ANALYTICAL SCIENCES FEBRUARY 2009, VOL.

25 2009 The Japan Society for Analytical Chemistry

255

Comparison of Plasmonic Sensing between Polymer- and Silica-coated Gold Nanorods

Naoya OMURA,* Ichiro UECHI,** and Sunao YAMADA*,**,***
* Department of Materials Physics and Chemistry, Graduate School of Engineering, Kyushu University, Moto-oka, Nishi, Fukuoka 8190395, Japan **Department of Applied Chemistry, Kyushu University, Moto-oka, Nishi, Fukuoka 8190395, Japan ***Center for Future Chemistry, Kyushu University, Moto-oka, Nishi, Fukuoka 8190395, Japan

Gold nanorods (AuNRs), rod-like gold nanoparticles, show strong plasmon bands in the near-infrared region, based on the longitudinal oscillation mode of surface plasmons. In this research, we have coated AuNRs with silica or poly(sodium 4styrenesulfonate) (PSS) in order to evaluate the effects of surface modifications as a sensing probe material of AuNR. Multilayered assemblies of the PSS-coated AuNRs have been successfully fabricated on the glass substrate by electrostatic layer-by-layer adsorption, and their plasmonic band profiles have been analyzed as a function of refractive index. The profiles showed satisfactory linear responses with the refractive indices. The effect of the thickness of silica layer on the shift of plasmon band with refractive index of the surrounding media has been investigated, along with, the effect of silica coating on the thermal stability of AuNR. The thermal stability of the silica-coated AuNR has also been investigated. The two types of the above-described coating approaches have been compared. (Received August 19, 2008; Accepted September 17, 2008; Published February 10, 2009)

Introduction
Gold nanoparticles with the diameters of the order of one to one hundred nanometers have a fascinating optical property that offers potential applications to analytical fields.13 This optical property originates from localized surface plasmon resonance (LSPR), leading to the appearance of characteristic absorption peaks as plasmon bands, in the visible to near-infrared region. The absorption band profile strongly depends on the size and shape of the nanoparticles and is sensitive to the dielectric constant of the surrounding medium. As to the shape of the gold nanoparticle, a gold nanoparticle (AuNR), rod-shaped gold nanoparticle, has two distinct plasmon bands, one appearing in the near infrared region (longitudinal plasmon oscillation mode: SPL) and the other in the visible region (transverse plasmon oscillation mode: SPT). It has also been well recognized experimentally and theoretically that the SPL band is much stronger and much more sensitive to the dielectric constant (refractive index) of the surrounding medium as compared with the SPT band.4 In addition, the SPL band is superior for analytical application in biomaterials, such as in-vivo imaging of biotissues,5 since it appears in the near-infrared region whose light has high permeability to biomaterials.6 Due to the recent progress on the preparation methods of AuNRs, such as electrochemical method,7 template method,8 seed-mediated method,9 photochemical methods,10 and the combination of chemical reduction method and photoirradiation method,11,12 studies on the optical and analytical applications of To whom correspondence should be addressed. E-mail: sunaotcm@mbox.nc.kyushu-u.ac.jp

AuNRs have been growing year to year. However, the problem of the dispersibility of AuNRs still restricts expansions of the utilization of AuNRs. In all preparation methods reported so far, cetyltrimethylammonium bromide (CTAB) has been essential for the generation of AuNRs, and thus AuNRs have been capped with CTAB molecules. Although capping with CTAB is important to produce and keep the rod-shape, it is still not enough to prevent flocculation of AuNRs in the colloidal solution. In addition, CTAB, a cationic surfactant, has cytotoxic activity. These aspects lead to the need for surface modification of AuNR with appropriate chemical species. It is also quite important to keep the spectral characteristics of AuNR itself (vide infra), which maybe degraded by proximity as well as flocculation of AuNRs. From these viewpoints, we have carried out two kinds of surface modification of AuNR: polymer- and silica-coating, by slight modifications of the previous methods of the electrostatic adsorption technique13 and of silica-coating.14 Colloidal stability, optical properties, and thermal stability of the as-prepared AuNRs were investigated and are compared with each other.

Experimental
Reagents and chemicals Milli-Q grade water was used to make up all solutions. The colloidal solution of gold (AuNRs) with the aspect ratio of ca. 4 was supplied from Dainippon Toryo Co., Ltd. The solution showed absorption maxima at 520 and 900 nm, corresponding to transverse and longitudinal oscillation modes, respectively. Poly(sodium 4-styrenesulfonate) (PSS, Mw 70000, Aldrich), poly(allylamine hydrochloride) (PAH, Mw 15000, Aldrich),

256

ANALYTICAL SCIENCES FEBRUARY 2009, VOL. 25

Fig. 2 Schematic representation for the procedures of silica-coated gold nanorods.

