Plasmonic Nanoparticles-Liquid Crystal Composites†

Saumyakanti Khatua, Pramit Manna, Wei-Shun Chang, Alexei Tcherniak, Eric Friedlander,Eugene R. Zubarev,* and Stephan Link*Department of Chemistry, Rice UniVersity, 6100 Main Street, Houston, Texas 77005

ReceiVed: August 16, 2009; ReVised Manuscript ReceiVed: NoVember 20, 2009

We report on the plasmonic properties of 6 nm gold nanoparticles that form highly stable solutions in thenematic liquid crystal 4-cyano-4-n-pentylbiphenyl (5CB). The nanoparticles were covalently functionalizedwith 4-sulfanylphenyl-4-[4-(octyloxy)phenyl]benzoate, which resembles the structure of the 5CB molecules.The solubility of these nanoparticles in 5CB was significantly higher than that of conventional alkanethiol-terminated nanoparticles. An 8 nm shift of the surface plasmon resonance was observed when the goldnanoparticles were dissolved in the nematic phase of 5CB, as compared to the isotropic solution in methylenechloride. Good agreement of the experimental surface plasmon resonance shift with Mie calculations usingan adjusted dielectric function for a reduced electron mean free path in small nanoparticles confirmed thatthe gold nanoparticles are solvated by the liquid crystal molecules. The stability of this composite was verifiedby repeated temperature cycling between the isotropic and nematic phases. We also investigated the nematic-to-isotropic phase transition temperature and the threshold voltage for the Freedericksz transition in gold-nanoparticle-doped and undoped liquid crystal devices.

Introduction

Noble metal nanoparticles are of great interest because oftheir intense tunable absorption and scattering resonances causedby collective oscillations of the conduction band electrons, whichare known as surface plasmons.1 The surface plasmon resonancemaximum is very sensitive to the dielectric constant of thesurrounding media.2-4 The large anisotropy of the liquid crystalrefractive index is therefore ideally suited for tuning of theplasmon resonance by electric field-induced switching of theliquid crystal director orientation. This has been demonstratedfor thin metal films,5,6 gold nanoparticle arrays,7 metallic holearrays,8 gold nanorods,9,10 and single gold nanospheres.11

Conversely, the shift in the surface plasmon resonance has beenexploited to measure the local orientation of the liquid crystalmolecules in close proximity to metal nanostructures.12-14 Forall these studies, the metal nanoparticles were first immobilizedon a solid surface and then covered by the liquid crystal solvent.However, it is also desirable to switch the plasmonic responseof anisotropic nanostructures by inducing an electric field-dependent nanoparticle reorientation in the liquid crystal matrix.To realize such a composite material, it is first necessary touniformly disperse metal nanoparticles within the bulk of anematic liquid crystal.

Composite materials consisting of liquid crystals doped withnanoparticles have indeed attracted much scientific and tech-nological interest, but mainly because the incorporation ofnanomaterials enhances the electro-optical properties of theliquid crystal itself. For example, doping with MgO and SiO2nanoparticles has been reported to decrease the threshold voltagefor reorientating the nematic liquid crystal director along anapplied electric field, known as the Freedericksz transition, dueto a decrease in order parameter.15 Suspending ferromagneticSn2P2S6 nanoparticles in a nematic liquid crystal host was foundto enhance the dielectric anisotropy of the liquid crystal, which

also led to a lower threshold voltage.16 Within a smallcomposition gap, multiwall carbon nanotubes increase thenematic-to-isotropic phase transition temperature because of anenhanced ordering of the liquid crystal molecules along thecarbon nanotubes.17

