A Tunable Plasmon Resonance in Gold Nanobelts - ACS Publications

ACS2GO © 2019. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to homescreen...
0 downloads 0 Views 4MB Size
LETTER pubs.acs.org/NanoLett

A Tunable Plasmon Resonance in Gold Nanobelts Lindsey J. E. Anderson,†,§ Courtney M. Payne,‡,§ Yu-Rong Zhen,†,§ Peter Nordlander,†,§ and Jason H. Hafner*,†,‡,§ †

Department of Physics and Astronomy, ‡Department of Chemistry, and §Laboratory for Nanophotonics, Rice University, 6100 Main Street, Houston, Texas 77005, United States

bS Supporting Information ABSTRACT: Plasmonic nanowires with sub-100-nm rectangular cross sections were found to exhibit a strong transverse plasmon peak at visible wavelengths. By correlating atomic force microscopy measurements of individual nanobelts with their dark-field scattering spectra, it is seen that the transverse peak tunes with cross-sectional aspect ratio. Simulations revealed that the scattering plasmonic modes are transverse antisymmetric excitations across the nanobelt width. Unlike larger diameter silver nanowires, these nanobelts exhibit sharp, tunable plasmon resonances similar to those of nanoparticles. KEYWORDS: Nanowire, nanobelt, nanoribbon, nanoparticle, localized surface plasmon resonance, surface plasmon polariton, plasmonics, nanophotonics, waveguide

N

oble metal nanostructures with subwavelength dimensions exhibit localized surface plasmon resonances when excited by light. These resonances lead to strong absorption, scattering, and near field enhancement at frequencies that depend on the nanoparticle size and shape.1 Over the past decade, a wide variety of plasmonic gold and silver structures have been fabricated to manipulate light at the nanometer scale where they are pursued for novel applications and basic research.2,3 For example, plasmon resonant optical absorption enables molecularly targeted heating for applications in nanomedicine such as photothermal therapy and drug delivery.4,5 Larger plasmonic nanostructures exhibit bright plasmon resonant scattering which may prove useful for microscopic imaging contrast and particle tracking.6,7 Nanoparticles with sharp tips and gaps between nanoparticles create regions of highly enhanced electromagnetic fields which amplify powerful spectroscopies for sensing applications.8 Plasmonic nanoparticles can be coupled to fluorescent dyes or quantum dots to manipulate radiative decay rates and directions, thus fundamentally manipulating optical emission.9,10 All of these applications rely to some extent on the ability to tune nanoparticle plasmon resonances to match a biological window of tissue transparency, to coincide with an emitter excitation, or simply to match a laser source. The capability to tune the plasmon resonance has played a crucial role in stimulating the current interest in plasmonics. Plasmonic effects are also studied in extended structures where they may be used to transport optical information at the nanometer scale. For example, light can be coupled into surface plasmon polaritons in two-dimensional films, and its propagation can be manipulated by nanostructures within the film.11,12 Recently there has been interest in plasmon propagation in onedimensional plasmonic nanowires.13 Here light can be coupled r 2011 American Chemical Society

into and out of the nanowires through symmetry broken sites such as tips, kinks, or nanoparticle nanowire junctions.14,15 The coupling between propagating nanowire plasmons and nearby quantum dots could enable the integration of plasmonic structures into nanometer scale optoelectronic devices.16 19 Nanowire propagating plasmon modes have been studied largely with silver nanowires that have cross sections of 75 260 nm diameter.14,20 22 While these efficiently couple, transport, and emit light, their large size leads to transport based on a mixture of higher order plasmonic modes and may limit the strength of interactions with nearby emitters and other plasmonic structures.23 Here we describe the optical properties of gold nanowires with rectangular cross sections referred to as gold nanobelts.24 The gold nanobelts described here have sub-100-nm cross-sectional dimensions and lengths greater than 10 μm. This structure combines the tunability and sharp resonances of subwavelength sized nanoparticles with the extended structure of nanowires. Our gold nanobelt synthesis followed a previously described method with some modifications.24 Briefly, HAuCl4 was added to an aqueous surfactant solution of 6.5:1 CTAB to SDS (cetyltrimethylammonium chloride to sodium dodecylsulfate). The gold solution was then reduced by ascorbic acid to initiate nanobelt growth and stored at 27 C for 12 h (see Supporting Information for details). During the growth reaction, the solution gradually became turbid and exhibited a reddish hue (Figure 1a). The color is typical of plasmon resonant gold nanoparticle solutions, and the turbidity was due in part to the formation of micrometer-scale surfactant crystals that are visible in an optical Received: September 5, 2011 Revised: October 4, 2011 Published: October 05, 2011 5034

