TiO2 Bowl‐Like Nanocavity With Inner Au Embedded for Visible‐Light Photocatalysis

TL;DR

A new TiO2 nanobowl with gold nanoparticles inside improves visible-light photocatalysis for degrading methylene blue.

Contribution

The first synthesis of Au nanoparticles embedded within TiO2 nanobowls for enhanced photocatalytic performance.

Findings

Au@@TiO2 shows markedly improved catalytic activity in visible-light degradation of methylene blue.

The TiO2 nanobowl structure provides a stable support for inner Au centers.

The method offers a new strategy for high-performance nanocomposite catalysts.

Abstract

In this work, we developed a novel method to synthesize Au nanoparticles (NPs) and a TiO2 nanocomposite catalyst with Au NPs embedded individually within the inner surface of a TiO2 bowl‐like nanostructure (nanobowl) for the first time (denoted as Au@@TiO2). This unique TiO2 nanobowl shell can serve as a stable support for inner Au centers, and the resulting Au@@TiO2 shows markedly improved catalytic activity in the visible‐light degradation of methylene blue (MB) compared with Au‐embedded TiO2 Hollow Spheres (HSs) and Au‐deposited commercial P25 (denoted as Au@TiO2 and Au‐P25, respectively). Our well‐established nanobowl provides a new strategy and pathway for the synthesis of high‐performance nanocomposite catalysts.

Au centres embedded into TiO2 hollow cavity and bowllike inner shell: this unique Au@@TiO2 photocatalyst was prepared successfully by a chemical template‐etching method for the first time. Experimental results showed excellent photoactivity.

Figures

Scheme of the synthesis and formation process of Au‐embedded TiO2 nanobowl. (a–d) TEM images reveal the morphological evolution from SiO2‐Au, SiO2‐Au@TiO2, Au@TiO2 to Au@@TiO2.

Schematic illustration of the synthesis process for the Au@@TiO2 catalyst.

TEM images of the as‐prepared TiO2 samples, (a) SiO2‐Au, (b) SiO2‐Au@TiO2 nanocomposite, (c) SiO2‐Au@TiO2 nanovoid, (d) Au@TiO2 nanostructure, (e) Au@@TiO2, (f) HRTEM of the Au@@TiO2.

(a) Relationship between MB concentration and reaction time of different TiO2 photocatalysts. (b) The corresponding pseudo‐first‐order kinetic rate plot.

Full text

Introduction

1

Photocatalysis has attracted extensive attention as a promising technology for environmental remediation and solar energy conversion due to its ability to drive redox reactions under light irradiation [1, 2, 3]. In traditional semiconductor photocatalysts such as Titanium dioxide (TiO_2_), the efficiency of photocatalytic reactions under visible light is limited by its wide bandgap and rapid electron‐hole recombination. As a result, only a low percentage of solar light (3%–5%) can be directly utilized by TiO_2_ because of its relatively wide bandgap (e.g., 3.2 eV for anatase TiO_2_; 3.0 eV for rutile TiO_2_) [4, 5, 6]. To overcome these limitations, a broad range of strategies has been explored, including heterojunction construction, defect and doping engineering, and surface modification to extend light absorption and improve charge separation [4, 7, 8].

Recent advances highlight the importance of interface and structural design in enhancing photocatalytic activity. Heterostructure engineering and Z‐scheme designs have shown remarkable improvements in radical generation and charge separation efficiency for organic pollutant degradation [9, 10, 11]. Similarly, composites with visible‐light responsive components such as g‐C_3_N_4_ and bismuth oxyhalides demonstrate strong synergistic effects in water purification applications, often driven by improved redox potentials and radical pathways [12, 13].

Another emerging direction is the incorporation of plasmonic metal nanoparticles, such as Au, which can induce localized surface plasmon resonance (SPR) effects and generate hot carriers under visible light, thereby enhancing photocatalytic performance beyond the intrinsic semiconductor absorption range [14, 15, 16]. The size, morphology, and spatial distribution of such plasmonic centers critically influence light harvesting and interfacial charge transfer [17, 18, 19]. Moreover, there is still a big challenge in the application of noble metals because of their poor structural stability (such as leaching, self‐agglomeration) during the reaction period. Despite these significant advances, achieving stable and efficient visible‐light photocatalysts remains challenging due to issues such as metal nanoparticle aggregation, insufficient light utilization, and limited control over reaction microenvironments [20, 21, 22]. In this work, we report a nanocavity‐embedded Au@@TiO_2_ nanobowl architecture that aims to integrate plasmonic sensitization and structural confinement to address these challenges and enhance photocatalytic degradation of organic pollutants under visible light.