On the other hand, solvent-induced anisotropic alignment ofcarbon nanotubes18,19 and rod-like polymer chains20,21 by ther-motropic liquid crystals has been shown to create high degreesof solute order that scales with the relative difference in sizebetween the solvent and the solute molecules.22 Because thealignment direction of the rod-like solutes is typically parallelto the nematic director, application of an external electric fieldcan be used to tune the anisotropic properties of the solutemolecules through a solvent-directed reorientation.19,23 Ananisotropic ordering of 2-3 nm gold nanocrystals into chain-like aggregates by discotic liquid crystals was also observedand greatly enhanced the conductivity of the composite materialindependent of the phase.24,25 The solubility of gold nanocrystalsin this size range can be enhanced by functionalizing the particlesurface with ligands that contain mesogenic units resemblingthe chemical structure of the liquid crystal solvent.26-31 Goldnanoparticles of 2-3 nm show only a very weak and broadsurface plasmon resonance because the strength of the plasmonoscillation strongly depends on the total number of electrons.1,28

However, the optical properties of larger gold nanoparticles withstrong plasmon resonances solvated in thermotropic liquidcrystals have not been studied in detail.

Here, we report on the plasmonic properties of 6 nm goldnanoparticles in 4-cyano-4-n-pentylbiphenyl (5CB). The com-monly used thermotropic liquid crystal 5CB forms a nematicphase in which the main molecular axis of each liquid crystalmolecule is oriented along a common direction, called thenematic director. An increase in temperature causes a transitionto the isotropic phase in which any orientational order is lost,whereas applying an electric field allows one to realign thenematic director. The solubility of the nanoparticles was greatlyenhanced after the surface capping material was changed to

† Part of the “Martin Moskovits Festschrift”.* Corresponding authors. E-mails: zubarev@rice.edu, slink@rice.edu.

J. Phys. Chem. C 2010, 114, 7251–7257 7251

10.1021/jp907923v 2010 American Chemical SocietyPublished on Web 12/11/2009

ligands that chemically resemble the liquid crystal molecules.Modeling of the surface plasmon resonance using Mie calcula-tions corrected for small nanoparticle size effects agreed wellwith the experimental spectra showing that the nanoparticlesare homogeneously solvated in the nematic liquid crystal phase.Optical absorption measurements as a function of temperatureand applied electric field confirmed the homogeneous solvationof the gold nanoparticles.

Experimental Section

General. Gold(III) chloride (HAuCl4 ·3H2O), 1-decanethiol,sodium borohydride, N,N-diisopropylcarbodiimide (DIPC) werepurchased from Aldrich Chemical Co. 4-Mercaptophenol and4′-octyloxybiphenyl-4-carboxylic acid were purchased from TCIAmerica and Alfa-Aesar, respectively. 4-(N,N-Dimethylami-no)pyridinium-4-toluenesulfonate (DPTS) was prepared bymixing saturated tetrahydrofuran (THF) solutions of 4-(N,N-dimethylamino)pyridine and p-toluenesulfonic acid at roomtemperature. Unless otherwise stated, all chemicals were usedwithout any further purification.

Synthesis of 1-Decanethiol-Capped Gold Nanoparticles.Decanethiol-functionalized gold nanoparticles (∼2 nm in di-ameter) were prepared by two-phase synthesis described byBrust and co-workers.32 In a typical synthesis, 120 mg ofgold(III) chloride was dissolved in 10 mL of DI water. In aseparate vial, 730 mg of tetraoctylammonium bromide wasdissolved in 27 mL of toluene. These two solutions were mixedand stirred vigorously until the gold chloride was completelytransferred into the organic layer. To this organic layer, 65 mgof 1-decanethiol was added, and then a freshly prepared aqueoussolution of sodium borohydride (123 mg in 8.5 mL water) wasadded slowly into the mixture. The organic layer turned darkbrown during the reduction. After 5 min of stirring, an excessof methanol was added, and the gold nanoparticles werecentrifuged at 1200 rpm for 5 min. Purified particles weredissolved in THF and kept as a stock solution.

Synthesis of Monodisperse 6 nm Gold Nanoparticles. Tenmilligrams of decanethiol-capped, 2 nm gold particles weredissolved in 3 g of neat 1-decanethiol. The solution wastransferred into an 8 mL vial and capped with a Teflon-coatedcap. The vial was heated to 175 °C and kept at that temperaturefor 1 h. During this annealing process, the solution became darkred, indicating the overall increase in the size of the goldnanoparticles. The solution was then treated with excessmethanol, and the particles were centrifuged at 1200 rpm for 5min. The precipitate containing decanethiol-capped 6 nm goldnanoparticles was dissolved in THF.

