Skip to main content
NCBI home page
RSC Advances logoLink to RSC Advances
. 2026 Oct 5;16(46):60072–60080. doi: 10.1039/d6ra08205j

Facet-preserving conversion of Pd@Pt core–shell rhombic dodecahedra into Pt nanocages with extended {110} facets

Hojin Ahn a,✉
PMCID: PMC13636253  PMID: 42835963

Abstract

Hollow nanostructures with well-defined facets are attractive for improving the utilization of active metals while maintaining facet-dependent surface properties. However, preserving such facets during the transformation of core–shell nanocrystals into hollow architectures remains a major synthetic challenge. Here, shape-retentive rhombic dodecahedral Pt nanocages exposing Pt{110} facets were constructed using Pd rhombic dodecahedra as removable templates. Synthetic control experiments identified the halide environment during Pt growth and the Pt shell thickness as key parameters governing successful hollow structure formation. Bromide-assisted growth produced a sufficiently continuous Pt shell, while a thickness of about 2 nm provided an effective balance between structural stability and accessibility of the Pd core, enabling reproducible formation of shape-retentive Pt nanocages. Core removal increased the electrochemically active surface area by 2.1-fold and the Pt mass activity for the oxygen reduction reaction by 2.2-fold, while the specific activity remained nearly unchanged. Moreover, the activity normalized to the total metal mass increased by more than 10-fold. These results demonstrate that the hollow architecture enhances active metal utilization while preserving the intrinsic activity of the Pt{110} facet, providing a rational strategy for constructing facet-controlled hollow nanostructures.

Shape-retentive core-removal increases Pt accessibility while maintaining Pt{110}-rich surface activity, resulting in an enhancement in ORR mass activity.graphic file with name d6ra08205j-ga.webp

Introduction

Platinum is the benchmark electrocatalyst for the oxygen reduction reaction (ORR), but its scarcity and high cost continue to motivate strategies that maximize the intrinsic activity of Pt surfaces and the fraction of Pt atoms participating directly in catalysis.1–3 Shape-controlled nanocrystals offer a particularly effective strategy because catalytic activity can be tuned through the atomic arrangement of the exposed surfaces, while hollow and open architectures can further increase metal utilization by reducing the number of buried atoms.4–6 Combining facet control with a hollow geometry is therefore attractive, yet synthetically demanding: the outer surface must retain a well-defined crystallographic orientation while the internal template is removed without inducing shell fracture, alloying, or surface reconstruction.

The low-index surfaces of face-centered-cubic (fcc) Pt display distinct atomic coordination environments and adsorption properties. Pt{111} and Pt{100} surfaces have been studied extensively, in part because octahedra, cubes, and related polyhedra enclosed by these facets can be synthesized with high morphological yields.7–9 By contrast, extended Pt{110} terraces are difficult to stabilize at the nanoscale because the open atomic arrangement of the {110} surface gives it a higher surface energy and a stronger tendency to reconstruct or be replaced by lower-energy facets.10,11 Because rhombic dodecahedra are enclosed by twelve {110} facets, they represent an ideal model geometry for investigating Pt{110} surfaces. Nevertheless, the shape-controlled synthesis of monometallic Pt rhombic dodecahedra remains considerably less established than that of Pt nanocubes and octahedra.

Templated core–shell synthesis offers a practical means of circumventing this limitation. Preformed Pd nanocrystals can serve as structural templates that impart their shape and facet arrangement to the Pt shell, while the small lattice mismatch and shared fcc crystal structure of Pd and Pt favor conformal and epitaxial Pt overgrowth. Pt overgrowth has previously been demonstrated on Pd octahedra, cubes, decahedra, icosahedra, and nanoplates.12–16 Selective removal of Pd cores from appropriately designed Pd@Pt structures has further yielded Pt-based hollow nanostructures, including nanocages and nanorings.16,17 Notably, Pt nanocages exposing well-defined {111} or {100} facets were obtained by coating Pd octahedra or cubes with only a few atomic layers of Pt and subsequently removing the Pd cores.5,17 These studies demonstrated that a sacrificial core can be used to combine ultrathin Pt walls with well-defined low-index surface orientations.

Rhombic dodecahedral Pt-based hollow structures have also been explored in multimetallic nanoframe systems.18–20 For instance, Pt–Ni rhombic dodecahedra can be converted into Pt-rich edge frameworks with three-dimensionally accessible catalytic surfaces. In these structures, however, removal of the interior is accompanied by the transformation of the broad rhombic faces into edge-dominated frameworks. Consequently, these approaches do not directly yield continuous Pt nanocages that retain extended {110} terraces.

A method was previously developed for the conformal growth of Pt on rhombic dodecahedral Pd seeds, yielding Pd@Pt core–shell nanocrystals enclosed by well-defined Pt{110} facets.21 In this system, the Pd core not only served as a structural template for the Pt shell but also imposed lattice strain, enabling investigation of the strain–activity relationship of Pt{110} surfaces. Removing the Pd core presents a conceptually distinct target: a self-supporting hollow Pt shell exposing well-defined {110} facets, with a greater fraction of accessible Pt atoms, including the Pt surface generated from the formerly buried Pd/Pt interface. Achieving this structure, however, is not merely a matter of etching away the core. If Pt deposition is discontinuous, locally intermixed with Pd, or insufficiently thick, the shell may rupture or collapse upon Pd dissolution. Conversely, an excessively thick shell may preserve the overall morphology but diminish the Pt-utilization advantage of the hollow architecture.

