Open Access
Add to BibSonomyAbstract
Nanoporous gold (NPG) is an advanced functional material with both propagating and localized surface plasmon resonance (PSPR and LSPR) effects. In this work, uniform NPG films with controlled thickness and small pore size were easily prepared by sputtering deposition followed by low-temperature dealloying. Using slab waveguide spectroscopy, the LSPR absorption peak of the NPG film was measured to shift from 566 nm to 586 nm by changing the surrounding refractive index from 1.333 to 1.368. Total internal reflection (TIR) SERS spectra for Rhodamine 6G (R6G) and Nile blue (NB) molecules adsorbed in the NPG film were investigated with a prism coupler. Upon increasing the incident angle (θ) from the critical value, the intensity of the SERS signal excited with the s-polarized laser beam of 532 nm wavelength gradually decreases but with the p-polarized laser beam exhibits a peak at θ ≈53°. The peak intensity is 2 times stronger than that excited with normal incidence of the same laser beam. The PSPR mode of the NPG film can be excited at 785 nm wavelength, which leads to a strong SERS signal of NB. In contrast, the LSPR effect at 785 nm wavelength is too weak to make the SERS signal undetectable with normal incidence. The findings indicate that the TIR geometry can bring about at least a twofold enhancement of SERS sensitivity of NPG film relative to the conventional normal incidence and that the contributions of LSPR and PSPR effects to the SERS sensitivity of NPG film are different at different incident wavelengths.
© 2016 Optical Society of America
1. Introduction
Nanoporous gold (NPG) is an ancient yet advanced functional material which has been used for a large variety of important applications, including catalysis, cantilever, lithium battery, electrochemical sensors, plasmonic sensors, and SERS substrates [1–6 ]. These applications are based on the following unique features of NPG: it is easy to fabricate and possesses large internal surface area, high connectivity of three-dimensional pores, good uniformity, excellent biocompatibility as well as remarkable electronic, mechanical, optical and plasmonic properties. The localized surface plasmon resonance (LSPR) effect of NPG makes it often used for fabricating both plasmonic sensors and SERS substrates. However, according to the published papers, the SERS enhancement factor of conventional NPG films measured with the normal incidence of excitation beam is quite low relative to other gold nanostructures such as nanoparticles. So far, much effort has been made to optimize the microstructure of NPG films for improving the SERS enhancement factor. The reported approaches to optimization of NPG microstructures include: (1) preparing small-pore NPG by chemical dealloying at a low temperature; (2) enlarging the gold ligaments by plating gold on the pore walls; (3) wrinkling NPG films by thermal contraction of the NPG/Polymer-substrate composites; (4) imprinting NPG films by direct imprinting of porous substrate (DIPS) [7–12 ]. These approaches indeed lead to significant improvement of SERS enhancement factor, while they also complicate the NPG-fabricating process, making NPG cost ineffective.
In addition to LSPR, NPG films have the propagating surface plasmon resonance (PSPR) effect. Unlike LSPR that can be easily observed by simple normal transmission method, the PSPR of NPG films is hard to be implemented because of the following three reasons: (1) the PSPR mode is an evanescent wave that needs to be excited with the Kretschmann configuration under the phase-matching condition, and this condition is closely related with the excitation wavelength, the incident angle, the NPG thickness and the surrounding refractive index; (2) NPG films are generally prepared using commercial AuAg alloy leaves of 100 nm thickness, and a lack of choice in thickness of alloy leaves makes optimization of NPG thickness for effective PSPR effect difficult to do; (3) complex refractive index of NPG films is an unknown function of wavelength, and it changes with the film porosity and the surrounding medium. For example, when a-50-nm-thick NPG film is exposed to air, a well-defined PSPR band can be detected in the wavelength range from visible to near infrared. However, after the NPG film is immersed in an aqueous solution, the PSPR band disappears in this spectral region. Owing to the above difficulties, the PSPR effect of NPG film and its application were rarely reported in the past decade since the first paper on the NPG’s PSPR property was published in 1996. To the best of our knowledge, up to date only the LSPR effect of NPG was utilized in SERS measurements with convention normal incidence of the laser beam, and the PSPR-based SERS application of NPG has not yet been reported.