Experiment

2

Material Preparation

2.1

Fabrication of Au Adhered to SiO2 Core (Denoted as SiO2‐Au) Nanostructure

2.1.1

All glassware was totally cleaned with aqua regia (3 parts HCl, 1 part HNO_3_) for 10 h and then washed with deionized water. Tetraethyl orthosilicate (TEOS, 99%, 0.86 mL), de‐ionized water (4.3 mL), ethanol (23 mL), and ammonia solution (26%, 0.62 mL) were mixed and stirred for 6 h, retaining at room temperature. To functionalize the silica template surface with amino (‐NH_2_) groups, 20 mL of the above SiO_2_ aqueous solution was mixed with 0.5 mL of 3‐aminopropyltriethoxysilane (APTES, 99%) and heated to 80°C for 2.5 h. The precipitated ‐NH_2_ treated SiO_2_ NPs were separated by centrifugation approach, rinsed 3 times with ethanol, and dried in vacuum at 60°C for the whole night, which was then re‐dispersed in 5 mL of ethanol. Sodium citrate (1%, Tianjin Chemical Reagent No. 1 Plant, 5 mL) was dispersed into a 50 mL aqueous solution of HAuCl_4_ · 3H_2_O (J&K) (4 × 10^−4 ^M) under sonication and heated to 95°C. 37.5 mL of 400 mM TTAB (Aldrich, 99%) was put into the mixture, which came into being burgundy and was kept heating for 15 min with vigorous stirring. TTAB was introduced as a cationic surfactant to cap and stabilize Au NPs, preventing aggregation and improving their dispersion in aqueous media. The TTAB‐capped Au colloids were subsequently immobilized onto APTES‐modified (‐NH_2_) SiO_2_ via interfacial adsorption driven by the affinity of amine‐functional surfaces toward Au NPs. At room temperature, the SiO_2_‐Au could be produced successfully by mixing the above TTAP‐capped Au aqueous solution (20 mL) with ‐NH_2_ modified SiO_2_ aqueous solution (20 mL) under sonication and stirring for 5 h. Subsequently, the precipitated SiO_2_‐Au NPs were separated from the mixture after centrifugation and washing with ethanol several times. Simultaneously, this powder was re‐dispersion in 5 mL of ethanol.

Preparation of Yolk‐Into‐shell Nanocatalysts

2.1.2

The above solution was redispersed in a mixture of hydroxypropyl cellulose (HPC, Tokyo Chemical Industry Co., Ltd., 0.1 g), ethanol (20 mL), and deionized water (0.1 mL). Stirring for 40 min, titanium tert‐butoxide (TBOT, 98%, Tianjin Chemical Reagent No 1 Plant, 1 mL), which was dispersed in 5 mL ethanol, was added into the solution at a constant rate of 0.5 mL·min^−1^. After addition, the temperature was controlled at 85°C, and the mixture was stirred at 900 rpm under refluxing conditions for 100 min. The ultimate product (SiO_2_‐Au@TiO_2_ nanocomposites) was isolated using a centrifugation method, washed with ethanol, and saved in 5 mL of ethanol. To cut off all organic compounds and crystallize the amorphous TiO_2_, the SiO_2_‐Au@TiO_2_ was calcined in air at 500°C (2°C/min) for 3 h. The treated samples were dispersed in 20 mL of water under sonication conditions, heated to the given temperature (50°C), and aqueous NaOH solution (2.5 M, Tianjin Guangfu Fine Chemical Institute, 1.5 mL) was added. At 70°C, adding another aqueous NaOH solution (2.5 M, 1 mL) for 6 h to remove the SiO_2_ template and obtain Au core embedded into TiO_2_ shell nanostructure.

Synthetization of Au@@TiO2 Nanohybrids

2.1.3

Finally, the Au@TiO_2_ products were redispersed in 20 mL of water under sonication. When the solution was heated to 70°C, another aqueous NaOH solution (2.5 M, 1 mL) was added for 6 h, which removed the partial TiO_2_ shell, and Au@@TiO_2_ was received.

Material Analysis

2.2

Nitrogen adsorption isotherms of the nanohybrid were detected using a Micromeritics Tristar3000 analyzer by nitrogen adsorption at 77 K. The value of specific surface areas can be evaluated based on the isotherms according to the BET method. The morphology of the nanohybrid was studied on a FEI Tecnai G2 F20 transmission electron microscope at 100 kV. The as‐synthesized powder was dispersed in ethanol by sonification. Several drops of the suspension were collected and dispersed onto a copper grid‐supported transparent carbon foil and dried naturally. Surface morphology‐SEM images could be assessed through JEM‐2100F. The Diffuse Reflectance Spectra (DRS) results of the nanohybrid were investigated, stemming from a SHIMADZU UV‐2550 spectrophotometer. A 60 nm diameter integrating sphere and BaSO_4_ used as the reflectance sample were installed.

