8934
Langmuir 2008, 24, 8934-8938
Enhancement at the Junction of Silver Nanorods Geun Hoi Gu and Jung Sang Suh* Nanomaterials Laboratory, Department of Chemistry, Seoul National UniVersity, Kwanak-ro 599, Kwanak-gu, Seoul 151-747, Republic of Korea ReceiVed March 18, 2008. ReVised Manuscript ReceiVed May 9, 2008 The enhancement of surface enhanced Raman scattering (SERS) at the junction of linearly joined silver nanorods (31 nm in diameter) deposited in the pores of anodic aluminum oxide templates was studied systematically by excitation with a 632.8 nm laser line. The single and joined silver nanorod arrays showed a similar extinction spectrum when their length was the same. Maximum enhancement was observed from the junction system of two nanorods of the same size with a total length of 62 nm. This length also corresponded to the optimum length of single nanorods for SERS by excitation with a 632.8 nm laser line. The enhancement at the junction was ∼40 times higher than that of the 31 nm single nanorod, while it was 4 times higher than that of the 62 nm single nanorod. The enhancement factor at the junction after oxide removal was ∼3.9 × 109.
Introduction Surface-enhanced Raman scattering (SERS) is emerging as a probing technique for biosensors due to its high sensitivity.1–8 SERS has the sensitivity to detect even single molecules.9–15 For the detection of single molecules, a very high surface enhancement up to 1014-1015 may be required. It is known that a very strong enhancement is obtained from particular sites, the so-called hot spots, which may be junctions between nanoparticles.1,16–19 The simplest junction system may be a joined system of two nanoparticles of the same size. Although the enhancement of this simplest junction system has been studied intensively theoretically and experimentally, the junction factor that is defined as the ratio of SERS intensity measured from a joined system of two nanoparticles to that measured from a single nanoparticle * Author to whom correspondence should be addressed. E-mail:
[email protected]. (1) Xu, H.; Bjerneld, E. J.; Kall, M.; Borjesson, L. Phys. ReV. Lett. 1999, 83, 4357. (2) Cao, Y. C.; Jin, R.; Mirkin, C. A. Science (Washington, DC, U.S.) 2002, 297, 1536. (3) Shafer-Peltier, K. E.; Haynes, C. L.; Glucksberg, M. R.; Van Duyne, R. P. J. Am. Chem. Soc. 2003, 125, 588. (4) Bell, S. E. J.; Sirmuthu, M. S. J. Am. Chem. Soc. 2006, 128, 15580. (5) Shanmukh, S.; Jones, L.; Driskell, J.; Zhao, Y.; Dluhy, R.; Tripp, R. A. Nano Lett. 2006, 6, 2630. (6) Habuchi, S.; Cotlet, M.; Gronheid, R.; Dirx, G.; Michiels, J.; Vanderleyden, J.; De Schryver, F. C.; Hofkens, J. J. Am. Chem. Soc. 2003, 125, 8446. (7) Grubisha, D. S.; Lipert, R. J.; Park, H.; Driskell, J.; Porter, M. D. Anal. Chem. 2003, 75, 5936. (8) Souza, G. R.; Levin, C. S.; Hajitou, A.; Pasqualini, R.; Arap, W.; Miller, J. H. Anal. Chem. 2006, 78, 6232. (9) Nie, S.; Emory, S. R. Science (Washington, DC, U.S.) 1997, 275, 1102. (10) Kneipp, K.; Kneipp, H.; Deinum, G.; Itzkan, I.; Dassari, R. R.; Feld, M. S. Appl. Spectrosc. 1998, 52, 175. (11) Etchegoin, P.; Mather, R. C.; Cohen, L. F.; Hartigan, H.; Brown, R. J. C.; Milton, M. J. T.; Gallop, J. C. Chem. Phys. Lett. 2003, 375, 84. (12) Graham, D.; Mallinder, B. J.; Smith, W. E. Angew. Chem., Int. Ed. 2000, 39, 1061. (13) Bizzarri, A. R.; Cannistraro, S. Appl. Spectrosc. 2002, 56, 1531. (14) Bjerneld, E. J.; Foldes-Papp, Z.; Kall, M.; Rigler, R. J. Phys. Chem. B 2002, 106, 1213. (15) Delfino, I.; Bizzarri, A. R.; Cannistraro, S. Biophys. Chem. 2005, 113, 4. (16) Talley, C. E.; Jackson, J. B.; Oubre, C.; Grady, N. K.; Hollars, C. W.; Lane, S. M.; Huser, T. R.; Nordlander, P.; Halas, N. J. Nano Lett. 2005, 5, 1569. (17) Imura, K.; Okamoto, H.; Hossain, M. K.; Kitajima, M. Nano Lett. 2006, 6, 2173. (18) Xu, H.; Aizpurua, J.; Kall, M.; Apell, P. Phys. ReV. E: Stat., Nonlinear, Soft Matter Phys. 2000, 62, 4318. (19) Johansson, P.; Xu, H.; Kall, M. Phys. ReV. B: Condens. Matter Mater. Phys. 2005, 72, 35427.
