Boron Incorporation Induces Lattice Expansion in Platinum Nanoparticles for Enhanced Oxygen Reduction Reaction

Abstract

The oxygen reduction reaction (ORR) is a key cathodic process in zinc‐air batteries, while its sluggish kinetics and the limited stability of Pt nanoparticles constrain practical performance. Achieving high activity, selectivity, and stability while minimizing Pt loading remains a key objective in electrocatalyst design. Herein, B‐doped Pt nanoparticles supported on carbon nanobowls (B‐Pt/CNBs) with an estimated Pt:B:C atomic ratio of 1.00:0.86:98.14 were synthesized by an ice‐bath reduction strategy. In alkaline media, B‐Pt/CNBs deliver an onset potential of 1.073 V versus RHE and a half‐wave potential of 0.966 V versus RHE, with an electron transfer number close to 4 and an H₂O₂ yield below 5%. The catalyst also shows enhanced methanol tolerance and durability, retaining 92.3% of its initial ECSA after 10 000 CV cycles. DFT free‐energy calculations show that B doping decreases the uphill free‐energy changes for the conversion of O* to OH* and OH* to H₂O, facilitating the protonation and removal of oxygen‐containing intermediates. This work demonstrates B doping as an effective strategy for designing active and durable low‐Pt ORR electrocatalysts.

Interstitial boron incorporation into Pt nanoparticles supported on carbon nanobowls induces Pt lattice expansion and regulates the local coordination environment of Pt active sites. The expanded Pt lattice modulates the Pt–O interaction and optimizes the adsorption/desorption energetics of oxygenated intermediates, thereby facilitating an efficient four‐electron oxygen reduction reaction (ORR) pathway in alkaline media.

1. Introduction

Among various promising energy conversion devices, zinc‐air batteries (ZABs) have received extensive research attention due to their sustainable advantages and high power density [1, 2]. As the global energy landscape undergoes a paradigm shift toward renewable and green technologies, ZABs have emerged as a frontrunner for next‐generation energy storage due to their remarkable theoretical energy density, low cost, and inherent operational safety [3]. Despite the unique properties of ZABs, their widespread commercialization has been hindered by the intrinsically sluggish kinetics of the oxygen reduction reaction (ORR). The cathodic ORR, characterized by a complex four‐electron (4e⁻) transfer process, typically exhibits a high overpotential, which drastically limits the round‐trip efficiency and power output of the battery system [4, 5]. Among various catalyst materials, noble metals such as Pd and Pt stand out owing to their favorable electronic structures and excellent catalytic properties [6, 7, 8, 9, 10]. In particular, Pt‐based materials are recognized as benchmark catalysts for the ORR because of their unique chemical, physical, and electronic properties. According to the well‐established Sabatier principle, Pt lies near the apex of the activity volcano plot, offering a near‐ideal balance between the adsorption and desorption of oxygen species [11, 12]. However, monometallic catalysts often struggle to simultaneously achieve high activity, favorable product selectivity, and long‐term structural and chemical stability under catalytic operating conditions [13]. Furthermore, the scarcity and escalating cost of Pt, coupled with its susceptibility to dissolution and particle coarsening in harsh alkaline environments, necessitate the development of more robust and atom‐efficient catalytic systems [14, 15, 16]. Therefore, improving the utilization of precious metals by using different methods based on reducing the metal dosage of Pt has become a hot spot in current research.

