Single-Particle Spectroscopy of Gold Nanorods beyond the Quasi-Static Limit: Varying theWidth at Constant Aspect Ratio

Liane S. Slaughter, Wei-Shun Chang, Pattanawit Swanglap, Alexei Tcherniak,Bishnu P. Khanal, Eugene R. Zubarev, and Stephan Link*Department of Chemistry, Rice UniVersity, Houston, Texas 77005

ReceiVed: February 9, 2010

We have examined how the surface plasmon resonances (SPRs) of chemically grown gold nanorods with tunablewidths and lengths evolve due to phase retardation. For nanorods with diameters d > 30 nm, the aspect ratio is nota sufficient parameter for determining the energy of the longitudinal SPR. To rigorously study the effects of thesize, we performed correlated scanning electron microscopy and single-particle spectroscopy on broad gold nanorodsthat were chemically grown wider (d > 100 nm) and longer while maintaining the surface chemistry andhemispherical end-cap geometry as the slim rods we compared them to (d < 30 nm). At a low aspect ratio of 2.2,the longitudinal SPR of the broad nanorods significantly red-shifted and broadened as the width increased. Inaddition, broad gold nanorods with d >100 nm displayed higher order plasmon modes that were not observed forslim nanorods of similar aspect ratio. To measure the full spectrum of the largest nanorods, we implemented anew strategy for acquiring single-particle extinction spectra with an extended window of 500-1700 nm by combininga Si CCD camera and an InGaAs array detector. This experiment revealed that changing the width from 25 to 120nm while maintaining an aspect ratio of only 3.1 caused the longitudinal dipole SPR to red shift 560 nm to 1300nm. The spectroscopic studies were complemented by theoretical modeling using the discrete dipole approximation.While we found excellent agreement between the measured and predicted maxima of the longitudinal dipole SPR,the intensities of the multipolar plasmon modes were significantly enhanced in the single-particle spectra comparedto calculations.

Myriad experiments have exploited the sensitivity and efficacyof the localized surface plasmon resonance (SPR) of a varietyof gold nanoparticles to detect changes in the surroundingmedium for chemical sensing or probing the dynamics ofbiological molecules.1-6 The longitudinal SPR of gold nanorods(AuNRs) in particular manifests itself as a sharp peak in theextinction spectrum. Decreased damping relative to sphericalnanoparticles leads to a significantly narrower peak7,8 which isvery sensitive to changes in the surrounding refractive indexand incident polarization, forming the basis for highly effectivenanostructure probes.9-11 Improvements in synthesis and puri-fication of AuNRs have enabled facile tuning of the longitudinalSPR maximum, λmax, by adjusting the length and hence aspectratio.10,12 Controlling λmax is important for optimizing the SPRresponse for specific applications. For example, the longitudinalSPR of AuNRs needs to fall within the optically transparentwindow of biological tissue (650-900 nm) during in vivoimaging.

Among the most common AuNR synthesis methods issolution phase chemical growth, which produces AuNRs withsurface chemistry which can be easily modified.3,13-16 Surfacefunctionalization enables one to target specific moleculesthrough selective binding. Examples include exploiting thestrong interactions between streptavidin and biotin3,6 or oligo-nucleotides.15 The combination of an anisotropic shape andadjustable surface chemistry imparts the additional ability totailor AuNRs as oriented building blocks for rational assemblyusing small functional molecules16 or macromolecules such asblock copolymers.14 Chemical preparation methods typicallyyield AuNRs with diameters, d ) 10-30 nm, aspect ratios <5,

and a narrow distribution of widths.3,17,18 In this size regime,the longitudinal dipole SPR λmax scales linearly with the aspectratio almost independently of the actual dimensions, therebyemphasizing the effect of the AuNR length.9 This fact isroutinely used in characterizing AuNRs based on their opticalextinction spectra and agrees well with calculations using thequasi-static, i.e., dipole, approximation.3,17

On the other hand, the similarity to radio antennas has sparkedconsiderable interest in using plasmonic AuNRs to concentrate andlocalize optical radiation thereby beating the diffraction limit ofconventional optics. For such device applications, template-assistedpreparations are typically employed yielding AuNRs with signifi-cantly larger dimensions,19,20 as the AuNR length has to approachabout half the wavelength of the light in order to act as a resonantantenna.21,22 For AuNRs with diameters of ∼80 nm but high aspectratios (>10), the length remains the more important parameter.23,24