Ligand Exchange of 6 nm Gold Nanoparticles. A concen-trated solution was prepared by dissolving 1 g of 4-mercap-tophenol in 2 mL of isopropyl alcohol. Gold nanoparticles (10mg) dissolved in 2 mL of THF were added dropwise withvigorous stirring. The mixture was stirred for 12 h, followedby addition of excess hexane and centrifugation at 1200 rpmfor 5 min. The precipitate was washed several times withmethylene chloride to remove the residual organic ligands. Theresulting material was treated with iodine in deuterated THF,and the supernatant was analyzed by NMR to determine theratio of mercaptophenol and residual decanethiol ligands.According to the integration of NMR signals, the nanoparticlescontained 70% mercaptophenol ligands in the organic shell.

Covalent Attachment of 4′-n-Octyloxybiphenyl-4-carboxy-lic Acid to Mercaptophenol-Functionalized 6 nm GoldNanoparticles. 4′-n-Octyloxybiphenyl-4-carboxylic acid (20mg) was dissolved in 3 mL of methylene chloride, followed by

addition of 10 mg DPTS and 10 drops of DIPC. After 5 min ofstirring at room temperature, the mixture became clear, and aconcentrated dimethylformamide (DMF) solution of nanopar-ticles (15 mg in 0.3 mL of DMF) was added to the reactionmixture. The color of the reaction mixture turned dark red,indicating the covalent coupling of nanoparticles with thecarboxylic acid. After 2 h, DPTS and DMF were removed bymultiple extractions with DI water. Methylene chloride wasevaporated under reduced pressure, and the residue was rinsedseveral times with DMF (poor solvent for the resulting nano-particles). The particles were dissolved in methylene chloride,and the residual DMF was extracted with DI water. Then themethylene chloride was evaporated, and the product wasdissolved in dry THF.

Synthesis of 4-Sulfanylphenyl-4-[4-(octyloxy)phenyl]ben-zoate (SOPB) Thiol. 4′-n-Octyloxybiphenyl-4-carboxylic acid(200 mg) and 387 mg of 4-mercaptophenol (5 equiv) weredissolved in 5 mL of methylene chloride. To this mixture, 200mg of DPTS was added, followed by a dropwise addition of0.5 mL of DIPC. The reaction was monitored by TLC (10 vol% THF in methylene chloride) in which the strongly UV activespot of the carboxylic acid anhydride (Rf ) 1) disappearedgradually, and the product spot appeared (Rf ) 0.6). After thecompletion of the reaction, the mixture was diluted three timeswith methylene chloride, and DPTS was removed by extractingwith DI water. The methylene chloride was evaporated underreduced pressure, and the residue was treated with excessmethanol. The precipitated product was isolated by centrifuga-tion and rinsed several times with methanol. The product wasdried under vacuum to yield 220 mg of a light brown powder(80% isolated yield). 1H NMR (400 MHz, CDCl3): δ 0.9 (t,3H), 1.30-1.35 (m, 11H), 1.80 (m, 2H), 3.5 (s, 1H), 4.02 (t,2H), 6.90 (d, 2H), 7.01 (d, 2H), 7.39 (d, 2H), 7.58 (d, 2H),7.68 (d, 2H), 8.08 (d, 2H).

