Enhanced Raman scattering when scatterer molecules located in TiO₂/Ag nanojunctions

Abstract

The TiO₂/4-mercaptopyridine (4-Mpy)/Ag sandwich structure has been fabricated by a self-assembly method and the enhanced Raman scattering of the molecule embedded in the nanojunction, known as surface-enhanced Raman spectroscopy (SERS), was studied. X-ray photoelectron spectroscopy (XPS) has been employed to investigate the formation process of the TiO₂/4-Mpy/Ag sandwich structure. Raman spectra of the TiO₂/4-Mpy/Ag sandwich structure were significantly enhanced relative to those observed on 4-Mpy adsorbed on Ag particles or TiO₂/4-Mpy. A possible distribution of 4-Mpy molecules in the fabrication of TiO₂/4-Mpy/Ag assemblies has been proposed. It shows that 4-Mpy molecules adsorb on the surface of the TiO₂ NPs, and the 4-Mpy molecules connect with Ag NPs via the S atom. The analysis of XPS confirms the result obtained from Raman spectra of TiO₂/4-Mpy/Ag assemblies. The SERS enhancement mechanism is discussed briefly in the article, and we ascribed it to the charge-transfer mechanism.

---

Introduction

Surface-enhanced Raman scattering (SERS) was first discovered in 1974 by Fleischmann.¹ In the same period, van Duyne reported that when pyridine was adsorbed on the surface of a rough Ag electrode the Raman signal was enhanced 10⁶ times in 1977.² Since then, many researchers have carried out scientific experiments and theoretical studies to validate this experimental fact. Compared to Raman spectroscopy, SERS signal enhancement can reach 10⁸ to 10¹⁴ times, which means the SERS technique can be used in the field of supersensitive analysis, such as chemical analysis, environmental testing, and so on.^3–9

When it comes to nanoscience and nanotechnology, SERS is often indispensable. The strongest SERS effects are usually observed in metal substrates, typically Ag and Au, that are very rough on a microscopic scale, such as electrode surfaces roughened by redox cycles, aggregated colloids, single ellipsoidal nanoparticles (NPs), arrays of metal NPs prepared by lithographic techniques, and metal nanoshells.^10–14 Since Quagliano first observed the SERS signal from the non-metallic InAs/GaAs substrate, scientists have shown increasing attention on the semiconductor SERS substrates.¹⁵ Our group has reported some SERS substrates and preliminary studied the SERS enhancement mechanism, such as ZnO, TiO₂, ZnS, CdTe, SnO₂.^16–20 Semiconductor–metal composite materials have attracted considerable attention due to their potential applications in diverse areas such as solar cells, optics and chemical sensing.^21–23 Further, we have studied the complex system of metal and semiconductor, such as ZnO/Ag, CuO/Ag and TiO₂/Ag, and observed the SERS signals when molecular adsorbed on the substrates.^24–26

In many SERS studies, the difficulty in experimentally gaining deep insight into the enhancement associated with the charge-transfer mechanism lies in the fact that these chemical enhancements (EM) are normally inextricably linked with electromagnetic (EM) enhancements.^27,28 In this work, TiO₂/4-Mpy/Ag composite was fabricated to investigate the SERS of molecules interconnecting of TiO₂ and Ag to study the CT effect in such connection. Experimental results showed that the SERS signal of this composite structure is much higher than the SERS signals of TiO₂/4-Mpy and Ag/4-Mpy separately. We found that, the Fermi level of Ag particles, the lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO) of 4-Mpy, the valence band (VB) and conduction band (CB) of TiO₂ NPs in this system are the main factors that affect the SERS enhancement, and finally composed an integrated charge-transfer contribution.

Experimental

Chemical reagents

4-Mercaptopyridine (4-Mpy) was purchased from Acros Organics Chemical Co. and used as received without further purification. The other reagents (AgNO₃, sodium citrate, tetrabutyltitanate, anhydrous ethanol and nitric acid) were purchased from Beijing Chemical Co. Triply distilled water was used for the entire synthesis.

Synthesis of silver colloid

Silver colloid was synthesized according to the literature.²⁹ Firstly, 0.036 g AgNO₃ was dissolved in 200 mL water and heated to boil under vigorous stirring. Then, a solution of 1% sodium citrate (4 mL) was added. The solution was kept boiling for about 1 h. Finally, the Ag colloid was obtained, which showed greenish yellow and displayed an absorption maximum at 420 nm.