As the overall size increases, however, the longitudinal dipole SPRλmax starts to depend sensitively on the width as well.22,25,26 Phaseretardation and dynamic depolarization effects27,28 become impor-tant, and a simple relationship between λmax and the aspect ratiotherefore no longer applies. Numerical simulations accounting forthe absolute dimensions of the AuNRs have predicted this departurefrom the quasi-static scaling of the longitudinal dipole SPR λmax

with aspect ratio.22,26,29,30 In particular, it was shown that changingthe AuNR diameter by only a factor of 2 from 40 to 80 nm whilemaintaining the aspect ratio as low as 3 red shifts λmax by as muchas 200 nm.22 Furthermore, with increasing size, higher order SPRmodes with both transverse and longitudinal polarization compo-nents are excited due to the fact that the phase of the interactingelectromagnetic field varies across the AuNRs.22,30

Experimentally, the red shift of the longitudinal dipole SPRλmax and the emergence of mulipolar SPR modes have been* Corresponding author, slink@rice.edu.

J. Phys. Chem. C 2010, 114, 4934–49384934

10.1021/jp101272w 2010 American Chemical SocietyPublished on Web 03/01/2010

demonstrated using ensemble vis-NIR extinction spectroscopyfor 85 nm wide AuNRs with lengths varying between 96 and1175 nm.31 These AuNRs were synthesized by electrochemicaldeposition of gold into anodized aluminum oxide templates.19,31,32

For AuNRs with d ) 83 nm and aspect ratios less than 3, thespectrum revealed a longitudinal dipole SPR λmax near 1100nm,32 which represents a drastic spectral shift from the 700 to750 nm range for AuNRs with d e 30 nm and similar aspectratios.3,17 Similar results were also obtained for AuNRs withwidths of 91 nm and lengths ranging between 90 and 1600 nm,which were prepared by electron-beam lithography on a glasssurface.33

Despite these important contributions, a detailed comparisonof the plasmonic properties of AuNRs over an extended rangeof widths is complicated by several factors. First, the typicaldiameters of chemically grown AuNRs differ from those oftemplate synthesized AuNRs by a factor of almost 3, for whichtheory predicts drastic shifts of the longitudinal dipole SPRλmax.22 Second and more importantly, to minimize the variablesthat affect the SPR other than the absolute dimensions, such astudy requires that all AuNRs have similar surface chemistryas well as geometry. While chemically grown AuNRs consistof pentahedrally twinned single crystalline domains,34 havenearly rounded end-caps, and have surfaces terminated bymolecular ligands,3,17,18 template-assisted preparation methodstypically yield AuNRs that are more polycrystalline, haveirregular end-cap geometries, or have no surface cappingmaterial.19,32 Although a method was recently developed forsmoothing the ends of AuNRs prepared from templates, theresulting products still retain surface features on the order of 5nm at the ends.20 Numerical calculations have explored thechanges induced by different end-cap geometries,26,35 but theyoften neglect other details or account for them with approxima-tions such as the use of an adjustable effective refractive indexto describe the AuNR surface chemistry and the surroundingmedium.35-37 It is therefore crucial to test how well theoreticalpredictions compare to actual experiments in which the AuNRwidth is changed while keeping the aspect ratio, the end-capgeometry, and surface chemistry constant.

Here, we present an investigation of low aspect ratio AuNRswith broadly tunable widths using correlated scanning electronmicroscopy (SEM) and single-particle spectroscopy. We mea-sured the scattering and extinction spectra of AuNRs that werechemically synthesized and then grown wider and longer whilepreserving the geometry and surface chemistry of the particle,hence overcoming the aforementioned difficulties. We showhow, with increasing width, the absolute dimensions and notjust the aspect ratio determine the plasmonic properties ofchemically grown AuNRs, which opens new possibilities fortuning AuNR-based chemical sensors. Furthermore, single-particle spectroscopy together with structural imaging allowedus to remove the size heterogeneity present among the AuNRsand compare the resonant modes for a particular AuNR widthand length to calculations using the discrete dipole approxima-tion (DDA). These measurements were enabled by an advancedsingle-particle extinction setup that combines a Si CCD camerawith an InGaAs array detector giving us access to an extendedspectral window from 500 to 1700 nm.