dx.doi.org/10.1021/nl203085t | Nano Lett. 2011, 11, 5034–5037

Nano Letters

LETTER

Figure 2. Dark field micrographs (left) and corresponding single nanobelt spectra (right). The given aspect ratios were determined by atomic force microscopy. The blue spectra are polarized transverse to the nanobelt, and the red spectra are polarized parallel to the nanobelt.

Table 1. The Nanobelt Sizes and Spectral Line Width from the Data in Figure 2

Figure 1. Gold nanobelts. The suspensions have (a) a pinkish turbid appearance and (b) a weak plasmon resonant spectral absorption peak at 520 nm. Electron micrographs illustrate (c) the overall ribbon-like morphology and (d) high-resolution crystalline structure. (e) Electron diffraction confirms the single-crystalline gold structure.

microscope. The spectral extinction exhibits a weak band at 520 nm (Figure 1b) which is also consistent with the presence of gold nanoparticles. Small aliquots of as-prepared nanobelt solutions were deposited onto glass substrates and electron microscopy grids, allowed to dry, and then rinsed with methanol to remove the excess surfactant. Figure 1 displays electron micrographs that illustrate the flat, elongated geometry of the structures. The high-resolution image and diffraction pattern are consistent with the previous report which concluded that the nanobelt structure is single-crystalline gold that grows along the Æ211æ direction and has a {111} plane on the top flat surface.24 Individual, isolated nanobelts on glass substrates were imaged with epi-illumination dark field microscopy (50/0.5 NA objective, Zeiss Axiovert 200M). As seen in Figure 2, the nanobelts exhibit a range of colors covering the visible spectrum from green to red. These same nanobelts were imaged by atomic force microscopy (AFM) to determine their height and width, provided in Table 1 (see the Supporting Information for AFM

x-AR

width (nm)

height (nm)

Γ (meV)

1.0

25

25

268

1.4

50

35

303

2.1 3.3

55 100

26 30

308 510

5.0

100

20

358

5.9

100

17

261

images and analysis). Here we define the cross-sectional aspect ratio, x-AR, as the nanobelt width divided by the height. Since these nanobelts are very long, we do not consider the length to width aspect ratio that has been addressed in previous studies.25,26 As seen in the dark-field images, the existence of a tunable plasmon resonance is apparent, as nanobelts with low cross-sectional aspect ratio appear green and those with higher aspect ratio appear red. The nanobelt scattering spectra were recorded from the center of each belt by an imaging spectrograph and CCD camera. White light was incident through the annular dark field objective and was unpolarized. A polarizer behind the objective analyzed the scattered light. As seen in Figure 2, the measurements reveal a plasmon mode polarized perpendicular to the nanobelt length which red shifts with increasing cross sectional aspect ratio. Also shown are scattering spectra polarized parallel to the nanobelts, which were relatively featureless. 5035

dx.doi.org/10.1021/nl203085t |Nano Lett. 2011, 11, 5034–5037

Nano Letters

LETTER

Figure 4. The calculated charge distribution of the zigzag mode at 624 nm excited by incident TM illumination.

Figure 3. (a) Gold nanobelt simulation geometry and (b) resulting spectra for cross-sectional aspect ratios that match Figure 2. (c) The calculated charge distribution of the scattering mode.