Photocatalytic Activities Measurements

2.3

The photocatalytic activity of these samples was detected by the degradation of methylene blue (MB, 98%, Tianjin Damao Chemical Co., Ltd.) dyes. First, the photo‐nanocatalyst (50 mg) was evenly put into a 100 mL quartz photoreactor containing 100 mL of 12 m mol L^−1 ^MB solution. Realized by a visible light source (780 nm ≥ λ ≥ 420 nm), the above solution was ultrasonicated for 2 min. At the same time, it was magnetically stirred for 30 min in the dark condition, which ensures good dispersion of the catalyst in the solution. At the same time, it will realize the adsorption‐desorption equilibrium between MB molecules and the varied catalyst. The reaction happened with flowing water in a quartz cylindrical jacket around the lamp, and the ambient temperature was maintained during the photocatalytic reaction. At a specific interval (30 min/time), every 2 mL analytical suspension was harvested from the mixture solution and subsequently centrifuged at 5000 rpm for 10 min to purify the product. Related to analyzing the concentration of the solution, the treated sample should be kept in a pipette (1 mL). The concentration of the filtrate was analyzed by recording the change from the maximum absorption peak (664 nm for MB) with a UV–vis spectrophotometer.

Probe Experiment

2.4

In such a scenario, 0.050 g Au@@TiO_2_ catalyst and 100 mL AgNO_3_ solvent (0.026 mol L^−1^) were mixed. 2.0 g of sodium citrate was then dispersed into the above mixture. After stirring for 0.5 h, the suspension was irradiated with a 300 W Xe lamp (λ > 420 nm) for 5 h at 80 mW/cm^2^. The solution was centrifuged and washed with deionized water three times.

Results and Analysis

3

The synthetic process of Au@@TiO_2_ and corresponding transmission electron microscopy (TEM) images are exhibited in Figure 1.

Scheme of the synthesis and formation process of Au‐embedded TiO2 nanobowl. (a–d) TEM images reveal the morphological evolution from SiO2‐Au, SiO2‐Au@TiO2, Au@TiO2 to Au@@TiO2.

There are three steps (Scheme 1) to synthesize the bowl‐like structured Au@@TiO_2_ NPs: (1) synthesis of SiO_2_‐Au NPs using the Stober method; (2) preparation of SiO_2_‐Au@TiO_2_ nanovoid via a template method and partial chemical etching process; (3) complete removal of the inner SiO_2_ core and partial TiO_2_ shell with NaOH treatment. Au NPs were successfully embedded into the TiO_2_ nanobowl inner shell, as shown in Figure 1d. It can be observed that almost all TTAB (myristyltrimethylammonium bromide)‐capped Au NPs were well deposited individually on the surface of SiO_2_‐NH_2_. The as‐prepared Au@@TiO_2_ NPs consisted of 10±5 nm Au cores, as exhibited in Figure S1. The average thickness and the diameter of anatase TiO_2_ shell embedded by the Au core are about 25±2 nm and 110±20 nm, respectively.

Schematic illustration of the synthesis process for the Au@@TiO2 catalyst.

The photocatalytic degradation reported in this work corresponds to Au NPs with an average size of ∼10 ± 5 nm (Figure S1). Although the activity of plasmonic photocatalysts can be influenced by Au particle size, the primary focus of this study is to highlight the structural advantage of the nanocavity‐embedded Au@@TiO_2_ nanobowl compared with conventional Au@TiO_2_ yolk‐shell and Au‐P25 systems. The enhanced performance is mainly attributed to improved light confinement/scattering and efficient charge separation enabled by the inner‐embedded nanocavity architecture, rather than a full size‐dependent optimization.

The morphologies of Au‐modified TiO_2_ HSs and nanobowl were shown in Figure S2. TEM images of SiO_2_‐Au, SiO_2_‐Au@TiO_2_ core‐shell, SiO_2_‐Au@TiO_2_ nanovoid, Au@TiO_2,_ and Au@@TiO_2_ are shown in Figure 2a–f. The dark dots are Au NPs. The TiO_2_ shell was coated on the SiO_2_‐Au surface and formed a sandwich structure after partial NaOH etching (Figure 2a–c). After calcination and a further NaOH etching process, the SiO_2_ core was completely removed, and anatase Au@TiO_2_ yolk‐embedded‐shell nanocomposite can be obtained (Figure 2d). Part of the TiO_2_ shell was removed, and Au was embedded into the inner part of the TiO_2_ nanobowl shell, which appeared after the second partial NaOH etching process (Figure 2e).

