Open Access
Add to BibSonomy
Abstract
An ultraviolet light-induced method is used to simultaneously grow silver nanoparticles (AgNPs) on the surface of titanium dioxide (TiO2) nanospheres and complete a self-cleaning function. By adjusting ultraviolet (UV) light-induced duration, TiO2/AgNPs composite samples were prepared as surface-enhanced Raman scattering (SERS) substrates. The electromagnetic distribution of TiO2/AgNPs was analyzed with FDTD Solutions simulation software, and the corresponding theoretical enhancement factor was calculated. Taking the Rhodamine 6G (R6G) molecule as an analyte, the experimental detection limit is lower than 10−12 mol/L under UV-induced duration of 10 min, and the analytical enhancement factor (AEF) is ∼ 6.8×1010. In addition, the UV light-induced used samples can show a self-cleaning function, and the samples can be used for 5 cycles, with certain stability and repeatability. Moreover, the samples’ performance of multi-molecule detection is experimentally carried on.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Surface-enhanced Raman scattering (SERS) is a powerful technology with high sensitivity, specificity, non-destructive testing and other advantages. It is widely used in biology, chemistry, food safety and other aspects [1–5]. Traditional SERS substrates are mostly composed of noble metal nanomaterials with high-sensitivity molecular detection ability. However, these types of metal nanostructures have the shortcomings of short storage time, instability and non-reusable, due to its easy oxidation property. In recent years, more and more researchers try to improve SERS substrates with advantages of high sensitivity, good stability and reusability, which has great development potential applications in food safety, immunoassay, and water purification detection [6–10]. Zhao et al. used a facile electrostatic self-assembly method to synthesis silver/silver chloride/graphene oxide (Ag/AgCl/GO), which can obtain detection limit of R6G as low as 1×10−7 mol/L. And the Ag/AgCl particles under visible light has excellence photocatalytic activity lead to an outstanding self-cleaning property of the Ag/AgCl/GO composite [11]. Du et al. prepared Fe3O4@TiO2@Ag substrates through a typical multi-step chemical reaction, which was used for an actual label-free detection of prostate specific antigen (PSA) in serum, showing a limit of detection as low as 16.25 pg/mL, and it can achieve a self-cleaning effect under 100 min of UV irradiation [12]. Cheng et al. synthesized Zinc oxide (ZnO) nanosheets by hydrothermal method, then deposited Ag nanoparticles on ZnO nanosheets by chemical method, and successfully prepared three-dimensional Ag/ZnO microspheres, which can detect 10−9 mol/L R6G molecule [13]. The prepared composite structure exhibited an excellent self-cleaning effect under UV irradiation.
Recently, semiconductor materials represented by ZnO and TiO2 have attracted widespread attention because of their low cost, high sensitivity, and uniformity. As one of the semiconductors, titanium dioxide has attracted widespread attention due to its convenient acquisition, low cost, non-toxicity, simple preparation. Chen et al. first synthesized a biological TiO2 with a butterfly wing structure through an improved sol-gel technology. After calcination, gold nanoparticles were deposited on the surface of the TiO2 matrix through a deposition-precipitation process. It has high catalytic activity when the ratio of gold nanoparticles to titanium dioxide is 8%, and completely decomposed methyl orange within 80 min of visible light [14]. Tao et al. used a two-step method based on electrochemical etching and sol-gel process to synthesize an ordered channel porous TiO2/Ag/silicon nanowires (TiO2/Ag/SiNWs) composite structure. The substrate can effectively degrade PNP molecules within 60 min under ultraviolet light [15]. Kai et al. used a two-step solvothermal method to continuously deposit TiO2 nanoparticles and silver nanoparticles on reduced graphene oxide to prepare a reproducible SERS substrate of Ag nanoparticles/titanium dioxide nanoparticles/reduced graphene oxide (Ag/TiO2/rGO), which can detect 10−14 mol/L 4-aminothiophenol (4-ATP). With the help of UV irradiation, it catalyzed degradation of 4-ATP molecules, and had good repeatability after 5 cycles [16]. Yang et al. synthesized TiO2 nanomaterials by hydrothermal method, and deposited Ag nanoparticles by electron beam evaporation. Its limit of detection of Malachite Green (MG) reached 4.47×10−16 mol/L. And under UV irradiation for 1 hour to degrade MG molecule, and it still had good Raman characteristics after three cycles [17]. Xie et al. used anodizing process to prepare TiO2 nanopore arrays through oxidation reaction on the surface of titanium plates, and then AgNPs were deposited on TiO2 nanohole array through the cyclic chemical reduction deposition process, and the detection limit of R6G was 10−8 mol/L. In addition, after 60 min of UV irradiation, the self-cleaning effect can be achieved [18]. Wang et al. used a hydrothermal method to prepare ordered TiO2 nanosheets on a carbonaceous surface, and deposited AgNPs in situ with an EF of approximately 2.52×108 for 4-mercaptobenzoic acid (4-MBA), and had a self-cleaning function [19]. According to analysis mentioned above, there are three facets to be improved: (1) Raman enhancement property; (2) the function of UV irradiation in preparation process; (3) multi-molecules detection.
