Journal of Colloid and Interface Science 461 (2016) 376–382

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

journal homepage: www.elsevier .com/locate / jc is

Controlling optical properties of metallic multi-shell nanoparticlesthrough suppressed surface plasmon resonance

http://dx.doi.org/10.1016/j.jcis.2015.09.0400021-9797/� 2015 Elsevier Inc. All rights reserved.

⇑ Corresponding author at: Department of Chemistry, Sungkyunkwan University,Suwon 440-746, South Korea

E-mail address: spark72@skku.edu (S. Park).

Jesus A. I. Acapulco Jr. a, Soonchang Hong a, Seong Kyu Kim a, Sungho Park a,b,⇑aDepartment of Chemistry, Sungkyunkwan University, Suwon 440-746, South KoreabDepartment of Energy Science, Sungkyunkwan University, Suwon 440-746, South Korea

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 15 April 2015Revised 27 August 2015Accepted 16 September 2015Available online 18 September 2015

Keywords:Multi-shell nanoparticlesNanorodsSurface plasmon resonanceSERS

a b s t r a c t

Herein, we report the surface plasmon resonance of plasmonic multi-shell nanoparticles compared tobimetallic Ag/Au hollow nanospheres of similar final size, shape, and percent composition. The surfaceplasmon resonance of solid and hollow nanoparticles exhibited a quadrupole mode that was particularlyprominent around the 100 nm size regime, while multi-shell nanoparticles did not show a quadrupolemode at a similar size. In the latter case, the quadrupole mode of the outermost nanoshell wassuppressed by the dipole modes of the inner shells, and the suppression of the quadrupole mode wasnot affected by the shape of the inner nanostructures. Light interaction of the multi-shell nanoparticlewas investigated through simulated electromagnetic field distribution obtained by finite-difference timedomain (FDTD) calculations which were in a good agreement with the results of surface-enhanced Ramanspectroscopy (SERS).

� 2015 Elsevier Inc. All rights reserved.

1. Introduction

Studies involving the plasmonic properties of metal nanoparti-cles have been increasing due to their wide variety of potential

applications such as in solar cells [1,2], electronics [3], and sensingdevices [4–6]. The root of these applications relies primarily on thelocalized surface plasmon resonance (LSPR) of the nanoparticle,which is sensitive to the surrounding chemical environment. LSPRtakes place when the plasmonic nanoparticle is irradiated withlight causing the electrons at the conduction band to oscillate,which is driven by the electric field [7]. In this way, LSPR resultsin the manipulation of photons to be absorbed and radiated as well

Fig. 1. Schematic illustration for the synthesis of (A) multi-shell Au core nanorodand reference sample of (B) multi-shell Au core nanosphere, and (C) hollownanosphere.

J.A.I. Acapulco Jr. et al. / Journal of Colloid and Interface Science 461 (2016) 376–382 377

as enhancement of the electromagnetic (EM) field around the plas-monic nanoparticle [8,9]. The practical metal nanoparticles relatedto this field of research consist of gold (Au) and silver (Ag), sincethey exhibit high optical features at visible to near-IR spectralregions. Dipole plasmon resonance mode is only observed whenthe diameter of nanoparticle is almost equal or less than thewavelength of incident light; however, higher order modes suchas quadrupole, octupole, and hexadecapole begin to appear asnanoparticle size increases [10]. Moreover, LSPR is influenced notonly by the size, but also the nanoparticle’s shape, composition,and dielectric environment, thereby providing researchers anadvantage in their ability to fully control and manipulate light[11–13].

Complex nanostructures have started to appear in the field ofsolution-based synthesis of metal nanoparticles, and thus far haveexhibited exceptional LSPR behavior. One such example in thiscategory is the asymmetric core-satellite nanoassemblies reportedby Yoon et al. [14], who demonstrated that LSPR coupling betweenthe core and satellite nanoparticle exhibits a quantum tunnelingeffect. In addition, pure Au nanoshells [15] have been reported toexhibit a wide LSPR tuning range, going as high as the NIRfrequency, which is controlled by the radius of the dielectric coreand the thickness of the gold nanoshell. Nanocups and nanoshellsthat are partially opened by an offset core have a unique spectralproperty of far field scattering that is strongly dependent on thepolarization of light [16].