Figure 5. The dependence of the plasmon resonance peak on crosssectional aspect ratio in the experiments (b) and simulations (O).

The perpendicular spectra were plotted versus energy and fitted to find their line widths (Γ), which are also given in Table 1. To understand the nature of the observed nanobelt plasmon resonance, finite-difference time-domain (FDTD) calculations were carried out for the nanobelt geometry using a measured dielectric function for gold.27 The simulation geometry, shown schematically in Figure 3a, was designed to match the experimental dark field spectroscopy measurements. Illumination was incident at 45 to the nanobelt long axis (as in the experiment), and the total scattering cross section for an infinitely long nanobelt was calculated. The infinite nanobelt was simulated by extending it through perfectly matched layer boundary conditions that minimize reflections.28 With the incident k-vector along the nanobelt length and TE polarization across the nanobelt as shown in Figure 3a, a scattering resonance was found (Figure 3b) that shifts with aspect ratio in a manner similar to the experimental measurement. Analysis of the charge and field distributions showed that the bright scattering resonance is a plasmon mode with antisymmetric charge alignment across the nanobelt width as depicted by the charge distribution plot in Figure 3c. This mode is analogous to the propagating HE1 plasmon mode for nanowires of cylindrical cross section.29 Note that the simulation results in Figure 3 do not include the substrate. However, a simulation for a finite nanobelt of cross-sectional aspect ratio 3.3 was run that included a glass substrate and it was found to have little effect, causing only an 18 nm red shift in the scattering peak (see Supporting Information). To more closely match the experimental measurement, FDTD simulations were also carried out for a finite nanobelt and the differential scattering cross section was calculated. Due to computational complexity, these

simulations were performed on a nanobelt of only 0.4 μm length, which resulted in strong Fabry Perot resonances that do not occur in our experiment because the nanobelt lengths are greater than 10 μm. We therefore chose to compare the experimental data to the total cross sections calculated for infinite nanobelts. Other plasmon modes were found in the simulations that were not evident in the experiments. For instance, for TM polarization, the calculations revealed an antisymmetric mode, with its dipole moment oriented in a zigzag pattern along the nanobelt length (Figure 4). This peculiar orientation of the dipole moment is due to an admixture of symmetric and antisymmetric plasmon modes on the upper and lower surfaces of the nanobelt.29 The zigzag mode was found to predominantly scatter in a direction outside the numerical aperture of the microscope objective, so it was not observed. The measured spectra of the high cross-sectional aspect ratio nanobelts appear to have some structure as though they are made up of multiple peaks. While such peaks can be attributed to substrate interactions,30 here they are artifacts of the data analysis. Different choices of the background subtraction parameters cause somewhat different spectral shapes in Figure 2, although they do not affect the overall tunability. In addition, the measured line widths given in Table 1 are similar to the homogeneous line widths of single gold nanoparticles.31 Spectral scattering peaks for both the experiments and simulations were fitted to analyze the tunability of the transverse plasmon resonance in the nanobelt structure. The results, plotted in Figure 5, agree well and reveal an approximately linear relationship between the cross-sectional aspect ratio and resonance wavelength, which is very similar to the behavior for elongated 5036