In this paper, an UV induced method was used to simultaneously grow silver nanoparticles on the surface of TiO2 (TiO2/AgNPs) and complete a self-cleaning function. Its Raman enhancement properties including AEF, stability, self-cleaning and multi-molecules detection ability are theoretically and experimentally analyzed.
2. Experiment
2.1 Fabrication of TiO2/AgNPs composite structure
8 mg of anatase phase titanium dioxide microsphere powder (average diameter is 260 nm, Tianjin Baima Technology Co., Ltd.) was dispersed in 10 mL of deionized water, to configure 10−2 mol/L aqueous solution. Then we use pipetting gun to absorb 3-5 µl solution to drop it on a cleaned silicon wafer, and dry it naturally for further use (Figs. 1(a1-a2)).
Fig. 1. Experimental flowchart: (a1-a4) preparation of TiO2/AgNPs; (b1-b2) Raman measurements; (b3-b4) self-cleaning of TiO2/AgNPs composite structure.
An UV induced method was used to deposit silver nanoparticles on the surface of TiO2 microspheres. The specific preparation process is shown in Figs. 1(a3-a4): (1) Prepare a silver nitrate (AgNO3) solution with a concentration of 10−3 mol/L. (2) soak the prepared TiO2 substrates in 40 mL AgNO3 solution, with UV irradiation duration of 5 min, 10 min, 15 min, 20 min, 25 min, or 30 min for different samples. (3) Then we take it out and wash the substrate 3 times with deionized water to remove the residual silver nitrate solution on the surface, and dry it naturally and store it in a vacuum bag.
Before Raman measurements, the substrate was immersed in R6G solution of 10−6 mol/L to 10−12 mol/L for 2 hours (shading treatment). So, R6G molecules were fully adsorbed on TiO2 microspheres. Finally, the substrate is air-dried naturally for further Raman measurements as shown in Figs. 1(b1-b2).
2.2 Self-cleaning process
3-5µl of deionized water was added on the surface of the used substrate to make it wet, then under UV lamp irradiation for about 20 min. Self-cleaning flow chart is shown in Figs. 1(b3-b4). Then, we made Raman measurements on the TiO2/AgNPs composite substrate.
2.3 Instruments
The morphology and distribution of the samples was characterized by environmental scanning electron microscope (ESEM, Quattro S) and energy dispersive spectrometer (EDS, Talos). Raman signals are recorded by Horiba's LabRAM HR confocal Raman spectrometer with an excitation wavelength of 532 nm, the power of 5 mW and the integration time of 5 s. Lapspec software is used to remove the baseline of Raman spectrum, and nano measurer software is used to count the diameter and size of particles. UV curing machine (KW-4AC) was used in the experiment, with a center wavelength of 254 nm and the power was 16 w. In order to reduce random error, each Raman signals are averaged by multi-measurements.
3. Results and discussion
3.1 SEM and EDS characterization
SEM images of our samples under different UV irradiation duration are shown in Fig. 2. Silver nanoparticles are deposited on TiO2 surface. The calculated diameters of silver nanoparticles are 69 nm (for 5 min (Figs. 2(a1-a2)), 99 nm (for 10 min (Figs. 2(b1-b2)), 121 nm (for 15 min (Figs. 2(c1-c2)), 134 nm (for 20 min (Figs. 2(d1-d2)), 140 nm (for 25 min (Figs. 2(e1-e2)), and 146 nm (for 30 min (Figs. 2(f1-f2)), respectively. With an increase of the irradiation duration, the diameter of AgNPs become larger, the number was increased firstly and then decreased, because the phenomenon of AgNPs agglomeration appeared. The UV induced process can be analyzed as follows. At the beginning, the silver particles are randomly distributed on the surface of TiO2 film and the gap between particles is large, which can inhibit their fusion growth. As the increase of illumination time, the amount of silver particles grows, the gap becomes smaller, and the contact probability increased. Then, the near silver particles may experience a process of light fusion and it may widen the size of the silver particles distributed on the surface of the titanium dioxide film and the number of AgNPs were decreased [20].
