Advanced Hub Main Navigation Menu
Self-Assembled Hollow Gyroids with Bicontinuous Mesostructures: A Highly Robust Electrocatalyst Fixation Platform
Abstract
The electrochemical degradation of Pt/C in commercial proton exchange membrane fuel cells (PEMFCs) is a major challenge that limits their durability and performance. This degradation mainly arises from carbon corrosion, which facilitates the detachment of electrocatalyst particles that are weakly bound to catalyst supports. Herein, unusually robust hollow gyroid morphologies designed for strong electrocatalyst fixation and extensive surface accessibility during oxygen reduction reactions (ORR) are reported. To obtain self-assembled gyroid nanostructures using a poly(styrene-b-2-vinylpyridine) (PS-b-P2VP) block copolymer, a solvent vapour treatment with dimethylformamide, which is highly selective for the P2VP block, is applied. It is discovered that retaining residual solvent in the gyroid-forming P2VP microdomain before carbonization is crucial for forming hollow gyroids with embedded electrocatalysts. These hollow gyroid carbon-Pt (HGC-Pt) nanostructures exhibit a 3.6-fold enhancement in electrochemically active surface area compared to solid gyroid carbon (SGC) counterparts. Based on systematic analyses, this exceptional electrochemical stability is attributed to greatly enhanced surface accessibility derived from the hollow geometry, uniform and robust catalyst embedding, and extensive pyridinic nitrogen doping from the P2VP block.
1 Introduction
The electrochemical degradation of carbon supports during the oxygen reduction reaction (ORR) in proton exchange membrane fuel cells (PEMFCs) is a critical issue that undermines the longevity and efficiency of the cells.[1] During the ORR in an acidic electrolyte, the carbon support is exposed to high potentials and a low pH environment, leading to its oxidation and the formation of carbon dioxide, a process catalyzed by Pt nanoparticles (PtNPs) and facilitated by the presence of oxygen and water.[2] Furthermore, weak binding between Pt and carbon supports often allows agglomeration, detachment, and dissolution of the PtNPs, which accelerates the loss of activity in the noble metal NPs.[3] The overall catalytic activity consequently diminishes, manifesting as a drop in fuel cell performance. This challenge in maintaining catalyst stability while preventing carbon support degradation is pivotal in enhancing the durability and operational viability of PEMFCs.[4]
In addition to the inhibition of degradation, for a more efficient ORR, the support for electrocatalysts must meet several critical requirements. It should possess high surface area to maximize the active sites available for the reaction, robust structural integrity to endure operational stresses, and excellent electrical conductivity to facilitate electron transfer.[5] Additionally, the support must enable efficient mass transport of reactants and products, to and from the active sites, to maintain high reaction rates.[6] Gyroid nanostructures, with their unique 3D, bicontinuous, and highly porous architecture, satisfy these requirements effectively.[7] Their long-range connectivity ensures continuous pathways for electron transport while the well-defined channels facilitate efficient mass transport of oxygen and reaction intermediates.[8] This combination of features can enhance the overall catalytic activity and durability of the electrocatalyst, making gyroid structures highly suitable as supports for ORR catalysts.