is not clear. The reported values are in a wide range between 2 and 20.16,20,21 This wide variation is probably due to uncontrolled experiments in the size of nanoparticles, number of molecules adsorbed, and adsorption sites. The resonance condition of the surface plasmon is critically affected by the size of nanoparticles.22 When two nanoparticles of the same size are joined, the total length doubles. Therefore, the resonance condition will be changed when two nanoparticles are joined. For a meaningful comparison, both single and joined nanoparticles must be in resonance condition for excitation with a laser line. Also, for a simple comparison, the number of molecules adsorbed on both single and joined nanoparticles must be the same, and for the joined nanoparticles, molecules must be adsorbed only at the junction. Silver nanowire arrays fabricated on anodic aluminum oxide (AAO) templates have been used as substrates for SERS,23–28 and it has been suggested that they are promising nanostructures for fabricating hot spots for SERS.28 In fabrication of nanowires using AAO templates, the diameter and length of the nanowires can be controlled easily. Recently, AAO templates have been used to fabricate molecular device arrays.29 In this fabrication, silver was deposited in the pores of AAO templates, molecules were self-assembled, and then silver was deposited again. In this case, molecules were adsorbed only at the junction. These metal-molecule-metal devices show a semiconductor behavior.29 A metal-molecules-metal system is fundamentally a junction system for SERS. Here, we fabricated well-controlled single and joined silver nanorod arrays for SERS using AAO templates and measured their optical properties and relative SERS intensities. (20) Svedberg, F.; Li, Z.; Xu, H.; Kall, H. Nano Lett. 2006, 6, 2639. (21) Zhou, Q.; Zhao, G.; Chao, Y.; Wu, Y. L.; Zheng, J. J. Phys. Chem. C 2007, 111, 1951. (22) McFarland, A. D.; Young, M. A.; Dieringer, J. A.; Duyne, R. P. V. J. Phys. Chem. B 2005, 109, 11279. (23) Joo, Y.; Suh, J. S. Bull. Korean Chem. Soc. 1995, 16, 808. (24) Suh, J. S.; Lee, J. S. Chem. Phys. Lett. 1997, 281, 384. (25) Ruan, C.; Eres, G.; Wang, W.; Zhang, Z.; Gu, B. Langmuir 2007, 23, 5757. (26) Broglin, B. L.; Andreu, A.; Dhussa, N.; Heath, J. A.; Gerst, J.; Dudley, B.; Holland, D.; El-Kouedi, M. Langmuir 2007, 23, 4563. (27) Wang, H. H.; Liu, C. Y.; Wu, S. B.; Peng, C. Y.; Chan, T. H.; Hsu, C. F.; Wang, J. K.; Wang, Y. L. AdV. Mater. 2006, 18, 491. (28) Lee, S. J.; Morrill, A. R.; Moskovits, M. J. Am. Chem. Soc. 2006, 128, 2200. (29) Kim, L.; Kim, J.; Gu, G. H.; Suh, J. S. Chem. Phys. Lett. 2006, 427, 137.
10.1021/la800845h CCC: $40.75 2008 American Chemical Society Published on Web 07/11/2008
Enhancement of Joined SilVer Nanorods
Figure 1. Schematics of fabrication of (top) Ag-SAM and (bottom) Ag-SAM-Ag systems using AAO templates: (a) deposition of silver in the pores, (b) self-assembling of molecules on the tip of silver nanorods deposited in the pores, (c) deposition of silver again, and (d) etching away of some oxide.