Crystal‐phase engineering, nanostructure design, anchoring on appropriate carriers, and alloying with other elements are conventional strategies for improving the performance of Pt‐based materials [17, 18, 19, 20]. The incorporation of light‐element dopants such as B or F can induce lattice distortion, tune the d‐band center, and redistribute the local charge density around Pt active sites, providing an efficient route to optimize the adsorption–desorption behavior of oxygenated intermediates [21, 22, 23]. Growing evidence indicates that incorporating light elements into noble‐metal nanostructures can improve catalytic stability and provide additional opportunities for regulating ORR kinetics, making light‐element doping a promising strategy for developing atom‐efficient Pt electrocatalysts [24, 25]. Manipulating the binding strength of surface adsorbates by stabilizing or destabilizing key intermediates is fundamental to electrocatalytic activity and selectivity [10, 11, 24]. Light‐element doping can expand the Pt group metal lattice, generating tensile strain and altering the electronic states of surface metal atoms [25, 26]. In a representative example, boron‐doped Pd nanostructures exhibited an enlarged Pd–Pd spacing and an optimized electronic configuration, which weakened the adsorption of oxygen‐containing intermediates and promoted ORR kinetics in alkaline media. In addition, charge transfer and p–d orbital hybridization between light dopants and noble‐metal hosts can subtly but effectively modulate the surface electronic structure [24, 26, 27]. Because small light elements can be incorporated into noble‐metal lattices, the parent metallic framework can be largely preserved even in the event of partial dopant loss. For example, Zhang and coworkers found that the ORR activity and stability of Pd metallene could be modulated by controlling the doping sites of nitrogen due to the different N dopant locations, which created distinct local coordination environments and charge‐redistribution patterns in the Pd lattice that can alter the adsorption behavior of oxygenated intermediates [28]. As a result, light‐element doping offers a dual‐modulation pathway that combines lattice‐strain engineering with electronic structure regulation, making it highly promising for simultaneously improving catalytic activity and structural durability. Heteroatom doping provides an effective route to regulate the local structure and surface reactivity of platinum group metal catalysts, especially because light‐element incorporation can modify the electronic properties, coordination environment, and catalytic stability of metal‐based electrocatalysts [28, 29, 30]. The incorporation of light nonmetal or metalloid dopants can largely preserve the primary metallic framework while introducing lattice distortion, enlarged M–M spacing, and modified surface reactivity [30, 31, 32]. Recent progress in low‐Pt ORR catalysts has further demonstrated that catalytic performance is closely related to atomic Pt utilization, local coordination regulation, interfacial stabilization, and optimized adsorption of oxygenated intermediates [33, 34, 35, 36, 37]. In this context, B incorporation into Pt nanoparticles offers a feasible strategy to induce Pt lattice expansion, regulate the Pt–O interaction, and promote ORR kinetics under alkaline conditions. Carbon nanobowls (CNBs) provide more than a conductive support for Pt‐based nanoparticles. Their open hollow architecture offers accessible inner and outer surfaces for metal nucleation, while the curved carbon framework helps restrict nanoparticle growth and migration [38, 39]. The internal void and open rim can also facilitate electrolyte penetration and O₂ transport, improving the accessibility of supported active sites [40]. Similar porous hollow carbon structures have been shown to suppress Pt‐based nanoparticle aggregation, maintain metal dispersion after high‐temperature treatment, and stabilize highly dispersed metal clusters during electrocatalysis [41].

In this work, boron‐doped Pt nanocrystals supported on CNBs, denoted as B‐Pt/CNBs, were successfully synthesized via an ice‐bath reduction strategy using dimethylamine borane (DMAB) as the reducing agent. By introducing B as a light‐element lattice modulator, the B‐Pt/CNBs couple lattice expansion with electronic structure regulation, modifying the interaction between Pt sites and oxygen‐containing intermediates during ORR. Benefiting from the lattice expansion induced by B doping, B‐Pt/CNBs exhibit an onset potential (E_onset) of 1.073 V versus RHE and a half‐wave potential (E_1/2) of 0.966 V versus RHE, both superior to those of Pt/CNBs and commercial Pt/C. The catalyst also follows a near‐ideal four‐electron ORR pathway, as evidenced by an electron‐transfer number close to 4 and a H₂O₂ yield below 5%, while delivering enhanced specific activity (SA) and mass activity (MA) relative to the comparison catalysts. Moreover, B‐Pt/CNBs show excellent durability, with only a 5 mV loss in E_1/2 after 3000 CV cycles, negligible current decay during prolonged chronoamperometric testing, and markedly improved methanol tolerance. These results demonstrate that B doping can enlarge Pt–Pt interatomic spacings and regulate the local electronic environment of Pt, reshaping the ORR free‐energy landscape and promoting the conversion of adsorbed oxygen‐containing intermediates.

