Unveiling the Photocatalytic Behavior of PNTP on Au-Ag Alloy Nanoshells Through SERS

Abstract

Au-Ag alloy nanoshells (ANSs) were synthesized via chemical reduction, exhibiting superior plasmonic photocatalytic performance. TEM revealed uniform hollow structures (~80 nm), while EDS mapping confirmed homogeneous Au-Ag distribution throughout the shell. According to EDX analysis, the alloy contained 71.40% Ag by weight. XRD verified the formation of a substitutional solid solution without phase separation. The photocatalytic activity was evaluated using p-nitrothiophenol (PNTP) to 4,4′-dimercapto-azobenzene (DMAB) conversion monitored by SERS. UV-Vis spectroscopy showed LSPR peaks of ANSs between Au and Ag NPs, confirming effective alloying. Kinetic studies revealed that ANSs exhibited reaction rates 250–351 times higher than those of Au NPs and 5–10 times higher than those of Ag NPs. This resulted from the synergistic catalysis of Au-Ag and enhanced electromagnetic fields. ANSs demonstrated dual functionality as SERS substrates and photocatalysts, providing a foundation for developing multifunctional nanocatalytic materials.

1. Introduction

Photocatalysis is an environmentally friendly and sustainable approach for chemical transformations with significant applications in environmental remediation [1,2,3], energy conversion [1,2,3,4,5,6,7], and organic synthesis [1,2,5]. Photocatalysis is a process that operates fundamentally when the energy in light excites the electrons of the photocatalyst, generating electron–hole pairs that drive redox reactions [1,2].

Noble metal nanomaterials have attracted significant attention in photocatalysis due to their unique localized surface plasmon resonance (LSPR) effect [8,9,10]. LSPR enhances light absorption and increases catalytic efficiency through the amplification of local electromagnetic fields [11]. However, monometallic nanomaterials often involve inherent trade-offs that limit their overall effectiveness. These can range from suboptimal optical properties hindering sensitivity in sensing applications [12], to a compromise between catalytic activity and the structural stability of the material itself [13]. Furthermore, challenges include limited selectivity [14], the high cost of precious metals [15], and inferior electronic properties such as higher reaction overpotentials [16]. Au nanoparticles (NPs) exhibit excellent chemical stability but low catalytic activity [17]. In contrast, Ag NPs show high catalytic activity and improved surface-enhanced Raman scattering (SERS) performance but poor stability and sensitivity [18]. Bimetallic and multimetallic nanomaterials have emerged as promising alternatives to overcome the limitations of monometallic systems. Tuning alloy composition and structural morphology enables the modulation of LSPR and synergistic effects between metals [19,20,21,22]. Au-Ag alloy nanoshells (ANSs) show particular promise for photocatalysis owing to their high specific surface area, tunable LSPR, and hollow nanostructures [23,24,25,26,27]. For example, while Au@Ag nanorods [28] achieve rate constants of 0.030 s⁻¹ and NPAS10 systems [29] reach 0.0068–0.0070 s⁻¹, our synthesized ANSs exhibit significantly higher rate constants of 0.0562–0.0650 s⁻¹ under similar laser conditions, demonstrating the exceptional catalytic efficiency of the hollow alloy architecture among multimetallic nanomaterials.

P-nitrothiophenol (PNTP) is widely used to assess the plasmonic properties of noble metal nanomaterials [30,31,32]. In this study, it is employed as a probe molecule to evaluate the plasmonic catalytic activity of these materials. On noble metal surfaces, PNTP undergoes reduction to p-aminothiophenol (PATP) via hot electron transfer or intermolecular coupling to form 4,4′-dimercapto-azobenzene (DMAB) [33]. Notably, plasmon-driven photocatalytic conversion eliminates the need for additional chemical reagents and enables real-time SERS monitoring.

Despite considerable attention to ANSs’ catalytic performance, comprehensive systematic studies remain scarce. Studies on ANSs’ kinetic behavior and reaction mechanisms in plasmonic photocatalysis remain limited, as do systematic comparisons with monometallic nanoparticles [25,34,35]. Understanding the photocatalytic properties and SERS enhancement mechanisms of ANSs is essential for elucidating synergistic effects in bimetallic nanomaterials.