TEM images of the as‐prepared TiO2 samples, (a) SiO2‐Au, (b) SiO2‐Au@TiO2 nanocomposite, (c) SiO2‐Au@TiO2 nanovoid, (d) Au@TiO2 nanostructure, (e) Au@@TiO2, (f) HRTEM of the Au@@TiO2.

After complete removal of the inner SiO_2_ core, the mesoporous anatase TiO_2_ shell was also partially etched away by the elevated NaOH etching treatment. The oligomeric species of sodium titanate were generated during subsequent NaOH etching and can be easily dissolved in aqueous solution. As a result, the intact hollow‐sphere morphology was completely transformed, and an unusually cuplike TiO_2_ shape was obtained. More details can be found in our prior works [12, 17].

The Au@@TiO_2_ shows a well‐defined bowllike morphology, and Au NPs with the small particle size were homogenously embedded into the inner surface of the TiO_2_ shell, as verified by Figures 1d and 2e with a specific surface area of 177 m^2^g^−1^ (in Table S1, ESI). High resolution transmission electron microscopy (HRTEM) image of Au@@TiO_2_ indicates that the lattice fringes with d = 0.2355 and 0.3545, which should be attributed to Au {111} and anatase TiO_2_ {101} shown in Figure S3, respectively. In addition, almost all the products remain in the nanobowl shape without a broken shell, and no Au aggregation occurred after the third degradation reaction, as shown in Figure S4. Although the major advantage of Ag is its cheaper cost compared with other noble metals, the combination of Ag with TiO_2_ usually leads to Ag leaching, which is particularly important upon visible light irradiation. From the colloidal nanoparticles in Figure S8, we could also notice that Ag nanoparticles are irregular and more difficult to protect by the TiO_2_ supporter. In addition, Ag has the problem of corrosion and oxidation. Therefore, Au is generally preferred for these studies and selected in this work. The typical UV–vis diffuse reflectance spectra (DRS) were applied to characterize the optical properties of the obtained catalyst (Figure S5). As expected, the absorption edges of the obtained photocatalysts are located at 600 nm, which indicates that the photocatalysts can be selected as excellent visible light photocatalysts. In contrast to the unconventional Au@@TiO_2_, the Au@TiO_2_ NPs were also tested. The intensity of reflectance spectra of the Au@@TiO_2_ is much higher than that of the Au@TiO_2_ yolk‐into‐shell, which means a better localized surface plasmon resonance (SPR) effect. This can be ascribed to the improved light harvesting and scattering ability and efficiency, as confirmed by reflectance spectroscopy.

To further evaluate the photocatalytic performance of the anatase Au@@TiO_2_, we decided to measure these TiO_2_ models by photodegradation of MB used as a probe molecule under visible light irradiation (λ ≥ 420 nm). As a control, the same Au‐P25 and pure TiO_2_ HSs were also used as the reference catalysts. When there were pure TiO_2_ HSs photocatalysts and no catalyst in the system, only a negligible number of MB could be degraded. Following the intensity of the 664 nm peak versus time, the photoactivities of various catalysts were represented in Figure 3a. According to Figure 3a, the photocatalytic performance under visible light irradiation displays an order of Au@@TiO_2_> Au@TiO_2_> Au‐P25 under the same reaction conditions. The results of control experiments indicate that the Au@@TiO_2_ NPs have the best visible‐light photocatalytic activity. Based on the SPR effect of Au NPs and suppressing the recombination efficiency of the photogenerated electron‐hole pairs, we can observe that the photocatalytic ability of Au attached TiO_2_ was significantly enhanced [23]. The reaction details can be described as follows [12, 20]:

(a) Relationship between MB concentration and reaction time of different TiO2 photocatalysts. (b) The corresponding pseudo‐first‐order kinetic rate plot.

Photo‐induced electrons generated by visible light transfer from the Au NPs to the TiO_2_ conduction band. Ultimately, positive charges will then remain in the Au NPs. Subsequently, oxidation and reduction reactions will be conducted at different active sites, resulting from this exceptional charge separation process. Electrons are depleted quickly on the semiconductor side, which is called the reduction of O_2_‐electron acceptor. Substrates‐electron donors were oxidized through concentrated positive charges on the inner Au centers. Further experiments were established to investigate the migration process of electrons from embedded Au to the outer TiO_2_ supporter. Ag^+^ ions and sodium citrate were selected as probe material and a positive charge scavenger. As appeared in Figure S6, Ag NPs were found to gather on the outer TiO_2_ semiconductor where electrons are assembled. This clearly demonstrates the transfer orientation of electrons. It can be apparently noticed that Ag NPs are only deposited on the outer TiO_2_ supporter rather than on the Au centers. On the other hand, positive charges were used up by sodium citrate.