We investigated individual AuNRs selected from two sampleswith the goal to compare similar aspect ratios over a range ofdiameters. Figure 1 shows SEM images of these AuNR sampleslabeled “slim” rods with d ) 25 ( 5 nm (Figure 1a) and “broad”rods with d ) 100 ( 20 nm (Figure 1b). Slim AuNRs stabilizedwith cetyltrimethylammonium bromide (CTAB) were synthe-

sized in solution using the standard seed-mediated procedure.3

Broad AuNRs were also formed in solution through a reversibletuning strategy previously published for AuNRs.38 Briefly, theywere synthesized by first amplifying pentahedrally twinned slimAuNRs through fast deposition of Au(I) ions to form longerand wider AuNRs followed by partially dissolving their endsin Au(III)/CTAB solution to give broad AuNRs with roundedends. Distinct from common methods for growing AuNRs, thisprocedure requires extra manipulation of the Au(I) and Au(III)/CTAB concentrations. Typically, modest amounts of Au(I) insolution are reduced and deposited only at the ends of theAuNRs because CTAB preferentially binds to the {111} sidesof the AuNRs.34 Increasing the AuNR diameter requires anexcess of Au(I), in which case growth occurs along all crystalfaces. One therefore cannot selectively amplify the diameter ofthe AuNRs without increasing their length. Following growthin both dimensions, we then selectively oxidized and hencedissolved gold from the ends of the AuNRs using a solution ofAu(III)/CTAB complex as previously reported.38 This approachallowed us to adjust the lengths of broad AuNRs to ∼300 nmafter first growing the slim AuNRs to ∼600 × 100 nm.38,39

Single AuNRs deposited on patterned glass substrates werelocated exactly in both an SEM, operating in wet mode to offsetcharging on the nonconductive substrate, and a home-builtsingle-particle spectrometer with a cooled Si CCD camera (seeSupporting Information).40 Spectra a-c of Figure 2 display theexperimental scattering spectra recorded for polarizations parallel(Φ|, red) and perpendicular (Φ⊥, blue) to the longitudinal axesof three single AuNRs, each having a similar aspect ratio of∼2.2 ((0.1) but increasing in overall size from spectrum a tospectrum c in Figure 2. For the smallest AuNR in Figure 2a(32 × 69 ((2) nm), only a narrow longitudinal dipole SPRdominates the spectrum and reaches its maximum at 620 nm.

Figure 1. SEM images at 80000× of (a) slim AuNRs and (b) broadAuNRs.

Figure 2. Scattering spectra of single AuNRs with (a) d ) 32 nm, (b)d ) 81 nm, and (c) d ) 100 nm, all having an aspect ratio of ∼2.2 (0.1, and the corresponding DDA calculations, (d), (e), and (f),respectively. The spectra were recorded for scattered light polarizedparallel (red) and perpendicular (blue) to the orientation of the mainrod axis. A fit to a sum of two Lorentzian curves is included as blacklines in (f). In all images, the scale bar represents 100 nm.

Plasmonic Properties of Gold Nanorods J. Phys. Chem. C, Vol. 114, No. 11, 2010 4935

The scattering from the transverse SPR was too weak to bedetected. In contrast, the two broad AuNRs reveal markedlydifferent scattering profiles. As the AuNR size increases to 84× 174 ((5) nm in Figure 2b, the longitudinal dipole SPR redshifts to 740 nm and broadens significantly due to a combinationof increased radiation damping and dynamic depolarization.28,41

In addition, a transverse SPR peaks at 600 nm because of thelarger width of the AuNR. The maximum of the longitudinaldipole SPR in Figure 2c increases even further to 900 nm forthe largest AuNR (100 × 227 ((5) nm), while an additionalshoulder appears at shorter wavelengths. Fitting the spectrumwith Φ| using two Lorentzians gives a high-energy peakcentered at 730 nm, which reaches its maximum for Φ| andminimum for Φ⊥, consistent with an oscillation mainly polarizedalong the main rod axis. The fact that the broad scattering featurerecorded for Φ⊥ in Figure 2c is red shifted to 660 nm comparedto the one in Figure 2b, however, suggests that this mode alsohas partial transverse character.22,23,26 We therefore assign theshoulder to a multipolar SPR mode.

To further rationalize these observations, we performed DDAcalculations using the DDSCAT code (see Supporting Informa-tion).42 AuNRs with the same dimensions as measured in theexperiment were modeled as round cylinders with hemisphericalend-caps. We used Johnson and Christy data for the golddielectric function43 and incorporated a size correction factor.44

The medium dielectric constant was approximated as the averagebetween the supporting substrate and air.37 The simulatedscattering spectra for the three AuNRs are shown in spectrad-f in Figure 2. To simulate parallel (Φ|, red) and perpendicular(Φ⊥, blue) polarization, the AuNR was rotated in the plane ofthe fixed incident wave vector and selected polarization.