Preparation of Gold Nanoparticle-Liquid Crystal Com-posites. Gold-nanoparticle-doped liquid crystal composites wereprepared by dissolving the nanoparticles and the nematic liquidcrystal 5CB (TCI America) in methylene chloride as a commonsolvent, followed by its slow evaporation.33 Liquid crystal cellswere fabricated from two indium tin oxide (ITO)-coated glassslides (Delta Technologies) that were separated by a 50-µm-thick, self-adhesive mylar spacer (McMaster-Carr) and sealedwith epoxy glue. PVA alignment layers were spin-cast onto theslides at 3500 rpm and exposed to unidirectional rubbing. Thegold nanoparticle-liquid crystal composite was injected intothe cell by capillary forces. To avoid a nonuniform liquid crystalalignment, both the cell and the composite were heated to atemperature above the nematic-to-isotropic phase transitiontemperature before injecting the composite into the cell.Formation of the nematic liquid crystal phase was confirmedby observing birefringence between crossed polarizers.

Optical Absorption Measurements. Optical absorptionmeasurements were performed with a home-built absorptionspectrometer consisting of a halogen lamp (Thorlabs OSL1) asa light source, a polarizer, a depolarizer, and a fiber-coupledspectrometer (Ocean Optics S1024DWX). Samples were mountedon a rotation stage to allow for convenient alignment of thenematic director with respect to the polarization axis. Spectrawere integrated for 100 ms and averaged 100 times. Becausethe scattering in the nematic phase varied slightly from sampleto sample, we prepared six blank 5CB samples and averagedtheir spectra to obtain a solvent correction file. The sameprocedure was also applied for measurements in the isotropic5CB phase. The isotropic phase was prepared by conductive

7252 J. Phys. Chem. C, Vol. 114, No. 16, 2010 Khatua et al.

heating with a variable DC voltage (2-5 V) from a powersupply (MPJA 14602PS). The temperature was monitored witha thermocouple and digital thermometer (Omega HH26K) andwas equilibrated for 10 min before taking a measurement.Homeotropic alignment (i.e., alignment perpendicular to theglass surfaces) was induced by applying an AC electric field(sinusoidal wave with frequency of 1 kHz and maximumamplitude of 50 V) across the two ITO coated slides using afunction generator (Stanford Research System DS335) andvoltage amplifier (Harrison 6824A). For measurements of thenematic-to-isotropic phase transition temperature, TIN, and thethreshold voltage for the Freedericksz transition, the sampleswere mounted such that the nematic director made a 45° anglewith the polarization axis of the incident light and were placedbetween crossed polarizers instead of the polarizer-depolarizercombination. A second fiber-coupled spectrometer (OceanOptics USB4000) was used to measure the relative transmit-tance. The relative transmittance was defined as the transmit-tance as a function of temperature or voltage integrated over aspectral range where the plasmon resonance gave only a minorcontribution (>625 nm) minus the transmittance at the maximumtemperature or voltage applied.

Results and Discussion

To study the plasmonic properties of liquid crystal-goldnanoparticle composites, we focused on the preparation ofrelatively large and near-monodisperse nanoparticles (5-6 nm)that exhibit a significantly larger and narrower absorption inthe visible range when compared with their smaller polydisperseanalogues (1-3 nm).32 Although the synthesis of alkanethiol-capped monodisperse gold nanoparticles is known,34 there areno examples of monodisperse gold particles coated withfunctional organic molecules. Here, we developed a two-stepsynthetic route that enabled us to transform decanethiol-coatedgold nanoparticles (Figure 1) into their functional analoguescarrying alkyloxybiphenyl thiols that resemble the structure ofthe nematic liquid crystal 5CB.

In the first step, outlined in Scheme 1, monodispersedecanethiol-coated gold nanoparticles synthesized by a literatureprocedure34 were treated with a concentrated isopropyl alcoholsolution of a low molar mass functional thiol 4-mercaptophenol.We found that under these conditions, nearly 70% of the

decanethiol ligands were exchanged for mercaptophenol mol-ecules, as determined by 1H NMR of the organic residue thatformed after complete dissolution of the gold cores in thepresence of iodine. The resulting nanoparticles became poorlysoluble in organic solvents, but retained their narrow sizedistribution, which was similar to that of the starting material(Figure 1 and Figure S1 of the Supporting Information).