The ability to control the composition and morphology of Aunanocages [17,18] allows the precise tuning of LSPR throughgalvanic replacement between Ag nanoparticle and Au ions. Suchhollow nanostructures have been extensively studied both experi-mentally and theoretically through their conversion to multilayernanoshells [19,20], also known as nanomatryushka. Specifically,it has been shown that the dielectric spacer layer between the coreand shell determines the coupling of LSPR resulting in next-leveltunability of optical responses at a greater spectral coveragethan for individual Au nanoshells. Compared to simple nanostruc-tures where the absorption and scattering efficiencies are primarilya function of nanoparticle size, such a clear dependency formultilayered nanostructures becomes more complex andgeometry-dependent. Nanomatryushka have been reported tohave a greater scattering efficiency under a thick Au shell conditionwhile greater absorption efficiency can be attained when the Aushell is thin, which accounts for the coupling effect of constituentplasmons of the nanoparticle [21]. The theoretical offset conditionsfor the cores of gold–silica–gold multilayer nanoshells wasreported by the Drezek group [20]. Specifically, they demonstratedthat as the core nanoparticle moves away from the center but doesnot touch the nanoshell, the dipolar interaction tends to undergo ared shift due to increased plasmon interaction between the coreand shell.

There is little information in the present literature regardingstudies of the optical properties of structures with more thantwo nanoshells, and as a result there is a limited understandingof the LSPR of these complex nanostructures. It has been suggestedthat multi-shell nanostructures can act as optical condensers tomultiplicatively collect and focus light toward the center of thestructure causing an exponential increase in near-field enhance-ment as the number of metal shells increases [21]. For example,a double shell of hollow nanoparticles made from SiO2/TiO2 exhi-bits a unique light scattering effect within the nanoparticle [22],with multiple scattering events occurring between the inner andouter shell affecting the reflectance of light. In addition, changingthe geometry of the core nanoparticle, notably those with anisotro-pic properties, may further contribute to the limited evidence ofthe effect of resonance frequency. Lastly, compared to nano-spheres, nanorods have two distinct dipole LSPR characteristics

consisting of the transverse mode, originating from the shorter axisoscillation of electrons, and the longitudinal mode, which dependson the aspect ratio (long axis length/short axis length) and is usuallyassociated with a more intense band with a longer wavelength.

In this work, we report the surface plasmon resonance behaviorof a complex multi-shell nanoparticle utilizing galvanic replace-ment [23–24] as shown in Fig. 1A and B. The optical propertiesof the multi-shell nanoparticle were investigated through compar-ison with hollow nanospheres of similar final dimensions (Fig. 1C).We found that the optical response of plasmonic nanoparticlescould be controlled by the presence of multiple nanoshells insidethe outer most nanoshell under the conditions of similar final sizeand percent composition. The controlled optical response was dueto the strong interaction of constituent surface plasmon involvedon the structure of the multi-shell nanoparticle. The observationswere further confirmed by performing simulation of EM fielddistribution and were supported by SERS spectroscopy.

2. Experimental section

2.1. Materials

Gold chloride (HAuCl4�3H2O) was purchased from KOJIMA.Sodium citrate was purchased from YAKURI. Sodium borohydride(NaBH4) and silver nitrate (AgNO3) were purchased from JUNSEI.Sodium iodide (NaI) and ascorbic acid were purchased fromSigma Aldrich. Sodium hydroxide (NaOH) was purchased fromSAMCHUN, and cetyltrimethylammonium bromide (CTAB) waspurchased from Fluka. All involved reagents were dissolved andcontainers were rinsed with distilled water (18.2 MΩ) preparedvia a Milli-Q water purification system.