2. Results and Discussion

2.1. Morphology and Structure

The synthesis of B‐Pt/CNBs consists of two steps. First, C substrates shaped like bowls were produced using the pre‐synthesis method of the group [42]. Then, potassium chloroplatinate (K₂PtCl₄) as the Pt precursor and H₃BO₃ as the B source were added to the CNB dispersion and thoroughly mixed under alkaline conditions. DMAB was then introduced as a reducing agent under an ice‐bath environment, enabling the controlled reduction of Pt species and the incorporation of B into the Pt lattice. In addition, Pt/CNBs without B doping were also synthesized as a control sample to evaluate the effect of B doping on the ORR performance of Pt‐based catalysts. The morphology of the CNB support was first examined by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Figures S1–S3, the as‐prepared CNBs exhibit a well‐defined bowl‐like morphology with an average diameter of approximately 100 nm. After Pt and B incorporation, the SEM image of B‐Pt/CNBs (Figure 1a) shows that the bowl‐like architecture is well preserved, indicating that the ice‐bath reduction process does not destroy the CNB framework. The TEM image in Figure 1b confirms that the metal nanoparticles are homogeneously distributed across the hollow CNB substrate. A more detailed view of an individual B‐Pt/CNB is presented in Figure 1c, further illustrating its distinct hollow nanobowl architecture. Furthermore, the particle size distribution map in Figure 1d reveals that the loaded nanoparticles possess a narrow size distribution centered at 2.75 ± 0.20 nm. Comparing these results with Figure S4, the morphologies of B‐Pt/CNBs and Pt/CNBs are found to be almost identical, indicating that the nanostructure of the catalysts was not significantly altered by the introduction of boron. The CNB support plays an important role in maintaining the dispersion and accessibility of Pt‐based nanoparticles. The bowl‐shaped hollow carbon framework provides abundant anchoring sites for metal nucleation, and the void confinement effects can confine the growth of Pt nanoparticles, leading to their homogeneous distribution on the CNB surface. In addition, the open hollow architecture can facilitate electrolyte penetration and oxygen mass transport, improving the accessibility of active Pt sites during the ORR. The carbon framework also provides conductive pathways for interfacial electron transfer and contributes to the high Pt utilization, as reflected by the higher ECSA of B‐Pt/CNBs. Moreover, the structural robustness of CNBs is beneficial for suppressing severe nanoparticle migration and aggregation during the electrochemical operation. Similar CNB‐supported Pt‐based catalysts have shown that the porous hollow nanobowl structure can preserve nanoparticle dispersion after high‐temperature treatment and long‐term cycling, highlighting the important role of CNBs in improving both catalytic activity and structural stability. As a comparison, TEM images of commercial Pt/C were collected, showing that small Pt nanoparticles were uniformly dispersed on the carbon support (Figure S5). The elemental composition of B‐Pt/CNBs was further analyzed by ICP‐OES and EDS (Table S1). The ICP‐OES results show that the Pt and B contents in B‐Pt/CNBs are 14.0819 and 0.6703 wt%, respectively, corresponding to an estimated Pt:B:C atomic ratio of 1.00:0.86:98.14. EDS analysis independently gives a Pt:B:C atomic ratio of 1.09:0.94:97.97, which is close to the ICP‐OES‐derived estimate. Owing to the weak response of B in EDS and its relatively low content, the B content was mainly quantified by ICP‐OES, while EDS was used to qualitatively confirm the elemental distribution of B. This result not only proves the effectiveness of the synthesis method in reducing the amount of precious metals but also significantly reduces the cost of the catalysts and improves their practicality and economy.

FIGURE 1.

(a) SEM image of B‐Pt/CNBs. (b) Low‐magnification TEM image of B‐Pt/CNBs. (c) TEM image of an individual B‐Pt/CNB. (d) Particle‐size distribution histogram of Pt nanoparticles in B‐Pt/CNBs. (e) HRTEM image of B‐Pt/CNBs. (f) Enlarged HRTEM image taken from the selected region in (e). (g–i) HAADF–STEM and corresponding FFT images of B‐Pt/CNBs. (j) EDS element mapping images of B‐Pt/CNBs.

