotropic
Etching of Silver Nanoparticles for Plasmonic Structures Capable of
Single-Particle SERS. J. Am. Chem. Soc. 2010, 132, 268−274.
understand the oxidative etching of Pd decahedral nanocrystals (4) Zhang, H.; Jin, M.; Wang, J.; Li, W.; Camargo, P. H. C.; Kim, M.
by the O2/I− pair under different conditions. Electron J.; Yang, D.; Xie, Z.; Xia, Y. Synthesis of Pd−Pt Bimetallic
microscopy analyses of the etched particles revealed significant Nanocrystals with a Concave Structure through a Bromide-Induced
rounding of the penta-twinned decahedra due to the removal of Galvanic Replacement Reaction. J. Am. Chem. Soc. 2011, 133, 6078−
atoms from the equatorial vertices and edges. It was found that 6089.
the etching began from the equatorial vertices and then (5) He, D. S.; He, D.; Wang, J.; Lin, Y.; Yin, P.; Hong, X.; Wu, Y.; Li,
proceeded along the equatorial edges by removing the low- Y. Ultrathin Icosahedral Pt-Enriched Nanocage with Excellent Oxygen
coordination Pd atoms until a rounded structure was attained. Reduction Reaction Activity. J. Am. Chem. Soc. 2016, 138, 1494−1497.
The etching behaviors in DEG and DI water differed (6) Zhang, L.; Roling, L. T.; Wang, X.; Vara, M.; Chi, M.; Liu, J.;
significantly, with the former producing rounded particles and Choi, S.-I.; Park, J.; Herron, J. A.; Xie, Z.; Mavrikakis, M.; Xia, Y.
the latter preserving the initial shape. The addition of a Platinum-based Nanocages with Subnanometer-thick Walls and Well-
defined, Controllable Facets. Science 2015, 349, 412−416.
sufficient amount of AA into the water-based system resulted in
(7) Wang, X.; Figueroa-Cosme, L.; Yang, X.; Luo, M.; Liu, J.; Xie, Z.-
the formation of rounded structures, revealing the need for an X.; Xia, Y. Pt-based Icosahedral Nanocages: Using a Combination of
adequately reducing environment in order to initiate oxidative {111} Facets, Twin Defects, and Ultrathin Walls to Greatly Enhance
etching. This phenomenon can be attributed to the presence of Their Activity toward Oxygen Reduction. Nano Lett. 2016, 16, 1467−
a protective oxide layer on the surface of the Pd decahedra, 1471.
which had to be removed first through reduction before etching (8) Wang, X.; Vara, M.; Luo, M.; Huang, H.; Ruditskiy, A.; Park, J.;
could occur. Finally, we utilized the rounded particles as seeds Bao, S.; Liu, J.; Howe, J.; Chi, M.; Xie, Z.; Xia, Y. Pd@Pt Core−Shell
for the growth of pentagonal Pd nanorods with asymmetric, Concave Decahedra: A Class of Catalysts for the Oxygen Reduction
tapered structures, due to the involvement of both axial and Reaction with Enhanced Activity and Durability. J. Am. Chem. Soc.
radial growth. 2015, 137, 15036−15042.

■
(9) McEachran, M.; Keogh, D.; Pietrobon, B.; Cathcart, N.;
ASSOCIATED CONTENT Gourevich, I.; Coombs, N.; Kitaev, V. Ultrathin Gold Nanoframes
through Surfactant-Free Templating of Faceted Pentagonal Silver
*
S Supporting Information Nanoparticles. J. Am. Chem. Soc. 2011, 133, 8066−8069.
This material is available free of charge at The Supporting (10) Chen, C.; Kang, Y.; Huo, Z.; Zhu, Z.; Huang, W.; Xin, H. L.;
Information is available free of charge on the ACS Publications Snyder, J. D.; Li, D.; Herron, J. A.; Mavrikakis, M.; Chi, M.; More, K.
website at DOI: 10.1021/acs.chemmater.7b01776. L.; Li, Y.; Markovic, N. M.; Somorjai, G. A.; Yang, P.; Stamenkovic, V.
R. Highly Crystalline Multimetallic Nanoframes with Three-dimen-
Schematic showing the definition of size of a decahedral
sional Electrocatalytic Surfaces. Science 2014, 343, 1339−1343.
nanocrystal; TEM image of the etching products (11) Wang, X.; Ruditskiy, A.; Xia, Y. Rational Design and Synthesis of
obtained under Ar gas protection; HRTEM and Noble-metal Nanoframes for Catalytic and Photonic Applications.
HAADF-STEM images of a rounded Pd decahedron Natl. Sci. Rev. 2016, 3, 520−533.
without overlay; XPS results obtained from the initial Pd (12) Wiley, B.; Herricks, T.; Sun, Y.; Xia, Y. Polyol Synthesis of Silver
decahedra and the final etching products in DEG and Nanoparticles: Use of Chloride and Oxygen to Promote the
aqueous environments (PDF) Formation of Single-Crystal, Truncated Cubes and Tetrahedrons.

