Trace Alloying Unleashed: AuPt Nanoalloys Strategy Preserves Plasmonic Properties on Au Nanobipyramids While Boosting Electron Transfer for Visible Light Ammonia Synthesis

ABSTRACT

Photocatalysis, leveraging the redox capabilities of photocatalysts under light irradiation, emerges as a promising approach for clean energy conversion and pollution control. In this study, we engineered Au nanobipyramids (Au NBPs) with trace amounts of AuPt alloy to enhance their photocatalytic efficiency for selective ammonia synthesis. By modulating the reduction kinetics and precursor ratios, we synthesized three distinct Pt configurations: dense Pt layers (Au NBPs@Pt_d), sparse Pt clusters (Au NBPs@Pt_s), and trace AuPt alloy (Au NBPs@Au/Pt_alloy). Among them, Au NBPs@Au/Pt_alloy exhibited superior performance in photoelectrocatalytic nitrite‐to‐ammonia conversion, achieving a 10‐fold increase in ammonia production compared to Au NBPs and a 1.26‐fold enhancement under illumination vs. dark conditions. Multimodal characterization revealed that the ultra‐low AuPt alloy loading preserved the intrinsic localized surface plasmon resonance (LSPR) properties of Au NBPs while enhancing hot carrier generation and interfacial electron transfer efficiency. Density functional theory (DFT) calculations further confirmed that AuPt alloying optimized reaction free energy profiles by reducing adsorption barriers for key intermediates. This work not only advances the rational design of plasmonic catalysts but also demonstrates the potential of trace metal alloying for achieving high selectivity in visible‐light‐driven catalytic reactions, holding significant implications for sustainable energy conversion.

Au nanobipyramids (Au NBPs) with distinct Pt configurations—dense Pt layers (Au NBPs@Pt_d), sparse Pt clusters (Au NBPs@Pt_s), and trace AuPt alloy (Au NBPs@Au/Pt_alloy) were engineered. Au NBPs@Au/Pt_alloy exhibited optimal performance in the photoelectrocatalytic nitrite‐to‐ammonia conversion. Multimodal characterization revealed that the ultra‐low AuPt alloy loading preserved the intrinsic localized surface plasmon resonance (LSPR) properties of Au NBPs while enhancing hot carrier generation and interfacial electron transfer efficiency.

1. Introduction

The increasing energy crisis has intensified the demand for efficient utilization of clean and renewable energy sources [1]. Solar energy, as a promising alternative to traditional fossil fuels, plays a pivotal role in advancing industrial chemical transformations [2, 3]. Photocatalysis, leveraging the redox capabilities of photocatalysts under light irradiation, has emerged as a critical approach for pollutant degradation [4], chemical synthesis [5], and energy conversion [6]. Noble metal nanoparticles (NPs), particularly those exhibiting localized surface plasmon resonance (LSPR) effect, have garnered significant attention due to their exceptional visible‐light harvesting efficiency, intense electromagnetic field confinement, and the capacity to generate high‐energy hot electron–hole pairs via non‐radiative decay [7, 8, 9, 10, 11]. These properties not only enhance photocatalytic efficiency but also enable precise control over reaction pathways and product selectivity, as demonstrated in CO₂ reduction [12], methanol‐to‐hydrogen conversion [13], ammonia synthesis [14, 15, 16, 17], Suzuki C–C coupling reaction [18], and so on [19, 20].

Despite these advantages, intrinsic limitations of plasmonic NPs—such as insufficient catalytic active sites and rapid recombination of photogenerated carriers—hinder their practical applications [21, 22, 23]. To address these challenges, heterometallic architectures combining plasmonic metals (e.g., Au) with catalytic metals (e.g., Pt and Ru) have been strategically designed. Such bimetallic systems circumvent Schottky barrier formation and synergistically integrate plasmonic “nanoantennas” for light harvesting with catalytic “nanoreactors” for surface reactions [24, 25]. This nanoantenna‐reactor paradigm enhances both catalytic efficiency and selectivity by spatially separating light absorption and catalytic activation [26, 27]. Notably, interfacial hybridization between plasmonic and catalytic metals induces hybrid surface states that enable chemical interface damping (CID), a mechanism distinct from conventional Landau damping. CID facilitates direct hot electron transfer to adsorbate orbitals while retaining hot holes in plasmonic metals, thereby minimizing energy loss and improving charge utilization efficiency [28, 29, 30]. However, inappropriate structural design in hybrid systems may suppress intrinsic LSPR absorption and weaken electromagnetic field enhancement, ultimately compromising hot carrier generation. Traditional ensemble‐level spectroscopic techniques, often plagued by inhomogeneous spectrum broadening due to polydisperse nanoparticles, further complicate mechanistic studies [31, 32, 33]. To overcome these limitations, single‐particle spectroscopy techniques have been developed to precisely correlate LSPR linewidth broadening with plasmon damping pathways, offering unprecedented insights into plasmon decay dynamics [34, 35].

In this work, we engineered Au NBPs with distinct Pt configurations—dense Pt layers (Au NBPs@Pt_d), sparse Pt clusters (Au NBPs@Pt_s), and trace AuPt alloy (Au NBPs@Au/Pt_alloy)—by modulating reduction kinetics and precursor ratios. Au NBPs@Au/Pt_alloy exhibited optimal performance in the photoelectrocatalytic nitrite‐to‐ammonia conversion, achieving 10‐fold enhancement in NH₃ generation than Au NBPs and a 1.26‐fold enhancement under illumination vs. dark conditions (Scheme 1). Unlike conventional Pt deposition methods that often quench LSPR, the alloying strategy maintains strong plasmonic absorption while enhancing hot electron transfer and catalytic selectivity. Multimodal characterization combining single‐particle spectroscopy and theoretical simulations revealed that ultra‐low AuPt alloy loading in Au NBPs@Au/Pt_alloy preserved the intrinsic LSPR properties of Au NBPs while enhancing hot carrier generation and interfacial electron transfer efficiency. Density functional theory (DFT) calculations further confirmed that AuPt alloying optimized reaction free energy profiles by reducing adsorption barriers for key intermediates. This work establishes a structure‐activity‐mechanism nexus, demonstrating that minimal yet strategic metal loading maximizes plasmonic catalytic performance without compromising light‐harvesting capabilities. Our findings advance the rational design of plasmonic catalysts with controlled reaction selectivity under visible‐light excitation. This work establishes a new paradigm for designing plasmonic catalysts through atomic‐level interface engineering, with broad implications for solar‐driven chemical synthesis.

SCHEME 1.

Schematic illustration of tracing AuPt alloy on Au nanobipyramids for the selective and highly efficient photoelectrocatalytic ammonia synthesis.

2. Results and Discussion

2.1. Bimetallic Nanobipyramids Preparation and Microscopic Characterizations

The synthesis procedures for Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy are depicted in Figure 1A. Small‐sized Au NBPs were fabricated via a thermal‐induced seed‐mediated approach, utilizing pentatwinned Au seeds [36]. This process yielded Au NBPs with a longitudinal LSPR wavelength in the near‐infrared region (814 nm). To shift the LSPR resonance wavelength to the visible region for single‐particle spectroscopic analysis, these small‐sized Au NBPs were further grown as seeds, resulting in Au NBPs with a longitudinal LSPR resonance peak at 638 nm. To deposit Pt onto the surface of Au NBPs, an Ag replacement strategy was adopted to facilitate the deposition of Pt (Figure 1A).