In the second step, the mercaptophenol-coated gold nano-particles were reacted with 4′-octyloxybiphenyl-4-carboxylicacid under mild esterification conditions using DIPC and DPTSas coupling agents. During the coupling reaction, the nanopar-ticles gradually dissolved and formed a stable, homogeneoussolution in methylene chloride, which was indicative of suc-cessful covalent attachment of the liquid-crystal-like 4′-octyl-oxybiphenyl-4-carboxylic acid ligands. The reaction mixture waspurified according to standard procedures published elsewhere,35

and the 6 nm nanoparticles functionalized with SOPB ligandswere isolated in the form of a dark red powder with an 80%yield. The purity of the product was confirmed by a combinationof TLC, NMR, and TGA using standard literature protocols (seethe Supporting Information).36 Figure 2 shows a moleculargraphics representation of a polyhedral gold nanoparticlefunctionalized by SOPB ligands and, for comparison, a moleculeof 5CB.

The absorption spectra of decanethiol (blue) and SOPB-functionalized (red) gold nanoparticles in 5CB are comparedin the inset of Figure 3. The spectra were taken with themaximum possible amount of gold nanoparticles dissolved inthe same volume of liquid crystal solvent. The solubility of thedecanethiol-coated nanoparticles was very poor because nodistinct plasmon peak is observed, suggesting that the nano-particles are actually not solvated inside the 5CB but, rather,aggregated at the glass interface. On the other hand, thesolubility of SOPB-functionalized gold nanoparticles was muchenhanced with a calculated maximum gold nanoparticle con-centration of 0.2 wt %. The calculation is based on theexperimental optical density (OD) at the surface plasmonresonance maximum and the known molar extinction coef-ficient,37 but was also verified experimentally for these nano-particles by preparing solutions from a measured amount of solidSOPB-functionalized gold nanoparticle powder. It should beemphasized that the observed optical densities of ∼0.1 cor-respond to a path length of only 50 µm. For a 1 mm path lengthcuvette, the OD of these nanoparticle-liquid crystal compositeswould be 2, or 99% of the incident light absorbed at the plasmonresonance maximum.

The increase in solubility can be explained by the structuralsimilarity of the biphenyl-containing ligands to the 5CB solvent.Complete nematic solvation is further confirmed by the 8 nmred shift of the surface plasmon resonance for SOPB-coated 6nm nanoparticles in 5CB compared to the isotropic methylenechloride solution (Figure 3). The shift of the plasmon resonancemaximum is caused by the increase in solvent refractiveindex,2-4 indicating the successful incorporation of the goldnanoparticles inside the nematic liquid crystal solvent. Forcomparison, a spectral shift of similar magnitude has beenreported for 5 nm alkanethiolate monolayer-protected goldclusters when the solvent refractive index was varied from 1.33to 1.55.3 A spectral shift of the plasmon resonance due to particleaggregation can be excluded as a possible explanation herebecause no distinct shoulder was observed on the red edge ofthe spectrum.38,39 In addition, we confirmed that the width ofthe surface plasmon resonance did not broaden as a function of

Figure 1. Transmission electron microscopy image of decanethiol-coated gold nanoparticles. The inset shows the nanoparticle sizedistribution.

Plasmonic Nanoparticles-Liquid Crystal Composites J. Phys. Chem. C, Vol. 114, No. 16, 2010 7253

nanoparticle concentration, indicating no aggregation in thesolution or at the glass interface.

The absorption spectra shown in Figure 3 were recorded withthe nematic director aligned parallel to the polarization axisof the incident light. Rotating the sample 90° to record spectrawith a perpendicular orientation of the nematic director as wellas electrically switching the director to a homeotropic orientationhad no effects on the absorption of the spherical nanoparticles.

Furthermore, as illustrated in Figure 4A, the absorption spectrumof SOPB-functionalized gold nanoparticles in 5CB showed onlya very small shift of

particles are homogeneously solvated by the 5CB molecules. Itis important to point out that, as shown by these results, theanisotropy of the liquid crystal refractive index in the nematicphase is not experienced by spherical particles that are isotro-pically surrounded by the liquid crystal, in contrast to previousexperiments in which the nanoparticles were immobilized on asubstrate.7,9,11 For the latter, the presence of a surface and theabsence of free nanoparticle motion hence cause a break insymmetry and allow for the refractive index anisotropy to beexperienced by the plasmon resonance.