Experimental Procedures A highly ordered porous AAO template was fabricated by using a two-step anodization technique.30 Clean aluminum sheets (99.999%, 0.5 mm thickness, Goodfellow Ltd.) were anodized in 0.5 M sulfuric acid at 10 °C and at a constant applied voltage of 25 V for 20 h after electropolishing. The resultant aluminum oxide film was subsequently removed by dipping the anodized sheet into an aqueous mixture of phosphoric acid (6 wt %) and chromic acid (1.8 wt %) for 16 h at 60 °C. The second anodization was performed for 3 min under the same conditions. The thickness of the prepared film was ∼800 nm. Pore widening was carried out in 0.1 M phosphoric acid at 30 °C to reduce the thickness of the bottom barrier layer. Silver nanorods were deposited in the pores of the films by applying an ac voltage of 16.60 V with a frequency of 200 Hz in an ethanol solution containing 0.05 M AgNO3 (99.9999%, Aldrich) at 5 °C (see Figure 1). The AAO films that deposited silver nanorods were immersed in ethanol solution containing 1.0 × 10-3 M benzenethiol for 20 h, washed with ethanol, and then dried. For the joined system, Ag nanorods were deposited again after washing with ethanol. Finally, these films were dipped into a 0.1 M H3PO4 solution at 30 °C to remove some oxide of the AAO films, washed with ethanol, and then dried for Raman measurements. Raman spectra were observed by using a micro-Raman system. The laser power incident on the sample was 1 mW, and the acquisition time was 1 s. The UV-vis extinction spectra of the silver nanorod arrays fabricated on AAO templates were measured by a reflection method using a Cary Varian 300 biodiffuse reflectance kit. The fabricated templates and nanorods were analyzed by using a scanning electron microscope (Philips FEG XL (30 kV)).
Results and Discussion The SEM images of the silver nanorods deposited in the pores of the AAO templates prepared in sulfuric acid are shown in Figure 2. In Figure 2a, which is the top view observed after removing some oxide by etching in H3PO4 solution, silver nanorods are deposited in all pores. The surface is relatively unclean. This may be due to the fact that the oxide has not been etched away evenly. The diameter of the nanorods is very uniform and is ∼31 nm, which corresponds to that of the pores of AAO templates used. In the top view, the nanorods exhibit a perfect two-dimensional array with a hexagonal pattern. The distance between the nanorods is very uniform at 65 nm. The nanorod density is ∼2.7 × 1010 nanorods/cm2. In the side views, some nanorods are not visible. They might have slipped out from the pores during sampling since silver particles are seen in all the pores in the top view. The alignment of the nanorods is perfectly (30) Zong, R. L.; Zhou, J.; Li, Q.; Du, B.; Li, B.; Fu, M.; Qi, X. W.; Li, L. T.; Buddhudu, S. J. Phys. Chem. B 2004, 108, 16713.
Langmuir, Vol. 24, No. 16, 2008 8935
Figure 2. SEM images of silver nanorods deposited in the pores of AAO templates: (a) top view after etching away some oxide and (b-d) side views of silver nanorods. The average lengths of the Ag nanorods in panels b-d are 31, 62, and 62 nm, respectively, and the diameters are all 31 nm. The nanorods in panels b and c are single nanorods, while those in panel d are joined ones that are Ag(31 nm)-SAM-Ag(31 nm) systems.
vertical with respect to the surface of the template. The lengths of the nanorods in Figure 2b-d are ∼31, 62, and 62 nm, respectively. Although the nanorods in Figure 2d are joined ones that are prepared by deposition of Ag nanorods of 31 nm, then self-assembling the molecules, and deposition of Ag nanorods of 31 nm again, they look like single particles. For Raman measurements, several different samples were prepared. They are symbolized as follows: Ag31-SAM, Ag62SAM, Ag124-SAM, Ag31-SAM-Ag31, Ag42-SAM-Ag20, and Ag36-SAM-Ag36 systems. The schematics are shown in Figure 1. The Ag62-SAM system stands for Ag nanorods 62 nm in length that were deposited in the pores of an AAO template with the benzenethiol molecules self-assembled on the tip of the Ag nanorods. In an Ag31-SAM-Ag31 system, Ag nanorods 31 nm in length were deposited, molecules self-assembled, and the 31 nm Ag nanorods were deposited again. In an Ag-SAM system, molecules self-assembled at the tip of the nanorods that are shaped like a hemisphere. In