The DDA calculations show good agreement with theexperimental scattering spectra considering the fact that noadjustable parameters were used to improve the match with eachof the three AuNRs. The major trends as functions of the overallAuNR size were all reproduced. In particular, the longitudinaldipole SPR red shifts with increasing AuNR dimensions from630 nm (Figure 2d) to 765 nm (Figure 2e) and 930 nm (Figure2f), which is accompanied by broadening of the resonance peak.The calculated spectra using perpendicular excitation further-more reveal transverse SPRs with maxima at 550 nm (Figure2e) and 570 nm (Figure 2f), confirming at least the assignmentof the band in the perpendicularly polarized spectrum of the 81× 174 nm AuNR.

For the largest AuNR in Figure 2, the calculated spectrum withparallel polarization also has a shoulder at shorter wavelengths anda fit yields a maximum at 760 nm for this multipolar SPR.However, the same mode is much more pronounced in thecorresponding experimental spectrum and also appears to contributeto the broad spectrum having perpendicular polarization in Figure2c. The different amplitudes for this band in Figure 2c comparedto Figure 2f partly result from the differences in scattering geometrybetween the experiment and theory. In the experiment, the AuNRswere excited with a circular cone of unpolarized light containingwave vectors from all directions in the sample plane as well asthose out-of-plane allowed by the dark-field excitation scheme. Thepolarizer in the detection path then selected either the parallel orperpendicular polarization component of the scattered light. Incontrast, the DDA calculations in Figure 2f only considered onefixed in-plane wave vector orthogonal to the long rod axis. Inaddition, the simulations completely neglected the glass substrate,subsequently ignoring the effects of the interface which canobservably affect the peak line width.45 Symmetry breaking dueto the substrate together with the coupling of multipolar modes to

their image charges are expected to greatly enhance their oscillatorstrengths, thereby increasing especially the relative enhancementof the quadrupole SPR compared to the dipole SPR.46,47

The drastic red shift of the dipole SPR in the experimentalscattering spectra as the size increases is consistent with previoustheoretical investigations, which showed that for AuNRs withd > 30 nm, the quasi-static approximation becomes invalid andthe dimensions of the AuNR critically affect the spectra.22,23

Notably, these calculations using the boundary element methodsimulated capsule-shaped AuNRs with rounded ends, whichclosely resemble the broad AuNRs studied here. Fully mappingthe dependence of the longitudinal dipole SPR λmax withincreasing width and length predicted drastic red shifts withoutchanging the aspect ratio.22,23 The measured scaling of thelongitudinal dipole SPR with size reported here agrees well withthis mapping as well as analytical expressions derived by Encinaand Coronado for the dependence of each multipolar SPR onthe AuNR size through an extensive series of DDA calcula-tions.29 A higher order mode with partial transverse characterwas also predicted to arise upon the onset of retardation effectsand is enhanced by rounded ends relative to sharp ends.25 Thismode is similar to the quadrupole mode of a sphere which, asthe aspect ratio increases, will gradually become obscured asthe particle gains asymmetry.41 On the basis of comparison tothese theoretical studies we assign the SPR at 730 nm in Figure2c to the quadrupole mode.

The AuNRs discussed so far have a low aspect ratio of 2.2.Increasing the aspect ratio of a slim AuNR to 3.1 (24 × 74 ((2)nm) caused the longitudinal dipole SPR λmax to red shift from 620to 740 nm. For a broad AuNR with a similar aspect ratio, however,the same SPR mode occurs well outside the experimental windowof our Si CCD camera based on the spectra in Figure 2 and ourDDA calculations. We thus required a different strategy to resolvethe full optical spectrum of broad AuNRs with larger aspect ratios.Figure 3 illustrates the results using this strategy for a broad AuNRwith dimensions of 120 × 370 ((5) nm and aspect ratio of 3.1.The extinction spectrum was captured in transmission geometry,

Figure 3. Unpolarized extinction spectrum (green) and polarizedbackscattering spectra (red and blue) (a) of a 120 × 370 nm AuNRand the extinction calculated by DDA (b) of the same size AuNRexcited with circularly polarized light. The scattering spectra in (a) arevertically offset from the extinction spectrum for clarity.

4936 J. Phys. Chem. C, Vol. 114, No. 11, 2010 Slaughter et al.

where the spectrum from 900 to 1700 nm was acquired from themicroscope with a fiber coupled to an InGaAs array spectrometer(see Supporting Information). We recorded the unpolarized extinc-tion spectrum in Figure 3 in transmission geometry instead ofcollecting the back scattering as was done for the data shown inFigure 2. This arrangement was more feasible for optimizing thesignal because the InGaAs array detector, despite being able todetect wavelengths up to 2.2 µm, is less sensitive than the Si CCDcamera. We also measured the extinction spectrum from 500 to1000 nm in transmission geometry using the same Si CCDspectrometer described above and then combined data from thetwo detectors to obtain the full extinction spectrum shown in Figure3a.