2.2. Synthesis of gold nanorods

Gold nanorods were prepared by following a similar procedurereported by Chon et al. [25]. First, a seed solution of gold was pre-pared by reducing 20 mmol of HAuCl4 in 10 mL of a 0.1 M CTABsolution with 600 lL of 6.0 mM NaBH4. Second, a growth solutionof Au was prepared by slowly adding 0.1 M ascorbic acid into100 mL of a 0.1 M CTAB solution containing 0.05 mmol HAuCl4and 6.0 mmol AgNO3. The orange color of the gold salt solution dis-appeared, indicating that the Au was reduced. Finally, 120 lL of theseed solution was added to the growth solution under mild

378 J.A.I. Acapulco Jr. et al. / Journal of Colloid and Interface Science 461 (2016) 376–382

stirring. After 12 h without agitation, the final solution was cen-trifuged twice to remove excess CTAB and re-dispersed with water.

2.3. Silver coating of nanocores and subsequent nanoshell

The gold core nanorods and subsequent nanoshells were coatedwith Ag by reducing AgNO3 with ascorbic acid. Briefly, 15 mL of theas-prepared Au nanorods were added in a vial containing 0.05 MCTAB. The solution was stirred for 10 min at 500 rpm to allowhomogenous distribution of Au nanorod and CTAB. Next, 375 lLof 100 mM NaOH and 250 lL of 10 mM ascorbic acid were addedinto the solution under continued stirring. Lastly, 3 mL of 1 mMAgNO3 containing 50 lM iodide was added drop-wise (0.45mL/min)into the solution and kept at room temperature for 3 h.

2.4. Synthesis of nanoshells

Nanoshells were produced by utilizing galvanic replacementbetween coated Ag solid and Au ions. To do so, 100 lL of a20 mM aqueous HAuCl4 solution was diluted up to 10 mL with a0.05 M CTAB solution and added to Ag coated Au nanorodsdispersed in an equivalent volume of CTAB with I�. The reactionthen stirred for 30 min at room temperature. Prior to the synthesisof nanoshells, the solution was purified by centrifugation at4500 rpm for 20 min and re-dispersed with an equivalent volumeof water and CTAB with I�. The number of nanoshells was con-trolled by repeated Ag coating and galvanic replacement. Hollownanospheres were synthesized using a similar procedure utilizingAg nanospheres as a sacrificial template.

2.5. Synthesis of gold nanospheres

The gold nanosphere solution was prepared using a seed medi-ated method. To synthesize the gold seeds (d � 13 nm), 100 mL of a1.0 mM aqueous HAuCl4�3H2O solution was added to 100 mL of tri-ply deionized water (Millipore), which was then boiled underreflux with vigorously stirred. Next, 10 mL of a 38.8 mM aqueoussolution of sodium citrate was added. After 20 min, the solutionwas cooled down to room temperature and confirmed by checkingfor the presence of a peak at 518 nm in the corresponding UV–Visspectra. The growth solution was prepared by mixing 167 mL ofH2O, 4 mL of 20 mM HAuCl4, and 400 lL of 10 mM AgNO3.Thegrowth solutions were then stirred for 10 min. Next, a specificamount of gold seeds were added to the growth solutions followedby drop-wise addition of 30 mL 0.1 M ascorbic acid. After thereaction was completed, the gold nanospheres were purified bycentrifugation and re-dispersed in H2O.

2.6. Synthesis of silver nanospheres

Solid silver nanospheres were synthesized by a seed-mediatedmethod described in our previous work [6]. In a typical procedure,a 20 mL glass vial containing 2 mL of 0.25 mM AgNO3 and 18 mL of0.1 M CTAB was stirred followed by the addition of 600 lL of10 mM NaBH4 cooled at 4 �C in a refrigerator for 15 min. The solu-tion was then stirred for 2 min and kept in a 28 �C oven for 1–2 huntil the color of the solution shifted from colorless to light yellow,which was indicative of the synthesis of Ag seeds. A round flaskcontaining 200 mL of 0.1 M CTAB, 5 mL of 0.01 M aqueous AgNO3,10 mL of 0.1 M aqueous ascorbic acid, and 2 mL of 0.1 M aqueousNaOH solution was then mixed. Finally, a desired amount ofas-prepared Ag seeds were added to yield the desired final sizeof Ag nanospheres.