dx.doi.org/10.1021/nl203085t |Nano Lett. 2011, 11, 5034–5037

Nano Letters nanoparticles such as gold nanorods.1 Interestingly, the nanobelt structure has a lower sensitivity to the cross-sectional aspect ratio, shifting only 100 nm (∼525 to ∼625 nm) with cross-sectional aspect ratio increasing from 1 to 5. In contrast, gold nanorods (with their aspect ratio defined as the ratio of the length and width) shift over a similar spectral range with an aspect ratio change of only 1 to 2.5.32 An additional increase in the nanobelt cross-sectional aspect ratio through synthesis may enable the plasmon resonance to be further tuned to the near-infrared region. The results presented here describe a simple synthetic route toward an extended structure which exhibits sharp plasmon resonances for antisymmetric excitation across the nanobelt width. The nanobelts are single crystal structures with smooth surfaces which should minimize loss in plasmon propagation.21 These propagating modes may be important in plasmonic waveguide applications. Since their energies are well-defined, it may be possible to optimize in- and out-coupling antennas for efficient light-to-plasmon conversion33 enabling both efficient subwavelength waveguides and remote plasmon-enhanced spectroscopies such as surface-enhanced Raman scattering.34,35 These structures may also find applications as subwavelength plasmonic bandpass filters, only allowing light of specific wavelengths determined by the cross sectional geometry to be transmitted. A more detailed investigation of the propagation losses and the potential applications of these structures in plasmonics circuit applications is in progress. Furthermore, gold nanobelts may be considered for biological applications that currently employ tunable nanoparticles. The larger size of the nanobelts will provide a larger scattering cross section for imaging contrast or detection, and nanobelts may offer unique interactions with cells and tissues due to their size and shape. In conclusion, we have developed an efficient chemical synthesis method for the fabrication of gold nanobelts with cross sectional dimensions smaller than 100 nm. We showed that the transverse plasmon modes of these extended structures are remarkably narrow and depend sensitively on the cross-sectional aspect ratio. These nanostructures may find applications in plasmonic circuits or nanomedicine.

’ ASSOCIATED CONTENT

bS

Supporting Information. Gold nanobelt synthesis details, atomic force microscopy measurements, and further simulation results. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT The authors acknowledge support from the National Science Foundation under Grant 103575 (J.H.H., C.M.P.) and the Robert A. Welch Foundation under Grants C-1761 (J.H.H., L.J.E.A.) and C-1222 (P.N., Y.R.Z.).