Fig. 1 Schematic representation for the procedures of polymercoated gold nanorods and subsequent layer-by-layer deposition on glass substrates.

poly(vinylpyrrolidone) (PVP, Mw 55000, Aldrich), and tetraethyl orthosilicate (TEOS, Wako Pure Chemicals) were used without further purification. Deionized water (resistivity = 18.1 MW) was obtained from a Millipore water purification system. Apparatus A JASCO V-570 spectrophotometer was used to obtain absorption spectra of the samples. Transmission electron microscope (TEM) images were measured with a JEOL JEM200CX microscope. Zeta-potential measurements of polymercoated gold nanorods in water were evaluated using a MALVERN Zetasizer Nano ZS. Annealing of the samples were carried out using an electrical muffle furnace (KDF 007EX). Preparation of PSS-coated AuNR on polyion-modified glass substrate Before performing the layer-by-layer adsorption process, we centrifuged the solution of colloidal AuNRs twice (7000g, 10 min) to remove excess CTAB. The preparation procedure of PSS-coated AuNR is illustrated in Fig. 1. As the first step, a PSS-modified AuNR was prepared as follows. First, 5 mL of the solution of PSS (2 mg/mL, 6 mM NaCl) was added into the solution of AuNRs (5 mL), and the mixture was stirred for 3 h at room temperature. Then, the solution was centrifuged two times (7000g, 10 min) to remove residual PSS, and redispersed into a 1 mL aliquot of water, giving PSS-modified AuNR, denoted as: PSS/AuNR. The zeta potential of PSS/AuNR is negatively charged (39.5 mV). As the second step, a polyion-modified glass substrate was prepared as follows. Before performing the layer-by-layer deposition procedure, we cleaned the glass substrate (dimension: ca. 3 1 cm) by treating with a mixed solution of aqueous hydrogen peroxide (H2O2, 31%), ammonia (28%), and H2O (1:1:1 v/v) at 70C for 30 min, followed by rinsing with water, and drying with the flow of nitrogen gas. This treatment has made the glass surface hydrophilic. The glass substrate was stored in water. The resultant glass substrate was used for subsequent layer-by-layer deposition procedures, as illustrated in Fig. 1. First, the above-described glass substrate was immersed into the solution of PAH (2 mg/mL) for 20 min at room temperature, giving the PAH-modified glass substrate denoted as: PAH/glass. Then, PAH/glass was immersed into the solution of PSS (2 mg/ mL) for 20 min at room temperature, to make the negative surface, denoted as: PSS/PAH/glass. Finally, PSS/PAH/glass was

immersed again into the above-described solution of PAH to superimpose a positive charge at the outermost surface, giving the glass substrate modified with PAH/PSS/PAH as the precursor film. As the third step, the PSS-modified AuNR, PSS/AuNR, was deposited on the polyion-modified glass substrate, PAH/PSS/ PAH/glass, by immersing PAH/PSS/PAH/glass into the solution of PSS/AuNR (5 mL) for 24 h at room temperature. Accordingly, we have obtained the AuNR-modified glass substrate, denoted as: PSS/AuNR/PAH/PSS/PAH/glass. Preparation of silica-coated AuNR The preparation procedure of silica-coated AuNR is illustrated in Fig. 2. First, 5 mL of PAH solution (2 mg/mL, 6 mM NaCl) was added into 5 mL of the above-described PSS/AuNR solution, followed by stirred for 3 h at room temperature. After that, the mixed solution was centrifuged twice (7000g, 10 min) to remove residual PAH, and the remaining material was redispersed into 5 mL of water, giving PAH-coated PSS/ (AuNR), denoted as: PAH/PSS/AuNR. Next, 5 mL of poly(vinylpyrrolidone) (PVP: 4 mg/mL) solution was added into 5 mL of PAH/PSS/AuNR solution and the mixture was stirred overnight. Then, the solution was centrifuged once (7000g, 10 min) to remove residual PVP, and the remaining material was redispersed into 0.2 mL of water, giving PVP-coated PAH/PSS/AuNR as: PVP/PAH/PSS/AuNR. Finally, 0.2 mL of PVP/PAH/PSS/AuNR solution, 0.46 mL of H2O, 1 mL of isopropanol solution for hydrolysis (containing 55 mL of 30% ammonia), and finally 0.4 mL of isopropanol solution containing tetraethoxysilane (6.4 mM) were added in this order, and the reaction mixture was stirred for 2 h at room temperature. Then, the reaction mixture was centrifuged (7000g, 10 min), and the remaining material was subsequently redispersed into 1 mL isopropanol, giving the solution containing silica-coated AuNR. All layer-by-layer reactions and deposition procedures were carried out at room temperature, unless otherwise noted.