High‐resolution transmission electron microscopy (HRTEM) images show clear lattice fringes for the different samples (Figures 1e,f and S6). The lattice spacing in B‐Pt/CNBs is 0.230 nm, corresponding to the face‐centered cubic Pt (111) facets as well as that of Pt/CNBs (0.226 nm). The average lattice spacing of B‐Pt/CNBs increased to 0.230 nm after doping with B compared to the average lattice spacing of the (111) facet in Pt/CNBs (0.226 nm). According to the equation for tensile strain, the equation for the tensile strain (τ) of B‐Pt/CNBs is defined as τ = (d_{(B‐Pt/CNBs)‐}d_(Pt/CNBs))/d_(Pt/CNBs), where d_{(B‐Pt/CNBs)} and d_(Pt/CNBs) represent the lattice spacing of B‐Pt/CNBs and Pt/CNBs, respectively. According to the calculation, the strain value τ defined above is about 1.8%. This result is further confirmed by the HRTEM line‐scan profiles in Figures S7 and S8, which show that the interplanar spacings of Pt/CNBs increase from 0.226 and 0.196 nm to 0.230 and 0.200 nm in B‐Pt/CNBs, corresponding to the Pt (111) and (200) planes, respectively. The above results indicate that the spacing between Pt and Pt atoms can be enlarged by incorporating B atoms into the Pt lattice, and the lattice extension in B‐Pt/CNBs reduces the adsorption strength between oxygen‐containing intermediates and Pt sites and accelerates the desorption process of H₂O. The atomic structure of B‐Pt/CNBs was further examined by HAADF–STEM. As shown in Figure 1g, clear atomic contrast can be observed, and the enlarged image of the selected region (Figure 1h) resolves the ordered arrangement of Pt atoms. Moreover, the corresponding FFT pattern (Figure 1i) is consistent with fcc Pt viewed along the [011] zone axis, with identifiable (111) and (200) reflections, confirming the crystalline Pt framework. In addition, EDS plots further verified the successful doping of B atoms. As shown in Figures 1j and S9, the elemental distribution of B is relatively weak due to the low B content, but the distribution of Pt, B, and C in B‐Pt/CNBs can still be observed. In comparison, the EDS mapping of Pt/CNBs is shown in Figure S10.

The crystal structure and lattice extension of B‐Pt/CNBs were probed by an X‐ray powder diffractometer (XRD). The B‐Pt/CNBs showed a set of distinct diffraction peaks at 39.7°, 46.2°, and 67.4° (Figure 2a,b), which corresponded to the Pt (111), (200), and (220) crystallographic planes, respectively. Both Pt/CNBs and B‐Pt/CNBs showed a typical face‐centered cubic (fcc) Pt structure (JCPDS No. 04‐0802), indicating that B incorporation does not change the primary crystalline framework of fcc Pt. However, after B incorporation, the diffraction peaks of B‐Pt/CNBs are shifted toward lower angles than those of Pt/CNBs, which suggests that B incorporation leads to Pt lattice expansion. This observation is in good agreement with the HRTEM analysis presented above, which indicates an enlarged Pt(111) interplanar spacing upon B doping. According to Bragg’s law (2dsinθ = λ, λ = 1.5406 Å), the corresponding Pt (111) diffraction positions are estimated to be about 39.86° for Pt/CNBs and 39.13° for B‐Pt/CNBs, respectively, consistent with the shifted peaks observed in the XRD pattern. The consistency between the XRD and HRTEM results further supports the lattice expansion effect induced by B incorporation. Specifically, the low‐angle shift of the Pt diffraction peaks in the XRD patterns reflects an increase in the average lattice spacing of the Pt crystalline phase, while the HRTEM images directly show enlarged local interplanar spacings for the Pt (111) and (200) planes. The increase in the Pt (111) spacing from 0.226 nm for Pt/CNBs to 0.230 nm for B‐Pt/CNBs and the corresponding shift of the Pt diffraction peaks toward lower angles are therefore mutually consistent. These results indicate that B incorporation enlarges the Pt lattice and increases the Pt–Pt interatomic distance without changing the primary fcc Pt structure. The lattice expansion can be understood from the incorporation of small B atoms into the Pt lattice. Because B has a much smaller atomic size than Pt, its incorporation can alter the local atomic arrangement of the fcc Pt framework. B incorporation into the Pt lattice can induce local lattice distortion and expand the Pt–Pt interatomic distance. Such B‐induced lattice expansion may introduce tensile strain into the Pt lattice, which can further influence the electronic structure of surface Pt atoms and regulate the adsorption behavior of oxygen‐containing intermediates during the ORR.

FIGURE 2.

(a) XRD patterns of B‐Pt/CNBs and Pt/CNBs. (b) Enlarged XRD patterns over the selected 2θ range marked in (a). (c) XPS survey spectra of B‐Pt/CNBs and Pt/CNBs. Deconvoluted high‐resolution XPS spectra of (d) C 1s and (e) Pt 4f for B‐Pt/CNBs and Pt/CNBs, and (f) B 1s for B‐Pt/CNBs.