Synthesis of TiO₂ NPs

Typically, TiO₂ NPs were synthesized by a sol-hydrothermal process using a previously reported procedure.¹⁷ First, a mixed solution of 5 mL of tetrabutyltitanate and 5 mL of anhydrous ethanol was added dropwise into another mixed solution, consisting of 20 mL of anhydrous ethanol, 5 mL of water and 1 mL of 70% nitric acid, which were roughly stirred to carry out hydrolysis at room temperature. Subsequently, the sol was obtained by continuously stirring for 1 h. Next, the as-prepared sol was sealed into a 50 mL Teflon-lined autoclave at 160 °C for 6 h, and then cooled to room temperature, followed by drying at 60 °C for 24 h. Finally, TiO₂ NPs were obtained by calcining the sol-hydrothermal production for 2 h at 500 °C.

The fabrication condition used in this experiment has been optimized. We have synthesized different size of TiO₂ NPs by changing the calcination temperature (400, 450, 500, 550 and 600 °C). The TiO₂ NPs were used as SERS active substrate, and observed that the maximum SERS signal appeared at the diameter is 10.9 nm (calcinated at 500 °C).

Synthesis of TiO₂/4-Mpy/Ag sandwich structure

Firstly, 20 mg of TiO₂ NPs was dispersed in 10 mL of 4-Mpy (10⁻³ M) ethanol solution and was stirred for 2 hours. Secondly, the 4-Mpy modified TiO₂ NPs was obtained by centrifugation and washing the precipitate three times. Thirdly, TiO₂/4-Mpy was then dispersed into 10 mL of Ag sol and vigorous stirred for 2 hours, centrifuged and washed three times with the deionized water. Finally, the obtained product was centrifuged and dried for 24 hours at 60 °C to obtain the TiO₂/4-Mpy/Ag sandwich complex.

Sample characterization

The crystal structure of TiO₂ NPs was characterized by X-ray diffraction using a Siemens D5005 X-ray powder diffractometer with a Cu Kα (λ = 1.5418 Å) radiation source at 40 kV and 30 mA. The surface morphology of the samples was measured on a Hitachi H-8100 transmission electron microscope (TEM) operated at an acceleration voltage of 200 kV. X-ray photoelectron spectra (XPS) were investigated by using a VG ESCALAB MK II spectrometer with an Mg KR excitation (1253.6 eV). The electronic absorption spectra were recorded on a Shimadzu UV-3600 UV-vis spectrophotometer. Raman spectra were measured with a Renishaw Raman system model 1000 spectrometer. The 514.5 nm line from a 20 mW air-cooled argon ion laser was used as exciting light. Data acquisition was the result of three 10 s accumulations. The resolution of the Raman instrument was ca. 4 cm⁻¹. The Raman band of the silicon wafer at 520.7 cm⁻¹ was used to calibrate the spectrometer.

Results and discussion

The properties of the TiO₂ NPs including crystal structure and composition were demonstrated through a series of experiments. XRD spectrum was used to identify the size and structure of TiO₂ NPs. Fig. 1 shows the XRD spectrum of TiO₂ NPs calcined at 500 °C. The crystallite sizes (D) were about 10.9 nm. The TEM results also confirm this (Fig. 2a and d). This value was estimated from the half-bandwidth of the corresponding X-ray spectral peak by the Scherrer formula: D = kλ/(β cos θ), where λ is the X-ray wavelength, β is the half width of the (110) peak, θ is the Bragg diffraction angle, and k is a correction factor which is taken as 0.89.¹⁷

Fig. 1 XRD spectrum of TiO₂ NPs.

Fig. 2 TEM images of TiO₂ NPs (a), TiO₂/4-Mpy (b), TiO₂/4-Mpy/Ag (c), and size distribution of TiO₂ (d).

Fig. 2 illustrates the TEM images of TiO₂ NPs, TiO₂/4-Mpy and TiO₂/4-Mpy/Ag sandwich structure. As can be seen from Fig. 2a, the diameter of TiO₂ NPs is about 10 nm. Also, as can be seen from Fig. 2b, the edge of the TiO₂ NPs is not as clear as Fig. 2a shows us, it appears due to the change of the surrounding with the TiO₂ NPs, and the result is consistent with the TiO₂/4-Mpy structure. Fig. 2c exhibited that the Ag NPs were added in the sandwich system. It is consistent with the structure we wish to fabricate. And the TiO₂ size distribution is shown in Fig. 2d. It can indicate that the TiO₂/4-Mpy/Ag sandwich structure assemblies have been constructed successfully by the self-assembly method.

Fig. 3 shows the UV-vis spectrum of silver colloid that we used in the experiment. The peak located at 420 nm indicated that the Ag NPs diameter is about 40 nm that Ag NPs are well known can exhibit the best SERS signals. The result of the diameter is also consistent with the TEM results shows in the Fig. 2c.