FIGURE 1.

(A) Schematic illustration of the synthesis of Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy nanostructures. (B) Extinction spectra of Au NBPs (red line), Au NBPs@Ag (purple line), Au NBPs@Pt_d (blue line), Au NBPs@Pt_s (green line), and Au NBPs@Au/Pt_alloy (orange line). (C–F) DFM images (I), the single‐particle scattering intensity distribution (II), TEM images (III), and SEM images (IV) of Au NBPs (C), Au NBPs@Pt_d (D), Au NBPs@Pt_s (E), and Au NBPs@Au/Pt_alloy (F), respectively. All images in the same row share identical scale bars. (G–I) Minor axis (G), major axis (H), and tip curvature diameter distributions (I) of Au NBPs statistically analyzed from TEM images, respectively. (J) Pt shell thickness distribution of Au NBPs@Pt_d. (K) Pt cluster size distribution on Au NBPs@Pt_s statistically analyzed from TEM images. Solid lines represent Gaussian fitting curves. (L–O) HAADF‐STEM and elemental mapping images of Au NBPs (L), Au NBPs@Pt_d (M), Au NBPs@Pt_s (N), and Au NBPs@Au/Pt_alloy (O), respectively. The lower right corner images show the overlapped Au and Pt elemental distributions. All elemental maps share identical scale bars with their corresponding HAADF‐STEM images.

The synthesis processes of Au NBPs@Pt_d and Au NBPs@Pt_s are jointly governed by both kinetic and thermodynamic factors. With the same amount of Pt added, when the molar ratio of AA to H₂PtCl₆ is 10:3, the content of AA is relatively low, and the reduction rate of H₂PtCl₆ is slow. In this case, the deposition rate of Pt atoms is lower than the surface diffusion rate, and the process is mainly thermodynamically controlled, forming a dense Pt layer on the surface of Au NBPs. Conversely, when the molar ratio of AA to H₂PtCl₆ is 50:1, H₂PtCl₆ is rapidly reduced, with the deposition rate of Pt atoms exceeding the surface diffusion rate. Under kinetic control, the generated Pt clusters predominantly deposit rapidly along the edges of Au NBPs, forming a sparse Pt loading. As shown in Figure 1B, regardless of whether Pt is loaded densely or sparsely on Au NBPs, redshifts and broadening of the LSPR peaks are observed, and followed with the decrease in extinction efficiency.

To mitigate the attenuation of the plasmon resonance intensity of Au NBPs caused by the above‐mentioned sparse or dense Pt loading while introducing the catalytic metal Pt, a trace amount of AuPt alloy was loaded onto Au NBPs, constructing the Au NBPs@Au/Pt_alloy structure. Its extinction spectrum is shown in Figure 1B. The LSPR peak intensity of Au NBPs with the alloy structure did not decrease significantly, and the slight redshift of the resonance peak preliminarily verifies the successful synthesis of this material. Dark‐field optical microscopy (DFM) analysis reveals that the prepared Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy particles exhibit excellent monodispersity and uniform scattering signals in solution (Figure 1C–F). By comparing the single‐particle scattering intensities of each material, it can be seen that the structures of Au NBPs@Pt_d and Au NBPs@Pt_s are darker than those of Au NBPs and Au NBPs@Au/Pt_alloy, and Au NBPs@Pt_s shows the darkest orange‐red color, which is consistent with the changes in the extinction spectra.

To determine the morphology and size of these four materials, transmission electron microscopy (TEM) and scanning electron microscopy (SEM) characterizations were conducted. As shown in Figure 1C–F, Au NBPs@Pt_d forms a dense shell outside Au NBPs, and the corners at both ends of the shell are slightly sharper than those of Au NBPs. Cluster‐like Pt NPs are uniformly distributed on Au NBPs in the Au NBPs@Pt_s structure. Due to the low loading amount, the size of Au NBPs@Au/Pt_alloy is nearly identical to that of Au NBPs. Statistical analysis of the size of Au NBPs and the Pt thickness of Au NBPs@Pt_d and Au NBPs@Pt_s was performed by TEM, as shown in Figure 1G–K. The synthesized Au NBPs have a short‐axis width of 43.7±2.01 nm, a long‐axis length of 73.9±11.27 nm, and a tip arc diameter of 22.3±1.55 nm. The Pt shell of Au NBPs@Pt_d is approximately 1.53±0.29 nm, and the size of Pt clusters in Au NBPs@Pt_s is about 3.62±0.46 nm.

Subsequently, high‐angle annular dark‐field scanning transmission electron microscopy (HAADF‐STEM) combined with energy‐dispersive X‐ray spectroscopy (EDS) was employed to analyze the elemental distribution of each material (Figure 1L–O). The results show that the Pt element is distributed on Au NBPs in all three Pt‐loaded materials, confirming the successful loading of Pt. In Au NBPs@Pt_d, each particle has an Au NBP core, and Pt is uniformly and densely distributed on the surface of Au NBPs. In the Au NBPs@Pt_s structure, a sparser distribution of Pt on Au NBPs compared to Au NBPs@Pt_d can be observed. Because the amount of Pt added in Au NBPs@Au/Pt_alloy is extremely low, only a trace amount of AuPt fusion can be observed in the overlapping image of Au and Pt. Line‐scan EDS analysis across a single Au NBPs@Au/Pt_alloy particle demonstrates Au and Pt signals located at the same position (Figure S1). The specific contents of Au and Pt elements in the four materials were determined by inductively coupled plasma mass spectrometry (ICP‐MS), and the results are presented in Table 1. With the same amount of H₂PtCl₆ added, the Pt reduction rate of Au NBPs@Pt_s is higher than that of Au NBPs@Pt_d, which is attributed to the higher amount of reducing agent AA added during the synthesis of Au NBPs@Pt_s, resulting in a deposition rate greater than the diffusion rate and more Pt clusters formed. Notably, the extremely low Pt content of Au NBPs@Au/Pt_alloy, accounting for only 0.57 wt.%, provides evidence for the speculation on the reason why its LSPR peak intensity is almost the same as that of Au NBPs.

TABLE 1.

Au and Pt mass concentrations in Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy determined by ICP‐Mass spectroscopy. wt.% is the mass percentage and at.% is the atomic percentage.

Material	Au wt.%	Pt wt.%	Au at.%	Pt at.%
Au NBPs	100	—	100	—
Au NBPs@Ptd	95.89	4.11	95.85	4.15
Au NBPs@Pts	89.30	10.70	89.20	10.8
Au NBPs@Au/Ptalloy	99.43	0.57	99.42	0.58

2.2. Electronic Structure and Crystallographic Analysis of Bimetallic Nanobipyramids

To gain a deeper understanding of the elemental valence states and crystal plane characteristics of these materials, X‐ray photoelectron spectroscopy (XPS) and X‐ray powder diffraction (XRD) were utilized for the characterization. As shown in Figure 2A–D, the high‐resolution XPS spectra of Au 4f for Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy all display distinct peaks at Au⁰ 4f_5/2 and Au⁰ 4f_7/2, indicating the existence of metallic gold. Similarly, the high‐resolution Pt 4f spectra exhibit peaks at Pt⁰ 4f_5/2 and Pt⁰ 4f_7/2 (Figure 2A,E–G), confirming the successful loading of metallic Pt in each material.

FIGURE 2.