To further confirm these conclusions, we calculated theabsorption spectra of gold nanoparticles in methylene chlorideand 5CB using Mie theory.1 Because of an enhanced electron-surface scattering in metal nanoparticles smaller than the electronmean free path, the surface plasmon resonance is stronglydamped for the 6 nm gold nanoparticles studied here.1,40,41 Toaccount for the increased plasmon line width, we included inthe calculations the effect of a reduced mean free path bymodifying the damping constant, γ, of the free electroncontribution to the dielectric function according to γ ) γbulk +AVF/R.1,40 γbulk represents electron scattering processes for thebulk metal, whereas the second term accounts for electron-surface scattering, where VF is the Fermi velocity of theelectrons, R is the radius of the nanoparticles, and A is adimensionless fitting parameter, which varies between 0 and 1and describes the nature of the electron-surface scattering.1 Thisapproach has been successfully used to explain experimentallyobserved surface plasmon resonance spectra of small goldnanoparticles.4,42

We also obtained very good agreement between the experi-mental (solid lines) and calculated spectra (dashed lines) for 6nm gold nanoparticles in methylene chloride (green) and 5CB(red), as shown in Figure 3. For the refractive index of thesolvent, we used 1.42 and 1.60 for methylene chloride and 5CB,respectively. Because of the spherical symmetry of the goldnanoparticles and consistent with the discussions above, weassumed an isotropic average of the refractive index of 5CBaccording to 〈niso〉 ) (ne + 2no)/3.43 The dielectric function ofthe gold was computed using the tabulated values reported byJohnson and Christy.44 For the A parameter, we found that avalue of 0.25 produced the best match with the experimentalspectra, in excellent agreement with recent single nanoparticlestudies.45,46 The calculations confirm that the gold nanoparticlesare homogeneously solvated. In addition, the calculations alsoindicate that the cosolvent methylene chloride used in thepreparation of the nanoparticle-liquid crystal composites (seethe Experimental Section) has been completely evaporated. Tofurther verify the absence of residual methylene chloride thatmight stabilize the solvation of the nanoparticles in the nematicphase, we also dissolved a dried powder of SOPB-functionalized6 nm gold nanoparticles directly in 5CB at similar concentra-tions. The resulting spectra were identical to the ones obtainedby using methylene chloride as a cosolvent.

In addition, we tested the stability of the gold nanoparticle-5CB composite by repeated temperature cycling between thenematic and isotropic phases. In Figure 4B, the peak opticaldensities at the plasmon resonance are plotted for nanoparticle-liquid crystal composites that were consecutively cycled fourtimes between the nematic and isotropic phases at T ) 25 and35 °C, respectively. The error bars were calculated from threeindependent measurements from three different samples con-taining the same amount of SOPB-coated gold nanoparticles.The measured optical densities remained unchanged during thetemperature cycling, confirming the high stability of the com-posite. On the other hand, an unstable mixture resulted in agradual decrease in absorbance upon temperature cycling, asobserved for the decanethiol-coated gold nanoparticles. Themaxima of the plasmon resonance for the same temperature-dependent measurements are shown in Figure 4C. The smallshift of

To investigate the electro-optical properties of gold-nano-particle-doped 5CB devices, we furthermore compared thethreshold voltages for the Freedericksz transition for doped andneat liquid crystal cells. The results are presented in Figure 6and show a decrease of the threshold voltage (Vth) for thenanoparticle-liquid crystal composite by about 0.2 V, whereVth is defined as the voltage for which the transmission of aliquid crystal cell placed between crossed polarizers decreasesby 10% from its maximum value.47 Previous studies onnanoparticle-doped liquid crystal devices also found a decreasein the threshold voltage,15,16 which was explained by a decreasein the order parameter for MgO and SiO2 nanoparticle solutes15

and an enhancement of the liquid crystal dielectric anisotropyin the case of Sn2P2S6 nanoparticles.16 Considering the slightincrease in the nematic-to-isotropic phase transition temperature,an enhancement of the dielectric anisotropy due to metallic 6nm particles is the more likely explanation for the observedtrend. To gain further insight into the thermal and electro-opticalproperties of the liquid crystal solvent in metal-nanoparticle-doped devices, future measurements will be aimed at directlydetermining the solvent order parameter.