Fig. 2. SEM images under different UV light time: (a1-a2) 5 min; (b1-b2) 10 min; (c1-c2) 15 min; (d1-d2) 20 min; (e1-e2) 25 min; (f1-f2) 30 min.
In order to better distinguish AgNPs and TiO2 nanoparticles on the substrate, we use EDS to characterize it. The analysis results of EDS are shown in Fig. 3, including the element types (Fig. 3(a)) and element distribution (Figs. 3(b1-b4)). The different elements with different color. We can clearly see that TiO2/AgNPs composite substrate is mainly composed of Ti, Si, O, Ag (Ag: yellow, Ti: red, O: blue), and the Si element is from silicon wafer. And the Ag elements are evenly distributed on the surface of TiO2. From the image of SEM, we can see clearly that the brighter particles were AgNPs and the darker particles were TiO2 nanoparticles.
Fig. 3. (a) EDS analysis of TiO2/AgNPs for element types and inset is SEM image of TiO2/AgNPs under UV induced duration of 10 min;(b1-b4) elements distribution of survey, Ag, Ti, O.
3.2 Raman measurement
Figure 4(a) shows SERS signals of 10−8 mol/L R6G absorbed on our samples under different UV irradiation times. Typical Raman peaks (610, 772, 1182, 1311, 1362, 1510, 1574, 1650 cm−1) of the probe molecule R6G can be detected, and the sample under UV irradiation for 10 min has the best Raman enhancement among our prepared samples. Figure 4(b) shows the Raman signal of 10−6 mol/L R6G on TiO2/AgNPs, AgNPs and TiO2, respectively. We can see clearly that Raman peak and Raman intensity from TiO2/AgNPs composite structure are clearer and higher than those of AgNPs and TiO2. This is because the Raman enhancement in TiO2/AgNPs not only comes from the electromagnetic induction between silver nanoparticles gaps, but also from the charge transfer between silver nanoparticles, TiO2 nanospheres and probe molecule [20–24].
Fig. 4. (a) SERS spectra of R6G on samples under different UV irradiation duration; (b)Raman intensity of 10−6 mol/L R6G absorbed on TiO2/AgNPs, AgNPs, TiO2, respectively.
In order to further evaluate the sensitivity of the substrate, samples under UV irradiation for 10 min were used to detect R6G aqueous solution with a series of concentrations ranging from 10−6 mol/L to 10−12 mol/L (Figs. 5(a-b)). The detection limit can be as low as the concentration of 10−12 mol/L. Figures 5(c-d) show the Raman mapping result with 10−8 mol/L R6G as analytes. The calculated relative standard difference (RSD) at 611 cm−1 is 20.4%.
Fig. 5. spectra of TiO2/AgNPs: (a-b) detection limit of R6G; (c) mapping of R6G; (d) RSD at 611 cm−1; (e) multi- molecule detection.
In addition, the sample was used for multi-molecular detection, and the probe molecules were 10−8 mol/L of R6G, 10−6 mol/L of crystal violet (CV) and 10−4 mol/L of MG. It can be seen from Fig. 5(e) that the unique Raman peaks of different probe molecules (R6G: 610 cm−1, CV: 916 cm−1, MG: 1216, 797 cm−1) can be detected in both single molecule detection and mixed solution detection, indicating that the substrate has good Raman performance in multi-molecule mixed solution. Moreover, the disappearance of some characteristic peaks (R6G:772 cm−1) and the fusion of some characteristic peaks (R6G:1182 cm−1, CV:1188 cm−1, MG:1170 cm−1 and R6G:1650 cm−1, CV:1621 cm−1, MG:1616 cm−1). These peaks merge to form a wider Raman shift, and the characteristic peak value is stronger.