Here, a shape-retentive core-removal strategy is developed to produce rhombic dodecahedral Pt nanocages (Pt RD nanocages) exposing {110} facets from rhombic dodecahedral Pd@Pt core–shell nanocrystals (Pd@Pt RDs). The halide additive used during Pt deposition and the Pt shell thickness were identified as two key parameters governing whether the shell retains its structural integrity during Fe(iii)-mediated Pd core removal. The resulting Pt hollow nanocages retained extended {110} terraces and exhibited improved Pt utilization toward the ORR. Rather than treating core removal merely as a means of increasing surface area, this work establishes a synthetic framework for preserving a high-energy crystallographic surface during the transformation of a core–shell nanocrystal into a hollow Pt architecture.

Experimental

Chemicals and materials

Potassium tetrachloropalladate(ii) (K2PdCl4, Sigma-Aldrich, 98%), sodium tetrachloroplatinate(ii) hydrate (Na2PtCl4 hydrate, ≤49.8% Pt), potassium chloride (KCl, ≥99.0%), potassium bromide (KBr, ≥99.0%), potassium iodide (KI, anhydrous, 99%), poly(vinylpyrrolidone) (PVP, average Mw = 55 000), cetyltrimethylammonium chloride solution (CTAC, 25 wt% in water), ethylene glycol (EG, 99.8%), iron(iii) chloride (FeCl3, 97%), hydrochloric acid (HCl, 37%), ethanol (99.5%), and Nafion perfluorinated resin solution (5 wt% in lower aliphatic alcohols and water, contains 15–20% water) were purchased from Sigma-Aldrich. l-Ascorbic acid (AA, ≥99.5%) was obtained from DaeJung Chemicals & Metals. Perchloric acid (HClO4, 70%) and commercial Pt/C (20 wt% Pt) were obtained from GFS Chemicals and Alfa Aesar, respectively. All chemicals were used as received without further purification. Ultrapure water with a resistivity of 18.2 MΩ cm was used for all aqueous preparations.

Synthesis of Pd rhombic dodecahedra (Pd RDs)

Pd RDs were prepared according to a previously reported procedure with minor modifications.22 Aqueous solutions of KI (1.0 mL, 50 mM), K2PdCl4 (1.0 mL, 5 mM), and AA (1.0 mL, 50 mM) were sequentially added to an aqueous CTAC solution (5.0 mL, 4 mM) in a 20 mL glass vial. The mixture was gently agitated and then maintained at 98 °C in a convection oven for 5 h. After cooling to room temperature, the Pd RDs were isolated by centrifugation at 4300g for 10 min and washed twice with a PVP-containing ethanol solution (5 mg mL−1). The particles were redispersed in the same solution and aged at 30 °C for 12 h to obtain PVP-capped Pd RDs. The products were subsequently collected by centrifugation and redispersed in EG for use as seeds.

Synthesis of Pd@Pt RDs

Conformal Pt overgrowth on the Pd RDs was carried out using a modified version of the previously reported procedure.21 AA (60 mg), KBr (60 mg), PVP (300 mg), a Pd RD seed dispersion (0.464 mL, 1.29 mgPd mL−1 as determined by inductively coupled plasma-optical emission spectroscopy (ICP-OES)), and EG (10.536 mL) were combined in a round-bottom flask. The mixture was stirred at 500 rpm and heated at 110 °C for 30 min. The flask was then transferred to an oil bath maintained at 145 °C. A Na2PtCl4 hydrate solution in EG (4.0 mL, 0.15 mg mL−1) and an AA solution in EG (4.0 mL, 10 mg mL−1) were simultaneously introduced through separate syringe pumps at 1 mL h−1 each. After completion of the injections, the reaction mixture was kept at 145 °C for an additional 30 min. The resulting Pd@Pt RDs were cooled to room temperature, collected by centrifugation at 4300g for 10 min, washed twice with ethanol and once with water, and finally redispersed in 0.5 mL of water.

Synthesis of Pt RD nanocages

FeCl3 (1.0 mg), PVP (5.3 mg), KBr (5.3 mg), concentrated HCl (6.42 µL, 37%) and water (1.5 mL) were mixed in a reaction vial. An aqueous dispersion of the Pd@Pt RDs (0.5 mL) was then added, and the resulting mixture was stirred at 300 rpm at 80 °C for 3 h. The products were collected by centrifugation at 3000g for 10 min, washed twice with ethanol and once with water, and redispersed in water.

Characterization

Scanning electron microscopy (SEM) images were acquired using a field-emission SEM (FEI-INSPECT F50). Conventional transmission electron microscopy (TEM), Fast Fourier transform (FFT), and high-resolution TEM (HRTEM) analyses were performed using an FEI Tecnai F30 Super-Twin operated at 300 kV. High-angle annular dark-field scanning TEM (HAADF-STEM) imaging and energy-dispersive X-ray spectroscopy (EDS) elemental mapping were conducted using a double-Cs-corrected FEI Titan Cubed G2 60-300 microscope operated at 300 kV. Powder X-ray diffraction (XRD) patterns were recorded using a Rigaku D/MAX-2500 diffractometer. The Pd and Pt contents were quantified by ICP-OES with an Agilent 5110 instrument.