an Ag-SAM-Ag system, molecules selfassembled at the tip of the nanorods, and then Ag was deposited again. Therefore, the number of molecules in all samples is almost the same. In the joint systems, molecules exist only at the junction, which is between the convex hemisphere of the tip of the bottom nanorod and the concave hemisphere of the tip of the upper nanorod. The bottom of the upper nanorod of a joined nanorod has a concave hemisphere since both single and joined nanorods have a convex hemisphere in the tip shape. The SEM images of some of these systems are shown in Figure 2. In the SEM images, there is no difference between Ag62-SAM and Ag31-SAMAg31 systems. However, in I-V measures, Ag nanowire-SAMAg nanowire devices fabricated on AAO templates showed a semiconductor behavior.29 This means that the two Ag nanowires are connected by the self-assembled molecules instead of being connected directly. The UV-vis extinction spectra of the single and joined silver nanorod arrays fabricated on AAO templates, measured by a reflection method, are shown in Figure 3. The diameter of all the single and joined nanorods is 31 nm. For single nanorods whose length is ∼31 nm and have almost a spherical shape, there is only a broadband whose maximum is at 390 nm. However, for single nanorods whose length is ∼62 nm, there are two bands at 377 and 631 nm. The former is due to the transverse mode of the surface plasmon, and the latter is due to the longitudinal mode. With increasing the length of single nanorods, the band of the
8936 Langmuir, Vol. 24, No. 16, 2008
Gu and Suh
Figure 4. SERS spectra of benzenethiol adsorbed on the tip of silver nanorods and at the junction of joined systems: (a) Ag(31 nm)-SAM, (b) Ag(40 nm)-SAM-Ag(22 nm), (c) Ag(31 nm)-SAM-Ag(31 nm), and (d) Ag(62 nm)-SAM. Table 1. Relative SERS Intensities of Several Systems
Figure 3. UV-vis extinction spectra of single and joined silver nanorods and deconvoluted spectra of them (blue and red lines): (a) Ag(31 nm)SAM, (b) Ag(31 nm)-SAM-Ag(20 nm), (c) Ag(62 nm)-SAM, (d) Ag(31 nm)-SAM-Ag(31 nm), (e) Ag(80 nm)-SAM, and (f) Ag(40 nm)-SAMAg(40 nm).
longitudinal mode is shifted to red, while that of the transverse mode is slightly blue-shifted. These behaviors are already known theoretically and experimentally.30 When the length of single and joined nanorods is the same, their extinction spectra are very similar to each other. For example, for the joined silver nanorods whose total length is ∼62 nm, there are two bands whose maxima are at 367 and 642 nm. With increasing the total length of the joined nanorods, the band of the longitudinal mode is also shifted to red, while that of the transverse mode is almost not shifted. By excitation with the laser line of 632.8 nm, the transverse mode of the single and joined nanorods may not contribute to the SERS intensity because the extinction maximum of this mode is far away from the laser line. The extinction maximum of the longitudinal mode is very close to the laser line of 632.8 nm when the length of the single and joined nanorods is ∼62 nm. Therefore, this length may correspond to the optimum length for SERS by excitation with the 632.8 nm laser line.31 The diameter of silver nanorods reported in ref 31is 37 nm, which is larger than that of silver nanorods studied here (31 nm). However, the transverse mode of silver nanorods whose diameter is 37 nm also did not contribute to the SERS intensity measured by excitation with the 632.8 nm laser line because the maximum of the transverse mode is far away from the laser line. It is at nearly 370 nm. Although molecules exist at the junction of the joined nanorods, the single and joined nanorods have a similar extinction spectrum when their length is the same. This may be due to a tunneling between two nanoparticles in the joined nanorods. The molecular length of benzenethiol used in the formation of SAM is ∼0.6 nm,32 which may correspond to the gap between two nanoparticles in the joined nanorods. This gap is narrow enough for tunneling to take place between two nanoparticles in the joined nanorods. By tunneling, two nanoparticles in the joined nanorods may act optically like a single nanoparticle. This should be studied in detail. (31) Gu, G. H.; Kim, J. R.; Kim, L.; Suh, J. S. J. Phys. Chem. C 2007, 111, 7906. (32) Ruths, M. Langmuir 2003, 19, 6788.