X‐ray Photoelectron Spectroscopy (XPS) was used to examine the surface composition and local electronic structure of the catalysts. The survey spectrum of B‐Pt/CNBs shows the characteristic C 1s, O 1s, Pt 4f, and B 1s signals (Figure 2c). As shown in Figure 2e, the peaks at 71.4 and 74.8 eV are assigned to Pt⁰ 4f7/2 and Pt⁰ 4f5/2, respectively, whereas those at 72.5 and 76.0 eV correspond to Pt²⁺ species. Compared with Pt/CNBs, the Pt 4f peaks of B‐Pt/CNBs exhibit a positive shift of approximately 0.11 eV, which indicates that B doping changes the local electronic environment of Pt and induces electron redistribution around the Pt sites. The higher Pt 4f binding energy is consistent with a decrease in the local electron density around Pt arising from the Pt–B electronic interaction. Such charge redistribution may alter the occupancy and energy distribution of the Pt d states, affecting the interaction between Pt sites and the adsorbed oxygen‐containing intermediates. Because the Pt d‐band center was not directly calculated, the possible d‐band modulation is discussed as a consequence of the observed electron redistribution rather than as a directly measured result [10, 24].

As shown in Figure 2f, the high‐resolution B 1s spectrum can be fitted with three peaks. The low‐binding‐energy peak at approximately 186.2 eV is assigned to Pt‐B‐related B species and/or B⁰. The peaks at approximately 188.5 and 191.3 eV are assigned to B─O and B─O/BO _x species, respectively. The oxidized B peaks may originate from the partial oxidation of surface‐exposed B during sample handling and air exposure. The B 1s spectrum identifies the presence and chemical states of B‐related species in the surface and near‐surface region rather than serving as standalone evidence for a specific crystallographic location of B. The C 1s spectrum of B‐Pt/CNBs and Pt/CNBs shows no pronounced difference in the surface carbon functionalities (Figure 2d).

2.2. Electrocatalytic Evaluation

The electrocatalytic ORR performance of B‐Pt/CNBs was systematically evaluated in an alkaline electrolyte, with Pt/CNBs, commercial Pt/C, and CNBs as reference catalysts. The CV curves recorded in O₂‐saturated 0.1 M KOH are shown in Figures 3a and S11, while the CV curves recorded in N₂‐saturated 0.1 M KOH are shown in Figure S12 [43, 44]. The electrochemically active surface area (ECSA) was calculated from the CV curves and is summarized in Figure S13. The ECSA values were 35.56, 28.57, and 2.32 m² g_Pt ⁻¹ for B‐Pt/CNBs, Pt/CNBs, and commercial Pt/C, respectively. The higher ECSA of B‐Pt/CNBs relative to that of Pt/CNBs suggests that B incorporation improves the accessibility of electrochemically active Pt sites and enhances Pt utilization. Consistent with the polarization results, B‐Pt/CNBs delivered the highest limiting current density among the tested catalysts, reaching 6.16 mA cm⁻², slightly higher than those of commercial Pt/C and Pt/CNBs (Figure S14).

FIGURE 3.

(a) CV curves of B‐Pt/CNBs and Pt/C catalysts in O₂‐saturated 0.1 M KOH solution. (b) LSV curves of B‐Pt/CNBs, Pt/C, Pt/CNBs, and CNBs. (c) Tafel plots of B‐Pt/CNBs, Pt/CNBs, CNBs, and Pt/C. (d) Nyquist plots of B‐Pt/CNBs, Pt/CNBs, CNBs, and Pt/C. (e) Specific activity (SA) and (f) mass activity (MA) of B‐Pt/CNBs, Pt/CNBs, and Pt/C.