Electrochemical measurements

All electrochemical measurements were conducted at room temperature in a three-electrode system connected to a CHI 760D potentiostat. A glassy carbon rotating disk electrode (RDE) with a diameter of 5 mm (Pine Research Instrument) was used as the working electrode, while a reversible hydrogen electrode (RHE) and a Pt wire served as the reference and counter electrode, respectively. Prior to electrode preparation, the Pd@Pt RDs and Pt RD nanocages were immobilized on Vulcan XC-72 carbon black to obtain a metal loading of about 25 wt%. Catalyst inks were prepared by dispersing the carbon-supported catalysts in a Nafion-containing solution, giving final Pt and Nafion concentrations of 0.05714 µg µL−1 and 0.000714%, respectively. An aliquot of 35 µL of the catalyst ink, corresponding to a Pt loading of 2 µg, was drop-cast onto the glassy carbon RDE and allowed to dry under ambient conditions. Before ORR measurements, the prepared working electrodes were electrochemically conditioned in N2-saturated 0.1 M HClO4 solution by repeatedly cycling the potential between 0.08 and 1.1 V versus RHE at 50 mV s−1. Potential cycling was continued until reproducible cyclic voltammograms (CVs) were obtained. This conditioning procedure was used to establish a reproducible electrochemical surface state before activity measurements. The electrochemically active surface area (ECSA) was determined from the charge associated with hydrogen desorption after correction for the double-layer contribution: ECSA (m2 gPt−1) = QH/(qHmPt), where QH is the integrated hydrogen desorption charge, qH is the charge density corresponding to desorption of a hydrogen monolayer, and mPt is the Pt mass loaded on the electrode. Charge densities of 147 µC cm−2 and 210 µC cm−2 were used for Pt{110}-dominated catalysts and polycrystalline Pt/C, respectively.23,24 ORR polarization curves were recorded in O2-saturated 0.1 M HClO4 solution by linear sweep voltammetry at a scan rate of 20 mV s−1 and an electrode rotation rate of 1600 rpm. The polarization data were corrected for background current and uncompensated solution resistance. The kinetic current density, jk, was obtained using the Koutecky–Levich relationship: 1/j = 1/jk + 1/jd, where j is measured current density and jd is the diffusion-limited current density. The resulting kinetic currents were normalized to the Pt loading and ECSA to calculate mass and specific activities, respectively.

Results and discussion

Synthesis and structural characterization

The synthetic sequence for Pt RD nanocages is summarized in Fig. 1a. Single-crystalline Pd RDs were first prepared as shape-directing templates. A low-magnification transmission electron microscopy (TEM) image revealed uniform, faceted particles with projections characteristic of rhombic dodecahedra and an average plane-to-plane distance (particle size) of 83.22 ± 4.71 nm (Fig. 1b). The Pd RDs were subsequently used as seeds for Pt overgrowth to produce Pd@Pt RDs in an EG solution containing PVP, KBr, AA, and Na2PtCl4. Following Pt deposition, the average particle size increased to 87.25 ± 4.07 nm, corresponding to an average radial increase of approximately 2.0 nm, while the overall rhombic dodecahedral morphology was retained (Fig. 1c). An atomic-resolution high-angle annular dark-field scanning TEM (HAADF-STEM) image revealed a continuous Pt overlayer comprising 14 atomic layers along the [110] direction, corresponding to a thickness of approximately 2.0 nm based on the d220 spacing between successive occupied atomic layers in fcc Pt (Fig. 1d). STEM-EDS elemental mapping image showed Pd concentrated in the particle interior and Pt distributed along the outer region, confirming the formation of a Pd@Pt core–shell architecture (Fig. 1e and f). The Pd cores were selectively removed by Fe(iii)-mediated oxidative etching with the addition of bromide ion. Fe(iii) acted as the oxidant, while bromide complexation stabilized the dissolved Pd(ii) species and facilitated their removal from the particle interior. Under the selected conditions, the Pt shells remained largely intact. Following etching, the particles exhibited a pronounced hollow contrast in TEM images, with a low-contrast interior surrounded by darker Pt-rich walls (Fig. 1g). The rhombic dodecahedral morphology was retained without severe structural collapse in representative nanocages (Fig. 1h), while a low-magnification SEM image further showed the presence of numerous hollow nanostructures over a wider field of view (Fig. S1). Particle-count analysis by TEM of three independent synthesis batches showed that 91.6 ± 0.8% of the particles exhibited an apparently completely etched hollow interior without discernible dense core contrast. Among these apparently completely etched particles, 91.2 ± 3.6% retained the rhombic dodecahedral morphology without severe folding, fragmentation, or collapse. When both complete etching and morphology retention were considered, 83.5 ± 3.9% of the particles formed shape-retentive Pt RD nanocages. An HRTEM image and corresponding FFT pattern of a Pt RD nanocage viewed along the [110] zone axis confirmed its fcc crystalline structure (Fig. S2). To more directly identify the crystallographic termination of the broad external faces, additional HRTEM and FFT analyses were performed on Pt RD nanocages viewed along the [001] and [110] zone axes (Fig. S3). For the nanocage viewed along the [001] zone axis, two broad external faces were observed in an edge-on configuration, with their surface normals parallel to the corresponding {220} reciprocal-lattice directions and therefore to the Pt 〈110〉 directions. A second nanocage viewed along the [110] zone axis exhibited the same crystallographic relationship for another broad external face. These observations directly identify the broad external surfaces as members of the {110} family and provide crystallographic evidence that the extended Pt{110} facets were retained after selective Pd-core removal. The resulting Pt RD nanocages had a mean thickness of 1.97 ± 0.16 nm with a relatively narrow distribution (Fig. 1i). Small openings observed in the nanocages likely provided exit pathways for dissolved Pd species during the etching process and may subsequently facilitate electrolyte access to the inner Pt surface. STEM image and corresponding elemental mapping image indicated a Pt-rich shell along the particle perimeter and a substantially reduced Pd signal within the hollow interior (Fig. 1j and k). ICP-OES analyses showed the Pd and Pt contents within Pt RD nanocages were 4.8 at% and 95.2 at%, respectively, indicating that most of the Pd has been removed. The residual Pd detected by ICP-OES may originate from trace Pd retained within the shells or incompletely etched particles.