(A) XPS spectra of Au NBPs@Au/Pt_alloy (orange line), Au NBPs@Pt_s (green line), Au NBPs@Pt_d (blue line), and Au NBPs (red line). (B) XPS spectra of Au NBPs@Pt_d (blue line) and Au NBPs (red line) at Au 4f region. (C) XPS spectra of Au NBPs@Pt_s (green line) and Au NBPs (red line) at Au 4f region. (D) XPS spectra of Au NBPs@Au/Pt_alloy (orange line) and Au NBPs (red line) at Au 4f region. (E–G) XPS spectra of Au NBPs@Pt_d, Au NBPs@Pt_s and Au NBPs@Au/Pt_alloy at Pt 4f region, respectively. (H) XRD patterns of Au NBPs@Au/Pt_alloy (orange line), Au NBPs@Pt_s (green line), Au NBPs@Pt_d (blue line), and Au NBPs (red line). (I–L) HR‐TEM images of Au NBPs (I), Au NBPs@Pt_d (J), Au NBPs@Pt_s (K), and Au NBPs@Au/Pt_alloy (L), respectively.

Notably, compared with pure Au NBPs, the peak positions of Au⁰ 4f_5/2 and Au⁰ 4f_7/2 in the Pt‐loaded materials show observable shifts. The Au 4f peaks of Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy have significant positive shifts relative to Au NBPs. Specifically, Au⁰ 4f_5/2 and Au⁰ 4f_7/2 (86.78 and 83.18 eV) in Au NBPs shift to 87.18 and 83.58 eV in Au NBPs@Pt_d, to 87.28 and 83.58 eV in Au NBPs@Pt_s, and to 87.08 and 83.48 eV in Au NBPs@Au/Pt_alloy. This change in the binding energy of Au 4f can be attributed to the enhanced electron interaction (charge transfer) between Au and Pt, which increases as the Pt content in the samples rises. In the Pt 4f energy‐level spectra, two prominent peaks of Pt⁰ 4f_5/2 and Pt⁰ 4f_7/2 are detected. For Au NBPs@Pt_d, the peak values are 73.88 and 70.38 eV; for Au NBPs@Pt_s, they are 73.98 and 70.58 eV; and for Au NBPs@Au/Pt_alloy, they are 73.58 and 70.28 eV. Significantly, these values in each material have negative shifts when compared to pure Pt⁰ (Pt 4f_5/2 at 74.53 eV and Pt 4f_7/2 at 71.2 eV). The positive shift of Au 4f and the negative shift of Pt 4f are ascribed to the electron‐transfer process, where Au serves as an electron donor, transferring electrons to the loaded Pt, which acts as an electron acceptor.

The XRD patterns of the materials, presented in Figure 2H, show characteristic peaks at 38.1°, 64.5°, and 77.5°, corresponding to the (111), (220), and (311) crystal planes of Au, respectively. However, the crystal planes of Pt are not clearly observable in these XRD patterns. This is likely because the amount of Pt added in the three Pt‐loaded materials is relatively low compared to Au, falling below the detection sensitivity of XRD. To accurately analyze the crystal planes of Pt in each material, high‐resolution transmission electron microscopy (HRTEM) measurements were carried out. The crystal planes were identified by calculating the lattice‐fringe spacing. As shown in Figure 2I, the HRTEM image of Au NBPs reveals the presence of the Au (111) crystal plane with a lattice spacing of 0.234 nm (measured using Gatan Digital Micrograph software). In the Au NBPs@Pt_d structure, the Pt (111) crystal plane with a lattice spacing of 0.22 nm is observed (Figure 2J), validating the deposition of Pt on Au NBPs. Additionally, at the outer edges of Au NBPs in the Au NBPs@Pt_s structure, the Pt (110) crystal plane with a lattice spacing of 0.275 nm is detected (Figure 2K), indicating the sparse loading of Pt clusters. In the Au NBPs@Au/Pt_alloy structure, a (111) crystal plane with a lattice spacing of 0.229 nm is identified (Figure 2L). This lattice spacing, which lies between that of Au (111) and Pt (111), indicates the formation of an AuPt alloy on Au NBPs. The random orientation of the lattice fringes further implies that the AuPt alloy has a polycrystalline nature.

2.3. Photoelectrochemical Performance of the Bimetallic Nanobipyramids

Upon illumination, the excitation of LSPR in Au NBPs can generate high‐energy hot carriers [37, 38]. The deposition of a secondary metal, Pt, on Au NBPs enables the transfer of generated hot electrons to Pt, improving the charge separation efficiency and concurrently enhancing the catalytic performance of Pt [25]. To elucidate the disparities in hot carrier generation capabilities and charge separation efficiencies among Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy, the photocurrent responses (i–t curves) of these four materials were investigated under xenon lamp illumination, as depicted in Figure 3A. All Pt‐loaded samples exhibited stronger photocurrents compared to pristine Au NBPs, indicating that the presence of Pt indeed boosts the charge separation efficiency of Au NBPs. Notably, Au NBPs@Au/Pt_alloy demonstrated the most pronounced photocurrent response, which can be attributed to multiple factors. The formation of AuPt alloy on the surface of Au NBPs@Au/Pt_alloy provides abundant pathways for hot electron transfer, facilitating efficient electron transport and significantly reducing the electron–hole recombination. Additionally, UV–vis extinction spectrum analysis reveals that Au NBPs@Au/Pt_alloy exhibits intense LSPR effect, leading to enhanced light absorption and utilization.

FIGURE 3.

(A–C) Photocurrents (A), OCVD (B) and average electron lifetimes (C) of Au NBPs (red line), Au NBPs@Pt_d (blue line), Au NBPs@Pt_s (green line), and Au NBPs@Au/Pt_alloy (orange line) under visible‐light irradiation (𝜆>420 nm) in Na₂SO₄ aqueous solution, respectively. (D) EIS plots of Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy.

To gain in‐depth insights into the charge recombination dynamics, the open‐circuit photovoltage decay (OCVD) process of these particles was meticulously studied [39]. The decay of open‐circuit voltage (V_oc) upon turning off the illumination under steady‐state conditions was monitored in situ, as shown in Figure 3B. When illumination ceases, V_oc decays promptly due to the recombination of electron–hole pairs. The decay rate of V _oc is closely related to the electron lifetime (τ _n), which can be calculated using equation:

$${\tau }_{\mathrm{n}}=-\frac{K_{\mathrm{B}}T}{e}{\left(\frac{d\left(V_{\mathrm{oc}}\right)}{\mathrm{d}t}\right)}^{-1}$$

where K _B T represents thermal energy and e is the elementary charge. Under illumination, the photovoltages of all three Pt‐loaded samples exceeded that of Au NBPs, with Au NBPs@Au/Pt_alloy exhibiting the highest photovoltage, consistent with the trends observed in the photocurrent curves. Furthermore, the analysis of measured electron lifetimes revealed that Au NBPs@Au/Pt_alloy possessed the longest electron lifetime (Figure 3C). This indicates that this structure can not only efficiently harness the LSPR of Au NBPs to absorb light energy and generate a larger quantity of hot carriers but also significantly enhance the carrier separation efficiency.

Electrochemical impedance spectroscopy (EIS) was further employed to understand the electrochemical processes occurring at the material interfaces [40]. EIS assesses the resistive and capacitive behaviors of an electrochemical system by applying alternating current potentials at varying frequencies and measuring the corresponding current responses [41]. The EIS measurements were conducted at open‐circuit potential, with a constant potential applied within the frequency range of 0.1 Hz to 100 kHz under illumination conditions. The radius of the semicircular Nyquist plot is indicative of the charge transfer process at the electrode/electrolyte interface; a smaller radius implies lower charge transfer resistance and more efficient charge transfer [42]. As evident from Figure 3D, Au NBPs@Au/Pt_alloy exhibited the smallest radius, signifying the most efficient interfacial charge transfer within this system.