Fig. 3 UV-vis spectrum of silver colloid.

To further examine the character of the complex, the state of Ag in the TiO₂/4-Mpy/Ag complex measured by XPS was obtained (Fig. 4). The Ag 3d 5/2 peak appears at a binding energy of 368.3 eV, and the splitting of the 3d doublet is 6.1 eV.³⁰ These data prove the formation of metallic silver, indicating the formation of Ag NPs in the TiO₂/4-Mpy/Ag complex.

Fig. 4 XPS spectrum of TiO₂/4-Mpy/Ag structure.

The coordination of Ag NPs with the sulfur of 4-Mpy molecules is confirmed by comparing the XPS spectra. Shifts in binding energies of the peaks for 2p 3/2 and N 1s are shown in Fig. 5 and 6. The first step is to attach 4-Mpy molecules on the TiO₂ nanoparticles to functionalize the substrate. The binding energy for the S 2p 3/2 transition appears at 161.35 eV. After the silver nanoparticles attached on the TiO₂/4-Mpy, the peak involving S 2p 3/2 shifts from 164.4 eV to 162.4 eV. Previous studies of alkanethiols and thiophenol on various metal surfaces show similar shift in binding energy due to the formation of thiolate species.³¹ Moreover, the intensity of the XPS results of S 2p 3/2 decreases a little after the silver nanoparticles are adsorbed. Therefore, there is no doubt that Ag NPs are mainly assembled through the S atom, which matches well with the previous conclusion. It is also very interesting, from Fig. 6 we can conclude that Ag bonded with the thio-group and TiO₂ bonded with the N. Moreover, the result indicates that it is prone to form the Ag–S bond instead of the Ag–N since the former system seems more stable.

Fig. 5 XPS spectra of S 2p of 4-Mpy, TiO₂/4-Mpy/Ag and TiO₂/4-Mpy.

Fig. 6 XPS spectra of N 1s of 4-Mpy, TiO₂/4-Mpy/Ag and TiO₂/4-Mpy.

Fig. 7 exhibits the SERS spectra of 4-Mpy molecules adsorbed on the TiO₂ NPs, Ag sol, and TiO₂/4-Mpy/Ag, respectively. As we can see from Fig. 7 the most intense SERS signal was observed when Ag NPs connected with the TiO₂/4-Mpy, and we ascribed the phenomenon to the charge-transfer mechanism. The SERS intensity of the TiO₂/4-Mpy/Ag complex is 7.4 times greater than that with the molecule adsorbed on Ag NPs. This complex structure play an important role in promoting the charge-transfer process, thus enhancing the SERS signals. Also we observe that the SERS signals of 4-Mpy molecule exhibit obvious differences not only in the intensity but also Raman frequency, as compared with that SERS enhancement in the Ag and TiO₂/4-Mpy. In Table 1, we list the observed SERS spectrum of 4-Mpy adsorbed on Ag and TiO₂/4-Mpy/Ag sandwich structure.

Fig. 7 SERS spectra of 4-Mpy on (a) TiO₂ NPs (b) TiO₂/4-Mpy/Ag assemblies and (c) Ag colloids.

Table 1 Raman shifts (cm⁻¹) and assignments of 4-Mpy molecule^a

Raman	SERS			Assignment
Bulk	On Ag(Au)	On TiO2	On TiO2/Mpy-Ag
a Assignments from ref. 18 and 32.
642	660	663	662	β(CCC)
720	710	727	709	β(CC)/(C–S)
786	791		812	β(CH)
785	1004	1021	1008	Ring breathing
1040	1049		1020	β(CH)
1076	1065	1063	1062	β(CH)
1100	1095	1119	1096	Ring breathing/C–S
1197	1217	1224	1200	β(CH)/(NH)
1247	1218	1240	1217	β(CH)
1285	1311	1320	1318	β(CH)
1392	1412	1426	1422	ν(CC)
1456	1454			ν(C=C/C=N)
1475	1478	1478	1478	ν(C=C/C=N)
1580	1580		1578	ν(C=C/C=N)
1595			1593	ν(CC)
1614	1615		1610	ν(CC)