One practical route to obtain a gyroid structure is block copolymer (BCP) self-assembly, which offers significant advantages. BCP self-assembly enables the precise control of nanoscale morphology, resulting in highly ordered and uniform structures. This method allows for the fine-tuning of pore sizes, surface areas, and overall architecture, which are critical parameters for optimizing the performance of electrocatalysts.[8] Furthermore, BCP-derived nanostructures can be fabricated through relatively simple and scalable processes, making them potentially practical for industrial applications. The versatility of BCP self-assembly also allows for the creation of various nanostructures, including gyroids, cylinders, and lamellae, providing a wide range of options for customizing the properties of the carbon support to meet specific requirements for different catalytic reactions.[9] The fabrication of well-ordered and well-defined mesoporous structures is crucial for various applications, particularly in the development of advanced energy conversion catalysts. Among the various fabrication methods, block copolymer (BCP) self-assembly offers a compelling approach for generating patterned nanostructures with high precision and scalability.[10] The inherent ability of BCPs to spontaneously form ordered structures simplifies the fabrication process, offering significant advantages over alternative techniques. For instance, nanoparticle-based approaches or electroreduction methods often encounter challenges in achieving well-defined 3D architectures.[11] These methods can be time-consuming, expensive, and may not consistently yield high-quality nanostructures. In contrast, the self-organizing nature of BCP self-assembly provides a facile and robust route to fabricate intricate mesoporous structures with well-defined morphologies, making it a highly attractive strategy for catalyst preparation. However, the typical strut widths (>10 nm) of BCP gyroids, generally considerably larger than the diameters (<5 nm) of commercial carbon-based support particles, restricts the specific surface area and thus the activity of the electrocatalysts they support.[12]
Here, we introduce a method to create internally hollow gyroid carbon (HGC) nanostructures that significantly enhance both the electrochemical performance and the durability of electrocatalysts even compared to their fully solid counterparts and commercial products. Through comparative analyses, we confirmed that the key step that determines whether a gyroid becomes solid or hollow is whether residual solvent is completely removed from the gyroid microdomain before subsequent steps of Pt salt incorporation and carbonization. We report that HGC nanostructures provide a 3.6-fold increase in electrochemically active surface area (ECSA) compared to solid gyroid carbon (SGC). Relative to commercial Pt/C, the HGC-Pt nanoparticle (NP) composite structures not only demonstrate significantly enhanced ORR mass activity, but also superior retention capability during accelerated stressed tests (ASTs) of 20 000 cycles. This performance boost is explained by several key features of the HGC-Pt NP composites: the hollow structure of the gyroids allows for more facile accessibility for reactants; simultaneous reduction of Pt salt and carbonization of BCPs ensures robust embedding of PtNPs and excellent physical binding; and the innate pyridinic nitrogen-doping effect from the nitrogen-containing P2VP block achieves outstanding electrochemical stability of HGC.[13]
2 Results and Discussion
To elucidate the mechanism underlying the formation of hollow channels within the gyroid structure, a comparative analysis of the processes involved in SGC and HGC was conducted, as depicted in Figure 1. In both cases, 3D gyroid-based Pt-carbon nanostructures were synthesized utilizing BCP self-assembly combined with Pt salt incorporation (Figures S1 and S10, Supporting Information). Poly(styrene-b-2-vinylpyridine) (PS-b-P2VP) BCP was selected as the framework due to its several advantageous characteristics. First, the distinct polarity contrast between the two polymer blocks facilitates the formation of well-defined self-assembled nanostructures.[14] Second, the selective salt binding capability of P2VP enables highly selective chemical treatments to load catalytic metals.[14, 15] Furthermore, the P2VP block was chosen for its pyridine units, which provide nitrogen doping during carbonization, thereby producing a robust pyridinic N-doped carbon support structure, which will be discussed in detail later.[4, 13, 16]
For self-assembly, the BCP was cast into micron-scale-thick films and subjected to annealing with dimethylformamide (DMF) solvent vapor, resulting in the generation of highly interconnected 3D networks of gyroidal P2VP domains in the PS matrix, as shown in Figure 1a.[17] DMF was selected as the vapor-annealing solvent due to its polar nature, which allows for more selective swelling of the P2VP block.[18] Additionally, the slow evaporation kinetics of DMF enables control of its selective residence in the P2VP block after the solvent vapor annealing process.[19]
In the case of SGC, the self-assembled BCP films treated with DMF vapor were completely dried in a vacuum oven and subsequently immersed in an acidic aqueous solution containing Pt salts (Figure 1b). The noble metal ions selectively bond to the P2VP chains,[14, 15] decorating the nanostructured gyroid domains. The metal-loaded BCP films were then heat-treated at 400–900 °C in an Ar atmosphere, which carbonized the P2VP with embedded PtNPs while thermally decomposing the PS matrix.[20] As a result, the pyrolysis produced 3D bicontinuous solid nanostructures with PtNPs uniformly embedded in the gyroids.