In order to introduce the PSPR effect into SERS application of NPG, the prism-coupled total internal reflection (TIR) geometry (namely, the Kretschmann configuration) is proposed. Raman scattering excited by TIR was first revealed in 1973 by lkeshoji et al. [13]. Use of the TIR geometry for SERS excitation has several merits, including good surface-selective detection capability, flexible adjustment of polarization state of the excitation beam, and controllable field enhancement factor [13,14 ]. For the NPG-based SERS measurements, the TIR method makes the simultaneous excitation of PSPR and LSPR possible. In this case, the field enhancement factor could be further increased through the coupling between PSPR and LSPR of the NPG film [15,16 ].
Considering that the PSPR and LSPR excitations are closely related to the geometry parameters of NPG film, a sputtering-dealloying method is employed instead of using commercial available AuAg alloy leaves. The precisely controlled thickness of NPG (~50 nm) leads to an efficient excitation of PSPR while the small pore size gives rise to a strong LSPR effect [8]. Both p-polarized and s-polarized evanescent field generated by attenuated total reflection (ATR) method are used as excitation: the SERS signal obtained with different polarized evanescent field excitation show great difference in intensity, indicating a higher SERS enhancement factor with the p-polarized evanescent field excitation. The p-polarized evanescent field excitation method not only improves the NPG Raman enhancement factor without complicating the fabrication method but also makes SERS detection compatible with the well-established PSPR technique which has a long history in sensing application.
2. Fabrication and characterization of NPG films
NPG films were prepared according to the following steps: (i) a 3 nm chromium layer and a 10 nm gold layer were successively sputtered on the glass substrate; (ii) a 50-nm-thick film of AuAg alloy (weight ratio of Au:Ag = 1:1) was sputtered on the gold layer, which is only half the thickness of commercial alloy leaves widely used in the reported papers [4,7–11 ]; (iii) the metal-coated glass substrate was immersed in 68 wt.% HNO3 to dissolve the Ag component of the alloy film. This process is referred to as dealloying. After dealloying, the glass substrate was rinsed with deionized water to remove the residual acid. Note that the chromium layer was used to improve the adhesion of the upper layers to the glass substrate while the gold layer plays a role of preventing the Cr dissolution in the dealloying process. To achieve small-pore NPG films, the dealloying was carried out at low temperature (−18°C) for 24h. The NPG film was also prepared by chemical dealloying at room temperature (20°C) for 1h. Surface morphologies of the two NPG films were investigated using a scanning electron microscope (SEM). According to the SEM images shown in Figs. 1(a) and 1(b) , the selective dissolution of Ag atoms and the diffusion-aggregation of the remaining Au atoms result in the formation of a nanoporous sponge composed entirely of Au atoms since no detectable Ag has been observed with energy-dispersive X-ray (EDX) spectroscopy in Fig. 1(c) and a much smaller pore size comes with the low-temperature dealloyed one, which leads to an enhanced LSPR effect [8]. The following experiments were conducted with the low-temperature dealloyed NPG films.
Fig. 1 (a) SEM image of the NPG film prepared by 1h dealloying at room temperature; (b) and (c) SEM image and EDX spectrum of the NPG film prepared by 24h dealloying at −18°C.