3.3 Self-cleaning properties
TiO2 is a good photocatalyst, which can effectively degrade organic molecules and achieve self-cleaning effect. As is shown in Fig. 6(a), the Schott barrier effect formed on the surface of TiO2/AgNPs can effectively reduce the recombination of electron-hole pairs in TiO2 under UV irradiation. Oxygen in solution obtains electrons to form a strong oxidant, which can degrade R6G molecule and decomposes them into oxidation products (CO2, H2O) to achieve the effect of self-cleaning. 3-5 µl of pure water was dropped on the TiO2/AgNPs substrate, and irradiated for 20 min under UV light to effectively degrade it. There was no obvious R6G Raman peak after irradiated. And 3-5µl of 10−8 mol/L R6G solution was added to the substrate, and an obvious R6G Raman peak appeared again. Shown in the Fig. 6(b), the characteristic peak intensity of R6G gradually decreased with the prolongation of the UV illumination time, and disappeared at approximately 20 min. Five cycles of SERS measurement and self-cleaning experiments were carried out on the TiO2/AgNPs composite structure (Fig. 6(c)). And Fig. 6(d) shows the Raman intensity (the RSD of peak intensity is 2%) at 1312 cm−1 under self-cleaning. TiO2/AgNPs are effectively eliminated the organic molecules on the surface and the reusable performance is realized under UV irradiation.
Fig. 6. (a) schematic of self-cleaning; (b) Photo-catalytic activities of TiO2/AgNPs as a function of time for degradation of R6G under UV-light illumination (c) Raman effect of self-cleaning experiment for 5 times; (d) Raman intensity at 1312 cm−1 under self-cleaning.
3.4 Mechanism
It is generally believed that the SERS effect has two interaction mechanisms, namely the electromagnetic mechanism (EM) and the charge transfer (CT) mechanism. The EM mechanism is caused by the localized surface plasmon resonance (LSPR) of metal nanoparticles, and the CT mechanism is caused by the charge transfer between the substrate and the probe molecule. The enhancement mechanism of the TiO2/AgNPs composite structure is the combined contribution of EM effect of silver nanoparticles and CT effect of TiO2/AgNPs composite structure and the probe molecules.
In order to verify the electromagnetic effect, FDTD Solutions software was used to calculate the electromagnetic field distribution of the samples. Based on SEM images, we set the thickness of Si and SiO2 of 300 nm, and the diameter of TiO2 of 260 nm. The diameter of silver nanoparticles of 100 nm, and the spacing between silver nanoparticles of 2 nm. The incident laser is a plane wave with a wavelength of 532 nm, which is polarized along the x axis and propagates in the negative direction of the z axis. Set periodic boundary conditions that extend indefinitely in the xy direction.
In order to analyze the different distributions of TiO2/AgNPs, three models are considered (Fig. 7). The strong electromagnetic effects are distributed between the silver nanoparticles, and this tiny gap is called a “hot spot”.
Fig. 7. Electromagnetic distribution of (a) single-layer titanium dioxide; (b) double-layer titanium dioxide; (c) double-layer split-layer titanium dioxide.
According to the electromagnetic simulation results, the approximate electromagnetic enhancement results can be calculated with the following:
${{E}_{0}}$= 1 V/m is the incident field intensity, ${{E}_{{out}}}\textrm{(}{{\omega }_{0}}{)}$ and ${{E}_{{out}}}{(}{{\omega }_{s}}{)}$ represent the local electric field intensity of incident light (frequency ${\omega _0}$) and Raman scattered light (frequency ${\omega _s}$) respectively.
Shown in Fig. 7(a), when only single titanium dioxide on the silicon wafer, the largest ${{E}_{{max}}}$ is 61.2 V/m, and ${E}{{F}_{{EM}}}$ is 1.4×107. When there are two layers of titanium dioxide, the maximum ${{E}_{{max}}}$ is 51.8 V/m (Fig. 7(b)) and ${E}{{F}_{{EM}}}$ is 7.1×106. The maximum ${{E}_{{max}}}$ is 61.3 V/m and ${E}{{F}_{{EM}}}$ is 1.4×107, when titanium dioxide is superimposed in a dislocation position (Fig. 7(c)).
As shown in Fig. 8, the CT mechanism between the TiO2/AgNPs composite structure and R6G molecules is mainly determined by the following two aspects: a laser with a wavelength of 532 nm can excite silver nanoparticles to generate hot electrons to transition to the lowest unoccupied molecular orbital (LUMO) of R6G. In addition, the “hot electrons” of silver nanoparticles can also drive the conduction band (CB) of titanium dioxide to generate photocurrent, and the “hot electrons” excited on the titanium dioxide are driven to the LUMO of R6G by visible light, thereby generating SERS signals.