In contrast, the process for forming HGC omits the evaporation of residual DMF within the P2VP segment. Immediately after retrieval from the solvent vapor annealing chamber, the self-assembled film is immersed into the aqueous metal precursor solution. The polar nature of DMF induces severe swelling of P2VP, resulting in an ≈48.9% increase in the thickness after the immersion process (Figure 2b). The solution captured in the gyroid block during self-assembly is a critical factor that determines whether the final gyroid structure will be hollow or solid. During the initial stage of the subsequent carbonization step, the captured solution within the P2VP microdomain vaporizes rapidly, building up pressure instantaneously inside the P2VP and compressing the P2VP block toward the PS matrix. The thermal robustness of PS, compared to that of P2VP, preserves the external structure of the gyroid until the hollow structure is formed (Figure S2, Supporting Information).[20, 21] Consequently, HGC exhibits a well-ordered structure comparable to that of SGC, as depicted in the Figure 1c.
The HGC-Pt composite nanostructures exhibit tunable properties through variations in carbonization temperature. As illustrated in Figure S3 (Supporting Information), transmission electron microscopy (TEM) characterization of the samples revealed significant increases in the average diameters of Pt nanoparticles (PtNPs) with escalating pyrolysis temperatures. Specifically, the average diameter of PtNPs increased from 1.3 to 15.3 nm as the carbonization temperature rose from 500 to 900 °C. This growth indicates a concomitant reduction in the specific surface area of PtNPs at higher pyrolysis temperatures. TEM analysis further revealed that higher carbonization temperatures lead to increased surface exposure of the catalytic nanoparticles. At lower temperatures, the smaller PtNPs were predominantly embedded within the carbon matrices. In contrast, at higher temperatures, a larger fraction of the larger PtNPs protruding from the gyroid carbon supports was observed. These competing effects suggest the existence of an optimal carbonization temperature that could maximize the accessible catalytic surface area,[22] which will be discussed in subsequent sections.
To elucidate the mechanism underlying hollow structure formation induced by residual solvent, we examined two key aspects: (i) the enhanced absorption of the aqueous solution into BCP films with residual DMF and (ii) the role of this absorption in generating vapor pressure during the early stages of carbonization. For comparison, we quantitatively measured the changes in the thickness of the BCP films (with and without residual DMF) before and after immersion in the aqueous solution. Optical reflectometry measurements revealed swelling ratios of 48.9% for samples with DMF and 26.7% for those without DMF (Figures 2a,b). These results indicate that the presence of polar DMF in the BCP significantly enhances the absorption of aqueous solutions, resulting in a higher swelling ratio for films with residual DMF. A significant fraction of the solvent molecules is expected to remain in the P2VP domain until the early stage of carbonization due to their low evaporation rate, which is attributed to their polar nature. It is hypothesized that, during carbonization, these residual solvent molecules may generate vapor within the P2VP domain, leading to the formation of hollow spaces in the gyroid carbon structure.
The geometric differences of the gyroid carbon nanostructures depending on the presence or absence of residual solvent in P2VP were confirmed through analyses using TEM, high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), and energy dispersive spectroscopy (EDS). As depicted in Figure 2c, the TEM images revealed a larger average radius (52.0 nm) in the HGC compared to the SGC (31.3 nm). This is attributed to higher vapor pressure generated in the P2VP domain exerting stronger pressure on the PS matrix.