2.7. Simulation method

Simulations were performed using a finite difference timedomain (FDTD) solver (CST Microwave Studio, CST, Framingham,MA, USA). Tetrahedral meshing was applied to extract the precisesimulation of field distribution. We utilized the derived Drudedielectric function [26], which was fitted to a particular frequencyrange from the Johnson and Christy [27] bulk dielectric data [28].The electric field component of the incident light was alignedparallel to the x-direction while the magnetic field componentwas aligned parallel to the y-direction. E-field and H-field monitorswere utilized to record the field distribution at the desiredwavelength.

2.8. Instrumentation

A kd Scientific syringe pump was used to control the rate ofaddition during Ag coating. The synthesized nanostructures werecharacterized using a JEM-2100F for high-resolution transmissionelectron microscopy (TEM) images, high angle annular dark-fieldscanning TEM (HAADF-STEM), and energy dispersive spectroscopy(EDS) line profile. A JEOL 7000F and JEOL 7600F were used toobtain field emission scanning electron microscopy (SEM) imagesand EDS elemental percentages. S-3100 Scinco and UV-3600Shimadzu spectrophotometers were used to acquire UV–vis–NIRextinction spectra. Raman spectroscopic studies were performedusing a micro-Raman set-up equipped with an Ar ion laser(Renishaw, Inc., New Mills, UK).

3. Results and discussion

As shown in Fig. 1, we utilized different shapes of Au nanopar-ticles as cores to synthesize complex multishell nanoparticles. Agoverlayer coating and subsequent galvanic replacement wereharnessed to control the number of layers and thickness of shells.Typically, the Au nanorods with a length (L) of 76.7 ± 4.1 nm wereconverted to confined Au nanorods with a single shell (thickness ofshell is 3.6 ± 0.3 nm) as shown in panels A&B of Fig. 2, the measure-ments for which were obtained using high angle annular darkfield scanning transmission microscopy (HAADF-STEM). It is note-worthy that the shell thickness and the distance between the coreand the shell could be tailored by adjusting the thickness of the Agoverlayer. Specifically, a thicker Ag overlayer coating led to thethicker shell and longer separation between cores and shells. Thiswas due to the Kirkendall effect, as well as epitaxial Au growth onthe Ag layer, through which Ag is continuously oxidized from theinner layer, dissolved in the solution, and finally reduced anddeposited on the outer Ag layer. Usually, complete replacementof Ag layer with Au ions is not achievable without affecting themorphology and instead results in the formation of an Au/Ag alloy.It has been suggested previously that alloy nanoshells are morefavorable due to the metal–metal bonding energy of Au to Ag(229 kJ mol�1), which is higher than that of Au to Au (226 kJ mol�1)[28]. The bimetallic state of the Ag/Au nanoshell was confirmedthrough line mapping energy dispersive spectroscopy (EDS) analy-sis (Supplementary information Fig. S1). Multi-shell nanostruc-tures were achieved by repeated Ag coating and galvanicreplacement (panel C&D for double and triple layers, respectively).Interestingly, the average aspect ratios decreased significantlyfrom that of the original nanorods (2.0) to 1.7 (for mono-), 1.5(for double-), and 1.4 (for triple-nanoshells). The central Au nanor-ods maintained their original morphology regardless of the totalnumber of replacement reactions.

As the aspect ratio of the multi-shell nanostructure decreased,the corresponding UV–visible and near-IR spectra exhibited unique

Fig. 2. HAADF-STEM images of (A) Au nanorod with increasing number ofnanoshells from (B) single, (C) double to (D) triple nanostructures. UV–vis–NIR(E) experimental and (F) DDA calculation spectra of corresponding nanostructures.