This study systematically investigates PNTP photocatalytic behavior on ANS surfaces using SERS as an in situ monitoring technique, demonstrating a reagent-free approach with real-time spectroscopic evaluation. Real-time monitoring of PNTP Raman spectral evolution is employed to elucidate the adsorption mechanism and reaction pathway on ANS surfaces. Kinetic analysis is conducted to quantify the catalytic performance of ANSs in comparison with monometallic Au and Ag NPs. The dual functionality of ANSs as both efficient photocatalysts and SERS substrates is demonstrated through comprehensive characterization and performance evaluation. This work provides fundamental insights into synergistic effects in bimetallic nanomaterials and offers experimental guidance for designing multifunctional nanocatalytic materials.

2. Results and Discussion

2.1. Microstructural Characterization

ANSs with an average particle size of approximately 80 nm were synthesized using chemical reduction. Figure 1a illustrates ANSs synthesis and structural evolution. Transmission Electron Microscopy (TEM) reveals the transformation of solid Ag nanoparticles into hollow alloy nanoshells. The hollow architecture enhances SERS and provides additional active sites for PNTP photocatalysis. TEM imaging, as shown in Figure 1b, reveals near-spherical ANSs with uniform size distribution and distinct hollow structures.

Elemental mapping analysis characterized the composition and spatial distribution of the ANSs. As shown in Figure 1c, energy-dispersive X-ray spectroscopy (EDS) analysis reveals that Au (green) and Ag (red) are concentrated in the nanoparticle shells with weak signals from the core. This confirms the hollow ANSs structure. The Au-Ag overlay image displays significant elemental overlap in the shell region, producing a continuous mixed signal. This indicates effective Au-Ag alloying at the nanoscale, forming a uniform alloy shell. The homogeneous elemental distribution throughout the shell is expected to promote high catalytic efficiency.

Energy-dispersive X-ray spectroscopy (EDX) analysis was performed at 200 kV to characterize the elemental composition of the sample (Figure 1d). Quantitative analysis reveals a binary Ag-Au alloy composition, comprising 71.40 ± 1.0 wt% (57.76 at%) Ag and 28.60 ± 1.0 wt% (42.24 at%) Au. The complete mass balance confirms the absence of detectable impurities within the analytical sensitivity limits. This composition represents a marked contrast to previously synthesized porous Au-Ag shells obtained through dealloying, which exhibited substantially lower silver content (9.2 ± 1.0 wt%, 15.6 at%) [29]. The high silver content in our ANSs (71.40 wt%) results from the precursor stoichiometry in our chemical reduction synthesis. During the synthesis, HAuCl₄ (0.9 mL, 50 mM) and AgNO₃ (0.9 mL, 100 mM) provide an Ag: Au molar ratio of 2:1. In citrate-mediated chemical reduction in Au-Ag alloys, the final composition is primarily determined by the initial precursor molar ratio, as both metals undergo simultaneous reduction under similar kinetic conditions.

As shown in Figure 1e, particle size distribution analysis was performed on 300 ANS particles using Nano Measurer software (version 1.2.5). The results revealed that ANS particles exhibited an average external diameter of 82.5 ± 19.0 nm with a characteristic unimodal distribution pattern. The results demonstrated that approximately 55% of the particles were concentrated within the 70–100 nm size range, with the predominant fraction (20.67%) distributed in the 70–80 nm interval. These findings indicate a relatively uniform particle size distribution, confirming the reproducibility of the synthesis methodology.