Fig. 1. (a) Schematic illustration of the formation of Pt RD nanocages. (b) TEM image of Pd RDs. The distribution of particle size is shown in the inset. (c) TEM image of Pd@Pt RDs. The distribution of particle size is shown in the inset. (d) High-resolution HAADF-STEM image of Pd@Pt RD. (e) HAADF-STEM image and (f) corresponding EDS elemental mapping image of Pd@Pt RD. (g) TEM image and (h) high-magnification TEM image of Pt RD nanocages. (i) High-resolution TEM image of the yellow dashed rectangle region in part h. The distribution of wall thickness is shown in the inset. (j) HAADF-STEM image and (k) corresponding EDS elemental mapping image of Pt RD nanocage. Particle-size statistics were obtained from 237 Pd RDs and 223 Pd@Pt RDs, and the wall thickness distribution was obtained from 54 measurements. All values are reported as mean ± standard deviation.

Fig. 1

Open in a new tab

Powder X-ray diffraction (XRD) was used to examine the crystal structure of the Pd RDs, Pd@Pt RDs, and Pt RD nanocages (Fig. S4). All three samples exhibited the characteristic reflections of a fcc structure, which could be indexed to the (111), (200), (220), (311), and (222) planes. No additional reflections attributable to crystalline impurities or secondary phases were detected. Since Pd and Pt have similar lattice parameters, their diffraction peaks substantially overlap. The reflections of the Pd@Pt RDs remained close to those of the Pd seeds, reflecting the dominant contribution of the large Pd core and the coherent epitaxial relationship between the Pd core and Pt shell. Following removal of the Pd core, the diffraction peaks of the Pt RD nanocages shifted slightly toward the reference positions of the bulk Pt, consistent with depletion of Pd and partial relaxation of the lattice constraint imposed on the Pt shell. The local lattice structure of the Pt RD nanocages was examined by HRTEM (Fig. S5). The lattice periodicities measured along the Pt[001] and Pt[01̄1] directions were approximately 0.3921 and 0.2763 nm, respectively, in close agreement with the corresponding bulk Pt values. These measurements indicate substantial release of the epitaxial lattice constraint after core removal, with no appreciable residual strain along the measured directions.

Halide- and thickness-dependent formation of hollow Pt structures

The structural integrity of the hollow Pt nanocages was strongly influenced by the halide present during Pt deposition. Halide ions can affect several coupled processes, including Pt(ii) precursor speciation, reduction kinetics, and the chemical stability of the Pd template.25–27 The halide additive therefore governs not only the rate at which Pt atoms are generated but also their distribution over the Pd surface and the ability of the resulting shell to withstand subsequent core removal.

To evaluate the role of halide ions during Pt deposition, a series of control experiments were performed under otherwise identical reaction conditions by varying only the halide additive. In the absence of an added halide, numerous small Pt particles were observed away from the Pd seeds, indicating extensive homogeneous nucleation and insufficient Pt overgrowth under the selected conditions (Fig. 2a). TEM-EDS analysis of the Pd@Pt RDs prepared with no added halide, excluding spatially separated Pt nanoparticles, gave a composition of 94.36 at% Pd and 5.64 at% Pt. The relatively low Pt fraction is consistent with substantial off-seed Pt nucleation. After etching, 96.6% of the particles exhibited apparent complete core removal, but none retained the original rhombic dodecahedral morphology. These defects were amplified during the etching process, producing deformed or folded structures rather than intact hollow shells (Fig. 2b). Addition of KCl gave a similar outcome. Although the Pd@Pt particles generally retained the overall morphology of the Pd templates before etching, the deposited Pt shells showed rough and locally heterogeneous surfaces (Fig. 2c). Following Pd removal, hollow structures were obtained, but pronounced wall deformation, folding, and partial collapse were frequently observed (Fig. 2d). Numerous openings exhibiting contrast in the form of bright spots in the TEM image were also observed, suggesting the presence of locally thin or weakly connected regions in the Pt shells. TEM-EDS analysis of the Pd@Pt RDs prepared with KCl gave 89.34 at% Pd and 10.66 at% Pt. After etching, 98.5% of the particles showed apparent complete core removal, but none of these particles retained the original rhombic dodecahedral morphology. These results indicate that efficient Pd removal alone is insufficient for shape retention when the Pt coating is rough or locally discontinuous. Because the Pt precursor was initially introduced as a chloroplatinate complex, additional chloride appears to have produced only a limited change in precursor reactivity under the present conditions. Chloride therefore did not provide sufficient control over Pt deposition to generate a uniformly interconnected Pt shell capable of retaining the original particle geometry during etching.

Fig. 2. TEM images of Pd@Pt RDs prepared with (a) no added halide, (c) an equimolar amount of KCl (38 mg), and (e) an equimolar amount of KI (84 mg) in place of KBr (60 mg) used in the optimized synthesis, together with TEM images of the corresponding structures after Fe(iii)-mediated etching (b, d, and f, respectively). All other Pt growth and etching conditions were kept identical.

Fig. 2

Open in a new tab

By contrast, KBr established a more favorable growth regime, as demonstrated by the conformal Pd@Pt core–shell structures and shape-retentive Pt nanocages shown in Fig. 1. TEM-EDS analysis showed that Pd@Pt RDs prepared with KBr contained 87.73 at% Pd and 12.27 at% Pt. Particle-count analysis further showed that 91.6% of the particles were apparently completely etched, and 91.2% of these apparently completely etched particles retained the rhombic dodecahedral morphology. Together with the conformal core–shell morphology, these results are consistent with efficient incorporation of Pt onto the Pd templates. Bromide can modify the coordination environment of the Pt(ii) precursor and retard its reduction by the reducing agent.25,26 Such bromide-mediated regulation is consistent with the suppressed homogeneous nucleation and conformal Pt deposition observed experimentally.