2.4. Plasmonic Photocatalysis for 4‐Nitrophenol Hydrogenation Reduction

Based on the aforementioned analyses, Au NBPs@Au/Pt_alloy is predicted to exhibit relatively higher plasmonic photocatalytic activity due to its high hot electron generation efficiency and effective charge transfer capability. To verify this argument, the photocatalytic activities of the catalyst were explored using the reduction of 4‐Nitrophenol (4‐NP) to 4‐Aminophenol (4‐AP) by sodium borohydride (NaBH₄) in an aqueous solution under illumination as a model reaction, Figure 4A [43]. First, the catalytic performances of Au NBPs@Au/Pt_alloy structures prepared with different Au:Pt ratios in the HAuCl₄ and H₂PtCl₆ mixture were investigated under the same conditions (Figure 4B). The results indicated that when the Au:Pt ratio in the mixture was 9:1, the prepared Au NBPs@Au/Pt_alloy structure showed the highest 4‐NP conversion rate, demonstrating optimal catalytic performance. The mass ratio of Au and Pt from Au NBPs@Au/Pt_alloy under this ratio is shown in Table 1.

FIGURE 4.

Photocatalytic activities of Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy heterostructures toward the selective photoreduction of 4‐NP. (A) Diagram of catalytic reaction. (B) 4‐NP reduction conversion rates of Au NBPs@Au/Pt_alloy prepared with different HAuCl₄:H₂PtCl₆ molar ratios in mixed solutions. (C) ln(C₀/C_t)‐t plots for 4‐NP reduction by different catalysts. (D) The reaction rate constants for 4‐NP reduction with different catalysts. (E) The 4‐NP conversion of different photocatalysts was compared under light and dark conditions, and the reaction time was 10 min. (F–I) Relationship between 4‐NP conversion (left side) and catalyst extinction spectra (right side) for each catalyst at the same reaction time for light source with wavelengths at 650, 575, 540, and 450 nm, (F) Au NBPs, (G) Au NBPs@Pt_d, (H) Au NBPs@Pt_s, and (I) Au NBPs@Au/Pt_alloy. Solid line is the extinction spectrum of the material. (J) Variation of 4‐NP conversion for each catalyst under the same reaction conditions when different kinds of hole sacrificers were added. (K,L) Effect of different concentrations of hole sacrificers Na₂SO₃ (K) and methanol (L) on 4‐NP conversion under the same reaction conditions, using Au NBPs@Au/Pt_alloy as catalysts.

Figure 4C illustrates the change in 4‐NP concentration over reaction time for each catalyst at 25°C under light illumination. The reaction follows first‐order reaction kinetics. According to the time‐dependent curves, the reaction rate constants of each catalyst were calculated (Figure 4D): k_water = 1 × 10⁻⁴, k_{Au NBPs} = 7.71 × 10⁻³, k_{Pt NPs} = 8.8 × 10⁻³, k_{Au NBPs+Pt NPs} = 1.47 × 10⁻², k_{Au NBPs@Ptd} = 2.12 × 10⁻², k_{Au NBPs@Pts} = 4.04 × 10⁻² and k_{Au NBPs@Au/Ptalloy} = 0.1316 min⁻¹. When no catalyst was present (i.e., replacing the catalyst with an equal volume of water), the reaction barely occurred, emphasizing the critical role of catalysts in initiating this reaction. Au NBPs and Pt nanoparticles (Pt NPs) exhibited low catalytic activities in the photocatalytic reduction of 4‐NP due to insufficient reactive sites and weak light response capabilities, respectively. The photocatalytic activity of the mixture of Au NBPs and Pt NPs remained relatively low, indicating that an interface for hot carrier transfer between Au and Pt is necessary. Although there is a shared interface between Au NBPs and Pt in the Au NBPs@Pt_d nanostructure, it showed lower photocatalytic activity compared to other Pt‐loaded materials, suggesting a close correlation between charge separation and material topology. In the Au NBPs@Pt_d structure, the dense Pt layer on the surface of Au NBPs may hinder the consumption of hot holes in Au NBPs by reactants, leading to the recombination of hot carriers. In contrast, in Au NBPs@Pt_s, Pt is sparsely loaded on the edges of Au NBPs in the form of Pt clusters, providing sufficient space for reactant diffusion and facilitating charge separation, thus exhibiting better catalytic performance than Au NBPs@Pt_d. However, Au NBPs@Au/Pt_alloy outperformed Au NBPs@Pt_s by more than three times. This superiority is mainly attributed to the maintenance of the light absorption capacity of Au NBPs while enhancing charge separation and catalytic activity through the addition of catalytic metal Pt, maximizing the synergistic effect of the bimetallic antenna (Au)—reactor (Pt) system. Under dark conditions, each catalyst showed certain intrinsic catalytic activity for the reduction of 4‐NP, and the trend of activity changes was similar to that under illumination, but the reaction efficiency was significantly lower (Figure 4E; Figure S2). This result indicates that although the catalysts possess some catalytic ability in the absence of light, illumination significantly enhances their catalytic performance, highlighting the crucial role of the light‐driven effect in improving catalyst performance.

To clarify the plasmon‐mediated light‐driven effect, the influence of incident light wavelength on the performance of each catalyst was explored. As shown in Figure 4F–I, catalytic experiments for the hydrogenation reduction of 4‐NP were conducted using four incident light wavelengths (650, 575, 540, and 450 nm) under the same reaction time, light intensity, and other conditions. Evidently, all four photocatalysts exhibited significant wavelength dependence. The maximum 4‐NP conversion rate was achieved when the incident light wavelength was closest to the LSPR peak of the material. This outcome is attributed to the highest light absorption efficiency of Au NBPs at the LSPR, enabling the generation of more hot carriers for catalysis. This indicates that the light‐driven effect of each catalyst is mainly influenced by the LSPR of Au NBPs, confirming the role of Au as a nanoantenna in the bimetallic Au‐Pt photocatalyst.

The thermal effect in this work is minimal because the 4‐NP conversion rate in the solution with higher temperature (42.5°C, in dark) is still comparable with that at 25°C without light irradiation (Figure S3). Furthermore, the presence of hole scavengers confirmed that the conversion of 4‐NP is a hot electron‐mediated reduction reaction (Figure 4J–L). When light irradiates the surface of Au NBPs@Au/Pt_alloy, Au NBPs act as nanoantennas to capture photons, generating enhanced local electromagnetic fields and producing hot carriers through non‐radiative decay. Subsequently, hot electrons generated by Au NBPs can transfer to Pt through the Au‐Pt interface, while hot holes remain on Au NBPs. Due to electrostatic adsorption, the ─NO₂ groups are enriched on the negatively charged Pt surface, and BH₄ ⁻ is adsorbed on the positively charged Au NBPs, leading to the cleavage of the B─H bond and the formation of Au‐H intermediates. Active hydrogen is then transferred from the Au‐H intermediates to 4‐NP adsorbed on the Pt surface. Under the catalytic action of H provided by NaBH₄ and Pt sites, 4‐NP is gradually reduced to 4‐AP by hot electrons, highlighting the significance of hot electrons in this catalytic reaction. The addition of hole scavengers (such as Na₂SO₃ and methanol) consumes the holes on Au NBPs, reducing the recombination of hot electrons and holes, significantly improving their separation efficiency, and further enhancing the catalytic activity of the reaction. However, the concentration of hole scavengers requires optimization, as both excessively low and high concentrations can have adverse effects on the reaction (Figure 4K,L).