3. LSPR absorption band of NPG measured by slab waveguide spectroscopy
LSPR effect of metallic nanostructures is responsible for not only plasmonic sensors but also electromagnetic enhancement of SERS signal. Ultraviolet-visible extinction spectroscopy is usually employed to determine the LSPR resonant band of SERS substrate. However, the LSPR resonant band of the NPG film used here cannot be directly measured by UV-Vis extinction spectroscopy (Shimadzu, UV-2600) owing to not only the weaker LSPR effect compared to metal nanoparticles but also the extremely small thickness. Therefore, the slab waveguide spectroscopy is introduced and the experimental setup is illustrated in Fig. 2 . This method is based on the liquid-clad-RI-variation induced resonant band shift of the evanescent-wave absorption of the NPG film deposited on a free-standing slab glass waveguide [17]. The resonant band can be readily determined by measuring the waveguide output intensity spectra before and after the clad exchange air for aqueous sodium chloride solutions of different concentrations. The CCD dark current [ID(λ)], the output light intensity spectrum with the empty chamber [IR(λ)] and liquid-filled chamber [IS(λ)] were measured, respectively. The absorbance [A(λ)] is determined by Eq. (1):
Fig. 2 Schematic diagram of the optical waveguide spectroscopy setup for measurement of LSPR absorption spectra of the NPG film.
Five aqueous sodium chloride solutions were prepared, and their concentrations in weight percent are 0, 5%, 10%, 15% and 20% and the corresponding refractive indexes are n = 1.3329, 1.3418, 1.3507, 1.3596 and 1.3684. Figure 3(a) shows the extinction spectra measured after immersion of the NPG film in different solutions. The LSPR extinction peak of the NPG film is clearly seen in each spectrum. According to Fig. 3(b), the LSPR peak linearly shifts from 566 nm to 586 nm with increasing the surrounding refractive index from 1.3329 to 1.3684, which lead to the refractive-index sensitivity of 556 nm/RIU. This sensitivity is 2 times higher than the largest sensitivity (270 nm/RIU) reported in ref. 18 in which the NPG films were prepared using commercial AuAg alloy leaves of 100 nm thickness.
Fig. 3 (a) LSPR absorption spectra of the NPG film covered with aqueous NaCl solution of different concentrations; (b) Linear dependence of the peak wavelength on refractive index of the bulk solution.
4. Prism coupled total internal reflection SERS application of NPG films
NPG does not show a promising SERS enhancement factor as compared with other nanostructured gold, such as gold nanoparticles, and this point has been verified by the weak LSPR absorption. Here the evanescent field excited SERS effect of NPG film with Kretschmann configuration is introduced and the experimental setup is shown by Fig. 4 . Before SERS detection, the NPG film was immersed in an aqueous solution of 1μM Rhodamine 6G (R6G) for five hours, until the surrounding environment of substrate was stable and the R6G molecules uniformly adsorbed on the exterior and interior surface of NPG, and then dried in air to immobilize the R6G molecules. P-polarized evanescent field (H y, E x, E z) and s-polarized evanescent field (E y, H x, H z) at 532 nm (a collimated light with equal power of 10 mw and the same spot size of 1mm in diameter) were used as excitations.
It was observed with the above experimental setup that the refracted laser beam just completely disappeared at an incident angle of θ ≈36.5°. With use of this angle as the critical angle, the real part of refractive index for the NPG film is calculated to be 0.9. This value was used for calculation of field enhancement factor (refractive index of glass substrate is 1.52). The SERS spectra for R6G molecules adsorbed inside the NPG film were measured at different angles of incidence. The first measurement was carried out by setting the incident angle to θ = 37° that is slightly larger than the critical value to make sure generation of a stronger evanescent field. Figure 5(a) shows the SERS spectra of R6G excited by s-polarized evanescent field. It is evident that the SERS signal is angle dependent and its intensity gradually decreases with increasing the incident angle. To interpret this experimental result, we calculate the s-polarized electric field (E y) enhancement factor at glass/NPG interface. Figure 5(b) displays the calculated curve. The electric field enhancement factor rapidly decreases from its peak value down to zero with increasing the incident angle from the critical angle to 90°, which is in good consistent with the experiment results.