Iodide exerted a stronger influence on both Pt deposition and the stability of the Pd template. Strong Pt–I coordination can influence Pt precursor reduction kinetics, while iodide can also promote oxidative etching of Pd.27,28 Although the outer surfaces of the Pd@Pt particles appeared relatively smooth, bright contrast void-like regions were observed within the particle interiors (Fig. 2e). These internal cavities suggest that partial oxidative etching of the Pd seeds occurred concurrently with Pt deposition, likely facilitated by iodide-promoted Pd dissolution and stabilization of dissolved Pd(ii) as iodopalladate complexes including [PdI4]2−.27,29 TEM-EDS analysis of the Pd@Pt RDs prepared with KI gave 88.08 at% Pd and 11.92 at% Pt, which is essentially comparable to the composition of the KBr-derived particles. Thus, the poor structural stability of the KI-derived particles cannot be explained simply by a lower overall Pt incorporation or a higher average Pd content. After etching, 74.1% of the particles exhibited apparent complete core removal, but none retained the original rhombic dodecahedral morphology. The internal void-like contrast observed before etching suggests that iodide perturbed the integrity of the Pd template during Pt deposition, likely through partial Pd corrosion. Such corrosion can alter the spatial distribution of Pd and Pt even when the average Pd/Pt composition remains similar. Upon subsequent etching, selective removal of Pd from locally Pd-containing or weakly connected regions can generate discontinuities in the shell, resulting in pronounced fragmentation and collapse (Fig. 2f). Therefore, the instability of the KI-derived particles is attributed primarily to the structurally nonuniform shell formed on the partially corroded Pd template, rather than to the overall Pd/Pt ratio alone.

Overall, the halide-dependent outcome was governed not only by the overall extent of Pt incorporation but also by the continuity and structural integrity of the deposited shell. The Pd@Pt RDs contained 5.64, 10.66, 12.27, and 11.92 at% Pt under no-halide, KCl, KBr, and KI conditions, respectively. The markedly lower Pt fraction in the no-halide sample is consistent with extensive off-seed Pt nucleation, whereas bromide promoted efficient Pt incorporation together with a conformal shell morphology. Notably, the similar average Pd/Pt compositions of the KBr- and KI-derived particles demonstrate that overall composition alone does not determine structural stability. In the iodide-containing system, perturbation of the Pd template and the resulting nonuniform shell structure led to severe deformation during subsequent Pd removal. Bromide therefore provided the most favorable balance between Pt incorporation, preservation of the Pd template, and formation of a continuous shell capable of surviving the core-removal process. These results demonstrate that the success of the core-etching step is largely predetermined during Pt-shell growth: Fe(iii)-mediated etching can preserve the rhombic dodecahedral morphology only when the shell is sufficiently uniform, continuous, and mechanically interconnected.

To quantitatively distinguish the effect of precursor loading from that of shell thickness, the average Pt shell thickness before etching was estimated from the plane-to-plane distances of the Pd seeds and Pd@Pt RDs measured from TEM images. Using an average Pd seed dimension of 83.22 nm, precursor volumes of 6.0, 4.0, 3.5, 3.0, 2.0, 1.0, and 0.75 mL gave estimated Pt shell thicknesses of 2.92, 2.01, 1.63, 1.34, 1.09, 0.65, and 0.47 nm, respectively. Thus, systematic variation of the Pt precursor volumes produced a monotonic shell thickness series. A clear thickness-dependent trade-off was observed between Pd-core removal and retention of the rhombic dodecahedral morphology (Fig. 3). At the largest estimated shell thickness of 2.92 nm (6.0 mL precursor), several particles retained a dark interior contrast after the etching process, indicating incomplete removal of the Pd core (Fig. 3a). The limited etching may arise because a thick, compact shell restricts the formation of pinholes and the transport of Fe(iii) into the particle interior. Moreover, excessive shell thickness increases the amount of Pt buried within the wall, thereby diminishing the Pt utilization advantage expected from a hollow architecture.

Fig. 3. TEM images of the products obtained after Fe(iii)-mediated etching of Pd@Pt RDs prepared using different injection volumes of a Na2PtCl4 hydrate solution in EG (0.15 mg mL−1): (a) 6.0 mL, (b) 3.5 mL, (c) 3.0 mL, (d) 2.0 mL, (e) 1.0 mL, and (f) 0.75 mL. All other synthesis and etching conditions were identical to those used for the optimized preparation, which employed 4.0 mL of the precursor solution.

Fig. 3

Open in a new tab

As the estimated shell thickness decreased below the optimized value, Pd-core removal became more apparent, but the structural stability of the resulting hollow particles progressively deteriorated (Fig. 3b–f). At an estimated shell thickness of 1.63 nm, hollow structures could still be formed, although wall deformation and localized folding were frequently observed (Fig. 3b). Further decreases in shell thickness to 1.34 and 1.09 nm resulted in increasingly wrinkled and collapsed structures after etching (Fig. 3c and d). At estimated shell thicknesses of 0.65 and 0.47 nm, continuous particle walls were no longer preserved, and the products consisted predominantly of highly porous or fragmented Pt networks (Fig. 3e and f). These observations indicate that locally thin or incompletely connected Pt shells are readily disrupted when the supporting Pd core is removed.