In addition to the dipole SPR of the AuNR at 1300 nm, anotherstrong peak occurs at 700 nm in the extinction spectrum. Thecorresponding scattering spectra for Φ| and Φ⊥ show that scatteringboth longitudinal and transverse to the rod axis contribute to thispeak, although it contains more transverse character. Because ofthe large line width of the 700 nm band, we assign this resonanceto the superposition of the transverse oscillation and multiple higherorder SPR modes. This assignment is further supported by DDAcalculations of the extinction spectrum for a AuNR with the samedimensions and circularly polarized excitation light shown in Figure3b. Spectra a and b in Figure 3 are in good agreement for thelongitudinal dipole SPR. However, apart from a featureless offsetfrom zero intensity between 600 and 1000 nm, the theoreticalspectrum only shows a weak and narrow SPR mode peaked around540 nm. The latter is the transverse SPR based on its polarizationdependence. Hence, the intensity of the peak at 700 nm in theexperimental spectrum is strongly enhanced, presumably againbecause of interactions with the surface and the specific geometryof the experimental setup. Small shape deviations from an idealsymmetric cylinder would also affect the charge distribution at thesurface and increase the intensity of otherwise barely detectablemodes.35 Consistent with an increased red shift of higher orderSPR modes for larger nanostructures,26,31 the quadrupole mode forthe 120 × 370 nm AuNR is expected to overlap with the onset ofthe longitudinal dipole SPR, while several multipolar modes ofeven higher order likely contribute to the broad band observed at700 nm. Future polarization-dependent extinction measurementsmight be able to resolve this issue.

Despite the less than perfect match between experiment andtheory for the higher order SPR modes, we found very goodquantitative agreement for the longitudinal dipole SPR λmax

based on our own DDA calculations and published scaling lawsas shown in Figure 4. It should be noted that the inputparameters were limited to the overall shape and dimensionsof the AuNRs and an effective medium dielectric constant. In

Figure 4 we have used an empirical formula derived by Encinaand Coronado.29 They obtained relationships to predict themaxima of multipolar SPRs for rods of varying aspect ratio Rfor a given diameter d and medium dielectric constant εm.29 Forthe dipole mode, l ) 1, the longitudinal SPR λmax is given by

The coefficients A, B, and C were derived from an extensiveseries of DDA calculations aimed at characterizing the nonlinearscaling of the resonance condition as each of the parameters d,R, and l change. Using these coefficients, the relationship wasgeneralized to the equation given above assuming that the realpart of the metal dielectric function is given according to theDrude-Sommerfeld model by: εm ) R - �2λ2, where R ) 9.8and � ) 7.3 µm-1 for gold. Figure 4 plots the predicted scalingof the longitudinal dipole SPR λmax with the aspect ratio for thewidths of each individual AuNR described herein. The measuredλmax values for the dipole mode are included as points. Withoutemploying any adjustable parameters, the experimentally ob-served red shift of the dipole peak as a function of both thediameter and the aspect ratio agrees very well with the predictedtrends.22,29 It should be noted that eq 1 was derived for rodswith diameters outside the quasi-static limit, but for d e 40nm. Our experiments suggest that eq 1 is also applicable toAuNRs with even larger diameters.

In summary, we have examined the plasmonic properties ofchemically grown AuNRs with nearly the same geometry andsurface capping material covering a large size range. For AuNRsmuch wider than 30 nm, the quasi-static approximation fails asthe longitudinal dipole SPR red shifts and broadens due to phaseretardation even as the aspect ratio remains constant. Weobserved here that an additional higher order multipole emergesfor AuNRs with aspect ratios as low as 2.2 but d g 100 nm,which we assign to the quadrupolar plasmon oscillation.Furthermore, we have presented the IR extinction spectrum ofa broad AuNR with a low aspect ratio of 3.1 but with d ) 120nm, so wide that the dipole SPR peaks at 1300 nm. Correlatedsingle-particle spectroscopy and SEM characterization further-more allowed for a detailed modeling of the shape-dependentSPR response. While we found very good agreement for thelongitudinal dipole SPR λmax with DDA calculations andpredicted scaling laws despite limiting the number of variables,other spectral features such as the relative intensities of multipoleSPR modes or linewidths require a more complex theoreticalapproach taking explicitly into account the experimental ge-ometry including the substrate. The comparison of the plasmonicproperties of low aspect ratio AuNRs with variable diameterswas made possible through a combination of a unique chemicalsynthesis and ultra-broad-band single-particle spectroscopy.Because of the common surface chemistry independent of theAuNR size, these results offer new possibilities for creatingsensitive and tunable chemical SPR probes with significantlyenhanced scattering cross sections.