The E_onset and E_1/2 of B‐Pt/CNBs were 1.073 and 0.966 V versus RHE, respectively, which were superior to those of Pt/CNBs (1.018 and 0.922 V vs. RHE) and commercial Pt/C (1.026 and 0.943 V vs. RHE) (Figures 3b and S15). The improved ORR performance can be ascribed to the synergistic effect of the increased accessibility of Pt active sites and the B‐induced modulation of the Pt lattice/electronic structure, which together optimized the binding of oxygenated intermediates and accelerated ORR kinetics [45]. The Tafel slopes derived from the LSV curves were 109.4, 138.0, 175.2, and 98.6 mV dec⁻¹ for B‐Pt/CNBs, Pt/CNBs, CNBs, and commercial Pt/C, respectively (Figure 3c). The smaller Tafel slope of B‐Pt/CNBs relative to Pt/CNBs and CNBs indicated faster ORR kinetics, although commercial Pt/C exhibited a slightly lower value. This improvement could be ascribed to B‐induced lattice/electronic modulation, which optimized the adsorption strength of oxygenated intermediates on the catalyst surface. Electrochemical impedance spectroscopy further supported this interpretation. The Nyquist plots (Figure 3d) showed that B‐Pt/CNBs exhibited a smaller semicircle diameter than Pt/CNBs and CNBs and a charge‐transfer resistance comparable to or lower than that of commercial Pt/C, indicating more favorable interfacial charge‐transfer kinetics after B doping.

To further investigate the ORR pathway, LSV measurements were conducted at different rotation rates. The limiting current density gradually increased with increasing rotation speed (Figure 4a), and the corresponding Koutecky–Levich (K–L) plots exhibited good linearity, with a calculated electron transfer number (n) close to 4 (Figure 4b), indicating that the ORR on B‐Pt/CNBs predominantly proceeded via a four‐electron pathway. This reaction route was further verified by RRDE measurements, which showed an H₂O₂ yield below 5%, lower than that of commercial Pt/C, together with an n value higher than 3.9 (Figure 4c). These results confirmed that B‐Pt/CNBs mainly catalyzed oxygen reduction through an efficient 4e⁻ pathway to OH⁻ in alkaline media. Furthermore, the kinetic current densities were normalized to the ECSA and Pt mass to evaluate the SA and MA at 0.85 and 0.90 V versus RHE. The SA values of B‐Pt/CNBs reached 3.90 and 3.68 mA cm⁻², respectively, both higher than those of Pt/CNBs (3.59 and 3.00 mA cm⁻²) and commercial Pt/C (2.73 and 2.45 mA cm⁻²) (Figure 3e). Meanwhile, the corresponding MA values of B‐Pt/CNBs were 2.77 and 2.61 A mg_Pt ⁻¹, respectively, which also exceeded those of Pt/CNBs and commercial Pt/C (Figure 3f). The enhanced ORR activity of B‐Pt/CNBs could be attributed to the synergistic effect of B‐induced lattice expansion and electronic structure modulation, which optimized the binding of key oxygenated intermediates and facilitated ORR kinetics.

FIGURE 4.

(a) LSV curves of B‐Pt/CNBs at different rotation rates. (b) Koutecky–Levich (K–L) plots of B‐Pt/CNBs. (c) Comparison of the electron transfer number (n) and H₂O₂ yield for B‐Pt/CNBs and Pt/C. (d) Chronoamperometric durability tests. (e) LSV curves before and after 3000 CV cycles. (f) Methanol tolerance tests of B‐Pt/CNBs and Pt/C.

In addition to catalytic activity, stability is another important parameter for evaluating the ORR performance. To assess the durability of the catalysts, chronoamperometric tests and LSV measurements before and after 3000 CV cycles were conducted. The LSV curves of B‐Pt/CNBs after 3000 CV cycles showed a small decrease in E_1/2 compared to the original curves (5 mV) and were significantly lower than those of the LSV curves of Pt/C (15 mV) after the multi‐turn CV cycling test (Figure 4e). Moreover, the stability of the samples was further investigated by prolonged i‐t tests. After 7200 s of the chronoamperometric test, the current density of B‐Pt/CNBs decreased by almost a negligible amount, while the current density of commercial Pt/C decayed by 25.70% (Figure 4d). These results demonstrated the superior ORR stability of B‐Pt/CNBs, which could be attributed to the structural stabilization induced by B doping. A more stringent accelerated durability test further confirmed this advantage [46,47]. After 10,000 CV cycles, the E_1/2 of B‐Pt/CNBs showed only a 5 mV negative shift, whereas commercial Pt/C exhibited a much larger loss of 21 mV (Figure S16), highlighting the remarkable long‐term electrochemical stability of the B‐doped catalyst. Post‐cycling structural characterization provided further evidence for the robustness of B‐Pt/CNBs. SEM and TEM images recorded after 10,000 CV cycles (Figures S17 and S18) show that the nanobowl morphology and the dispersion of Pt nanoparticles were largely preserved, indicating that the catalyst framework remained stable during prolonged ORR operation. The EDS elemental mapping images of B‐Pt/CNBs after 10,000 CV cycles further reveal the retained distribution of C, Pt, and B throughout the catalyst (Figure S19), suggesting that the elemental distribution did not undergo obvious segregation after the durability test. In addition, the particle size distribution of Pt nanoparticles after cycling is centered at approximately 2.96 nm (Figure S20), only slightly larger than that of the fresh B‐Pt/CNBs, indicating that severe Pt nanoparticle aggregation was effectively suppressed. By contrast, the SEM and TEM images of Pt/CNBs after the stability test (Figures S21 and S22) show more obvious structural deterioration and less uniform particle dispersion, suggesting that the undoped catalyst is less resistant to structural evolution under electrochemical operating conditions. Moreover, the XRD pattern of B‐Pt/CNBs collected after the stability test remained essentially unchanged compared with that before the reaction (Figure S23), further confirming the preservation of the crystalline Pt framework during ORR electrocatalysis. Furthermore, the ECSA of B‐Pt/CNBs after 10,000 CV cycles was further evaluated to assess the retention of electrochemically active Pt sites. The corresponding ECSA retention is further discussed in Figure 5e.