We further investigated the PtNP distribution along the cross-sections of the respective struts of the gyroid carbons. If the hollow structure is formed by compressing the P2VP chains, PtNPs should be more densely concentrated at the edges of HGC, compared to SGC, since Pt salts preferentially attach to the pyridine parts of P2VP. Pt ions preferentially attach to the nitrogen sites of P2VP, because the nitrogen atoms in the pyridine rings coordinate with Pt ions through their lone pair electrons.[23] Thus, the density of PtNPs can be correlated with the density of P2VP domains. To confirm the formation of hollow structures induced by the build-up of vapor pressure within compressed P2VP domains, the resulting increase in P2VP domain density was investigated. However, during thermal annealing, Pt tends to agglomerate and form Pt particles, with their radius increasing as the temperature increases, as shown in Figure S3 (Supporting Information). To minimize the effects of agglomeration and ensure the density of Pt accurately reflects the density of P2VP, we selected 400 °C as the annealing temperature. This is the minimum temperature required for HGC formation, as the PS matrix begins to thermally decompose between 350 and 400 °C. Annealing at higher temperatures leads to coarsening of PtNPs, which is not ideal for the characterization of Pt distribution in HGC. Thus, a temperature of 400 °C was selected to facilitate characterization and elucidate the mechanism of hollow structure formation. The annealing temperature for electrochemical characterization was optimized experimentally (Figures S4 and S8, Supporting Information). Relative carbon and Pt densities were assessed via EDS mapping, with line profiles extracted along the cross-section edges of the struts. The resulting line profiles (illustrated in Figure 2e,g) revealed that the average atomic percentage (a.t.%) of Pt (≈74.1%) in the HGC shell significantly surpasses that (≈40.5%) in the SGC. This comparison supports that the hollow structure of the gyroid originates from the increased swelling of the gyroids due to the retained solvent, which subsequently transforms into vapor, driving the formation of empty spaces within the gyroids.
We next present a comparison of electrochemical characteristics with a commercially available carbon-supported Pt catalyst having a comparable Pt composition (40 wt.% Pt), denoted as Pt/C hereafter. Cyclic voltammetry (CV) and linear sweep voltammetry (LSV) characterizations were performed in half-cell configurations to evaluate the electrochemically active surface area (ECSA), half-wave potential (E1/2), mass activity (MA), and specific activity (SA). ECSA quantifies the area of the reacting interface and is a key parameter used to evaluate catalytic performance.[24] The E1/2 value indicates the capability of catalysts to reduce the ORR over-potential and reveals their catalytic activity.[25] The ECSA of the catalysts was determined from the half-cell CV curves by integrating the hydrogen underpotential deposition (HUPD) peaks observed in the potential range of 0.05-0.4 VRHE, which revealed the charge associated with desorption of hydrogen on the Pt surface.[26] MA and SA were calculated from the kinetic current density at 0.9 VRHE, normalizing the measured current densities by the Pt mass and ECSA, respectively.[27] These values were measured and compared with those of the commercially available carbon-supported Pt catalyst with a corresponding Pt composition (40 wt.% Pt), denoted as Pt/C hereafter. In order to compare the ECSA, we analyzed the CV characterization results of SGC-Pt and HGC-Pt (Figure 3a). The HGC-Pt sample exhibited a 3.6-fold higher ECSA value, which validates that the hollow channels in HGC greatly improves the accessibility and active surface area of PtNPs embedded in the gyroid support structures. In contrast, the SGC-Pt samples showed inferior ORR performance compared to HGC-Pt because they lack the hollow structure (Figure S7, Supporting Information).
In addition, the CV curve of HGC-Pt was compared with that of a commercially available Pt/C catalyst (Figure 3b). Despite the embedded configuration of PtNPs in HGC, which provides superior fixation capability, there is also a concern regarding a possible detrimental effect of shielding the active surface area of PtNPs. Nonetheless, as demonstrated in Figure 3b, HGC-Pt exhibited an ECSA of 67.5 m2g−1, which is even greater than that of the Pt/C (63.1 m2g−1). Furthermore, the cathodic scans in the CV curves demonstrate that HGC-Pt exhibited a greater Pt oxide reduction potential (≈0.77 VRHE) than that of commercial Pt/C (≈0.73 VRHE). The oxide reduction peaks are inversely proportional to the adsorption strength of oxygenated species on the Pt surface, which block the active sites for catalytic activities.[27, 28] Thus, the higher peak potential indicates lower surface coverage of surface oxides on PtNPs hosted in HGC. This was obtained by the high-temperature heat treatment under an inert condition and leads to greater catalytic contributions from the pristine Pt surface.