Results and Discussion

During the applications of AuNRs as spectrometric probe materials, surface modifications using appropriate chemical species are important to make the most of the optical characteristics of AuNRs. However, the dispersibility of AuNRs also should be taken into consideration for practical applications. In the present study, we have tried two approaches of surface modification: polyion- and silica-coating. Figure 3a shows absorption spectra of {PSS(AuNR)}n/PAH/ PSS/PAH/glass (n = 1, 2, 3). The characteristic plasmon bands of AuNRs appeared and increased as the number of layer-bylayer adsorption treatments increased (see Fig. 1). Some fluctuating of baseline was unavoidable. Figure 3b shows spectral shifts of SPL bands in different solvents; their refractive

ANALYTICAL SCIENCES FEBRUARY 2009, VOL. 25

257

Fig. 5 TEM images of 5 (a) and 10 (b) layers of {PSS(AuNR)}n/ PAH/PSS/PAH/glass.

Fig. 3 Absorption spectra in air (a) and absorption peaks in different solvents (b) for {PSS(AuNR)}n/PAH/PSS/PAH/glass (n = 1, 2, 3). Fig. 6 Absorption spectral changes of PSS(AuNR)/PAH/PSS/PAH/ glass after annealing at several temperatures for 60 min: a, 25; b, 50; c, 100; d, 150; e, 200; f, 280; g, 300C.

Fig. 4 Absorption peaks in air (a) and photographs (b) of {PSS(AuNR)}n/PAH/PSS/PAH/glass (n = 1 10) in air.

indices are: 1.33 (water), 1.36 (acetone), 1.43 (cyclohexane), 1.45 (chloroform). It is clear that the SPL bands show red-shifts with increasing refractive index, as has been verified experimentally15 and theoretically.1618 As shown in Fig. 4a, subsequent layer-by-layer deposition was successful up to ten cycles, in which the peak intensity of the SPL band increased roughly linearly with increasing the number of deposition cycles. Figure 4b shows photographs of the AuNR-deposited glass plates from 1 to 10 deposition cycles. The

degree of deposition was fairly homogeneous over the glass plate. As can be recognized from the SEM images shown in Fig. 5, PSS/AuNR is distributed randomly on polyion-modified glass substrate without appreciable tight aggregation. This is quite important in processing the spectral sharpness of the SPL AuNR. The thermal stability of PSS/AuNR was investigated by annealing it in an electrical furnace. Figure 6 shows spectral changes of PSS/AuNR/PAH/PSS/PAH/glass annealed at different temperatures (50 300C) for 1 h. The SPL band shows degradation and blue-shift with increasing the annealing temperature above 150C. In addition, the SPT band around 520 nm gradually enlarged. As AuNR, the SPL band is known to show blue-shifting with decreasing the aspect ratio.16,19,20 Accordingly, these spectral changes are due to the shape transformation of AuNR into thermodynamically more stable ellipsoidal particles and further into spherical particles. Thus, the surface modification with polyion is not effective in terms of keeping thermal stability. As an alternative method, silica-coating of AuNR was successfully carried out; the degree of silica layer was changed by changing the concentration of TEOS. Figure 7 shows TEM images of as-prepared silica-coated AuNR with different concentrations of TEOS. The thicknesses of silica layers are roughly homogeneous, and only one AuNR is incorporated in the silica layer. The thicknesses of the silica layers were 20, 10, and 3 nm, for the corresponding TEOS concentrations of 6.4, 0.64, and 0.064 mM, respectively. In order to investigate the spectral responsibility of silicacoated AuNRs, we dispersed them on the surfaces of glass plates by casting 1 mL of the colloidal solution of silica-coated AuNR,

258

ANALYTICAL SCIENCES FEBRUARY 2009, VOL. 25

Fig. 8 Absorption peaks of AuNR after annealing at different temperatures: a, 100; s, 150; f, 200C.

structures of the silica layers. In any case, the thermal stability of AuNR was better for the silica-coated AuNR as compared with the PSS-coated AuNR, though the spectroscopic sensitivity to the refractive index is relatively poorer.

Conclusions
Alternative processes for surface modifications of AuNRs have been achieved. In the electrostatic layer-by-layer approach, the PSS-modification has substantially improved the dispersibility of AuNR, and has enabled multilayer assembling on the glass substrate. In addition, the deposited AuNR showed a clear response to the refractive index of the surrounding medium. In the silica-coating, on the other hand, the spectral response to the refractive index was possible for less than 10 nm of layer thickness, yet the thermal stability was somewhat improved. In terms of surface modification, increases in the spectral sensitivity and in the thermal stability seem to be contradictory issues. However, a skillful surface modification must overcome the above problem. Further studies for such surface modifications are now in progress.