---

At present, there are two mechanisms describing the SERS phenomenon: electromagnetic enhancement mechanism and chemical enhancement mechanism.^33–35 Typically, the chemical enhancement is explained via the CT mechanism.³⁶ At a CT resonance, however, a considerably larger fraction of the enhanced intensity can be attributed to chemical effects. For semiconductor nanomaterials, the surface plasmon resonance frequency is located far from the laser line we choose in our experiment (514.5 nm). Therefore, we can assume that the chemical enhancement is the main factor that enhanced the SERS signals. The chemical enhancement functions as an electronic resonance process, where the charge-transfer occurs among the LUMO/HOMO of molecules, VB/CB of semiconductor, and Fermi level of the metal substrate, which increases the effective polarizability of the molecules. This not only enhances the intensity of Raman scattering, but also changes the peak positions. Previously, we have reported that for the pure TiO₂ nanoparticles SERS substrate that the SERS mechanism is attributed to the TiO₂-to-molecule charge-transfer mechanism. In the present experiment, however, the Ag NPs in the TiO₂/4-Mpy/Ag assembly most likely play a similar role in TiO₂-to-molecule charge-transfer process. Fig. 8 shows the energy levels diagram of the TiO₂/4-Mpy/Ag assembly, the energy levels list in the figure was reported in the literature.^37,38 When the laser irradiates the sample surface, the energy of the laser line of 514.5 nm (2.41 eV) used in the experiment has sufficient energy to induce electrons transfer from the Fermi level of Ag NPs to the LUMO level of the 4-Mpy molecules. The electrons can be ejected to the VB level of TiO₂ NPs, thus complete the charge-transfer process. This process produces a maximum SERS signal from the TiO₂/4-Mpy/Ag composite.

Fig. 8 Energy level diagram of the TiO₂/4-Mpy/Ag sandwich structure. All levels are measured from the vacuum.

In order to measure the degree of the charge-transfer contribution in the TiO₂/4-Mpy/Ag sandwich structure, we need to obtain the ration (R) of intensity of a non-totally symmetric line b₂ to that of a totally symmetric line a₁.³³ They derive most of their intensity from Franck–Condon factors (the Albrecht A-term), whereas the non-totally symmetric lines are derived from vibronic coupling (B-terms) that is highly sensitive to charge-transfer contributions to the enhancement. Use of the ratio also has the advantage of making our measure of charge-transfer independent of other effects on the total enhancement factor. The degree of charge-transfer (ρ_CT) is then defined as

ρ_CT = R/(1 + R)

When R is zero, there is presumably no charge-transfer (ρ_CT = 0), whereas for large R, the degree of charge-transfer approaches 1.³⁹ And in this system, the calculated charge-transfer contribute for the v₈ (b₂) line to the v₁ (a₁) is about 72%, also confirmed that the charge-transfer mechanism is the main contribution in the TiO₂/4-Mpy/Ag system.

Conclusions

In this work, we fabricated TiO₂/4-Mpy/Ag composite as SERS active substrate, and studied the SERS mechanism. Structure is an important parameter affecting the Raman intensity. Due to the complex substrate, the SERS signal was strongly enhanced, XPS results were used to analysis the bonding state between the TiO₂, 4-Mpy, and Ag NPs. It can be ascribed to the synergetic contribution of metal and semiconductor to SERS. The 4-Mpy molecules form a bridge between TiO₂ and Ag NPs, and promote charge-transfer, enhancing the SERS signal. It is expected that this work will not only be helpful to design semiconductor based SERS substrates and deeply understand the CT mechanism, but also be valuable for practical application of SERS technology in the nanomaterial field.

Acknowledgements

Supported by the Natural Science Foundation of China (Grant No. 21103062, 21273091, 21221063, 201327803), Specialized Research Fund for the Doctoral Program of Higher Education of China (Grant No. 20110061110017, 20100061120087), the Development Program of the Science and Technology of Jilin Province (No. 20140101160JC).