These results show that successful formation of a shape-retentive Pt nanocage requires an intermediate shell thickness. Under the present synthesis and etching conditions, an estimated pre-etching Pt shell thickness of approximately 2 nm provided the most favorable balance between Pd-core accessibility and mechanical integrity. A thicker shell of 2.92 nm hindered complete core removal, whereas shells thinner than approximately 1.6 nm progressively lost structural integrity during Pd dissolution. Thus, the optimum near 2 nm reflects the thickness required to maintain a continuous three-dimensional Pt framework while still allowing effective removal of the Pd template.

Electrocatalytic ORR performance

The electrocatalytic consequences of removing the Pd core while preserving the Pt{110}-rich rhombic dodecahedral nanocages were evaluated using the ORR. The Pd@Pt RDs and Pt RD nanocages were deposited onto carbon supports before evaluation of their electrocatalytic properties (Fig. S6). Commercial Pt/C was also examined as a benchmark catalyst under identical electrochemical conditions. CVs recorded in N2-saturated 0.1 M HClO4 solution displayed facet-sensitive hydrogen adsorption and desorption features consistent with Pt{110}-rich surfaces (Fig. S7).24 ECSAs of the catalysts were determined from the charges associated with the hydrogen desorption region in the CVs. The ECSA values of Pt/C, Pd@Pt RDs, and Pt RD nanocages were 69.08 ± 3.88, 29.10 ± 1.38, and 62.17 ± 3.24 m2 gPt−1, respectively. Notably, the Pt RD nanocages exhibited an ECSA value 2.1 times higher than that of the Pd@Pt RDs, indicating that a greater fraction of the Pt wall became electrochemically accessible, most likely through exposure of the inner surface via openings generated during etching.

The ORR polarization curves in Fig. 4a were obtained at room temperature in O2-saturated 0.1 M HClO4 solution. The polarization curve of Pt RD nanocages was shifted positively relative to those of Pd@Pt RDs and commercial Pt/C, demonstrating more favorable ORR kinetics. The mass and specific activities of the catalysts were calculated by normalizing kinetic currents at 0.9 V versus RHE to Pt masses and ECSAs of the catalysts, respectively (Fig. 4b). The mass activity of Pt RD nanocages reached 0.785 ± 0.015 A mgPt−1, compared with 0.358 ± 0.011 A mgPt−1 for Pd@Pt RDs and 0.203 ± 0.018 A mgPt−1 for commercial Pt/C. These values correspond to 2.2- and 3.9-fold enhancements over Pd@Pt RDs and Pt/C, respectively. To further evaluate the utilization of the total noble metal content, the kinetic currents were also normalized to the combined mass of Pt and Pd (Fig. S8). On this basis, the activity of the Pt RD nanocages was 0.764 ± 0.015 A mgmetal−1, compared with 0.0752 ± 0.0024 A mgmetal−1 for the Pd@Pt RDs, corresponding to a 10.2-fold enhancement. This comparison further highlights the improved noble metal utilization achieved by removing the largely buried Pd core. The specific activities of Pt RD nanocages and Pd@Pt RDs were comparable, at 1.26 ± 0.04 mA cm−2 and 1.23 ± 0.10 mA cm−2, respectively. This result does not contradict the strain–activity relationship reported previously for Pd@Pt RDs,21 because the strain effect on the exposed outer Pt surface was most pronounced for thinner Pt shells. In the present Pd@Pt RDs with a relatively thick Pt shell of approximately 14 atomic layers, the outer Pt surface is expected to experience a substantially smaller core-induced strain, whereas the Pt layers directly adjacent to the Pd core may remain more strongly constrained. These interfacial Pt layers are buried and electrochemically inaccessible before core removal. Upon removal of the Pd core, the formerly buried Pt/Pd interface becomes a newly accessible inner Pt surface while simultaneously being released from the lattice constraint imposed by Pd. Thus, core removal does not simply relax the strain of an identical population of active surface sites, but instead exposes additional Pt surface area. Consistently, the 2.2-fold increase in mass activity closely matched the 2.1-fold increase in ECSA, whereas the specific activity changed only marginally, indicating that the dominant electrochemical effect of core removal is enhanced Pt accessibility rather than a substantial change in the average intrinsic ORR activity. Although the nanocages contain 4.8 at% residual Pd, a minor contribution from this Pd cannot be completely excluded; however, the nearly unchanged specific activity indicates that residual Pd does not appear to dominate the observed activity enhancement. Overall, removal of the Pd core increased the fraction of electrochemically accessible Pt atoms while retaining the Pt{110}-rich surface characteristics.

Fig. 4. (a) Representative ORR polarization curves measured in O2-saturated 0.1 M HClO4 solution at a scan rate of 20 mV s−1. (b) Mass and specific activities of different catalysts at 0.9 V versus RHE. Data in (b) are presented as mean ± standard deviation from three independent measurements.

Fig. 4

Open in a new tab

Conclusion

The present results show that successful conversion of a faceted core–shell nanocrystal into a shape-retentive hollow structure is determined largely by the structural quality of the shell formed before core removal. In the Pd@Pt rhombic dodecahedral system, regulation of Pt growth by the halide environment and control of the shell thickness provided the structural continuity required to remove the Pd core while preserving the extended Pt{110} surfaces. Hollowing increased the Pt-mass activity for ORR by 2.2-fold while maintaining nearly unchanged specific activity, demonstrating that the improvement arose predominantly from enhanced Pt accessibility and utilization rather than from a change in intrinsic surface activity. Normalization to the total metal mass revealed a 10.2-fold activity enhancement, further highlighting the benefit of removing the largely buried Pd core. Overall, this work provides a synthetic framework for constructing shape-retentive hollow nanostructures that preserve controlled surface orientations while enabling more efficient utilization of active metals. Extending this design principle to other metals and crystallographic orientations may enable the construction of a broader range of facet-controlled hollow nanostructures.