single nanorods
relative intensity
joined nanorods
relative intensity
Ag31-SAM Ag42-SAM Ag62-SAM Ag124-SAM
1 2 10 0.5
Ag31-SAM-Ag20 Ag31-SAM-Ag31 Ag40-SAM-Ag22 Ag36-SAM-Ag36 Ag62-SAM-Ag62
11 40 15 12 1
The SERS spectra of several different systems are shown in Figure 4. They were measured by excitation with a 632.8 nm laser line after removing some oxide by etching in H3PO4 solution. The spectra are all the same except for their relative intensity. The Raman peaks correspond to the modes of the benzene ring.33 For example, the peak at 1573 cm-1 is due to ν8a, and that at 1000 cm-1 is due to ν12. The peak at 417 cm-1 is due to ν7a and contributions from the C-S stretching vibration (νCS). SERS spectra were measured using a micro-Raman system in the normal way by laser beam irradiation perpendicular to the surface of the sample template placed on the sample stage. The alignment of the Ag nanorods is perfectly vertical with respect to the surface of the template. In this case, the polarization direction of the laser beam is parallel to the short axis of the nanorods, while it is perpendicular to the long axis. Therefore, in this geometry, the transverse mode of the surface plasmon can be excited by irradiation, but the longitudinal mode cannot be excited. However, for our silver nanorods, the contribution of the transverse mode may be very small or negligible because the maximum of the transverse mode is far away from the laser line as mentioned previously. Further, the nanorods are well-separated from each other, and the gap between them is ∼34 nm. Therefore, there is almost no enhancement from a lateral interaction like that in bundling. For the micro-Raman system, a lens having a very short focal length was used to focus the laser beam and to simultaneously detect the Raman signals. In this case, the part of the laser beam at the edge refracts with a wide angle. The part of the refracted laser beam irradiates the nanorods at some angle. Therefore, the longitudinal mode of the surface plasmon of the nanorods can be excited since the refracted laser beam has a polarization component parallel to the long axis of the nanorods. Evidence of this argument was discussed in detail in our previous paper.31 The relative intensity of SERS spectra measured from the different systems is summarized in Table 1. Here, we set the relative intensity of the Ag62-SAM system as 10. It is known that a 62 nm Ag nanorod is the optimum length for SERS by (33) Szafranski, C. A.; Tanner, W.; Laibinis, P. E.; Garrell, R. L. Langmuir 1998, 14, 3570.
Enhancement of Joined SilVer Nanorods
excitation with a 632.8 nm laser line.31 When the length of nanorods is shorter or longer than 62 nm, the intensity decreases significantly. For example, the relative intensity of the Ag31SAM system is ∼1, while that of the Ag42-SAM system is ∼2. For the Ag124-SAM system, it is smaller than 1. However, when molecules exist at a junction of two nanorods, the SERS intensity increases greatly. The relative intensity of the Ag31-SAM-Ag31 joined system is ∼40. It is ∼4 times higher as compared to that of the Ag62-SAM single nanorod system, while it is 40 times higher as compared to that of the Ag31-SAM system. The total length of the Ag31-SAM-Ag31 joined system is 62 nm, which corresponds to the optimum length for SERS by excitation with a 632.8 nm laser line.31 The total length of a joined system affects the enhancement critically. For example, the relative intensity of the Ag31-SAM-Ag20 joined system, whose total length is shorter than 62 nm, is ∼11. This value is much smaller than that of the Ag31-SAM-Ag31 system, 40. When the total length of the joined system becomes longer than 62 nm, the relative intensity decreases significantly with increasing total length. The relative intensity of the Ag36-SAM-Ag36 system is ∼12, while that of the Ag62-SAM-Ag62 system is ∼1, which is somewhat higher than that of Ag124-SAM but much lower than that of the Ag62-SAM single nanorod system, 10. If the SERS intensity of a joined system is determined by the length of each nanoparticle, the relative intensity of the Ag62-SAMAg62 joined system should be at least similar to that of the Ag62-SAM system. However, the former is much weaker than the latter. The Ag31-SAM-Ag31 joined system shows the highest intensity. The total length, 62 nm, corresponds to the optimum length for SERS by excitation with a 632.8 nm laser line.31 Therefore, it is concluded that the resonance condition for the joined system is determined by the total length of two joined nanoparticles, not each nanoparticle or nanorod. This means that the optimum length of the joined systems is the same as that of the single systems, ∼62 nm by excitation with a 632.8 nm laser line. This agrees well with the result of the UV-vis extinction spectra. Another interesting point is that the relative intensity of the Ag42-SAM-Ag20 system is ∼15, which is considerably smaller than that of the Ag31-SAM-Ag31 system, 40. Their total length is the same (62 nm), which corresponds to the optimum length. However, their location of junction is different from each other. For the Ag31-SAM-Ag31 system, the junction exists in the middle, while for the Ag42-SAM-Ag20 system, it is off from the middle. This may mean that the enhancement is affected by the location of molecules existing in the joined systems. The enhancement at the junction of two joined nanoparticles was studied theoretically and experimentally. However, these studies did not seem to define or control the size of the nanoparticles. According to our results, the junction factor is critically affected by the size of the nanoparticles. The relative intensity of the Ag31-SAM-Ag31 system is 40, while that of the Ag31-SAM system is 1. In comparison to these two systems, the junction factor is ∼40. However, it is ∼7.5 in comparison to the Ag42-SAM-Ag20 and Ag42-SAM systems since the relative intensity of the Ag42-SAM-Ag20 and Ag42-SAM systems is ∼15 and 2, respectively. In these samples, the total length of the junction system is 62 nm, which is the optimum length for enhancement under our experimental conditions. When the total length of the junction systems is off from the resonance condition, the intensity is decreased dramatically, and the junction factor can become