The enhanced catalytic performance is further supported by the half-wave potential, mass activity, and specific activity measurements obtained from the ORR polarization curves.[29] First, as shown in Figure 3c, HGC-Pt displayed an E1/2 of 934.3 mV, which is 31.4 mV higher than that of Pt/C, indicating superior ORR activity. Second, as depicted in Figure 3f, HGC-Pt showed a mass activity of 397.5 mA mgpt−1and a specific activity of 5.9 A mPt⁻2, which are 2.7 and 2.5 times higher, respectively, than those of Pt/C (mass activity: 149.0 mA mgpt−1 and specific activity: 2.4 A mPt⁻2). The notable specific activity of HGC-Pt is attributed to the significantly enhanced ionic accessibility due to the open structure of the hollow morphology.[30]
The electrochemical durability of the fabricated nanostructure was evaluated using an accelerated degradation test (ADT) of 10 000 cycles over a potential range of 0.6-1.1 VRHE. The repeated cycles simulated load and start-stop cycles in PEMFCs, inducing degradation of the catalytic materials.[31] The LSV curves characterized before and after the ADT are shown in Figure 3d,e, respectively, for HGC-Pt and Pt/C. Pt/C demonstrated significant decreases in both parameters, with 70.1% loss in the ECSA and a 50.4 mV negative shift in the E1/2. HGC-Pt, on the other hand, exhibited superior durability with only 24.8% loss in the ECSA and a 7.4 mV shift in the E1/2 (Figure S6, Supporting Information). HGC-Pt demonstrated more stable CV curves in the double-layer capacitance region (0.4 to 0.6 VRHE) throughout the ADT, indicating excellent structural stability of the 3D hollow carbon support. From the MA and SA values in Figure 3f and Table 1 the Pt/C exhibited substantial loss of activity due to ADT, with 77.4% and 22.6% decreases in MA and SA, respectively, while HGC-Pt exhibited superior durability with a 21.1% decrease in MA and even a 5.0% increase in SA after the ADT. These results, summarized in Table 1, confirm the remarkable capability of HGC to securely embed PtNPs.
| Electrochemically Active Surface Area [m2 g−1] | Half-wave Potential [mV] | Mass Activity [mA mgpt−1] | Specific Activity [A mpt−2] | |||||
|---|---|---|---|---|---|---|---|---|
| Initial | After 10k cycles | Initial | After 10k cycles | Initial | After 10k cycles | Initial | After 10k cycles | |
| Commercial Pt/C | 63.1 | 18.4 | 902.9 | 852.4 | 149.0 | 33.64 | 2.36 | 1.83 |
| HGC-Pt | 67.5 | 50.7 | 934.3 | 926.9 | 397.5 | 313.6 | 5.89 | 6.18 |
In order to verify the superior durability of HGC, TEM was utilized to characterize and compare the electrocatalysts before and after the ADT. As exhibited in Figure 3h, before the ADT, the pristine commercial Pt/C demonstrated few-nm-scale PtNPs dispersed on irregularly shaped activated carbon supports. The TEM image of HGC-Pt in Figure 3j, on the other hand, exhibited more uniformly embedded PtNPs in hollow carbon support structures. After the ADT, the characterizations clearly revealed the size growth of the PtNPs for the commercial Pt/C (Figure 3h). The repeated cycles sintered the noble metal and formed large Pt particles in gaps between the carbon supports, which is linked with the substantial loss of ECSA and activity, as described in Table 1.[3] HGC-Pt, on the other hand, maintained the pristine structure with no significant changes in the particle size and distribution. This is attributed to the stronger support-catalyst interactions in the embedded morphology (Figure 3i,j). The electrocatalytic performance and TEM characterizations observed throughout the ADT revealed that the embedded structure of HGC-Pt was highly effective in preventing the degradation of the PtNPs and retaining the catalytic performance.