J.A.I. Acapulco Jr. et al. / Journal of Colloid and Interface Science 461 (2016) 376–382 379

surface plasmon resonance features (panel E in Fig. 2). Specifically,the original nanorods exhibited a transverse dipole mode centeredat 520 nm and a longitudinal dipole mode at 642 nm (black trace)that are dependent on the aspect ratio. Furthermore, as the numberof outer shells increased, the transverse mode gradually becamered-shifted and the longitudinal mode shifted to a shorter wave-length. In addition, when the aspect ratio reached 1.4 with threeshells, the two well-separated peaks merged to one peak centeredat 618 nm with a shoulder peak at the shorter wavelength. Thisobservation was not surprising given that the two distinct opticalfeatures originating from the anisotropic shape will eventuallyconverge toward the profile of an isotropic Au nanoparticle(i.e. spherical Au nanoparticles). After coating with a forth layer,the average aspect ratio reached 1.1, which is indicative of anevolution of the shape from elongated nanorods to isotropicsemi-spherical nanoparticles. By considering the correspondingouter physical dimensions and assuming as solid nanoparticles,we carried out discrete dipole approximation (DDA) theoreticalcalculation and the results are plotted in panel F in Fig. 2. In gen-eral, the theoretical calculation results are in a good agreementwith experimental observation, showing the convergence of trans-verse and longitudinal dipole modes as the aspect ratio decreases.

HAADF-STEM images revealed the morphology of the nanopar-ticles after the formation of a fourth shell, indicating a high homo-geneity in terms of the size and number of shells including cores asshown in panel A of Fig. 3. The corresponding UV–visible and near-IR spectra exhibited a symmetric single band centered at 597 nm(black solid trace in panel D). For comparison purposes, the dashed

black traces represent the surface plasmon profile of the originalAu nanorods. In order to analyze the observed optical features,we synthesized two other comparable nanoparticles consisting ofa multi-shell nanosphere with the same number of nanoshellsand bimetallic Ag/Au of hollow nanosphere. In Fig. 3, panels Athrough C show the HAADF-STEM images of the synthesizedmulti-shell nanorod, multi-shell nanosphere, and hollownanosphere with comparable final average diameter of 105.3(±5.4) nm, 111.1(±7.2) nm, and 108.8(±4.3) nm, respectively. TheUV–vis–NIR spectra of each nanostructure are shown in panel Dcorresponding to black, red, and blue solid lines, respectively,while the dashed lines depict the spectrum of the starting materialused to synthesize the corresponding nanostructures. It should benoted here that the higher order mode of LSPR started to appearwhen the size was significantly increased in addition to the dipolemode. This was observed from the solid Ag nanosphere (dashedblue line), which was the starting material for the hollow nano-sphere without inner nanostructures. The two peaks present forthe hollow nanosphere (blue solid line in panel D) could beassigned to the quadrupole and dipole modes centered at 528and 643 nm, respectively. Specifically, the quadrupole modeoriginates from phase retardation of oscillating surface electrons;however, it was not effectively operative with complexmulti-shell nanostructures and did not exhibit any clear featuresassociated with a higher LSPR mode for multi-shell nanostructurewith nanorods or nanosphere cores. The percent composition forall nanostructures was obtained under EDS analysis (lower leftinset in each HAADF-STEM images), which indicated that theiratomic compositions were very similar to each other.