Figure 1f presents a comparative X-ray diffraction (XRD) analysis of ANSs, pure Au NPs, and pure Ag NPs. All three samples exhibit characteristic face-centered cubic (fcc) crystal structures, with primary diffraction peaks observed at approximately 38°, 44°, 65°, and 78° (2θ), corresponding to the (111), (200), (220), and (311) planes of the fcc lattice, respectively. The magnified (111) diffraction profiles reveal a systematic shift in peak positions: the Ag NPs peak appears at the lowest angle, while the Au NPs peak appears at the highest angle. In contrast, the ANSs’ peak is positioned between them. This intermediate peak location demonstrates the formation of an Au-Ag alloy solution in the ANSs. According to Bragg’s law, in cubic crystals with fixed wavelength (λ) and Miller indices (hkl), the diffraction angle is inversely related to the lattice parameter [36]; thus, the intermediate peak position of ANSs between pure Au and Ag confirms the formation of a homogeneous substitutional solid solution rather than a phase-separated structure [37]. All samples display single, sharp, and symmetric diffraction peaks without any evidence of peak splitting or shoulder features, further validating the uniformity of the alloy phase. These XRD results provide conclusive confirmation of the successful synthesis of Ag-rich Au-Ag alloy nanostructures with excellent crystallinity and homogeneous solid-solution characteristics.

2.2. UV-Vis Spectroscopy

UV-Vis (ultraviolet–visible) absorption spectra of Ag NPs, Au NPs, and ANSs were recorded in aqueous media from 350 to 800 nm. Figure 2a shows LSPR peaks at 421 nm for Ag NPs and 542 nm for Au NPs, consistent with values found in the literature [38,39,40,41]. The LSPR peak of the ANSs is located at 576 nm, exhibiting a broader plasmonic resonance, which suggests potential electronic coupling between Au and Ag and confirms the successful formation of bimetallic nanoparticles [42]. The pronounced red-shift from Au NPs (542 nm) to ANSs (576 nm) and the broader resonance bandwidth can be attributed to the distinct structural and compositional features of the synthesized material. As confirmed by XRD analysis, the formation of a homogeneous Au-Ag substitutional solid solution modifies the electronic structure compared to pure metals. Additionally, the hollow shell architecture (~80 nm) and Ag-rich composition (71.40 wt% Ag) contribute to the altered optical response, distinguishing ANSs from conventional solid nanoparticles.

ANS substrates were immersed in different solvents, including ethanol (n = 1.360), ethanol–toluene mixtures at 3:1, 1:1, and 1:3 v/v ratios (n = 1.390, 1.429, and 1.462, respectively), and toluene (n = 1.495), and UV-Vis spectra were recorded from 400 to 800 nm to investigate the effect of solvents on optical properties. Previous studies have shown that solvent refractive index changes typically cause LSPR peak shifts in nanoparticles [43]. Figure 2b shows that increasing the solvent refractive index from 1.360 to 1.495 significantly varied the ANS absorption intensity without shifting the peak position. This indicates that ANS LSPR characteristics likely depend primarily on particle size, morphology, and elemental composition, rather than solvent refractive index.

2.3. Photocatalytic Process

SERS enables the evaluation of plasmon-driven chemical reactions due to its non-destructive nature, high sensitivity, and molecular recognition capabilities. PNTP was selected as a probe molecule to assess the photocatalytic performance of ANSs through real-time monitoring of PNTP to DMAB conversion [44,45]. ANS substrates were immersed in 0.1 mM PNTP ethanol solution for 2 h, then rinsed with ultrapure water to remove physically adsorbed molecules. PNTP molecules bond to ANS substrates via Au–S or Ag–S bonds, forming a platform for photocatalytic evaluation.

Figure 3 shows PNTP photocatalytic behavior on ANS surfaces. The process involves the reductive coupling of PNTP, with each molecule accepting four electrons to form DMAB, which contains an azo bond (–N=N–). Sulfur anchoring sites remain stable throughout the reaction, indicating that the transformation occurs in the adsorbed state. This surface adsorption mechanism enhances electron transfer from the metal substrate to adsorbed molecules.

PNTP converts to DMAB through two approaches: conventional chemical reduction and plasmon-driven photocatalysis [46,47,48]. Chemical reduction utilizes sodium borohydride (NaBH₄) to reduce PNTP to PATP, followed by oxidative coupling to DMAB [49]. The plasmonic-driven photocatalytic pathway offers a significant advantage, as it eliminates the need for additional chemical reagents, with the conversion solely relying on hot electrons generated by LSPR [28].