Conclusions

We have shown that a two-step synthetic replacement ofdecanethiol with SOPB functional ligands renders 6 nm goldnanoparticles soluble in the nematic phase of 5CB. Wedemonstrated spectroscopically that the nanoparticles are,indeed, dissolved rather than dispersed and explain the enhancedsolubility by the chemical similarity between the surface cappingmaterial and the liquid crystal solvent. The stability of thiscomposite was further verified by cycling between the isotropicand nematic phases of the liquid crystals. In addition, we founda small increase in the nematic-to-isotropic phase transition

temperature of 0.4 °C and a desirable decrease in the thresholdvoltage by almost 25% for an electric-field-induced homeotropicdirector alignment when a maximum nanoparticle concentrationof 0.2 wt % was added to 5CB. These results present animportant step toward dissolving nanoparticles with size andshape tunable plasmonic properties inside liquid crystal solvents.

Acknowledgment. S.L. thanks the Robert A. Welch Founda-tion (C-1664) and 3M for a Nontenured Faculty Grant. E.R.Z.acknowledges the financial support by NSF (DMR-0547399),the Robert A. Welch Foundation (C-1703), and the Alfred P.Sloan Foundation. W.S.C. acknowledges support from theRichard E. Smalley Institute for a Peter and Ruth Nicholasfellowship.

Supporting Information Available: Characterization of theSOPB-functionalized gold nanoparticles using TEM, TLC, 1HNMR, and TGA. This material is available free of charge viathe Internet at http://pubs.acs.org.

References and Notes

(1) Kreibig, U.; Vollmer, M., Optical Properties of Metal Clusters;Springer: Berlin, 1995.

(2) Ghosh, S. K.; Nath, S.; Kundu, S.; Esumi, K.; Pal, T. J. Phys. Chem.B 2004, 108, 13963.

(3) Templeton, A. C.; Pietron, J. J.; Murray, R. W.; Mulvaney, P. J.Phys. Chem. B 2000, 104, 564.

(4) Underwood, S.; Mulvaney, P. Langmuir 1994, 10, 3427.(5) Wang, Y. Appl. Phys. Lett. 1995, 67, 2759.(6) Wang, Y.; Russell, S. D.; Shimabukuro, R. L. J. Appl. Phys. 2005,

97, 023708.(7) Kossyrev, P. A.; Yin, A.; Cloutier, S. G.; Cardimona, D. A.; Huang,

D.; Alsing, P. M.; Xu, J. M. Nano Lett. 2005, 5, 1978.(8) Dickson, W.; Wurtz, G. A.; Evans, P. R.; Pollard, R. J.; Zayats,

A. V. Nano Lett. 2008, 8, 281.(9) Chu, K. C.; Chao, C. Y.; Chen, Y. F.; Wu, Y. C.; Chen, C. C.

Appl. Phys. Lett. 2006, 89, 103107.(10) Evans, P. R.; Wurtz, G. A.; Hendren, W. R.; Atkinson, R.; Dickson,

W.; Zayats, A. V.; Pollard, R. J. Appl. Phys. Lett. 2007, 91, 043101.(11) Mueller, J.; Soennichsen, C.; von Poschinger, H.; von Plessen, G.;

Klar, T. A.; Feldmann, J. Appl. Phys. Lett. 2002, 81, 171.(12) Evans, S. D.; Allinson, H.; Boden, N.; Flynn, T. M.; Henderson,

J. R. J. Phys. Chem. B 1997, 101, 2143.(13) Koenig, G. M., Jr.; Meli, M.-V.; Park, J.-S.; de Pablo, J. J.; Abbott,

N. L. Chem. Mater. 2007, 19, 1053.(14) Koenig, G. M., Jr.; Gettelfinger, B. T.; de Pablo, J. J.; Abbott, N. L.

Nano Lett. 2008, 8, 2362.(15) Haraguchi, F.; Inoue, K.; Toshima, N.; Kobayashi, S.; Takatoh, K.