Notes and references

1. M. Fleischmann, P. J. Hendra and A. J. McQuillan, Chem. Phys. Lett., 1974, 2566, 163.
2. D. L. Jeanmaire and R. P. van Duyne, J. Electroanal. Chem., 1977, 84, 1.
3. M. T. Sun, Z. L. Zhang, P. J. Wang, Q. Li and F. C. Ma, Light: Sci. Appl., 2013, 2, 112.
4. P. Wang, Y. P. Wang and L. T. Tong, Light: Sci. Appl., 2013, 2, 102.
5. Q. Sun, K. Ueno, Y. Han, A. Kubo and Y. Matsuo, Light: Sci. Appl., 2013, 2, 118.
6. A. E. Cetin, A. F. Coskun, B. C. Galarreta, M. Huang and D. Herman, Light: Sci. Appl., 2014, 3, 122.
7. K. Kneipp, H. Kneipp, I. Itzkan, R. R. Dasari and M. S. Feld, Chem. Rev., 1999, 99, 2957.
8. G. A. Baker and D. S. Moore, Anal. Bioanal. Chem., 2005, 382, 1751.
9. T. Murphy, H. Schmidt and H. D. Kronfeldt, Appl. Phys. B: Lasers Opt., 1999, 69, 147.
10. M. F. Mrozek, Y. Xie and M. J. Weaver, Anal. Chem., 2001, 73, 5953.
11. Y. Cui, B. Ren, J. L. Yao, R. A. Gu and Z. Q. Tian, J. Phys. Chem. B, 2006, 110, 4002.
12. M. Futamata, Y. Maruyama and M. Ishikawa, J. Phys. Chem. B, 2003, 107, 7607.
13. M. Panagopoulou, N. Pantiskos, P. Photopoulos, J. Tang and D. Tsoukalas, Nanoscale Res. Lett., 2011, 6, 629.
14. M. F. Cardinal, B. Rodriguez-Gonzalez, R. A. Alvarez-Puebla, J. Perez-Juste and L. M. Liz-Marzan, J. Phys. Chem. C, 2010, 114, 10417.
15. L. G. Quagliano, J. Am. Chem. Soc., 2004, 126, 7393.
16. Y. Wang, W. Ruan, J. Zhang, B. Yang and W. Xu, J. Raman Spectrosc., 2009, 40, 1072.
17. X. Xue, W. Ji, Z. Mao, H. Mao and Y. Wang, J. Phys. Chem. C, 2012, 116, 8792.
18. Y. Wang, Z. Sun, H. Hu, S. Jing and B. Zhao, J. Raman Spectrosc., 2007, 38, 34.
19. Y. Wang, J. Zhang, H. Jia, M. Li and J. Zeng, J. Phys. Chem. C, 2008, 112, 996.
20. J. L. Hou, X. F. Jia, X. X. Xue, L. Chen and W. Song, Chem. J. Chin. Univ., 2012, 33, 139.
21. I. M. Dharmadasa, J. D. Bunning, A. P. Samantilleke and T. Shen, Sol. Energy Mater. Sol. Cells, 2005, 86, 373.
22. E. Shaviv, O. Schubert, M. Alves-Santos, G. Goldoni and R. Di Felice, Acs Nano, 2011, 5, 4712.
23. B. K. Duan, J. Zhang and P. W. Bohn, Anal. Chem., 2012, 84, 2.
24. W. Song, X. X. Han, L. Chen, Y. M. Yang and B. Tang, J. Raman Spectrosc., 2010, 41, 907.
25. W. Song, Y. Wang and B. Zhao, J. Phys. Chem. C, 2007, 111, 12786.
26. Y. F. Wang, W. Song, W. D. Ruan, J. X. Yang and B. Zhao, J. Phys. Chem. C, 2009, 113, 8065.
27. W. Ji, X. X. Xue, W. D. Ruan, C. X. Wang, N. Ji, L. Chen, Z. S. Li, W. Song, B. Zhao and J. R. Lombardi, Chem. Commun., 2011, 47, 2426.
28. L. B. Yang, X. Jiang, W. D. Ruan, J. X. Yang, B. Zhao, W. Q. Xu and J. R. Lombardi, J. Phys. Chem. C, 2009, 113, 16226.
29. P. L. Lee and D. Melsel, J. Phys. Chem., 1982, 86, 3391.
30. I. Lopez-Salido, D. C. Lim and Y. D. Kim, Surf. Sci., 2005, 588, 6.
31. G. Xue, M. Ma, J. F. Zhang, Y. Lu and K. T. Carron, J. Colloid Interface Sci., 1992, 150, 1.
32. Y. Wang, H. Hu, S. Jing, Y. Wang and Z. Sun, Anal. Sci., 2007, 23, 787.
33. J. R. Lombardi and R. L. Birke, J. Phys. Chem. C, 2008, 112, 5605.
34. H. Xu and M. Kall, Surface-Enhanced Raman Scattering: Physics and Applications, 2006, vol. 103, p. 87.
35. S. J. Lee, Z. Guan, H. Xu and M. Moskovits, J. Phys. Chem. C, 2007, 111, 17985.
36. L. B. Yang, X. Jiang, W. D. Ruan, B. Zhao and W. Q. Xu, J. Phys. Chem. C, 2008, 112, 20095.
37. X. Xue, W. Ji, Z. Mao, Z. Li and W. Ruan, Spectrochim. Acta, Part A, 2012, 95, 213.
38. X. Q. Fu, T. S. Jiang, Q. Zhao and H. B. Yin, J. Raman Spectrosc., 2012, 43, 1191.
39. X. Fu, T. Jiang, Q. Zhao and H. Yin, J. Raman Spectrosc., 2012, 43, 1191.

---