Author contributions

H. Ahn: conceptualization, methodology, investigation, writing.

Conflicts of interest

There are no conflicts to declare.

Supplementary Material

RA-016-D6RA08205J-s001

Acknowledgments

This paper was supported by Konkuk University in 2025.

Data availability

All data relevant to this study are provided within the article and its accompanying supplementary information (SI). Supplementary information: additional structural and electrochemical characterization data. See DOI: https://doi.org/10.1039/d6ra08205j.

References

  1. Wang Y.-J. Long W. Wang L. Yuan R. Ignaszak A. Fang B. Wilkinson D. P. Unlocking the door to highly active ORR catalysts for PEMFC applications: polyhedron-engineered Pt-based nanocrystals. Energy Environ. Sci. 2018;11:258–275. doi: 10.1039/C7EE02444D. [DOI] [Google Scholar]
  2. Shao M. Chang Q. Dodelet J.-P. Chenitz R. Recent Advances in Electrocatalysts for Oxygen Reduction Reaction. Chem. Rev. 2016;116:3594–3657. doi: 10.1021/acs.chemrev.5b00462. [DOI] [PubMed] [Google Scholar]
  3. Zhang L. Doyle-Davis K. Sun X. Pt-Based electrocatalysts with high atom utilization efficiency: from nanostructures to single atoms. Energy Environ. Sci. 2019;12:492. doi: 10.1039/C8EE02939C. [DOI] [Google Scholar]
  4. Shi Y. Lyu Z. Zhao M. Chen R. Nguyen Q. N. Xia Y. Noble-Metal Nanocrystals with Controlled Shapes for Catalytic and Electrocatalytic Applications. Chem. Rev. 2021;121:649–735. doi: 10.1021/acs.chemrev.0c00454. [DOI] [PubMed] [Google Scholar]
  5. Zhao M. Wang X. Yang X. Gilroy K. D. Qin D. Xia Y. Hollow Metal Nanocrystals with Ultrathin, Porous Walls and Well-Controlled Surface Structures. Adv. Mater. 2018;30:1801956. doi: 10.1002/adma.201801956. [DOI] [PubMed] [Google Scholar]
  6. Wang J. Cui Y. Wang D. Design of Hollow Nanostructures for Energy Storage, Conversion and Production. Adv. Mater. 2019;31:1801993. doi: 10.1002/adma.201801993. [DOI] [PubMed] [Google Scholar]
  7. Kang Y. Pyo J. B. Ye X. Diaz R. E. Gordon T. R. Stach E. A. Murray C. B. Shape-Controlled Synthesis of Pt Nanocrystals: The Role of Metal Carbonyls. ACS Nano. 2013;7:645–653. doi: 10.1021/nn3048439. [DOI] [PubMed] [Google Scholar]
  8. Jung N. Chung D. Y. Ryu J. Yoo S. J. Sung Y.-E. Pt-based nanoarchitecture and catalyst design for fuel cell applications. Nano Today. 2014;9:433–456. doi: 10.1016/j.nantod.2014.06.006. [DOI] [Google Scholar]
  9. Poerwoprajitno A. R. Gloag L. Cheong S. Gooding J. J. Tilley R. D. Synthesis of low- and high-index faceted metal (Pt, Pd, Ru, Ir, Rh) nanoparticles for improved activity and stability in electrocatalysis. Nanoscale. 2019;11:18995–19011. doi: 10.1039/C9NR05802H. [DOI] [PubMed] [Google Scholar]
  10. Vitos L. Ruban A. V. Skriver H. L. Kollár J. The surface energy of metals. Surf. Sci. 1998;411:186–202. doi: 10.1016/S0039-6028(98)00363-X. [DOI] [Google Scholar]
  11. Kuntze J. Huck M. Rauch T. Bömermann J. Speller S. Heiland W. The Pt(110) surface studied by STM and RHEED. Surf. Sci. 1996;355:L300–L304. doi: 10.1016/0039-6028(96)00509-2. [DOI] [Google Scholar]
  12. Xie S. Choi S.-I. Lu N. Roling L. T. Herron J. A. Zhang L. Park J. Wang J. Kim M. J. Xie Z. Mavrikakis M. Xia Y. Atomic Layer-by-Layer Deposition of Pt on Pd Nanocubes for Catalysts with Enhanced Activity and Durability toward Oxygen Reduction. Nano Lett. 2014;14:3570–3576. doi: 10.1021/nl501205j. [DOI] [PubMed] [Google Scholar]
  13. Park J. Zhang L. Choi S.-I. Roling L. T. Lu N. Herron J. A. Xie S. Wang J. Kim M. J. Mavrikakis M. Xia Y. Atomic Layer-by-Layer Deposition of Platinum on Palladium Octahedra for Enhanced Catalysts toward the Oxygen Reduction Reaction. ACS Nano. 2015;9:2635–2647. doi: 10.1021/nn506387w. [DOI] [PubMed] [Google Scholar]
  14. Wang X. Vara M. Luo M. Huang H. Ruditskiy A. Park J. Bao S. Liu J. Howe J. Chi M. Xie Z. Xia Y. Pd@Pt Core–Shell Concave Decahedra: A Class of Catalysts for the Oxygen Reduction Reaction with Enhanced Activity and Durability. J. Am. Chem. Soc. 2015;137:15036–15042. doi: 10.1021/jacs.5b10059. [DOI] [PubMed] [Google Scholar]