Fig. 5 Raman signals excited by (a) s-polarized and (c) p-polarized evanescent field at different incident angles; (b) s-polarized electric field enhancement at glass/NPG interface (black line) and the Raman intensity of 614cm−1 from R6G (blue dots); (d) p-polarized electric field enhancement (red line represents for the x-component and black line represents for the z-component) at glass/NPG interface and the Raman intensity of 614cm−1 from R6G (blue dots).
The evanescent field generated by the p-polarized laser beam includes two electric field components: E x and E z. The angle dependence of E z is similar to that observed with the s-polarized laser beam while E x has a complicated relationship with the incident angle. As the incident angle increases from the critical value, the enhancement factor of E x initially increases and reaches maximum at θ = 46° and then decreases. Interestingly, the intensity of SERS signal measured with the p-polarized laser beam presents an angle dependence similar to that for the enhancement factor of E x. The only difference is that the angle corresponding to the strongest SERS signal is θ = 53°, larger than the calculated maximum angle of θ = 46°. This may be attributed to two reasons: the first reason is that the NPG film is infinite in the xy-plane but very limited in the z direction, making its lateral LSPR effect quite different from the vertical LSPR effect, i.e., more “hot spots” have been utilized in the lateral direction than in the vertical direction [17]; the second reason is that at a large incident angle PSPR of the NPG film may be excited to yield a contribution to the measured SERS signal. Considering that the resonance angle of PSPR is quite narrow (referring to Fig. 7(a)), we believe that the two mechanisms work together in this case. The SERS signal excited with normal incidence of the same laser beam was also obtained for comparison (note that the laser power is 10 mw and the diameter of laser spot is 1mm). As shown in Fig. 6 , the TIR-SERS signal excited at θ = 53° with the p-polarized laser beam is 2 times stronger than that excited with normal incidence of the same laser beam. The comparison indicates that the TIR method can lead to at least twofold increase in the SERS enhancement factor relative to the conventional normal incidence.
Fig. 6 TIR-SERS spectrum for R6G molecules adsorbed in the NPG film (the spectrum excited by normal incidence was also shown for comparison)
Near infrared excitation is preferred for SERS measurement to avoid unwanted background noise from fluorescent compounds and photodegradation; this is especially the case in light of biosample analysis. Here a 785 nm laser was used for excitation of TIR-SERS signal from Nile Blue (NB) molecules adsorbed in the NPG film from an aqueous solution of 1μM dye. The first TIR-SERS measurement was carried out by setting the laser beam in the s-polarization state. No SERS signal was detected at any incident angle beyond the critical value, implying the weak LSPR effect of NPG film at 785 nm wavelength. After resetting the laser beam to p-polarization state, the TIR-SERS signal of NB was observed in the case of θ ranging from 39° to 44°, and the signal intensity reaches maximum at θ = 40º. This result is attributed to the PSPR effect of the NPG film used. Our previous work has demonstrated that a room-temperature dealloyed NPG film allows to excite the PSPR mode in the wavelength range from 600 nm to 800 nm using the same experimental setup [13]. The excitation of PSPR mode at the NPG/air interface can generate a greatly enhanced electric field at the interface to interact with the molecules adsorbed on the external surface [20]. In addition, the evanescent field confined inside the NPG film enables to interact with those molecules adsorbed on the internal surface of NPG. As a result, the PSPR effect of NPG films can be used to achieve a large SERS enhancement factor. On the other hand, a concentrated solution (10 μM NB) was prepared for sufficient adsorption of dye molecules on a 50-nm-thick dense gold film, the SERS signal of NB was not detected even if the PSPR mode of the dense gold film was excited at 785 nm [21]. The comparison indicates that the PSPR effect of NPG can lead to a much higher SERS sensitivity than that of a dense gold layer. The SERS spectrum for NB molecules adsorbed in the NPG film was also measured with the normal incidence of 785 nm laser beam. As shown in Fig. 7(b) , no SERS signal was detected in this case. The findings offer further evidence that the LSPR effect of NPG is very weak at 785 nm wavelength.