Acknowledgment. Support was provided by the Robert A.Welch Foundation (C-1664 to S.L., C-1703 to E.R.Z.), 3M(Non-Tenured Faculty Grant to S.L.), and NSF (DMR-0547399and CBET-0506832 to E.R.Z.). L.S.S. acknowledges supportfrom an NSF IGERT fellowship and Britain Willingham forinsightful discussions. W.S.C. thanks the Smalley Institute for

Figure 4. Scaling of the longitudinal dipole SPR λmax with the aspectratio simulated using a medium dielectric constant of 1.26. Theexperimentally measured values for the individual AuNRs discussedhere are included as points.

λmax (µm) )[( R2

l2π2(Ad2 + B) + C)εm + R]1/2

�(1)

Plasmonic Properties of Gold Nanorods J. Phys. Chem. C, Vol. 114, No. 11, 2010 4937

a Peter and Ruth Nicholas fellowship. We thank ChristianSchoen from Nanopartz for providing the slim AuNRs.

Supporting Information Available: Details of the single-particle spectroscopy experiments and DDA calculations for a120×370 nm AuNR. This material is free of charge via theInternet at http://pubs.acs.org.

References and Notes

(1) Haes, A. J.; Zou, S.; Schatz, G. C.; Van Duyne, R. P. A NanoscaleOptical Biosensor: The Long Range Distance Dependence of the LocalizedSurface Plasmon Resonance of Noble Metal Nanoparticles. J. Phys. Chem.B 2003, 108, 109–116.

(2) Jain, P. K.; Lee, K. S.; El-Sayed, I. H.; El-Sayed, M. A. CalculatedAbsorption and Scattering Properties of Gold Nanoparticles of DifferentSize, Shape, and Composition: Applications in Biological Imaging andBiomedicine. J. Phys. Chem. B 2006, 110, 7238–7248.

(3) Murphy, C. J.; Sau, T. K.; Gole, A. M.; Orendorff, C. J.; Gao, J.;Gou, L.; Hunyadi, S. E.; Li, T. Anisotropic Metal Nanoparticles: Synthesis,Assembly, and Optical Applications. J. Phys. Chem. B 2005, 109, 13857–13870.

(4) West, J. L.; Halas, N. J. Engineered Nanomaterials for BiophotonicsApplications: Improving Sensing, Imaging, and Therapeutics. Annu. ReV.Biomed. Eng. 2003, 5, 285–292.

(5) Sonnichsen, C.; Reinhard, B. M.; Liphardt, J.; Alivisatos, A. P. AMolecular Ruler Based on Plasmon Coupling of Single Gold and SilverNanoparticles. Nat. Biotechnol. 2005, 23, 741–745.

(6) Baciu, C. L.; Becker, J.; Janshoff, A.; Sonnichsen, C. Protein-Membrane Interaction Probed by Single Plasmonic Nanoparticles. NanoLett. 2008.

(7) Sonnichsen, C.; Franzl, T.; Wilk, T.; von Plessen, G.; Feldmann,J.; Wilson, O.; Mulvaney, P. Drastic Reduction of Plasmon Damping inGold Nanorods. Phys. ReV. Lett. 2002, 88, 077402/077401-077402/077404.

(8) Novo, C.; Gomez, D.; Perez-Juste, J.; Zhang, Z. Y.; Petrova, H.;Reismann, M.; Mulvaney, P.; Hartland, G. V. Contributions from RadiationDamping and Surface Scattering to the Linewidth of the LongitudinalPlasmon Band of Gold Nanorods: A Single Particle Study. Phys. Chem.Chem. Phys. 2006, 8, 3540–3546.

(9) Miller, M. M.; Lazarides, A. A. Sensitivity of Metal NanoparticleSurface Plasmon Resonance to the Dielectric Environment. J. Phys. Chem.B 2005, 109, 21556–21565.

(10) Link, S.; El-Sayed, M. A. Spectral Properties and RelaxationDynamics of Surface Plasmon Electronic Oscillations in Gold and SilverNanodots and Nanorods. J. Phys. Chem. B 1999, 103, 8410–8426.