During plasmon-mediated catalysis, PNTP adsorbs onto ANS surfaces via Au–S or Ag–S bonds. Under laser irradiation, the LSPR effect of ANSs generates high-energy hot electrons and hole pairs [50]. Hot electrons are injected into PNTP nitro groups (-NO₂), initiating electron transfer and molecular rearrangements. Electron injection facilitates the stepwise reduction in the nitro group and the formation of an azo bond between adjacent molecules, resulting in DMAB [44,48]. This surface-mediated pathway achieves selective reactions while avoiding solution-phase side reactions. The reduction and coupling process [28] follows:

2PNTP + 8e⁻ + 8H⁺ → DMAB + 4H₂O

Plasmon-driven photocatalysis differs fundamentally from semiconductor photocatalysis. In semiconductor systems, photogenerated holes directly participate in oxidation reactions. In plasmonic systems, however, electron backfilling neutralizes most hot holes, suppressing oxidation side reactions. Some hot holes oxidize water molecules to generate protons and oxygen, with protons providing hydrogen for reduction. Hot electrons reduce PNTP to DMAB throughout the response, with product desorption regenerating active sites.

The surface-mediated mechanism, combined with ANS electromagnetic field enhancement, achieves efficient and selective photocatalytic conversion. Figure 3 validates the feasibility of plasmonic photocatalysis and demonstrates how designed metal nanostructures modulate reaction selectivity. This provides a theoretical foundation for developing high-performance plasmonic photocatalysts.

2.4. SERS Monitoring and Kinetic Analysis

Figure 4a shows the SERS spectra of PNTP molecules on the ANS substrate as the illumination time changes. Initial spectra display three prominent peaks at 1076, 1340, and 1574 cm⁻¹, corresponding to C-S stretching, -NO₂ symmetric stretching, and benzene ring vibrations of PNTP. After 4 s of irradiation, new peaks emerge at 1141, 1391, and 1441 cm⁻¹. These correspond to C–N symmetric stretching, N=N stretching, and C–H in-plane bending of DMAB. With increasing irradiation time, the DMAB peaks intensify while the PNTP peaks diminish, indicating a conversion of PNTP to DMAB. The SERS spectra also exhibit characteristic peaks in the 1000–1050 cm⁻¹ region, which correspond to the benzene ring breathing vibration (1006 cm⁻¹) and in-plane bending vibration of the benzene ring (1025 cm⁻¹) of the aromatic framework. Based on the essentially stable peak intensities during the irradiation process, these two peaks can be classified as spectator bands.

Catalytic activities of Ag NPs (Figure 4b), Au NPs (Figure 4c), and film NPs (Figure 4d) were tested under identical conditions for comparison. Photocatalytic rates on Ag and Au NPs are significantly lower than on ANSs. DMAB peaks appear later with a slower increase in intensity. Film NPs show barely detectable DMAB peaks after 31 min irradiation, indicating minimal catalytic activity. This likely results from the lack of plasmonic hot spots in film NPs, which limits electromagnetic field enhancement and hot electron generation necessary for photocatalysis. ANSs exhibit excellent photocatalytic activity. This synergistic effect makes them an ideal material for monitoring surface photocatalytic reactions.

To quantitatively measure the photocatalytic conversion kinetics of PNTP on different nanostructure substrates, this study established a kinetic expression based on a first-order reaction model. Under the assumption of a first-order reaction, it was found that the relative amount of the product generated shows a clear functional relationship with the consumption of the reactant. By establishing a linear relationship between the intensity ratio of the DMAB and PNTP Raman peaks and the reaction time, the following kinetic equation [51] can be derived:

$${\displaystyle \frac{x}{a - 2x}} \propto {\displaystyle \frac{I_{\mathit{DMAB}}}{I_{\mathit{PNTP}}}} \mathrm{or} \mathit{ln}\left(2{\displaystyle \frac{I_{\mathit{DMAB}}}{I_{\mathit{PNTP}}}} + 1\right) = \mathrm{k}\mathrm{t} \left(k \propto a{k}_0\right)$$

I_DMAB and I_PNTP represent Raman intensities for DMAB and PNTP peaks. k₀ is the first-order rate constant; k is proportional to k₀.