Fig. 7 (a) SERS spectra excited at different TIR angles using the p-polarized laser beam of 785 nm wavelength (the PSPR mode was excited at 40° TIR angle, making the SERS signal stronger); (b) Comparison between the PSPR-excited SERS signal and that excited with a free-space laser beam of same wavelength and equal power.
5. Conclusions
Ultrathin and small-pore NPG films with the two interesting plasmonic properties (LSPR and PSPR) were prepared by a combination of sputtering deposition and low-temperature chemical dealloying. The LSPR resonance wavelength of the NPG film surrounded with an aqueous solution was determined by broadband slab waveguide spectroscopy and the LSPR sensitivity to the surrounding refractive index was achieved to be 556 nm/RIU. The NPG films were used for TIR-Raman applications based on the LSPR effect at 532 nm excitation wavelength and the PSPR effect at 785 nm excitation wavelength. The experimental results indicate that the TIR-Raman signal of R6G excited at 532 nm wavelength is highly dependent on the incident angle and the angle dependence of the signal intensity measured with the s-polarized laser beam is quite different from that with the p-polarized beam. The reason for this difference was theoretically analyzed. The findings reveal that at 532 nm excitation wavelength the TIR method can make the SERS enhancement factor of NPG film increase at least 2 times relative to the conventional normal incidence. Using a 785 nm p-polarized laser beam for excitation, the TIR-SERS signal of NB can be observed only in the case of θ ranging from 39° to 44° and the signal reaches maximum at θ = 40º. This result is attributed to the PSPR effect of the NPG film. At 785 nm wavelength the LSPR effect of NGP film is too weak to make a detectable Raman signal. The work demonstrated that the TIR-Raman technique is advantageous over the conventional laser beam excitation method for the SERS application of NPG films.
Acknowledgments
This work was supported by the National Key Basic Research Program of China (No. 2015CB352100), the National Natural Science Foundation of China (No. 61377064), the Major National Scientific Instrument and Equipment Development Project of China (No. 2011YQ0301240802), and the Research Equipment Development Project of Chinese Academy of Sciences (No. YZ201508).
References and links
1. D. Garoli, G. Ruffato, P. Zilio, E. Calandrini, F. D. Angelis, F. Romanato, and S. Cattarin, “Nanoporous gold leaves: preparation, optical characterization and plasmonic behavior in the visible and mid-infrared spectral regions,” Opt. Mater. Express 5(10), 2246–2256 (2015). [CrossRef]
2. S. O. Kucheyev, J. R. Hayes, J. Biener, T. Huser, C. E. Talley, and A. V. Hamza, “Surface-enhanced Raman scattering on nanoporous Au,” Appl. Phys. Lett. 89(5), 053102 (2006). [CrossRef]
3. G. Ruffato, F. Romanato, D. Garoli, and S. Cattarin, “Nanoporous gold plasmonic structures for sensing applications,” Opt. Express 19(14), 13164–13170 (2011). [CrossRef] [PubMed]
4. F. Yu, S. Ahl, A. M. Caminade, J. P. Majoral, W. Knoll, and J. Erlebacher, “Simultaneous excitation of propagating and localized surface plasmon resonance in nanoporous gold membranes,” Anal. Chem. 78(20), 7346–7350 (2006). [CrossRef] [PubMed]