  15. Wang X. Choi S.-I. Roling L. T. Luo M. Ma C. Zhang L. Chi M. Liu J. Xie Z. Herron J. A. Mavrikakis M. Xia Y. Palladium-platinum core-shell icosahedra with substantially enhanced activity and durability towards oxygen reduction. Nat. Commun. 2015;6:7594. doi: 10.1038/ncomms8594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Wang X. Luo M. Huang H. Chi M. Howe J. Xie Z. Xia Y. Facile Synthesis of Pt–Pd Alloy Nanocages and Pt Nanorings by Templating with Pd Nanoplates. ChemNanoMat. 2016;2:1086–1091. doi: 10.1002/cnma.201600238. [DOI] [Google Scholar]
  17. Zhang L. Roling L. T. Wang X. Vara M. Chi M. Liu J. Choi S.-I. Park J. Herron J. A. Xie Z. Mavrikakis M. Xia Y. Platinum-based nanocages with subnanometer-thick walls and well-defined, controllable facets. Science. 2015;349:412–416. doi: 10.1126/science.aab0801. [DOI] [PubMed] [Google Scholar]
  18. Ren Y. Li C. Li B. Gao F. Zhang X. Yang X. Li L. Lu Z. Yu X. PtPd nanoframes derived from Pd@PdPt core–shell rhombic dodecahedrals with excellent catalytic performance toward methanol oxidation. Inorg. Chem. Front. 2021;8:2280. doi: 10.1039/D1QI00081K. [DOI] [Google Scholar]
  19. Becknell N. Son Y. Kim D. Li D. Yu Y. Niu Z. Lei T. Sneed B. T. More K. L. Markovic N. M. Stamenkovic V. R. Yang P. Control of Architecture in Rhombic Dodecahedral Pt–Ni Nanoframe Electrocatalysts. J. Am. Chem. Soc. 2017;139:11678–11681. doi: 10.1021/jacs.7b05584. [DOI] [PubMed] [Google Scholar]
  20. Niu Z. Becknell N. Yu Y. Kim D. Chen C. Kornienko N. Somorjai G. A. Yang P. Anisotropic phase segregation and migration of Pt in nanocrystals en route to nanoframe catalysts. Nat. Mater. 2016;15:1188–1194. doi: 10.1038/nmat4724. [DOI] [PubMed] [Google Scholar]
  21. Ahn H. Ahn H. An J. Kim H. Hong J. W. Han S. W. Role of Surface Strain at Nanocrystalline Pt{110} Facets in Oxygen Reduction Catalysis. Nano Lett. 2022;22:9115–9121. doi: 10.1021/acs.nanolett.2c03611. [DOI] [PubMed] [Google Scholar]
  22. Kim M. Kim Y. Hong J. W. Ahn S. Kim W. Y. Han S. W. The facet-dependent enhanced catalytic activity of Pd nanocrystals. Chem. Commun. 2014;50:9454. doi: 10.1039/C4CC02494J. [DOI] [PubMed] [Google Scholar]
  23. Biegler T. Rand D. A. J. Woods R. Limiting oxygen coverage on platinized platinum; Relevance to determination of real platinum area by hydrogen adsorption. J. Electroanal. Chem. Interfacial Electrochem. 1971;29:267–277. doi: 10.1016/S0022-0728(71)80089-X. [DOI] [Google Scholar]
  24. Marković N. M. Grgur B. N. Ross P. N. Temperature-Dependent Hydrogen Electrochemistry on Platinum Low-Index Single-Crystal Surfaces in Acid Solutions. J. Phys. Chem. B. 1997;101:5405–5413. doi: 10.1021/jp970930d. [DOI] [Google Scholar]
  25. Ghosh S. Manna L. The Many “Facets” of Halide Ions in the Chemistry of Colloidal Inorganic Nanocrystals. Chem. Rev. 2018;118:7804–7864. doi: 10.1021/acs.chemrev.8b00158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Zhou M. Wang H. Vara M. Hood Z. D. Luo M. Yang T.-H. Bao S. Chi M. Xiao P. Zhang Y. Xia Y. Quantitative Analysis of the Reduction Kinetics Responsible for the One-Pot Synthesis of Pd–Pt Bimetallic Nanocrystals with Different Structures. J. Am. Chem. Soc. 2016;138:12263–12270. doi: 10.1021/jacs.6b07213. [DOI] [PubMed] [Google Scholar]
  27. Ruditskiy A. Vara M. Huang H. Xia Y. Oxidative Etching of Pd Decahedral Nanocrystals with a Penta-twinned Structure and Its Impact on Their Growth Behavior. Chem. Mater. 2017;29:5394–5400. doi: 10.1021/acs.chemmater.7b01776. [DOI] [Google Scholar]
  28. Yin J. Wang J. Li M. Jin C. Zhang T. Iodine Ions Mediated Formation of Monomorphic Single-Crystalline Platinum Nanoflowers. Chem. Mater. 2012;24:2645–2654. doi: 10.1021/cm300056h. [DOI] [Google Scholar]
  29. Huang H. Ruditskiy A. Choi S.-I. Zhang L. Liu J. Ye Z. Xia Y. One-Pot Synthesis of Penta-twinned Palladium Nanowires and Their Enhanced Electrocatalytic Properties. ACS Appl. Mater. Interfaces. 2017;9:31203–31212. doi: 10.1021/acsami.7b12018. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

RA-016-D6RA08205J-s001
RA-016-D6RA08205J-s001.pdf (3.6MB, pdf)

Data Availability Statement

All data relevant to this study are provided within the article and its accompanying supplementary information (SI). Supplementary information: additional structural and electrochemical characterization data. See DOI: https://doi.org/10.1039/d6ra08205j.

Articles from RSC Advances are provided here courtesy of Royal Society of Chemistry

RESOURCES