Figure 5a shows that ANSs exhibit enhanced Raman signals at 1076 and 1574 cm⁻¹, surpassing those of Au NPs, Ag NPs, and film NPs substrates, confirming superior SERS enhancement. Figure 5b shows the temporal evolution of ln(2I_DMAB/I_PNTP + 1) on ANSs at 1141, 1391, and 1441 cm⁻¹. Initially, the steep slope indicates a rapid conversion from PNTP to DMAB. After 50 s, the curve levels off and saturates, indicating complete PNTP to DMAB transformation. Figure 5c compares kinetic curves at 1441 cm⁻¹ for different substrates. ANSs show the fastest rate, followed by Ag NPs, and Au NPs. Finally, all samples approach saturation, likely due to atmospheric O₂ inhibiting the coupling reaction. Near equilibrium, increased electron–hole recombination restricts catalytic conversion.

Rate constants were determined by kinetic fitting of Raman peaks for ANSs, Au NPs, and Ag NPs. For ANSs, rate constants at 1141, 1391, and 1441 cm⁻¹ were 0.0562, 0.0452, and 0.0650 s⁻¹. For Au NPs, rate constants at 1141, 1391, and 1441 cm⁻¹ were 0.00016, 0.00014, and 0.00026 s⁻¹. For Ag NPs, rate constants at 1141, 1391, and 1441 cm⁻¹ were 0.0079, 0.0078, and 0.0062 s⁻¹. Figure 5d compares the rate constants for different substrates. The rate constants for ANSs are 250–351 times higher than those for Au NPs and 5–10 times higher than those for Ag NPs, demonstrating a synergistic enhancement in plasmon-driven photocatalysis.

Table 1. Comparison of photocatalytic performance for PNTP → DMAB conversion on different noble metal substrates.

Catalyst	Rate Constant	Reaction Order	Laser/Power
ANSs	0.0562–0.0650 s−1	Second-order	633 nm, 0.42 mW
Au NPs	0.00016–0.00026 s−1	Second-order	633 nm, 0.42 mW
Ag NPs	0.0062–0.0079 s−1	Second-order	633 nm, 0.42 mW
Au@Ag NRs [28]	0.030 s−1	first-order	633 nm, 1.7 mW
NPAS10 [29]	0.0068–0.0070 s−1	Second-order	633 nm, 0.17 mW

further compares our results with recent literature data, confirming that ANSs achieve the highest rate constants among reported noble metal catalysts for PNTP → DMAB conversion. The quantitative comparison validates the effectiveness of our hollow alloy design and establishes a benchmark for future plasmonic photocatalysis studies.

3. Materials and Methods

3.1. Materials

Hydrogen tetrachloroaurate hydrate (HAuCl₄·4H₂O), p-nitrothiophenol (C₆H₅NO₂S, PNTP, 96%), silver nitrate (AgNO₃, 99.8%), and ethanol (AR) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Trisodium citrate dihydrate (C₆H₅N₃O₇·2H₂O, 99%) was from Sigma (St. Louis, MO, USA). 3-aminopropyltrimethoxysilane (C₆H₁₇NO₃Si, APTMS, 99%), hydrogen peroxide (H₂O₂, 30%), and sulfuric acid (H₂SO₄, 98%) were from Aldrich (St. Louis, MO, USA). All reagents were used directly without further purification. Aqueous solutions were prepared using deionized water (≥18 MΩ·cm) from an ELGA system.

3.2. Synthesis of ANSs

The experiment was conducted in a 250 mL three-neck round-bottom flask equipped with a condenser and a magnetic stirrer. All glassware, including reagent bottles and beakers, was pretreated by soaking in aqua regia (a mixture of concentrated HCl and HNO₃ in a 3:1 volume ratio) for 30 min to remove metallic ion contamination thoroughly. Subsequently, the glassware was repeatedly rinsed with deionized water (resistivity ≥ 18 MΩ·cm) until a neutral pH was achieved, and then dried under a nitrogen flow before use.