5. Y. Jiao, J. D. Ryckman, D. S. Koktysh, and S. M. Weiss, “Controlling surface enhanced Raman scattering using grating-type patterned nanoporous gold substrates,” Opt. Mater. Express 3(8), 1137–1148 (2013). [CrossRef]
6. D. Kramer, R. N. Viswanath, and J. Weissmüller, “Surface-stress induced macroscopic bending of nanoporous gold cantilevers,” Nano Lett. 4(5), 793–796 (2004). [CrossRef]
7. L. H. Qian and M. W. Chen, “Ultrafine nanoporous gold by low-temperature dealloying and kinetics,” Appl. Phys. Lett. 91(8), 083105 (2007). [CrossRef]
8. L. H. Qian, X. Q. Yan, T. Fujita, A. Inoue, and M. W. Chen, “Surface enhanced Raman scattering of nanoporous gold: Smaller pore sizes stronger enhancements,” Appl. Phys. Lett. 90(15), 153120 (2007). [CrossRef]
9. X. Y. Lang, L. Y. Chen, P. F. Guan, T. Fujita, and M. W. Chen, “Geometric effect on surface enhanced Raman scattering of nanoporous gold: Improving Raman scattering by tailoring ligament and nanopore ratios,” Appl. Phys. Lett. 94(21), 213109 (2009). [CrossRef]
10. L. Zhang, X. Lang, A. Hirata, and M. Chen, “Wrinkled nanoporous gold films with ultrahigh surface-enhanced Raman scattering enhancement,” ACS Nano 5(6), 4407–4413 (2011). [CrossRef] [PubMed]
11. H. Liu, L. Zhang, X. Lang, Y. Yamaguchi, H. Iwasaki, Y. Inouye, Q. Xue, and M. Chen, “Single molecule detection from a large-scale SERS-active Au₇₉Ag₂₁ substrate,” Sci. Rep. 1(1), 112 (2011). [PubMed]
12. Y. Jiao, J. D. Ryckman, P. N. Ciesielski, C. A. Escobar, G. K. Jennings, and S. M. Weiss, “Patterned nanoporous gold as an effective SERS template,” Nanotechnology 22(29), 295302 (2011). [CrossRef] [PubMed]
13. Z. Zhang, D. F. Lu, and Z. M. Qi, “Surface plasmon resonance sensing properties of nanoporous gold thin films,” Wuli Huaxue Xuebao 29(4), 867–873 (2013).
14. T. Ikeshoji, Y. Ono, and T. Mizuno, “Total reflection Raman Spectra; Raman scatting due to the evanescent wave in total reflection,” Appl. Opt. 12(10), 2236–2237 (1973). [CrossRef] [PubMed]
15. R. Iwamoto, M. Miya, K. Ohta, and S. Mima, “Total internal reflection Raman spectroscopy,” J. Chem. Phys. 74(9), 4780–4790 (1981). [CrossRef]
16. Y. Liu, S. Xu, H. Li, X. Jian, and W. Xu, “Localized and propagating surface plasmon co-enhanced Raman spectroscopy based on evanescent field excitation,” Chem. Commun. (Camb.) 47(13), 3784–3786 (2011). [CrossRef] [PubMed]
17. S. X. Huo, Q. Liu, S. H. Cao, W. P. Cai, L. Y. Meng, K. X. Xie, Y. Y. Zhai, C. Zong, Z. L. Yang, B. Ren, and Y. Q. Li, “Surface plasmon-coupled directional enhanced Raman scatting by means of the reverse Kretschmann configuration,” J. Phys. Chem. Lett. 6(11), 2015–2019 (2015). [CrossRef] [PubMed]
18. Z. M. Qi, S. Xia, and H. Zou, “Slow spontaneous transformation of the morphology of ultrathin gold films characterized by localized surface plasmon resonance spectroscopy,” Nanotechnology 20(25), 255702 (2009). [CrossRef] [PubMed]
19. X. Lang, L. Qian, P. Guan, J. Zi, and M. Chen, “Localized surface plasmon resonance of nanoporous gold,” Appl. Phys. Lett. 98(9), 093701 (2011). [CrossRef]
20. J. Zhang, L. Zhang, and W. Xu, “Surface plasmon polaritons: physics and applications,” J. Phys. D Appl. Phys. 45(11), 1311–1318 (2012). [CrossRef]
21. Q. Zhao, D. F. Lu, D. L. Liu, C. Chen, D. B. Hu, and Z. M. Qi, “Study of total internal reflection SERS based on self-assembled gold nanoparticle monolayer film,” Wuli Huaxue Xuebao 30(7), 1201–1207 (2014).