It was also anticipated that doping the hollow carbon support with N-functionalities, derived from the pyridine unit in the P2VP block, would further enhance the electrochemical durability of carbon,[4, 5, 13, 16] as well as the binding energy between the carbon and NPs in catalytic systems. Notably, due to the pyridine derived characteristic of nitrogen, the majority of N-doping takes place in the pyridinic site (Figure 3l), as verified via X-ray photoelectron spectroscopy (XPS), leading to efficient ionomer distribution and higher activity.[13] Additionally, by introducing pyridinic nitrogen atoms into the carbon lattice, the intrinsic stability of the carbon is improved, making it more resistant to oxidative degradation and corrosion during repeated ORR cycles[16] (Figure S5, Supporting Information). Moreover, the higher electronegativity of nitrogen creates stronger bonding sites for metal nanoparticles such as Pt. This increases the binding energy and securely anchors the nanoparticles to the carbon support, preventing NP migration, agglomeration, and detachment during cycling.[32] Nitrogen-doping also introduces defect sites in the carbon lattice, enhancing the dispersion and stability of the PtNPs.[33] These combined effects result in a more robust and durable catalyst system, which is essential for efficient and long-lasting performance in PEMFCs and other electrochemical applications.
To evaluate the electrocatalytic performance of the fabricated composites under more practical operating conditions, the catalysts were prepared in single-cell configurations using membrane electrode assemblies (MEAs) and subjected to electrochemical characterizations, which reflect the catalytic performance in realistic scenarios (Figure 4).[34] Accelerated stress tests (ASTs) were conducted for the cathode catalysts, HGC-Pt and commercial Pt/C, both using the same anodes with 40 wt.% commercial Pt/C. HGC-Pt and commercial Pt/C were employed as cathode catalysts, with both cells utilizing commercial Pt/C as the anode catalyst and Nafion 212 as the electrolyte. As shown in Figure 4, the HGC-Pt cell demonstrates significantly enhanced performance, achieving a maximum power density of 0.63 W cm−2, which is ≈43% higher than the 0.44 W cm−2 achieved by the Pt/C cell with the same Pt loading. To thoroughly investigate this performance enhancement, electrochemical impedance spectroscopy (EIS) was conducted at various current densities ranging from 0.2 to 1.2 A cm−2 and analyzed using the equiv. circuit model (Figure S9, Supporting Information).[35] HGC-Pt exhibits remarkably lower mass transport resistance (Rmt) compared to Pt/C at all current densities, except at 0.2 A cm−2 where both cells show negligible Rmt (Figure 4c). Moreover, as current density increases, Rmt of Pt/C rises more steeply than that of HGC-Pt. These results highlight the structural advantage of HGC-Pt, particularly in mitigating mass transport limitations at high current densities. Free from mass transport limitations (i.e., at 0.2 A cm−2), HGC-Pt demonstrates lower charge transfer resistance (Rct) compared to Pt/C (Figure S9, Supporting Information), consistent with the superior ORR activity (Figure 3c) and single cell mass activity (Figure S11, Supporting Information). Consequently, the HGC-Pt cell outperforms the Pt/C cell due to both the superior ORR catalytic activity, which is clearly evident at low current densities (i.e., 0.2 A cm−2, Figure 4), and the efficient mass transport facilitated by the hollow structure of HGC-Pt, which becomes more pronounced at higher current densities (i.e., 0.8 A cm−2). Accelerated durability tests (ASTs) were conducted by cycling each cell between 0.6 and 0.95 V, holding at each voltage for three seconds. Over 20 000 cycles, at a current density of 0.8 A cm−2, the voltage of the HGC-Pt cell decreases from 588 to 487 mV (−17.2%) while that of the Pt/C cell drops from 532 to 300 mV (−43.6%), indicating the superior durability of HGC-Pt compared to Pt/C.