Although other noble metals (e.g., Ag or Pt) have also been explored as cocatalysts for TiO_2_‐based photocatalysis, Au nanoparticles were selected in this work due to their strong localized surface plasmon resonance (SPR) effect under visible‐light irradiation and superior chemical stability. Control experiments were performed using Au@@TiO_2_ nanobowls, conventional Au@TiO_2_ yolk‐shell structures, Au‐P25, and pure TiO_2_ hollow spheres. The photocatalytic activity followed the order of Au@@TiO_2_ > Au@TiO_2_ > Au‐P25, confirming that the nanocavity‐embedded configuration provides the most efficient visible‐light degradation performance. In contrast, Ag NPs synthesized under identical conditions exhibited irregular morphology and are more susceptible to corrosion and leaching during photocatalysis (Figure S8), which further supports Au as a preferred candidate for stable nanocavity‐confined photocatalysts. Simultaneously, the reaction followed the first‐order rate kinetics, and it can be defined as the pseudo‐first‐order reaction mode: Ln(C_0_/C) = kt+A, where C means the concentration in the solution upon irradiation, C_0_ is the original concentration before irradiation, and k is the reaction rate constant. The kinetics of the degradation process can be described by a Langmuir‐Hinshelwood model in Figure 3b, due to low MB concentration. The photodegradation rate constants k value of Au P25, Au@TiO_2_ and Au@@TiO_2_ are 0.59 × 10^−2^, 1.13 × 10^−2^, 1.67 × 10^−2^ min^−1^, respectively. The catalytic activity improvement of Au@@TiO_2_ derives from several possible aspects. First, the surface area of the nanobowl NPs is higher than that of commercial P25, benefiting from the anatase TiO_2_ shell structure, which remarkably produces more active sites for a rapid consumption of electrons. Second, the highly active anatase (101) facet and good water dispersity also play a great role in the photocatalytic reaction according to our prior work [4, 20, 24]. The crystallinity of Au@@TiO_2_ was characterized by X‐ray diffraction (XRD) (as shown in Figure S7). Although Au@TiO_2_ yolk–shell possesses a slightly higher BET surface area than Au@@TiO_2_ nanobowls, the photocatalytic performance is not governed by surface area alone. The enhanced activity of Au@@TiO_2_ is mainly attributed to its nanocavity‐embedded architecture, which improves visible‐light harvesting through multiple reflection/scattering inside the bowl‐like cavity and strengthens the localized SPR effect of the inner Au centers. Moreover, the confined configuration facilitates efficient charge separation and suppresses electron‐hole recombination, leading to higher catalytic efficiency despite the slightly lower surface area. The cyclic stability of the Au@@TiO_2_ nanobowl photocatalyst was evaluated by repeated MB degradation experiments. The catalyst retains high photocatalytic activity even after three consecutive cycles (Figure S9), indicating good reusability under visible light irradiation. Moreover, TEM characterization of the used catalyst confirms that the nanobowl morphology remains intact without noticeable shell collapse, and no significant Au nanoparticle aggregation is observed after recycling (Figure S4). These results demonstrate the structural robustness and practical stability of the Au@@TiO_2_ nanocavity photocatalyst. Thereby, this well‐established design of Au yolk into TiO_2_ nanobowl shell can be regarded as the well‐optimized photocatalytic configuration.

Conclusions

4

In summary, a high‐performance anatase Au@@TiO_2_ photocatalyst with well‐dispersed Au NPs embedded into the inner surface has been successfully developed by a facile silica template‐coating method and chemical etching treatment. The as‐synthesized bowl‐like anatase TiO_2_‐based catalyst not only preserves the original advantages of conventional yolk‐shell photocatalysts, but also effectively prevents the aggregation of active Au by providing unique Au centers and a TiO_2_ shell in full contact. Additional synergistic effects became more stressed due to the interfacial chemical interactions [5, 17].

This novel Au‐embedded TiO_2_ bowl‐like architectural design method can potentially provide a strategy for various fields, for instance, dye‐sensitized solar cells and photoelectrochemical water splitting. Notably, the Au attached TiO_2_ model synthesis method can also be extended to many other types of advanced oxide‐based micro‐/nano‐semiconductor, such as ZnO, NiO, BiVO_4,_ and SnO_2_ [25, 26, 27].

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: gch270097‐sup‐0001‐SuppMat.docx.