2.5. Electromagnetic Field Distribution and Hot Electron Generation Rate Analysis

The catalytic activity of plasmonic photocatalysts is intricately linked to the efficiency of hot electron generation. To analyze the hot electron generation rates of four structures, we conducted simulations using finite‐difference time‐domain (FDTD) [44, 45]. First, we performed FDTD simulations on the optical response characteristics of each structure (Figure S4). The extinction spectra obtained from the simulations matched well with the experimental results in terms of peak positions and shapes (Figure 5A–D), validating the reliability of the simulation results and providing a solid theoretical basis for subsequent analyses of the electric field and hot carrier generation efficiency. As shown in Figure S4, all four structures exhibited higher light absorption than scattering, indicating that the designed photocatalyst structures possess high light utilization capabilities and low light losses. Notably, Au NBPs@Au/Pt_alloy demonstrated the largest absorption cross‐section, suggesting its superior light‐harvesting ability.

FIGURE 5.

(A–D) Comparison of the longitudinal LSPR extinction spectra of Au NBPs (A), Au NBPs@Pt_d (B), Au NBPs@Pt_s (C), and Au NBPs@Au/Pt_alloy (D) simulated by FDTD with the experimentally measured extinction spectra. (E) From left to right are the FDTD‐simulated electromagnetic field distributions of Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy, respectively. (F) From left to right, the COMSOL simulated distributions of hot electron generation rates along Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@ Au/Pt_alloy, respectively. (G–J) Hot electron generation rates distributed along the long axis of Au NBPs (G), Au NBPs@Pt_d (H), Au NBPs@Pt_s (I), and Au NBPs@Au/Pt_alloy (J) simulated by COMSOL.

Figure 5E illustrates the electromagnetic field distributions of Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy. Au NBPs, benefiting from the tip effect, exhibited significant local field enhancements at both ends, making them excellent candidates as “nanoantennas.” In the Au NBPs@Pt_d structure, the dense Pt layer on the surface of Au NBPs facilitated electron transfer at the Au‐Pt interface, weakening the plasmon resonance and resulting in an electromagnetic field distribution lower than that of Au NBPs. When sparse Pt clusters were loaded on the edges of Au NBPs (Au NBPs@Pt_s), the higher Pt loading induced a plasmonic shielding effect, suppressing the LSPR of Au NBPs and leading to a lower electromagnetic field intensity. Conversely, for the Au NBPs@Au/Pt_alloy structure, the minimal Pt loading did not inhibit the plasmon resonance of Au NBPs.

In noble metal materials, the strong local electromagnetic field enhancement induced by plasmon resonance can excite electrons from the ground state to high‐energy states, generating non‐equilibrium hot electrons. Evidently, there is a direct positive correlation between the enhancement of the local electromagnetic field and the efficiency of hot electron generation. Figure 5F–J shows the hot electron generation rate distributions of the four materials obtained from COMSOL simulations, which align with the electromagnetic field enhancement patterns, following the order: Au NBPs@Au/Pt_alloy >Au NBPs >Au NBPs@Pt_d >Au NBPs@Pt_s. The superior photocatalytic activity of Au NBPs@Au/Pt_alloy is primarily attributed to its excellent hot electron generation capability. The presence of a small amount of Pt enables efficient transfer of generated high‐energy hot electrons to Pt, facilitating rapid reduction reactions at Pt sites. This not only improves the separation efficiency of hot electrons and holes but also enhances the effective utilization of hot electrons.

2.6. Plasmonic Energy Decay Analysis via Single‐Particle Spectroscopy

To give a quantitative picture on the scenario described above, we further performed single‐particle spectroscopic analysis to explore the details of plasmonic resonance damping pathways (see Supporting Information) [29], which is crucial for the understanding of energy transfer mechanisms of hot carriers from these nanostructures, Figure 6. During plasmon decay, part of the energy dissipates through radiative damping (Γ_rad), while the remaining energy generates hot carriers via interband (Γ_IB) and intraband (γ_b) transitions and transfers to the molecular orbitals of reactants: Γ_total = Γ_rad  +  Γ_nonrad = Γ_rad  +Γ_bulk+Γ_surface [46, 47]. Bulk damping (Γ_bulk) consists of intraband damping (γ_b) and interband damping (Γ_IB), which dissipate plasmonic energy as electron–hole pairs (Landau damping). It is related to the properties of plasmonic particles and can be described by the metal dielectric function: ${\mathrm{\Gamma }}_{\mathrm{bulk}}={\gamma }_{\mathrm{b}}+{\mathrm{\Gamma }}_{\mathrm{IB}}={\gamma }_{\mathrm{b}}+{\epsilon }_{2,\mathrm{IB}}·\frac{{{\mathrm{E}}_{\mathrm{res}}}^3}{{(\hslash {\omega }_{\mathrm{p}})}^2}$. For gold nanoparticles, γ_b = 73 meV, ℏω_p = 9 eV, and ε_2,IB (the imaginary part of the dielectric function for interband transitions) is calculated by:

$${\epsilon }_{2,\mathrm{IB}}=\mathrm{Im}\left({\epsilon }_{\mathrm{JC}}\right)-\frac{({\gamma }_{\mathrm{b}}{\hslash }^2{{\omega }_{\mathrm{p}}}^2)}{{{\mathrm{E}}_{\mathrm{photon}}}^3}$$

where Im(ε_JC) is the experimental measurement data of Johnson and Christy at the corresponding energy (E_photon). Using these formulas and single‐particle spectroscopic analysis, we obtained Γ_rad and Γ_bulk distributions of Au NBPs with different sizes (see Supporting Information and Figure S5). As the particle size increased, the proportion of Γ_bulk gradually decreases, while the contribution of Γ_rad increases. Surface damping (Γ_surface) includes electron surface scattering (Γ_e‐surface) and chemical interface damping (Γ_CID). Electron surface scattering occurs when the nanoparticle size is smaller than its mean free path (L_eff) (the mean free path of Au is around 42 nm), and Γ_e‐surface is inversely proportional to L_eff—larger particles exhibit smaller Γ_e‐surface. Γ_CID is an energy dissipation pathway caused by the participation of molecules or other media adsorbed on the particle surface in plasmon energy transfer, which is considered as one of the most effective energy transfer pathways for hot carriers in photocatalytic reactions. It is well‐known that the plasmon decay process is extremely fast, typically on the order of 5–20 fs. Such rapid relaxation makes it highly challenging to accurately measure the decay dynamics in a time‐resolved way. Single nanoparticle scattering spectroscopy provide a viable alternative to time‐resolved measurements because the plasmon line width Γ is directly proportional to the plasmon decay rate as noted above [29, 47, 48].

FIGURE 6.

(A–D) Dark‐field images of Au NBPs (A), Au NBPs@Pt_d (B), Au NBPs@Pt_s (C), and Au NBPs@Au/Pt_alloy (D), respectively. (E–H) The single‐particle scattering spectrum corresponding to the single particles circled in the dark‐field images above. (I) Damping distribution diagram of plasmon for four materials. (J) Surface damping (Γ_surface) and (K) proportion of surface damping to total damping (Q_surface) of Au NBPs (red), Au NBPs@Pt_d (blue), and Au NBPs@Pt_s (green) and Au NBPs@Au/Pt_alloy (orange). (L) Hot electron transfer mechanism in Au NBPs. (M) Hot electron transfer mechanism at Au‐Pt interfaces for Pt‐modified Au NBPs.