Although each of the three nanoparticle types was very similarin shape, composition, and size, there were distinctive differenceswith respect to quadrupole mode. Specifically, when there wasno core, the quadrupole mode was observable with hollow nano-spheres. In contrast, when there was either a spherical or elon-gated core, the quadrupole mode was significantly suppressed.Based on these observations, we deduced a mechanism involvingthe suppression of the higher order quadrupole mode by themulti-shell nanostructure, illustrated figuratively in Fig. 4. Specifi-cally, at conditions where a higher order mode appears for solidnanostructures, the retardation of LSPR for a solid nanostructurefollowed the surface charge distribution shown in panel A. In thecase of hollow nanostructures (panel B), the quadrupole moderetains to be observable. Previously, it has been reported that theinner nanostructure affects the optical response in gold–silica–goldmultilayer nanoshells [19,29], indicating that the LSPR of both theinner and outer nanostructure exhibits a coupling effect. For multi-shell nanostructures (panel C), contribution of individual LSPR ofnanostructures inside the outer most shell should be realized; itis activated by a coupling effect through the penetrating lightand respective difference in charge distribution. It is noteworthythat only a dipole mode occurs for the inner nanostructures, sincethe condition for the appearance of quadrupole mode was not sat-isfied due to their insufficient size. Indeed, penetrated light willonly induce dipole oscillation of the inner shells, and the dipolewill subsequently induce an opposite charge distribution on thenearby outer shell, which in turn would suppress the quadrupolemode and enhance the appearance of the dipole mode. It is note-worthy that the distance between core and shell will be importantto induce the image charge as described in the scheme. The longerthe separation distance would induce the less degree of the imagecharge. Under the given experimental conditions (gap distance�ca. 3 nm), we could not observe the quadruple mode of outershell in the case of multishell nanoparticles.

To further confirm the suppression of quadrupole mode, theo-retical simulations for EM field distribution by finite differencetime domain (FDTD) were performed by adopting the measured

Fig. 3. HAADF-STEM images of (A) nanorod and (B) nanosphere cores of quadruple nanoshells while (C) Ag/Au hollow bimetallic nanosphere having comparable final shape,composition and size. Inset shows the EDS composition analysis. (D) UV–vis–near IR extinction spectra corresponding to the synthesized nanostructures shown fromHAADF-STEM images black (A), red (B), and blue line (C) while the dashed lines are assigned to their starting material. (For interpretation of the references to colour in thisfigure legend, the reader is referred to the web version of this article.)

Fig. 4. Deduced charge distribution of (A) solid, (B) hollow, and mechanism of thedamping of quadrupole mode occurring in a (C) multi-shell nanoparticle.

380 J.A.I. Acapulco Jr. et al. / Journal of Colloid and Interface Science 461 (2016) 376–382

dimensions from the HAADF-STEM images, the results of which arerepresented in Fig. 5. Each nanostructure was irradiated under twowavelength of incident light for dipole and quadrupole mode nearfield enhancement visualization, left and right panel respectively.Due to the similar final size of all the nanostructures, the quadru-pole plasmon wavelength of the hollow nanosphere was used bothfor the multi-shell nanorod and nanosphere. However, there wasno observation of the quadrupole mode feature for both multi-shell nanostructures. In addition, the report demonstrating thatlight is able to penetrate the outer most nanoshells suggests thatlight can be confined to the hollow space between the inner andouter nanostructure using an enhanced EM field [30]. This providesadditional insights regarding surface plasmonic shell-to-shellinteractions showing that ‘‘hot-spots” are generated at the gap hol-low spaces of the multi-shell nanoparticles. Among the multi-shellnanoparticles, the multi-shell nanorod exhibited the strongestenhancement due to the strong field enhancement on its tips,

which was associated with the longitudinal dipole mode thatwas more concentrated compared to the dipole mode of the nano-sphere. The short distance between shells allows the strong EMfield generation at the gaps.

Experimentally, it is difficult to confirm the light penetrationdepth with surface plasmon profiles obtained from UV–visible,near-IR experiments. Thus, in order to qualitatively estimate lightpenetration into shells, we performed surface enhanced Ramanscattering (SERS) measurements with the three nanostructures. Itis known that plasmonically active metal nanoparticles exhibitsstrong confinement of electromagnetic field around its surfaceleading to a significant yield of SERS intensity [31].