■
Nano Lett. 2004, 4, 1733−1739.
(13) Wiley, B. J.; Xiong, Y.; Li, Z.-Y.; Yin, Y.; Xia, Y. Right
AUTHOR INFORMATION Bipyramids of Silver: A New Shape Derived from Single Twinned
Corresponding Author Seeds. Nano Lett. 2006, 6, 765−768.
*E-mail: [email protected]. (14) Wang, Z.; Wang, H.; Zhang, Z.; Yang, G.; He, T.; Yin, Y.; Jin, M.
Synthesis of Pd Nanoframes by Excavating Solid Nanocrystals for
ORCID Enhanced Catalytic Properties. ACS Nano 2017, 11, 163−170.
Aleksey Ruditskiy: 0000-0002-1146-827X (15) Yang, Y.; Zhang, Q.; Fu, Z.-W.; Qin, D. Transformation of Ag
Madeline Vara: 0000-0001-8968-8734 Nanocubes into Ag−Au Hollow Nanostructures with Enriched Ag
Younan Xia: 0000-0003-2431-7048 Contents to Improve SERS Activity and Chemical Stability. ACS Appl.
Mater. Interfaces 2014, 6, 3750−3757.
Notes (16) Ye, X.; Jones, M. R.; Frechette, L. B.; Chen, Q.; Powers, A. S.;
The authors declare no competing financial interest.

■
Ercius, P.; Dunn, G.; Rotskoff, G. M.; Nguyen, S. C.; Adiga, V. P.;
Zettl, A.; Rabani, E.; Geissler, P. L.; Alivisatos, A. P. Single-particle
ACKNOWLEDGMENTS Mapping of Nonequilibrium Nanocrystal Transformations. Science
This work was supported in part by a grant from the NSF 2016, 354, 874−877.
(DMR-1506018) and startup funds from the Georgia Institute (17) Marks, L. D.; Peng, L. Nanoparticle Shape, Thermodynamics
and Kinetics. J. Phys.: Condens. Matter 2016, 28, 053001.
of Technology. A.R. was also supported by a Graduate Research (18) Choi, S.-I.; Herron, J. A.; Scaranto, J.; Huang, H.; Wang, Y.; Xia,
Fellowship from the NSF.

■
X.; Lv, T.; Park, J.; Peng, H.-C.; Mavrikakis, M.; Xia, Y. A
Comprehensive Study of Formic Acid Oxidation on Palladium
REFERENCES Nanocrystals with Different Types of Facets and Twin Defects.
(1) Zheng, Y.; Zeng, J.; Ruditskiy, A.; Liu, M.; Xia, Y. Oxidative ChemCatChem 2015, 7, 2077−2084.
Etching and Its Role in Manipulating the Nucleation and Growth of (19) Luo, M.; Huang, H.; Choi, S.-I.; Zhang, C.; Silva, R. R.; Peng,
Noble-Metal Nanocrystals. Chem. Mater. 2014, 26, 22−33. H.-C.; Li, Z.-Y.; Liu, J.; He, Z.; Xia, Y. Facile Synthesis of Ag Nanorods
(2) Long, R.; Zhou, S.; Wiley, B. J.; Xiong, Y. Oxidative Etching for with No Plasmon Resonance Peak in the Visible Region by Using Pd
Controlled Synthesis of Metal Nanocrystals: Atomic Addition and Decahedra of 16 nm in Size as Seeds. ACS Nano 2015, 9, 10523−
Subtraction. Chem. Soc. Rev. 2014, 43, 6288−6310. 10532.

5399 DOI: 10.1021/acs.chemmater.7b01776

Chem. Mater. 2017, 29, 5394−5400
Chemistry of Materials Article

(20) Luo, M.; Ruditskiy, A.; Peng, H.-C.; Tao, J.; Figueroa-Cosme,
L.; He, Z.; Xia, Y. Penta-Twinned Copper Nanorods: Facile Synthesis
via Seed-mediated Growth and Their Tunable Plasmonic Properties.
Adv. Funct. Mater. 2016, 26, 1209−1216.
(21) Huang, H.; Zhang, L.; Lv, T.; Ruditskiy, A.; Liu, J.; Ye, Z.; Xia,
Y. Five-fold Twinned Pd Nanorods and Their Use as Templates for
the Synthesis of Bimetallic or Hollow Nanostructures. ChemNanoMat
2015, 1, 246−252.
(22) Griffith, W. P.; Robinson, S. D.; Swars, K. In Gmelin Handbook
of Inorganic Chemistry, 8th ed.; Supplement Vol. B 2, Pd Compounds;
Griffith, W. P., Swars, K., Eds.; Springer-Verlar: Berlin/Heidelberg,
1989; pp 208−210.
(23) Sneed, B. T.; Young, A. P.; Tsung, C.-K. Building Up Strain in
Colloidal Metal Nanoparticle Catalysts. Nanoscale 2015, 7, 12248−
12265.
(24) Yamamoto, H.; Tokunaga, J. Solubilities of Nitrogen and
Oxygen in 1,2-Ethanediol + Water at 298.15 K and 101.33 kPa. J.
Chem. Eng. Data 1994, 39, 544−547.
(25) Skrabalak, S. E.; Wiley, B. J.; Kim, M.; Formo, E. V.; Xia, Y. On
the Polyol Synthesis of Silver Nanostructures: Glycolaldehyde as a
Reducing Agent. Nano Lett. 2008, 8, 2077−2081.
(26) Ruditskiy, A.; Zhao, M.; Gilroy, K. D.; Vara, M.; Xia, Y. Toward
a Quantitative Understanding of the Sulfate-Mediated Synthesis of Pd
Decahedral Nanocrystals with High Conversion and Morphology
Yields. Chem. Mater. 2016, 28, 8800−8806.
(27) Zemlyanov, D.; Klotzer, B.; Gabasch, H.; Smeltz, A.; Ribeiro, F.
H.; Zafeiratos, S.; Teschner, D.; Schnorch, P.; Vass, E.; Havecker, M.;
Knop-Gericke, A.; Schlogl, R. Kinetics of Palladium Oxidation in the
mbar Pressure Range: Ambient Pressure XPS Study. Top. Catal. 2013,
56, 885−895.

5400 DOI: 10.1021/acs.chemmater.7b01776

Chem. Mater. 2017, 29, 5394−5400