The analytical enhancement factor of the experimental results is calculated:
${AEF}$ Represents the analytical enhancement factor, ${{I}_{{SERS}}}$ and ${{C}_{{SERS}}}$ respectively represent the Raman intensity and R6G concentration of the substrate under SERS, ${{I}_{{RS}}}$ and ${{C}_{{RS}}}$ respectively represent the Raman intensity and R6G concentration of the substrate under non-SERS.
According to the experimental results in Fig. 5(b), ${AEF}$ is calculated as ∼6.8×1010. It shows that the SERS effect on the surface of titanium dioxide is jointly determined by the EM mechanism and the CT mechanism.
4. Conclusion
TiO2/AgNPs were prepared by the method of ultraviolet induction, and the experiment showed that the substrates with 10 min under UV irradiation have better Raman enhancement performance (AEF > 1010) and self-cleaning properties (cycle 5 times). It is expected to have potential applications in low-cost, reusable, and high-sensitivity SERS sensors.
Funding
National Natural Science Foundation of China (62175023, 61875024); Chongqing Outstanding Youth Fund (cstc2019jcyjjqX0018).
Acknowledgments
We would like to thank Dr. Gong Xiangnan at Analytical and Testing Centre of Chongqing University for his help in Raman measurement.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. K. Kullavadee, “Gap-dependent surface-enhanced Raman scattering (SERS) enhancement model of SERS substrate–probe combination using a polyelectrolyte nanodroplet as a distance controller,” Langmuir 37(36), 10776–10785 (2021). [CrossRef]
2. Y. Zhang, L. H. Yao, L. X. Liu, Z. H. Peng, Z. P. Li, H. Wang, Q. J. Zhou, T. Wei, Y. R. Guo, X. Yang, and L. Xuan, “Shape-diversified silver nanostructures on Al foil fabricated in micellar template for high performance surface enhanced Raman scattering applications,” Opt. Mater. 121, 111629 (2021). [CrossRef]
3. D. Y. Lin, Y. He, R. L. Dong, W. Li, F. L. Meng, Y. F. Zhang, and L. B. Yang, “The rapid SERS detection of succinylcholine chloride in human plasma is based on the high affinity between quaternary ammonium salt structures,” Spectrochim. Acta, Part A 263, 120172 (2021). [CrossRef]
4. M. Pannico and P. Musto, “pH activated colloidal nanospheres: a viable sensing platform for the theralpeutic drug monitoring of the anticancer drug 6-Mercaptopurine,” Appl. Surf. Sci. 570, 151232 (2021). [CrossRef]
5. Y. S. Chen, Q. F. Chen, and X. Wei, “Separable surface enhanced Raman spectroscopy sensor platformed by HPTLC for facile screening of malachite green in fish,” Microchem. J. 170, 106694 (2021). [CrossRef]
6. D. Xu, L. F. Duan, W. L. Jia, G. H. Yang, and Y. Q. Gu, “Fabrication of Ag@Fe2O3 hybrid materials as ultrasensitive SERS substrates for the detection of organic dyes and bilirubin in human blood,” Microchem. J. 161, 105799 (2021). [CrossRef]
7. D. X. Huo, B. Chen, M. T. Li, G. W. Meng, Y. Lei, and C. H. Zhu, “Template-assisted fabrication of Ag nanoparticles@ZnO-nanorods array as recyclable 3D surface enhanced Raman scattering substrate for rapid detection of trace pesticides,” Nanotechnology 32(14), 145302 (2021). [CrossRef]
8. S. K. Gahlaut, P. Devi, and J. P. Singh, “Self-sustainable and recyclable Ag nanorods for developing Ag-Ag2S nanoheterostructures using sewage gas: applications in photocatalytic water purification, hydrogen evolution, SERS and antibacterial activity,” Appl. Surf. Sci. 528, 147037 (2020). [CrossRef]
9. Q. Chen, C. L. Shi, L. X. Qin, S. Z. Kang, and X. Q. Li, “A low-cost 3D core-shell nanocomposite as ultrasensitive and stable surface enhanced Raman spectroscopy substrate,” Sens. Actuators, B 327, 128907 (2021). [CrossRef]