Typical SERS spectra of benzenethiol adsorbed on a multi-shellnanorod (black line), multi-shell nanosphere (red line), and hollownanosphere (blue line) under 633 nm laser excitation are shown inFig. 6. We found that the multi-shell nanostructures (nanorodand nanosphere as core) exhibited a stronger SERS intensitythan the hollow nanosphere observed from all bands associatedwith the SERS signal characteristic of benzenethiol’s S–H bending +in-plane ring deformation (995 cm�1), in-plane ring deformation(1020 cm�1), C–S stretching + in-plane ring deformation(1072 cm�1), and C–C stretching (1572 cm�1) [32]. Among thenanostructures we investigated, the multi-shell nanorod showedthe strongest SERS signals. This finding was consistent with thewell-known lightening rod effect with elongated nanostructures,whereby large fields become more concentrated on highly curvedtips as compared to dull surfaces. The LSPR of the inner core andshells affects the overall LSPR of multi-shell nanostructures withrespect to EM field enhancement. Taken together, these results also

Fig. 5. Local EM field distribution of (A) multi-shell nanorod, (B) multi-shellnanosphere, and (C) hollow nanosphere simulated by finite-deference time domain(FDTD). Left panel shows excitation at dipole mode of corresponding nanoparticlewhile right panel shows incident wavelength excitation.

Fig. 6. SERS spectra of benzenethiol adsorbed on (black line) multi-shell nanorod,(red line) multi-shell nanosphere, and (blue line) hollow nanosphere. (For inter-pretation of the references to colour in this figure legend, the reader is referred tothe web version of this article.)

J.A.I. Acapulco Jr. et al. / Journal of Colloid and Interface Science 461 (2016) 376–382 381

suggested that multi-shell nanostructures can serve as optical con-densers that collect and focus light to increase the EM near-field.

4. Conclusions

We investigated the surface plasmon resonance behavior ofmulti-shell nanoparticles synthesized through repeated Ag layer

deposition with galvanic replacement. A reference sample ofbimetallic Ag/Au hollow nanosphere with comparable final size,composition, and shape was used to quantify the effect of nano-shells located inside the outer most nanoshell by UV–vis–NIR spec-trophotometry. Due to the gradual change from anisotropic toisotropic geometry of the Au nanorod as a core for multi-shellnanoparticle, we were also able to investigate the effect of the coreshape. The presence of inner nanoshells and geometrical shape ofthe core showed the damping of quadrupole mode at 100 nm sizeregime leading to cancellation of the higher order mode whetherspherical or elongated structure. Through the strong influence ofdipole mode of the inner nanostructures, the higher order modeof the outer most nanoshell was dampened, thereby offeringanother strategy to control the optical response of the plasmonicnanostructure. These findings were further confirmed by theoreti-cal calculation of electromagnetic (EM) field distribution byfinite-difference time domain (FDTD) that aided the visualizationof corresponding plasmonmode features. Surface-enhanced Ramanscattering (SERS) spectroscopy analysis was in a good agreementwith obtained simulated data. In further conclusion, we believe thatthis work will contribute to a better understanding of plasmonicmaterials, especially in multi-shell nanostructure systems thatcan be applied in photonics, sensing devices, and catalysts.

Acknowledgements

This work was supported by the National Research Foundationof Korea (National Leading Research Lab: 2011-0027911) and theKorea government Ministry of Trade, Industry and Energy andthe Agency for Defense Development through Chemical and Bio-logical Defense Research Center (UD140017ID).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.jcis.2015.09.040.

References

[1] P. Reineck, G.P. Lee, D. Brick, M. Karg, P. Mulvaney, U. Bach, Adv. Mater. 24(2012) 4750–4755.

[2] S.H. Yoo, L. Liu, T.W. Ku, S.C. Hong, D. Whang, S. Park, Appl. Phys. Lett. 103(2013) 143101–143104.

[3] Z. Fang, Y. Wang, Z. Liu, A. Schlather, P.M. Ajayan, F.H.L. Koppens, P.Nordlander, N.J. Halas, ACS Nano 6 (2012) 10222–10228.

[4] J.N. Anker, W.P. Hall, O. Lyandres, N.C. Shah, J. Zhao, R.P. Van Duyne, Nature 7(2008) 442–453.