(11) Sonnichsen, C.; Alivisatos, A. P. Gold Nanorods as Novel Non-bleaching Plasmon-Based Orientation Sensors for Polarized Single-ParticleMicroscopy. Nano Lett. 2005, 5, 301–304.

(12) Jana, N. R.; Gearheart, L.; Murphy, C. J. Wet Chemical Synthesisof High Aspect Ratio Cylindrical Gold Nanorods. J. Phys. Chem. B 2001,105, 4065–4067.

(13) Khanal, B. P.; Zubarev, E. R. Rings of Nanorods. Angew. Chem.,Int. Ed. 2007, 46, 2195–2198.

(14) Nie, Z.; Fava, D.; Kumacheva, E.; Zou, S.; Walker, G. C.;Rubinstein, M. Self-Assembly of Metal-Polymer Analogues of AmphiphilicTriblock Copolymers. Nat. Mater. 2007, 6, 609–614.

(15) Rosi, N. L.; Giljohann, D. A.; Thaxton, C. S.; Lytton-Jean, A. K. R.;Han, M. S.; Mirkin, C. A. Oligonucleotide-Modified Gold Nanoparticlesfor Intracellular Gene Regulation. Science 2006, 312, 1027–1030.

(16) Thomas, K. G.; Barazzouk, S.; Ipe, B. I.; Joseph, S. T. S.; Kamat,P. V. Uniaxial Plamon Coupling through Longitudinal Self-Assembly ofGold Nanorods. J. Phys. Chem. B 2004, 108, 13066–13068.

(17) Nikoobakht, B.; El-Sayed, M. A. Preparation and Growth Mech-anism of Gold Nanorods Using Seed-Mediated Growth Method. Chem.Mater. 2003, 15, 1957–1962.

(18) Kim, F.; Song, J. H.; Yang, P. Photochemical Synthesis of GoldNanorods. J. Am. Chem. Soc. 2002, 124, 14316–14317.

(19) Foss, C. A.; Hornyak, G. L.; Stockert, J. A.; Martin, C. R. Template-Synthesized Nanoscopic Gold Particles: Optical Spectra and the Effects ofParticle Size and Shape. J. Phys. Chem. 1994, 98, 2963–2971.

(20) Banholzer, M. J.; Li, S.; Ketter, J. B.; Rozkiewicz, D. I.; Schatz,G. C.; Mirkin, C. A. Electrochemical Approach to and the PhysicalConsequences of Preparing Nanostructures from Gold Nanorods withSmooth Ends. J. Phys. Chem. C 2008, 112, 15729–15734.

(21) Novotny, L. Effective Wavelength Scaling for Optical Antennas.Phys. ReV. Lett. 2007, 98, 266802/266801–266802/266804.

(22) Bryant, G. W.; Garcia de Abajo, F. J.; Aizpurua, J. Mapping thePlasmon Resonances of Metallic Nanoantennas. Nano Lett. 2008, 8, 631–636.

(23) Aizpurua, J.; Bryant, G. W.; Richter, L. J.; Garcia de Abajo, F. J.;Kelley, B. K.; Mallouk, T. Optical Properties of Coupled Metallic Nanorodsfor Field-Enhanced Spectroscopy. Phys. ReV. B 2005, 71, 235420–235413.

(24) Gomez-Medina, R.; Yamamoto, N.; Nakano, M.; Garcia de Abajo,F. J. Mapping Plasmons in Nanoantennas Via Cathodoluminescence. NewJ. Phys. 2008, 105009.

(25) Myroshnychenko, V.; Rodriguez-Fernandez, J.; Pastoriza-Santos,I.; Funston, A. M.; Novo, C.; Mulvaney, P.; Liz-Marzan, L. M.; Garcia deAbajo, F. J. Modelling the Optical Response of Gold Nanoparticles. Chem.Soc. ReV. 2008, 37, 1792–1805.

(26) Khlebtsov, B. N.; Khlebtsov, N. G. Multipole Plasmons in MetalNanorods: Scaling Properties and Dependence on Particle Size, Shape,Orientation, and Dielectric Environment. J. Phys. Chem. C 2007, 111,11516–11527.

(27) Kuwata, H.; Tamaru, H.; Esumi, K.; Miyano, K. Resonant LightScattering from Metal Nanoparticles: Practical Analysis Beyond RayleighApproximation. Appl. Phys. Lett. 2003, 83, 4625–4627.

(28) Meier, M.; Wokaun, A. Enhanced Fields on Large Metal Paricles:Dynamic Depolarization. Opt. Lett. 1983, 8, 581–583.

(29) Encina, E. R.; Coronado, E. A. Resonance Conditions for MultipolePlasmon Excitations in Noble Metal Nanorods. J. Phys. Chem. C 2007,111, 16796–16801.