FIGURE 5.

(a) XPS survey spectrum of B‐Pt/CNBs after 10 000 CV cycles. Deconvoluted high‐resolution XPS spectra of (b) C 1s, (c) Pt 4f, and (d) B 1s after 10 000 CV cycles. (e) ECSA values of B‐Pt/CNBs before and after 10 000 CV cycles. (f) ORR free‐energy profiles on Pt(111) and B‐Pt(111) at U = 1.23 V versus RHE. (g) Optimized surface structures and adsorption configurations of ORR intermediates on Pt(111) and B‐Pt(111).

In practical applications, methanol tolerance is also an important criterion for ORR catalysts. The methanol resistance of B‐Pt/CNBs and Pt/C was tested by adding a methanol solution in 0.1 M KOH. As shown in Figure 4f, the current density of B‐Pt/CNBs decreased by only 7.44% after the addition of methanol, whereas that of Pt/C decreased by 24.16%, demonstrating the superior methanol tolerance and practical potential of B‐Pt/CNBs [48]. Overall, the outstanding methanol tolerance, high activity, and excellent durability of B‐Pt/CNBs highlight their strong potential for practical ORR applications.

To further evaluate the structural stability and mechanistic origin of the enhanced ORR performance, post‐cycling XPS analysis, ECSA retention, and DFT calculations were performed. As shown in Figure 5a, the survey XPS spectra of B‐Pt/CNBs before and after 10,000 CV cycles still show the characteristic signals of C, O, Pt, and weak B‐related species, indicating that the main elemental composition of the catalyst is largely retained after long‐term electrochemical cycling. The high‐resolution C 1s spectra (Figure 5b) show that the C─C peak remains dominant after cycling, while the C─O and O─CO peaks exhibit only slight changes, suggesting that the CNB support is largely preserved with limited surface oxidation. In the Pt 4f spectra (Figure 5c), Pt⁰ remains the dominant species after 10,000 CV cycles, accompanied by a slight increase in Pt²⁺ peaks, which may be associated with mild surface oxidation or adsorbed oxygenated species formed during the ORR operation. Importantly, the B 1s signal remains detectable after cycling (Figure 5d), although it is still relatively weak because of the low B content and the limited XPS response of boron. The retained B‐related peaks suggest that B species are not completely lost during the durability test, while the slight variation in oxidized B species may originate from partial oxidation of surface‐exposed B during electrochemical cycling and air exposure. Furthermore, the ECSA of B‐Pt/CNBs after 10,000 CV cycles was further evaluated to assess the retention of electrochemically active Pt sites. As shown in Figure 5e, the ECSA of B‐Pt/CNBs slightly decreased from 35.56 to 32.81 m² gPt⁻¹ after 10,000 CV cycles, corresponding to an ECSA retention of approximately 92.3%, further indicating that most active Pt sites were preserved during the long‐term ORR durability test.