Initially, 100 mL of 1 mM AgNO₃ solution was added to the flask. The flask was then placed in an oil bath and heated to 100 ± 2 °C while maintaining magnetic stirring at 600 rpm. After temperature stabilization, 1 mL of 0.034 M C₆H₅N₃O₇·2H₂O was rapidly injected into the reaction system, which was kept under continuous stirring. Within 10–15 min, the solution color gradually changed from colorless to pale yellow, and finally to gray, indicating the initial formation of silver nanoparticles. To ensure complete reaction, the solution was continuously heated and stirred at this temperature for 1 h, ultimately yielding silver nanoparticles with an average diameter of approximately 60 nm.

Under continuous stirring, 0.9 mL of 50 mM HAuCl₄ and 0.9 mL of 100 mM AgNO₃ were simultaneously injected into the reaction system, followed immediately by the addition of 2.1 mL of 0.034 M C₆H₅N₃O₇·2H₂O. The mixed solution continued to react at 100 °C for 1 h. During this period, the system color underwent a gradual transition from purple to reddish-brown, and finally to green within 20–30 min.

After the reaction was completed, heating was discontinued, and the system was allowed to cool naturally to room temperature (approximately 40 min). Finally, deionized water was added to precisely dilute the solution to a total volume of 105 mL, thereby reducing residual ion concentration and inhibiting particle aggregation. This method ultimately produced gold-silver alloy nanoshells with an average diameter of approximately 80 nm.

3.3. Preparation of SERS-Active ANS Substrates

Figure 6a illustrates SERS substrate fabrication. Glass slides were cleaned with detergent and water, then dried with nitrogen. This removes organic contaminants and particles.

Slides were hydroxylated using Piranha solution (a mixture of concentrated sulfuric acid and hydrogen peroxide in a 2:1 volume ratio). This activates the surface and introduces hydroxyl groups for silanization. Slides were then rinsed with water and dried with nitrogen.

For silanization, slides were immersed in 0.5 M APTMS for 24 h under sealed conditions. APTMS forms siloxane bonds with surface hydroxyls, exposing amino groups. Slides were washed with detergent and water to remove unreacted reagents, then dried. To prevent particle interactions, the diluted ANSs colloidal solution was added to the surface of the functionalized glass slides and left for 12 h. ANSs attach through interactions with amino groups. After rinsing and drying, SERS-active ANS substrates were obtained. As shown in Figure 6b, ANSs exhibit a relatively uniform distribution on the glass substrate, with moderate particle size and regular morphology. This suggests a stable deposition process and strong adhesion of the alloy to the glass surface.

3.4. Microstructural Characterization and Raman Spectroscopy

UV-Vis spectra were recorded using a UV-2550 spectrophotometer (Shimadzu, Kyoto, Japan). Microstructure was characterized by SEM (JEM-7600F) and TEM (JEM-2100). Both devices were manufactured by JEOL Ltd. EDX analysis was performed using JEM-2100F (JEOL Ltd., Akishima, Tokyo, Japan). XRD patterns were collected using a Philips PW-1830 diffractometer (JEOL Ltd., Japan). Raman spectra were collected using a Renishaw inVia system with 633 nm excitation (Melles Griot, Carlsbad, CA, USA). Acquisition time was 4 s. Laser power was controlled to prevent thermal effects.

4. Conclusions

This study demonstrates that ANSs exhibit exceptionally high photocatalytic activity, with performance enhancements of up to two orders of magnitude compared to monometallic nanoparticles. Their hollow alloy architecture enables reaction rates in plasmon-driven PNTP conversion that are 250–351 times higher than those of gold nanoparticles and 5–10 times higher than those of silver nanoparticles. This remarkable improvement is likely due to the synergistic electronic coupling between Au and Ag, along with the unique hollow nanostructure, which enhances electromagnetic fields and facilitates hot electron transfer.

The dual functionality of ANSs—efficient photocatalysis combined with vigorous SERS activity—enables real-time monitoring of the reaction, revealing the rapid formation of DMAB within seconds of laser irradiation. This research establishes bimetallic hollow nanostructures as a robust platform for plasmonic catalysis, highlighting the potential of rational nanostructure design in developing multifunctional catalytic materials.