3 Conclusion
In this research, we have developed novel pyridinic carbon nanostructures with hollow gyroid morphologies designed to firmly fix Pt catalysts and maximize surface accessibility during ORR. Utilizing self-assembled PS-b-P2VP BCP, we performed solvent vapor treatment with DMF, which preferentially swells the P2VP block. This process was followed by Pt salt incorporation and pyrolysis, resulting in the carbonization of the gyroid domain and embedding of PtNPs within the pyridinic gyroid carbon framework. The various analysis results obtained in this study suggest that the presence of residual solvent in the P2VP domains prior to carbonization is critical for forming uniformly hollow gyroids with embedded PtNPs. The HGC structures exhibited a 3.6-fold increase in ECSA compared to SGC structures. In addition, MEA data, including EIS measurements, demonstrate the superior activity and efficient mass transport of HGC-Pt. These improvements are attributed to the improved reactant accessibility afforded by the hollow architecture, the stable and uniform fixation of PtNPs, and pyridinic nitrogen doping of the P2VP block, which combined provide excellent electrochemical stability. Our hollow gyroid nanostructure platform is expected to provide a promising solution for simultaneously improving the durability and performance of various electrochemical energy conversion devices such as fuel cells, supercapacitors, lithium-air batteries, etc., making them highly suitable for practical applications. Several different metal salts have also been tested for the hollow gyroid nanostructure platform beyond Pt, including HGC-Ru, HGC-Rh, HGC-Ir, and HGC-Os (Figure S12, Supporting Information). The versatility of the HGC platform can be utilized not only for energy conversion catalysts, such as the oxygen evolution reaction, but also in other applications requiring 3D bicontinuous mesoporous structures.
4 Experimental Section
Materials
All materials were used as purchased. Poly(styrene-b-2-vinylpyridine) (PS-b-P2VP) polymer with number average molecular weight (Mn) of 183-b-52 kg mol−1 was obtained from Polymer Source, Inc. All chemicals, including N,N’-Dimethylformamide (DMF, anhydrous, 99.8%), hydrochloric acid (HCl, 37%), perchloric acid (HClO4, 70%), and Nafion solution (5 wt.%) were obtained from Sigma Aldrich and stored in a vacuum desiccator cabinet. Chloroplatinic acid hexahydrate (H2PtCl6∙6H2O, 99.9%) and commercially available Pt/C (40 wt.% Pt) were also obtained from Sigma Aldrich and stored in a vacuum desiccator cabinet.
Catalyst Preparation
PS-b-P2VP was dissolved in DMF to 10 wt.% and cast into films using bar-coating technique. Bar coating was carried out using an automated film coating device (MRX-TMDH150) with its platform set at 25 °C and coating speed set at 5 cm s−1. The films were dried under ambient conditions for one hour and then baked for two hours in a vacuum oven set at 60 °C. The polymer films were then placed in DMF annealing chambers for six hours to induce self-assembly. To prepare the composite nanostructures with SGC, the annealed films were baked for 30 min in a vacuum oven set at 60 °C before being submerged in aqueous incorporation solutions of 5 mM H2PtCl6 and 1% HCl. To prepare the composite nanostructures with HGC, the polymer films were submerged in the incorporation solutions as soon as they were retrieved from the annealing chambers. The films were retrieved after 24 h of immersion and thoroughly rinsed with deionized water (DIW). The films were then carbonized in Ar-filled tube furnaces heated at a ramp of 5 °C min−1 and maintained for four hours at pyrolysis temperatures specified in the study.
Physicochemical Characterizations
The composite nanostructures were characterized using a field emission scanning electron microscope (FE-SEM, Hitachi S-4800) and a transmission electron microscope (TEM, FEI Company Tecnai F20 and Thermo Fisher Talos F200X G2). The fabricated catalysts were further analyzed with a high resolution thermogravimetric analyzer (TGA, NETZSCH TG209 F1 Libra) and X-ray photoelectron spectroscope (XPS, Thermo Fisher Scientific K-alpha). The thickness of the PS-P2VP film was measured via reflectometry (F20, Filmetrics).