The C 1s spectrum changes only slightly after cycling, with no distinct new species being detected (Figure 5b). More evident changes are found in the Pt 4f region. The Pt⁰ 4f7/2 and Pt⁰ 4f5/2 peaks remain clearly resolved, whereas the relative contribution of Pt²⁺ increases moderately compared with that of the initial spectrum (Figure 5c). This change is more consistent with the partial oxidation of surface Pt and the adsorption of oxygen‐containing species during ORR, followed by air exposure after the test. Since XPS reflects the surface chemical state rather than dissolved Pt species, the increase in Pt²⁺ is not taken as direct evidence of Pt dissolution. The continued presence of the metallic Pt peaks is also consistent with the relatively small decrease in the ECSA after cycling. The B 1s spectrum provides further information on the chemical evolution of B species (Figure 5d). The low‐binding‐energy peak assigned to Pt‐B‐related species or B⁰ remains discernible after 10 000 cycles, together with the B─O and BO _x peaks. Compared with the initial spectrum in Figure 2f, the oxidized B contributions become relatively more pronounced. This variation can be attributed to the partial oxidation of surface‐exposed B during electrochemical cycling and subsequent sample handling in the air. The retained low‐binding‐energy peak indicates that B‐related species associated with Pt are still present after the durability test, although the remaining B content cannot be quantified from these spectra alone.

DFT calculations were further performed to compare the four‐electron ORR free‐energy profiles on Pt(111) and B‐Pt(111). As shown in Figure 5f, the calculated pathway at U = 1.23 V follows O₂ + * → OOH* → O* → OH* → H₂O + *. As shown in Figure 5f, the Gibbs free‐energy changes for the four elementary steps on Pt(111) at U = 1.23 V are 0.29, −1.20, 0.46, and 0.45 eV, respectively. The corresponding values on B‐Pt(111) are 0.34, −1.14, 0.38, and 0.41 eV, respectively. Although OOH* formation is slightly more endergonic on B‐Pt(111), the subsequent proton‐coupled electron‐transfer steps are thermodynamically more favorable. Specifically, the free‐energy change for the conversion of O* to OH* decreases from 0.46 eV on Pt(111) to 0.38 eV on B‐Pt(111), while that for the conversion of OH* to H₂O decreases from 0.45 to 0.41 eV.

Consequently, the largest uphill free‐energy change is reduced from 0.46 eV on Pt(111) to 0.41 eV on B‐Pt(111). The corresponding optimized surface models and intermediate configurations are shown in Figure 5g. These results indicate that B doping modifies the ORR thermodynamic landscape and facilitates the later conversion and removal of oxygen‐containing intermediates, providing thermodynamic support for the improved ORR activity of B‐Pt/CNBs.

3. Conclusion

In summary, a boron‐doped Pt catalyst supported on CNBs (B‐Pt/CNBs) is successfully constructed using DMAB as the reducing agent, affording an ultralow‐Pt catalyst with excellent ORR performance in alkaline media. Boron doping not only induces lattice expansion but also modulates the local electronic structure of Pt. DFT free‐energy calculations further show that the largest uphill free‐energy change decreased from 0.46 eV on Pt(111) to 0.41 eV on B‐Pt(111), facilitating the later conversion of oxygen‐containing intermediates. The lattice expansion and electron redistribution induced by B doping reshape the ORR free‐energy landscape, facilitating the protonation and removal of O and OH and promoting the overall ORR process. As a result, B‐Pt/CNBs delivered an E_1/2 of 0.966 V versus RHE, outperforming commercial Pt/C under alkaline conditions. In addition, both the MA and SA were markedly higher than those of commercial Pt/C at 0.85 and 0.90 V versus RHE, respectively. The obtained B‐Pt/CNBs also exhibited excellent durability, with only a 5 mV loss in E_1/2 after 3000 CV cycles and negligible current decay during a 7200 s chronoamperometric test. Even after 10 000 CV cycles, B‐Pt/CNBs retained 92.3% of the initial ECSA and showed only limited particle growth without a major crystalline phase change, indicating their outstanding structural stability. The consistency between the calculated free‐energy profiles and the experimental results further supports the role of B doping in facilitating the conversion of oxygen‐containing intermediates during the ORR. Overall, this B‐doping strategy provides a promising route for the rational design of high‐performance, low‐Pt ORR electrocatalysts for fuel cells and related sustainable energy technologies.

Funding

This work was supported by the National Natural Science Foundation of China (52402273, 52072197, and 52274308), Youth Innovation Team Development Program of Shandong Higher Education Institutions (2022KJ155), and Taishan‐Scholar‐Young‐Talent Program (tsqn201909114).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Data Availability Statement

Research data were not shared.