[5] O.R. Bolduc, J.F. Masson, Anal. Chem. 83 (2011) 8057–8062.[6] Y. Choi, J.-H. Choi, L. Liu, B.K. Oh, S. Park, Chem. Mater. 25 (2013) 919–926.[7] X. Lu, M. Rycenga, S.E. Skrabalak, B. Wiley, Y. Xia, Annu. Rev. Phys. Chem. 60

(2009) 167–192.[8] R. Jin, Y. Cao, C.A. Mirkin, K.L. Kelly, G.C. Schatz, J.G. Zheng, Science 294 (2001)

1901–1903.[9] J. Theiss, P. Pavaskar, P.M. Echternach, R.E. Muller, S.B. Cronin, Nano Lett. 10

(2010) 2749–2754.[10] A.S. Kumbhar, M.K. Kinnan, G. Chumanov, J. Am. Chem. Soc. 127 (2005)

12444–12445.[11] E. Ringe, J.M. McMahon, K. Sohn, C. Cobley, Y. Xia, J. Huang, G.C. Schatz, L.D.

Marks, R.P. Van Duyne, J. Phys. Chem. C 114 (2010) 12511–12516.[12] K.L. Kelly, E. Coronado, L.L. Zhao, G.C. Schatz, J. Phys. Chem. B 107 (2003) 668–

677.[13] S. Hong, K.L. Shuford, S. Park, Chem. Mater. 23 (2011) 2011–2013.[14] J.H. Yoon, J. Lim, S. Yoon, ACS Nano 6 (2012) 7199–7208.[15] S.J. Oldenburg, R.D. Averitt, S.L. Westcott, N.J. Halas, Chem. Phys. Lett. 288

(1998) 243–247.[16] J.B. Lassiter, M.W. Knight, N.A. Mirin, N.J. Halas, Nano Lett. 9 (2009) 4326–

4332.[17] Y. Sun, B.T. Mayers, Y. Xia, Nano Lett. 2 (2002) 481–485.[18] E. González, J. Arbiol, V.F. Puntes, Science 334 (2011) 1377–1380.[19] R. Bardhan, S. Mukherjee, N.A. Mirin, S.D. Levit, P. Norlander, N.J. Halas, J. Phys.

Chem. C 114 (2010) 7378–7383.[20] Y. Hu, S.J. Noelck, R.A. Drezek, ACS Nano 3 (2010) 1521–1528.[21] H. Xu, Phys. Rev. B 72 (2005) 073405.

382 J.A.I. Acapulco Jr. et al. / Journal of Colloid and Interface Science 461 (2016) 376–382

[22] J. Lee, S.H. Hwang, J. Yun, J. Jang, A.C.S. Appl, Mater. Interfaces 6 (2014) 15420–15426.

[23] Y. Sun, B. Mayers, Y. Xia, Adv. Mater. 15 (2003) 641–646.[24] S. Hong, J.A. Acapulco Jr., H.Y. Jang, S. Park, Chem. Mater. 26 (2014) 3618–3623.[25] P. Zijlstra, C. Bullen, J.W.M. Chon, M. Gu, J. Phys. Chem. B 110 (2006) 19315–

19318.[26] C. Oubre, P. Nordlander, J. Phys. Chem. B 46 (2004) 17740–17747.

[27] P.B. Johnson, R.W. Christy, Phys. Rev. B 6 (1972) 4370–4379.[28] X. Xia, Y. Wang, A. Ruditskiy, Y. Xia, Adv. Mater. 25 (2013) 6313–6333.[29] Y. Hu, R.C. Fleming, R.A. Drezek, Opt. Express 16 (2008) 19579–19591.[30] D.K. Lim, K.S. Jeon, J.H. Hwang, H. Kim, S. Kwon, Y.D. Suh, J.M. Nam, Nat.

Nanotechnol. 6 (2011) 452–460.[31] M.G. Blaber, G.C. Schatz, Chem. Commun. (2011) 3769–3771.[32] H.Y. Jung, Y. Park, S. Park, S.K. Kim, Anal. Chim. Acta 602 (2007) 236–243.