When Pt is loaded onto Au NBPs, hybrid electron energy levels formed at the interface, providing an additional pathway for plasmonic energy decay which can be clearly observed from the single‐particle scattering spectrum, Figure 6A–H and Figure S6. After LSPR excitation, the rapid recombination of electron–hole pairs on Au NBPs surface significantly limits the number of hot carriers transferred to reactant molecular orbitals, thereby reducing the catalytic activity [49]. When Pt or AuPt alloy is loaded on the surface of Au NBPs, the generated hot electrons can transfer to Pt not only indirectly but also directly through the strong interaction at the Au‐Pt interface (Figure 6L,M). This interaction induces Γ_CID, enabling direct electron transfer from Au to Pt while leaving hot holes in Au, thus reducing the recombination rate of electron–hole pairs and enhancing the utilization efficiency of hot electrons. This direct electron transfer process after Pt deposition is well reflected in the increase of the fraction of Γ_surface, Table 2. Consequently, these structures exhibit higher catalytic activities than pristine Au NBPs. Moreover, Au NBPs@Pt_s features a higher direct electron transfer process compared to Au NBPs@Pt_d, (Figure 6I–K), which more efficiently promotes hot electron transfer from Au to Pt and suppresses electron–hole recombination, well explaining the better catalytic performance over Au NBPs@Pt_d.

TABLE 2.

Damping calculations for four materials.

Material	λ (nm)	E (eV)	Γtotal (meV)	Γbulk(meV)	Γrad(meV)	Γsurface (meV)	Qbulk	Qrad	Qsurface
Au NBPs	642	1.93	166.72	99.23	41.67	25.82	0.60	0.25	0.15
Au NBPs@Ptd	656	1.89	274.09	73.09	61.80	139.2	0.27	0.23	0.5
Au NBPs@Pts	660	1.88	406.82	73.09	67.73	266	0.18	0.17	0.65
Au NBPs@Au/Ptalloy	645	1.92	171.61	73.09	42.29	56.22	0.43	0.25	0.32

It is worth of noting that both sparse and dense core–shell structures significantly affect the plasmon resonance effect of Au NBPs, thereby influencing the generation efficiency of hot carriers (Figure 1B). When a trace amount of AuPt alloy structure is loaded on the particle surface, the extinction performance of Au NBPs@Au/Pt_alloy remains unchanged, maintaining good extinction efficiency. In contrast to pure Au NBPs, a significant increase in surface damping is observed from Au NBPs@Au/Pt_alloy, which can be attributed to the effective CID effect from AuPt alloy, Table 2.

2.7. Photoelectrocatalytic Reduction of Sodium Nitrite to Ammonia

From the 4‐NP reduction experiments and theoretical analysis, we explored the photocatalytic performance differences among four plasmonic catalysts (Au NBPs, Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy), revealing the synergistic regulation mechanism of hot electron generation and transfer capabilities on catalytic activity. To further explore the application of alloy structure in plasmon‐driven catalytic reactions, we studied the performance in the photoelectrochemical reduction of nitrite to ammonia [50, 51]. The photoelectrocatalytic reduction of sodium nitrite to ammonia (NO₂RR) is a complex reaction involving multiple electron transfer steps and intermediate adsorption processes (e.g., NO₂ ⁻→NO→NH₂OH→NH₃, Figure 7A). Pure Au NBPs catalysts struggle to drive N─O bond cleavage and N─H bond formation due to a lack of active sites. Therefore, we employed these three supported catalysts with well‐defined active sites (such as Pt and AuPt alloy) to evaluate their applicability in this complex reduction system and further elucidate the selectivity enhancement induced by plasmon excitation.

FIGURE 7.

(A) Mechanistic scheme for the reduction of NO₂ ⁻ to NH₃ under alkaline conditions. (B) LSV curves of different catalysts with/without NO₂ ⁻. (C) LSV curves with/without light illumination. Color code: Au NBPs@Au/Pt_alloy (orange), Au NBPs@Pt_s (green), Au NBPs@Pt_d (blue). (D) NH₃ production yields and (E) Faradaic efficiencies of catalysts under identical reaction conditions with/without illumination. (F) Linear relationship between current density and scan rate for each catalyst. The slope represents double‐layer capacitance (C_dl). (G) Electrochemically active surface area (ECSA) calculated from C_dl. (H) ECSA‐normalized NH₃ production yields with/without light illumination.

First, linear sweep voltammetry (LSV) curves of Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy were measured in 0.1 M NaOH electrolyte solution with and without the addition of 10 mM NaNO₂ to assess their sodium nitrite reduction activities (Figure 7B) [52]. The LSV curves showed significant changes in current density after adding NaNO₂, indicating the activity of each catalyst in the sodium nitrite reduction reaction. Among them, Au NBPs@Au/Pt_alloy exhibited the most substantial current change, suggesting it possesses the highest NO₂RR activity. Subsequently, LSV curves of the three catalysts were tested in a mixed electrolyte solution of 0.1 M NaOH and 10 mM NaNO₂ under both illuminated and dark conditions (Figure 7C). Evidently, the current density responses under illumination were stronger than those in the dark, indicating that light irradiation enhances the performance of the catalysts.

To study the reaction selectivity, the amount of generated NH₃ was quantified using the indophenol blue method (Figure S7). As depicted in Figure 7D, the selectivity of Au NBPs@Au/Pt_alloy for NH₃ after 1 h of electrolysis at −1.6 V (vs. SCE) increased from 116.2 in dark to 146.52 mg L⁻¹ h⁻¹ mg_cat ⁻¹ under photoexcitation, while Au NBPs@Pt_d (14.4 in dark and 37.3 mg L⁻¹ h⁻¹ mg_cat ⁻¹ under illumination) and Au NBPs@Pt_s (19.1 in dark and 61.3 mg L⁻¹ h⁻¹ mg_cat ⁻¹ under illumination) showed relatively low NH₃ yields regardless of illumination conditions. Without Pt loading, Au NBPs exhibit very low NH₃ yields even under illumination (2.4 in dark and 14.3 mg L⁻¹ h⁻¹ mg_cat ⁻¹ under illumination). This can be attributed to the stronger light absorption of Au NBPs@Au/Pt_alloy and the efficient transfer of hot carriers to Pt‐based sites, leading to enhanced activity and thus facilitating higher NH₃ formation. Additionally, the Pt centers with lower coordination numbers and Pt─O bonds on Au NBPs@Au/Pt_alloy may also contribute to the rapid hydrogenation to form ammonia, as these centers have a lower energy barrier for hydrogen evolution compared to metallic Pt centers. The calculated Faraday efficiencies were relatively low because protons (H*) in aqueous solution may be reduced to hydrogen (H₂), consuming parts of electrons and causing some charge to be unused for nitrite reduction (Figure 7E). Quantitative analysis of H₂ content in gaseous products revealed that the amount of H₂ generated accounts for merely 0.7% of the total NH₃ yield when using Au NBPs@Au/Pt_alloy as the catalyst (Figure S8). The high selectivity achieved herein confirms that our catalyst can effectively suppress such side reaction pathways. As summarized in Table S1, Au NBPs@Au/Pt_alloy demonstrates competitive or superior performance in NH₃ yield and Faradaic efficiency compared to other recently reported plasmonic NO₂RR catalysts under visible‐light illumination [50, 51, 53, 54, 55].