(30) Pelton, M.; Aizpurua, J.; Bryant, G. Metal-Nanoparticle Plasmonics.Laser Photonics ReV. 2008, 2, 136–159.

(31) Payne, E. K.; Shuford, K. L.; Park, S.; Schatz, G. C.; Mirkin, C. A.Multipole Plasmon Resonances in Gold Nanorods. J. Phys. Chem. B 2006,110, 2150–2154.

(32) Kim, S.; Shuford, K. L.; Bok, H.-M.; Kim, S. K.; Park, S.Intraparticle Surface Plasmon Coupling in Quasi-One-Dimensional Nano-structures. Nano Lett. 2008, 8, 800–804.

(33) Schider, G.; Krenn, J. R.; Hohenau, A.; Ditlbacher, H.; Leitner,A.; Aussenegg, F. R.; Schaich, W. L.; Puscasu, I.; Monacelli, B.; Boreman,G. Plasmon Dispersion Relation of Au and Ag Nanowires. Phys. ReV. B2003, 68, 155427.

(34) Johnson, C. J.; Dujardin, E.; Davis, S. A.; Murphy, C. J.; Mann, S.Growth and Form of Gold Nanorods Prepared by Seed-Mediated, Surfactant-Directed Synthesis. J. Mater. Chem. 2002, 12, 1765–1770.

(35) Noguez, C. Surface Plasmons on Metal Nanoparticles: The Influenceof Shape and Physical Environment. J. Phys. Chem. C 2007, 111, 3806–3819.

(36) Jain, P. K.; El-Sayed, M. A. Surface Plasmon Coupling and ItsUniversal Size Scaling in Metal Nanostructures of Complex Geometry:Elongated Particle Pairs and Nanosphere Trimers. J. Phys. Chem. C 2008,112, 4954–4960.

(37) Novo, C.; Funston, A. M.; Pastoriza-Santos, I.; Liz-Marzan, L. M.;Mulvaney, P. Influence of the Medium Refractive Index on the OpticalProperties of Single Gold Triangular Prisms on a Substrate. J. Phys. Chem.C 2007, 112, 3–7.

(38) Khanal, B. P.; Zubarev, E. R. Purification of High Aspect RatioGold Nanorods: Complete Removal of Platelets. J. Am. Chem. Soc. 2008,130, 12634–12635.

(39) Tsung, C.-K.; Kou, X.; Shi, Q.; Zhang, J.; Yeung, M. H.; Wang,J.; Stucky, G. D. Selective Shortening of Single-Crystalline Gold Nanorodsby Mild Oxidation. J. Am. Chem. Soc. 2006, 128, 5352–5353.

(40) Chang, W.-S.; Slaughter, L. S.; Khanal, B. P.; Manna, P.; Zubarev,E. R.; Link, S. One-Dimensional Coupling of Gold Nanoparticle Plasmonsin Self-Assembled Ring Superstructures. Nano Lett. 2009, 9, 1152–1157.

(41) Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C. The OpticalProperties of Metal Nanoparticles: The Influence of Size, Shape, andDielectric Environment. J. Phys. Chem. B 2002, 107, 668–677.

(42) Draine, B. T.; Flatau, P. J. Discrete-Dipole Approximation forScattering Calculations. J. Opt. Soc. Am. A 1994, 11, 1491–1499.

(43) Johnson, P. B.; Christy, R. W. Optical Constants of the NobleMetals. Phys. ReV. B 1972, 6, 4370.

(44) Coronado, E. A.; Schatz, G. C. Surface Plasmon Broadening forArbitrary Shape Nanoparticles: A Geometrical Probability Approach.J. Chem. Phys. 2003, 119, 3926–3934.

(45) Hovel, H.; Fritz, S.; Hilger, A.; Kreibig, U.; Vollmer, M. Width ofCluster Plasmon Resonances: Bulk Dielectric Functions and ChemicalInterface Damping. Phys. ReV. B 1993, 48, 18178.

(46) Bedeaux, D.; Vlieger, J. Optical Properties of Surfaces; ImperialCollege Press: London, 2002.

(47) Knight, M. W.; Wu, Y.; Lassiter, J. B.; Nordlander, P.; Halas, N. J.Substrates Matter: Influence of an Adjacent Dielectric on an IndividualPlasmonic Nanoparticle. Nano Lett. 2009, 9, 2188–2192.

JP101272W

4938 J. Phys. Chem. C, Vol. 114, No. 11, 2010 Slaughter et al.