Fig. 7 (a) (d) TEM images of silica-coated AuNRs prepared by different concentrations of TEOS: (a) 6.4, (b) 0.64, (c) and (d) 0.064 mM, and (e) absorption peaks of AuNR in solutions as a function of refractive indices: a, 6.4; S, 0.64; d, 0.064 mM.

followed by drying in air. Figure 7b shows the relationships between the peak positions of the SPL bands and the refractive indices of the solvents. It is clear that the peak shift is observed for thinner- and middle-silica layers and not for the thicker-silica layer. There seem to be two reasons for the peak shift. One is that the plasmonic electric field based on LSPR experiences the solvent molecules present outside of the silica layer of less than ~10 nm, while the other reason is that the solvent molecules permeate into the silica layer when the silica layer is thinner than ~10 nm. As can also be recognized from Fig. 7(e), the peak shift fluctuated considerably with decreasing the thickness of silica layer, though the reason is not clear at this stage. In any case, the silica-coated AuNR can sense the refractive index of the solvent when the thickness is less than ~10 nm. The thermal stability of silica-coated AuNR was also investigated. No appreciable spectral change were observed for annealing at 100, 150, and 200C for 1 h. Moreover, the spectral shifts of the annealed sample showed similar tendencies, as shown in Fig. 8. In all samples, the SPL peak tended to show a slight red-shift with increasing refractive index. Thus, the annealing procedure certainly changes the microscopic

Acknowledgements
The present work is supported by the Special Project of Molecular and System Life Science in the Grant of Promotion of Novel Interdisciplinary Fields Based on Nanotechnology and Materials, and by Grant-in-Aid for Scientific Research (No. 19049012) on Priority Area Strong Photon-Molecule Coupling Fields (470), from the Ministry of Education, Culture, Spotrts, Science and Technology (MEXT) of Japan. The authors thank Prof. Y. Katayama for the use of MALVERN Zetasizer Nano ZS. The authors also thank Dainippon Toryo Co., Ltd. for the supply of AuNRs.

References
1. A. J. Haes and R. P. Van Duyne, Anal. Bioanal. Chem., 2004, 379, 920. 2. I. Uechi and S. Yamada, Anal. Bioanal. Chem., 2008, 391, 2411. 3. S. Kawata and H. Masuhara (ed.), Handai Nanophotonics, 2006, Elsevier, Amsterdam.

ANALYTICAL SCIENCES FEBRUARY 2009, VOL. 25

4. Y.-Y. Yu, S.-S. Chang, C.-L. Lee, and C. R. C. Wang, J. Phys. Chem. B, 1997, 101, 6661. 5. T. Niidome, Y. Akiyama, K. Shimoda, T. Kawano, T. Mori, Y. Katayama, and Y. Niidome, Small, 2008, 4, 1001. 6. J. F. Frangioni, Curr. Opin. Chem. Biol., 2003, 7, 626. 7. B. M. I.van der Zande, M. R. Bhmer, L. G. J. Fokkink, and C. Schnenberger, Langmuir, 2000, 16, 451. 8. N. R. Jana, L. Gearheart, and C. J. Murphy, J. Phys. Chem. B, 2001, 105, 4065. 9. F. Kim, J. H. Song, and P. Yang, J. Am. Chem. Soc., 2002, 124, 14316. 10. Y. Niidome, K. Nishioka, H. Kawasaki, and S. Yamada, Chem. Commun., 2003, 2376. 11. H. Takahashi, Y. Niidome, and S. Yamada, Chem. Commun., 2005, 2247. 12. K. Nishioka, Y. Niidome, and S. Yamada, Langmuir, 2007,

259
23, 10353. 13. G. Decher, Science, 1997, 277, 1232. 14. I. Pastoriza-Santos, J. Prez-Juste, and L. M. Liz-Marzn, Chem. Mater., 2006, 18, 2465. 15. S. Vial, I. Pastoriza-Santos, J. Prez-Juste, and L. M. LizMarzn, Langmuir, 2007, 23, 4606. 16. S. Link and M. A. El-Sayed, J. Phys. Chem. B, 2005, 109, 10531. 17. P. K. Jain, E. Susie, and M. A. El-Sayed, J. Phys. Chem. B, 2006, 110, 18243. 18. M. D. Malinsky, K. L. Kelly, G. C. Schatz, and R. P. Van Duyne, J. Am. Chem. Soc., 2001, 123, 1471. 19. M. B. Mohamed, K. Z. Ismail, S. Link, and M. A. El-Sayed, J. Phys. Chem. B, 1998, 102, 9370. 20. S.-S. Chang, C.-W. Shih, C.-D. Chen, W.-C Lai, and C. R. C. Wang, Langmuir, 1999, 15, 701.