To further understand the reasons for the high NO₂RR activity of Au NBPs@Au/Pt_alloy, the electrochemically active surface areas (ECSA) of Au NBPs@Pt_d, Au NBPs@Pt_s, and Au NBPs@Au/Pt_alloy were measured by the electrochemical double‐layer capacitance method (Figure 7F; Figure S9) [56]. There is a direct proportional relationship between the double‐layer capacitance value (C_dl) and ECSA. The ECSA of Au NBPs@Au/Pt_alloy (50.965 cm²) was larger than that of Au NBPs@Pt_d (25.9 cm²) and Au NBPs@Pt_s (37.955 cm²). Due to the surface area effect, Au NBPs@Au/Pt_alloy exhibited higher catalytic activity for NO₂RR. Notably, the ECSA of Au NBPs@Au/Pt_alloy was only 1.34 times larger than that of Au NBPs@Pt_s, yet the NH₃ yield of Au NBPs@Au/Pt_alloy was 2.4 times higher than that of Au NBPs@Pt_s at −1.6 V (vs. SCE) (Figure 7G). This phenomenon indicates that Au NBPs@Au/Pt_alloy has a higher intrinsic activity for NO₂RR than Au NBPs@Pt_s. The specific activity of the catalyst, normalized by ECSA, can effectively reflect the intrinsic catalytic activity of the material. After normalizing the NH₃ yield by ECSA, the NH₃ yield of Au NBPs@Au/Pt_alloy remained higher than that of Au NBPs@Pt_s and Au NBPs@Pt_d, further confirming the higher intrinsic activity of Au NBPs@Au/Pt_alloy (Figure 7H). It is worth to note that the catalyst (Au NBPs@Au/Pt_alloy) exhibited excellent recyclability, retaining >90% activity over five cycles (Figure S10), which is attributed to the stable morphology and elemental states, as verified by post‐reaction TEM and XPS analysis (Figure S11).

2.8. Elucidating the Reaction Mechanism With DFT Calculations

To elucidate the reaction mechanism of the NO₂ reduction reaction (NO₂RR) and identify its rate‐determining step, DFT calculations were performed to investigate the charge density distribution, free energies, and adsorption energies of reaction intermediates [57, 58]. First, the d‐orbital projected density of states (PDOS) of Au NBPs@Au/Pt_alloy and Pt was calculated (Figure 8A; Figure S12A). The results revealed that Au NBPs@Au/Pt_alloy exhibited a lower d‐band center (d_c, −3.32 eV) compared to Au NBPs (−3.13 eV), indicating that even at low concentrations, Pt can modulate the electronic structure at the alloy interface. Additionally, work function calculations of Au NBPs@Au/Pt_alloy showed that the surface AuPt alloy had a lower Fermi level (E_f) and a larger work function (5.189 eV, Figure 8B,C). The contact between Au and the AuPt alloy induced charge redistribution around the interface, causing electrons to transfer from Au to AuPt until their work functions equilibrated. Consequently, hot electrons excited by LSPR accumulated at the AuPt sites, while holes concentrated on the Au side. Similarly, work function calculations were performed for pure Pt, revealing that Pt (5.656 eV) possesses a higher work function compared to Au (4.928 eV, Figure S12B). This indicates the existence of electron transfer from Au to Pt in both Au NBPs@Pt_d and Au NBPs@Pt_s structures. This charge separation suppressed electron–hole recombination, thereby enhancing the utilization efficiency of hot electrons.

FIGURE 8.

DFT theoretical simulations. (A) Projected density of states (PDOS) calculations for d‐orbitals of Au NBPs and Au NBPs@Au/Pt_alloy. (B,C) Work function calculations of Au (B) and AuPt alloy (C) in Au NBPs@Au/Pt_alloy. (D) Differential charge density distribution for *NO adsorption at Pt top site in Au NBPs@Au/Pt_alloy model (red: electron accumulation; blue: electron depletion. (E) Adsorption energies of *NO₂ ⁻, *NO and *NH₃ on (111) surfaces of Au NBPs and Au NBPs@Au/Pt_alloy model nanoparticles. (F) Free energy diagrams of NO₂RR intermediates on (111) surfaces of Au NBPs, Pt, and Au NBPs@Au/Pt_alloy models. (G) Schematic diagram of the photocatalytic mechanism of the Au NBPs@Au/Pt_alloy. (H) Reaction mechanism of sodium nitrite reduction to ammonia catalyzed by the Au NBPs@Au/Pt_alloy.

Figure 8D illustrates the charge density difference distribution of *NO adsorption on the Pt top surface within the Au NBPs@Au/Pt_alloy model. The adsorption of *NO led to significant local charge redistribution at the active Pt center, with electron density migrating from Pt to the adsorbed *NO species. Moreover, the adsorption energy of *NO and *NO₂ on Au NBPs@Au/Pt_alloy was notably higher than that on Au NBPs, indicating a stronger affinity of *NO and *NO₂ toward Au NBPs@Au/Pt_alloy (Figure 8E). Figure 8F presents the Gibbs free energy profiles of various reaction intermediates during the NO₂RR process based on the data from Table S2. For the Au NBPs, Pt, and Au NBPs@Au/Pt_alloy, the deoxydation of *NO₂ to *NO were identified as the rate‐determining step (RDS). Among them, Au NBPs@Au/Pt_alloy exhibited the lowest energy barrier (0.43 eV), compared to 0.60 eV for Au NBPs and 0.52 eV for Pt. This barrier could be further reduced under LSPR excitation. Additionally, the adsorption energy of *NH₃ on the catalyst surface showed that Au NBPs@Au/Pt_alloy had the lowest value, suggesting that *NH₃ could desorb more readily from its surface, contributing to higher catalytic activity. Therefore, both the electronic structure of Au NBPs@Au/Pt_alloy and its interaction with LSPR excitation collectively contribute to its superior NO₂RR activity and selectivity toward NH₃ (Figure 8G,H). In contrast to the NO₂RR pathway, the competing pathway for N₂ formation exhibits progressively higher energy barriers, indicating its negligible feasibility (Figure S13). This is further confirmed by in situ Fourier transform infrared spectroscopy (FT‐IR), where the intermediates associated in the proposed reaction processes are detected (Figure S14), such as the peaks attributed to ammonium ions (NH₄ ⁺, 1428 cm^{− 1}), nitric oxide (NO, 1540 cm^{− 1}), and hydroxylamine (NH₂OH, 1560 cm^{− 1}).

3. Conclusion

In this work, we present a robust plasmonic photocatalyst through the rational design of Au nanobipyramids (Au NBPs) with atomically engineered AuPt alloy interfaces. The key innovation lies in the development of a trace alloying strategy, where precisely controlled incorporation of AuPt alloy sites (Au NBPs@Au/Pt_alloy) achieves unprecedented synergy between plasmonic excitation and catalytic functionality. Unlike conventional Pt deposition methods that compromise plasmonic properties, our approach maintains strong LSPR while creating optimized active sites, as verified by multimodal characterizations. The alloyed interface demonstrates three distinct features: (1) nearly 100% preservation of plasmonic absorption from Au NBP, (2) enhancement in hot electron generation compared to Pt monolayer counterparts, and (3) selective activation of the nitrite‐to‐ammonia pathway with greatly improved NH₃ selectivity. This work establishes a new paradigm for designing plasmonic catalysts through atomic‐level interface engineering, with broad implications for solar‐driven chemical synthesis and environmental remediation.