LETTER

(3) Rycenga, M.; Cobley, C. M.; Zeng, J.; Li, W. Y.; Moran, C. H.; Zhang, Q.; Qin, D.; Xia, Y. Chem. Rev. 2011, 111, 3669–3712. (4) Liao, H. W.; Nehl, C. L.; Hafner, J. H. Nanomedicine 2006, 1, 201–208. (5) Hirsch, L. R.; Gobin, A. M.; Lowery, A. R.; Tam, F.; Drezek, R. A.; Halas, N. J.; West, J. L. Ann. Biomed. Eng. 2006, 34, 15–22. (6) Lee, S. E.; Lee, L. P. Curr. Opin. Biotechnol. 2010, 21, 489–497. (7) Schl€ucker, S. ChemPhysChem 2009, 10, 1344. (8) Golightly, R. S.; Doering, W. E.; Natan, M. J. ACS Nano 2009, 3, 2859–2869. (9) Giannini, V.; Fernandez-Dominguez, A. I.; Heck, S. C.; Maier, S. A. Chem. Rev. 2011, 111, 3888–3912. (10) Muskens, O. L.; Giannini, V.; Sanchez-Gil, J. A.; Rivas, J. G. Nano Lett. 2007, 7, 2871–2875. (11) Bouhelier, A.; Huser, T.; Tamaru, H.; Guntherodt, H. J.; Pohl, D. W.; Baida, F. I.; Van Labeke, D. Phys. Rev. B 2001, 63, 155404. (12) Barnes, W. L.; Dereux, A.; Ebbesen, T. W. Nature 2003, 424, 824–830. (13) Halas, N. J.; Lal, S.; Chang, W. S.; Link, S.; Nordlander, P. Chem. Rev. 2011, 111, 3913–3961. (14) Sanders, A. W.; Routenberg, D. A.; Wiley, B. J.; Xia, Y.; Dufresne, E. R.; Reed, M. A. Nano Lett. 2006, 6, 1822–1826. (15) Li, Z. P.; Hao, F.; Huang, Y. Z.; Fang, Y. R.; Nordlander, P.; Xu, H. X. Nano Lett. 2009, 9, 4383–4386. (16) Liu, M. Z.; Lee, T. W.; Gray, S. K.; Guyot-Sionnest, P.; Pelton, M. Phys. Rev. Lett. 2009, 102, 107401. (17) Akimov, A. V.; Mukherjee, A.; Yu, C. L.; Chang, D. E.; Zibrov, A. S.; Hemmer, P. R.; Park, H.; Lukin, M. D. Nature 2007, 450, 402–406. (18) Fedutik, Y.; Temnov, V. V.; Schops, O.; Woggon, U.; Artemyev, M. V. Phys. Rev. Lett. 2007, 99, 136802. (19) Chang, D. E.; Sorensen, A. S.; Hemmer, P. R.; Lukin, M. D. Phys. Rev. B 2007, 76, 035420. (20) Dickson, R. M.; Lyon, L. A. J. Phys. Chem. B 2000, 104, 6095–6098. (21) Ditlbacher, H.; Hohenau, A.; Wagner, D.; Kreibig, U.; Rogers, M.; Hofer, F.; Aussenegg, F. R.; Krenn, J. R. Phys. Rev. Lett. 2005, 95, 257403. (22) Knight, M. W.; Grady, N. K.; Bardhan, R.; Hao, F.; Nordlander, P.; Halas, N. J. Nano Lett. 2007, 7, 2346–2350. (23) Solis, D.; Chang, W.-S.; Khanal, B. P.; Bao, K.; Nordlander, P.; Zubarev, E. R.; Link, S. Nano Lett. 2010, 10, 3482–3485. (24) Zhao, N.; Wei, Y.; Sun, N. J.; Chen, Q. J.; Bai, J. W.; Zhou, L. P.; Qin, Y.; Li, M. X.; Qi, L. M. Langmuir 2008, 24, 991–998. (25) Payne, E. K.; Shuford, K. L.; Park, S.; Schatz, G. C.; Mirkin, C. A. J. Phys. Chem. B 2006, 110, 2150–2154. (26) Bryant, G. W.; De Abajo, F. J. G.; Aizpurua, J. Nano Lett. 2008, 8, 631–636. (27) Johnson, P. B.; Christy, R. W. Phys. Rev. B 1972, 6, 4370–4379. (28) Berenger, J. P. J. Comput. Phys. 1994, 114, 185–200. (29) Zhang, S. P.; Wei, H.; Bao, K.; Hakanson, U.; Halas, N. J.; Nordlander, P.; Xu, H. X. Phys. Rev. Lett. 2011, 107, 096801. (30) Sherry, L. J.; Chang, S. H.; Van Duyne, R. P.; Schatz, G. C.; Wiley, B. J.; Xia, Y. N. Nano Lett. 2005, 5, 2034–2038. (31) Sonnichsen, C.; Franzl, T.; Wilk, T.; von Plessen, G.; Feldmann, J. New J. Phys. 2002, 4. (32) Brioude, A.; Jiang, X. C.; Pileni, M. P. J. Phys. Chem. B 2005, 109, 13138–13142. (33) Fang, Z.; Fan, L.; Lin, C.; Zhang, D.; Meixner, A. J.; Zhu, X. Nano Lett. 2011, 11, 1676–1680. (34) Hutchison, J. A.; Centeno, S. P.; Odaka, H.; Fukumura, H.; Hofkens, J.; Uji-i, H. Nano Lett. 2009, 9, 995–1001. (35) Fang, Y.; Wei, H.; Hao, F.; Nordlander, P.; Xu, H. X. Nano Lett. 2009, 9, 2049–2053.

’ REFERENCES (1) Link, S.; El-Sayed, M. A. Int. Rev. Phys. Chem. 2000, 19, 409–453. (2) Cortie, M. B.; McDonagh, A. M. Chem. Rev. 2011, 111, 3713–3735. 5037

dx.doi.org/10.1021/nl203085t |Nano Lett. 2011, 11, 5034–5037