10. T. Lee, S. Kwon, H. J. Choi, H. Lim, and J. Lee, “Highly sensitive and reliable microRNA detection with a recyclable microfluidic device and an easily assembled SERS substrate,” ACS Omega 6(30), 19656–19664 (2021). [CrossRef]
11. N. Zhao, X. Fei, X. N. Cheng, and J. Yang, “Synthesis of silver/silver chloride/graphene oxide composite and its surface-enhanced Raman scattering activity and self- cleaning property.and its surface-enhanced Raman scattering activity and self-cleaning property,” IOP Conf. Ser.: Mater. Sci. Eng. 242, 012002 (2017). [CrossRef]
12. Y. Y. Du, H. M. Liu, Y. Chen, Y. R. Tian, X. L. Zhang, C. J. Gu, T. Jiang, and J. Zhou, “Recyclable label-free SERS-based immunoassay of PSA in human serum mediated by enhanced photocatalysis arising from Ag nanoparticles and external magnetic field,” Appl. Surf. Sci. 528, 146953 (2020). [CrossRef]
13. Y. Cheng, W. Z. Wang, L. Z. Yao, J. Wang, H. S. Han, T. Y. Zhu, Y. J. Liang, J. L. Fu, and Y. N. Wang, “3D Ag/ZnO microsphere SERS substrate with ultra-sensitive, recyclable and self-cleaning performances: application for rapid in site monitoring catalytic dye degradation and insight into the mechanism,” Colloids Surf., A 607, 125507 (2020). [CrossRef]
14. J. J. Chen, H. L. Su, F. Song, W. J. Moon, Y. S. Kim, and D. Zhang, “Bioinspired Au/TiO2 photocatalyst derived from butterfly wing (Papilio Paris),” J. Colloid Interface Sci. 370(1), 117–123 (2012). [CrossRef]
15. B. R. Tao, F. J. Miao, and P. K. Chu, “Fabrication and photoelectrochemical study of vertically oriented TiO2/Ag/SiNWs arrays,” J. Alloys Compd. 635, 112–117 (2015). [CrossRef]
16. K. Hsu and D. Chen, “Highly sensitive, uniform, and reusable surface-enhanced Raman scattering substrate with TiO2 interlayer between Ag nanoparticles and reduced graphene oxide,” ACS Appl. Mater. Interfaces 7(49), 27571–27579 (2015). [CrossRef]
17. W. Y. Yang, J. Q. Tang, Q. H. Ou, X. Q. Yan, L. Liu, and Y. K. Liu, “Recyclable Ag-deposited TiO2 SERS substrate for ultrasensitive malachite green detection,” ACS Omega 6(41), 27271–27278 (2021). [CrossRef]
18. Y. B. Xie, “Fabrication of highly ordered Ag/TiO2 nanopore array as a self-cleaning and recycling SERS substrate,” Aust. J. Chem. 74(10), 715–721 (2021). [CrossRef]
19. Y. Wang, M. S. Li, D. G. Wang, C. P. Han, J. J. Li, C. Y. Wu, and K. Xu, “Fabrication of highly uniform Ag nanoparticle-TiO2 nanosheets array hybrid as reusable SERS substrates,” Colloid Interface Sci. Commun. 39, 100324 (2020). [CrossRef]
20. D. Riassetto, F. Roussel, L. Rapenne, H. Roussel, S. Coindeau, O. Chaix, F. Micoud, M. Chatenet, and M. Langlet, “Synthesis and functionalities of noble metal nanoparticles formed through simple all-inorganic photochemical procedures,” J. Exp. Nanosci. 5(3), 221–243 (2010). [CrossRef]
21. X. Zhao, W. W. Zhang, C. C. Peng, Y. J. Liang, and W. Z. Wang, “Sensitive surface-enhanced Raman scattering of TiO2/Ag nanowires induced by photogenerated charge transfer,” J. Colloid Interface Sci. 507, 370–377 (2017). [CrossRef]
22. J. Fujisawa, N. Kaneko, and M. Hanaya, “Interfacial charge-transfer transitions in TiO2 nanoparticles adsorbed with 4-Mercaptobenzenoic acid: carboxy versus thiol anchor and adsorbate-to-TiO2 versus TiO2-to-adsorbate charge transfer,” J. Phys. Chem. C 125(24), 13534–13541 (2021). [CrossRef]
23. S. Q. Huang, C. J. Wu, Y. Y. Wang, X. Yang, R. Yuan, and Y. Q. Chai, “Ag/TiO2 nanocomposites as a novel SERS substrate for construction of sensitive biosensor,” Sens. Actuators, B 339, 129843 (2021). [CrossRef]
24. M. Jakob, H. Levanon, and P. V. Kamat, “Charge distribution between UV-irradiated TiO2 and gold nanoparticles: determination of shift in the Fermi level,” Nano Lett. 3(3), 353–358 (2003). [CrossRef]