4. Experimental Section

4.1. Synthesis of Gold Nanobipyramids (Au NBPs)

4.1.1. Synthesis of Au Seeds

The synthesis of Au NBPs adopted the seed growth method [25]. Pentatwinned Au Seeds were prepared initially. In a 10 mL glass vial, H₂O (1.125 mL), CTAC (0.1 M, 2.5 mL), HAuCl₄ (10 mM, 125 µL), and citric acid (0.02 M, 1.25 mL) were added in sequence. Subsequently, freshly prepared NaBH₄ (25 mM, 125 µL) in ice water was added under vigorous stirring. When the color of the mixed solution changed from light yellow to brown, the glass vial was placed in a water bath at 80°C and heated for 90 min. After cooling to room temperature, the solution was stored at 4°C.

4.1.2. Synthesis of Au NBPs_{814 nm}

AA (0.1 M, 80 µL) was added to a mixed solution of CTAB (0.1 M, 10 mL), HAuCl₄ (10 mM, 500 µL), AgNO₃ (10 mM, 100 µL), and HCl (2 M, 200 µL). The color of the solution changed from yellow to colorless. Then, 100 µL of purified Au seed was added to this mixed solution. Since Au NBPs are anisotropic nanomaterials and need to grow statically, the reaction was allowed to proceed statically at 30°C for 3 h. After the reaction, the synthesized Au NBPs were centrifuged, redispersed in deionized water, and stored at 4°C for use.

4.1.3. Synthesis of Au NBPs_{638 nm}

AA (1 mM, 25 µL) and HAuCl₄ (1 mM, 25 µL) were added to the synthesized Au NBPs_{814 nm} (10 mL) in sequence. The mixture was placed in a 37°C water bath and left to stand until the solution changed from reddish‐brown to peacock blue, and then the reaction was stopped. The reaction between AA and HAuCl₄ caused gold particles to deposit on the sides of Au NBPs, leading to further anisotropic growth of Au NBPs and a blueshift of its LSPR peak wavelength to 638 nm. After the reaction, the product was centrifuged, and the precipitate was dispersed in water and stored at 4°C.

4.2. Synthesis of Pt‐Loaded Composite Structures

4.2.1. Synthesis of Au NBPs@Ag

First, the purified Au NBPs_{638 nm} were dispersed into a CTAC solution (10 mM, 4 mL). Then, under stirring, AA (10 mM, 32 µL), AgNO₃ (1 mM, 20 µL), and NaOH (100 mM, 80 µL) were added in sequence. The mixed solution was reacted at 30°C for 30 min.

4.2.2. Synthesis of Au NBPs@Pt_d

AA (16.75 mM, 80 µL) and H₂PtCl₆ (2.5 mM, 160 µL) were added in sequence to 4 mL of Au NBPs@Ag, and the mixture was reacted at 30°C for 30 min. After the reaction, the obtained sample was centrifuged and dispersed in deionized water and stored at 4°C for later use.

4.2.3. Synthesis of Au NBPs@Pt_s

AA (0.1 M, 200 µL) and H₂PtCl₆ (10 mM, 40 µL) were added in sequence to 4 mL of Au NBPs@Ag, and the mixture was reacted at 30°C for 30 min. After the reaction, the obtained sample was centrifuged and dispersed in deionized water and stored at 4°C for later use.

4.2.4. Synthesis of Au NBPs@Au/Pt_alloy

AA (50 mM, 300 µL) was added to the purified Au NBPs_{638 nm} soluiton (dispersed in 4 mL 10 mM CTAC solution) and followed by the addition of a 9:1 mixed solution of HAuCl₄ (0.1 mM, 0.673 mL) and H₂PtCl₆ (0.1 mM, 0.075 mL). The reaction was carried out at 30°C for 60 min.

4.3. Photoelectrochemical Performance Testing

4.3.1. Fabrication of Catalyst‐Modified Indium Tin Oxide (ITO) Working Electrodes

20 µg of the catalyst was evenly dropped onto the ITO electrode (with an effective area of 0.25 cm²) in small portions multiple times. After natural drying, 5 µL of Nafion was evenly dropped onto the ITO electrode and left to dry naturally.

4.3.2. Testing of Photocurrent Response (i–t), Transient Open‐Circuit Voltage Decay (OCVD), and Electrochemical Impedance Spectroscopy (EIS)

The i–t and OCVD tests were conducted in a 0.5 M Na₂SO₄ electrolyte using a standard three‐electrode cell system on a Chenhua CHI 660E electrochemical workstation. The EIS test used a 1 mM potassium ferricyanide solution as the electrolyte. During the testing process, a Pt plate, a saturated calomel electrode, and the catalyst‐modified ITO electrode were used as the counter electrode, reference electrode, and working electrode, respectively. A xenon lamp (HSX‐F300) was used as the light source, and the photocurrent and photovoltage responses with and without illumination were recorded. The power of the xenon lamp was set to 20 mW/cm². In the i–t test, illumination was alternated on and off at 2.3 s intervals. The OCVD test lasted for a total of 200 s, and visible light was turned on and off at 50 and 150 s after the start, respectively. In the EIS test, under the irradiation (20 mW/cm²), the EIS curves in the frequency range of 0.1 Hz to 100 kHz were measured.

4.4. Photocatalytic Hydrogenation Reduction Experiment of 4‐Nitrophenol (4‐NP)

The photocatalytic performance of different structures was verified by the photocatalytic hydrogenation reduction of 4‐nitrophenol (4‐NP) to 4‐aminophenol (4‐AP) at room temperature and atmospheric pressure. A catalyst with an absorbance of 1 (1 cm optical path) at the LSPR absorption peak was added to a mixed solution containing 4‐NP (1 mM, 25 µL), NaBH₄ (100 mM, 25 µL), and 325 µL of deionized water, giving a total reaction volume of 500 µL. Subsequently, a white LED lamp (with a power meter reading of 5 mW) was used as the light source to irradiate the solution for a certain period of time. The change in the 4‐NP absorption peak in the range of 250–550 nm was monitored by UV–vis spectroscopy, and the catalytic performance of each catalyst was evaluated based on the change in the 4‐NP absorption peak. In the control experiment with and without a catalyst, an equal volume of water was used instead of the catalyst, and the experiment was carried out under the same conditions. To explore the wavelength dependence of the catalytic efficiency of each catalyst, filters of 450, 540, 575, and 650 nm were used to simulate light sources of different bands under a white‐light source. During the experiment, the light intensity under each light source was ensured to be the same, and the influence of light irradiation at different wavelengths on the catalytic reaction rate was observed. When exploring the influence of different hole scavengers on the catalytic reaction rate of 4‐NP, 25 µL of different concentrations of Na₂SO₃ or methanol were added respectively. To maintain a total volume of 500 µL in the catalytic reaction system, the amount of deionized water added was changed to 300 µL, and the addition amounts of other reagents and the reaction conditions remained the same.

Funding

This work was supported by the National Natural Science Foundation of China (NSFC, Project No. 22174079), Natural Science Foundation of Hunan Province (2024JJ3030).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: advs73365‐sup‐0001‐SuppMat.docx.

Data Availability Statement

All data needed to evaluate the conclusion in the paper are present in the paper and/or the Supplementary Materials.