Half-Cell Electrochemical Characterizations
All half-cell analyses were conducted in a 0.1 M HClO4 electrolyte using a glassy carbon working electrode, platinum counter electrode, and Ag/AgCl reference electrode. The catalyst materials were loaded using ink solutions prepared with 2 mg of the catalysts, 200 µL of isopropyl alcohol (IPA), 780 µL of DIW, and 20 µL of 5 wt.% Nafion solution. Subsequently, 10 µL of the solution was dropped onto a glassy carbon rotating disk electrode (RDE) and dried in the ambient atmosphere with the electrode rotating at 500 rpm. CV and linear-sweep voltammetry (LSV) characterizations were performed using an electrochemical workstation (WonaTech ZIVE SP1) with a modulated speed rotator (Pine Instruments) and RDE set at 1600 rpm. The CV characterizations were carried out in an Ar-saturated electrolyte in a potential range of 0.05-1.2 VRHE at a scan rate of 20 mV s−1. The LSV characterizations were performed in an O2-saturated electrolyte in a potential range of 0.05-1.2 VRHE at a scan rate of 5 mV s−1. The accelerated degradation tests (ADTs) were conducted in Ar-saturated electrolytes in a potential range of 0.6-1.1 VRHE at a scan rate of 50 mV s−1.
Single-Cell Electrochemical Characterizations
HGC-Pt was evaluated as a cathode catalyst, in comparison to commercial Pt/C (40 wt.%, Alfa Aesar). Catalyst inks were first prepared by dispersing 4.4 mg of catalyst powder and 7.6 µL of Nafion ionomer (20 wt.%, Sigma Aldrich) in a 4 mL mixture of DI water and n-propanol (70:30 volumetric ratio). Anode-coated Nafion 212 membranes, with a Pt loading of 0.1 mgPt cm−2, were purchased from CNL. The prepared catalyst inks were sprayed onto the uncoated side of the Nafion 212 membrane, achieving a Pt loading of 0.15 mgPt cm−2, and then dried under air at 80 °C. The active electrode area was 5 cm2. The prepared membrane-electrode assemblies (MEAs) were then assembled with gas diffusion layers (SGL SIGRACET, 36BB, 280 µm), gaskets, graphite bipolar plates with a surpentine flow field, and current collectors.
The performance of the prepared single cells was evaluated by supplying fully humidified hydrogen (H2) and air to the anode and cathode at flow rates of 200 and 600 sccm, respectively, with a back pressure of 1.5 bar. The cell temperature was maintained at 80 °C. EIS was conducted at current densities of 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 A cm−2, with an excitation current amplitude of 0.4 mA and a frequency range from 1 Hz to 100 kHz. The cathode functioned as the working electrode while the anode served as both the counter and reference electrode. For accelerated durability test (AST), fully humidified H2 and nitrogen (N2) were supplied to the anode and cathode, respectively, at a flow rate of 200 sccm. Square voltage cycling was conducted between 0.95 and 0.6 V, with 3-s holds at each voltage at 80 °C. A fuel cell test station (CNL) and a frequency response analyzer (Biologic HCP-803) were used for the measurements.
Statistics
ImageJ was used to perform statistical analysis on image data obtained using TEM (FEI Company Tecnai F20). A total of 100 samples were measured to determine the cross-sectional radii of both HGCs and SGCs. The average size of the HGCs was 52.0 nm with a standard deviation of 5.34 nm while the average size of the SGCs was 31.3 nm with a standard deviation of 4.49 nm.
Acknowledgements
G.H.L., S.C., and H.Y. contributed equally to this work. This work was supported by a Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korean government (MOTIE) (No. 20214000000650, Energy Innovation Research Center for Fuel Cell Technology). This work was also supported by the Nano & Material Technology Development Program through the National Research Foundation of Korea (NRF) funded by Ministry of Science and ICT (RS-2024-00450102).
Conflict of Interest
The authors declare no conflict of interest.
Open Research
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