Jpn. J. Appl. Phys., Part 2 2007, 46, L796.(16) Reznikov, Y.; Buchnev, O.; Tereshchenko, O.; Reshetnyak, V.;

Glushchenko, A.; West, J. Appl. Phys. Lett. 2003, 82, 1917.(17) Duran, H.; Gazdecki, B.; Yamashita, A.; Kyu, T. Liq. Cryst. 2005,

32, 815.(18) Lynch, M. D.; Patrick, D. L. Nano Lett. 2002, 2, 1197.(19) Dierking, I.; Scalia, G.; Morales, P.; LeClere, D. AdV. Mater. 2004,

16, 865.(20) Link, S.; Hu, D.; Chang, W. S.; Scholes, G. D.; Barbara, P. F.

Nano Lett. 2005, 5, 1757.(21) Barbara, P. F.; Chang, W. S.; Link, S.; Scholes, G. D.; Yethiraj,

A. Annu. ReV. Phys. Chem. 2007, 58, 565.(22) Tcherniak, A.; Solis, D., Jr.; Khatua, S.; Tangonan, A. A.; Lee,

T. R.; Link, S. J. Am. Chem. Soc. 2008, 130, 12262.(23) Chang, W.-S.; Link, S.; Yethiraj, A.; Barbara, P. F. J. Phys. Chem.

B 2008, 112, 448.(24) Holt, L. A.; Bushby, R. J.; Evans, S. D.; Burgess, A.; Seeley, G.

J. Appl. Phys. 2008, 103, 063712.(25) Kumar, S.; Lakshminarayanan, V. Chem. Commun. 2004, 1600.(26) Qi, H.; Kinkead, B.; Marx, V.; Zhangi, H.; Hegmann, T. Chem-

PhysChem 2009, 10, 1211.(27) Kumar, S.; Pal, S. K.; Kumar, P. S.; Lakshminarayanan, V. Soft

Matter 2007, 3, 896.(28) Marx, V. M.; Girgis, H.; Heiney, P. A.; Hegmann, T. J. Mater.

Chem. 2008, 18, 2983.(29) Frein, S.; Boudon, J.; Vonlanthen, M.; Scharf, T.; Barbera, J.; Suss-

Fink, G.; Burgi, T.; Deschenaux, R. HelV. Chim. Acta 2008, 91, 2321.(30) In, I.; Jun, Y.-W.; Kim, Y. J.; Kim, S. Y. Chem. Commun. 2005,

800.

Figure 5. Nematic-to-isotropic phase transition of neat 5CB (red) and5CB doped with SOPB-functionalized gold nanoparticles (green). Thetemperature dependence of the relative transmittance (data points) wasfitted with a sigmoidal function (solid lines). Error bars were calculatedfrom three independent measurements.

Figure 6. Freedericksz transition of nanoparticle-doped (green) andundoped (red) liquid crystal devices. The lines connect the experimentalpoints and are included as guides for the eye only. Error bars werecalculated from two independent measurements.

7256 J. Phys. Chem. C, Vol. 114, No. 16, 2010 Khatua et al.

(31) Zeng, X.; Liu, F.; Fowler, A. G.; Ungar, G.; Cseh, L.; Mehl, G. H.;Macdonald, E. J. AdV. Mater. 2009, 21, 1746.

(32) Brust, M.; Fink, J.; Bethell, D.; Schiffrin, D. J.; Kiely, C. J. Chem.Soc., Chem. Commun. 1995, 1655.

(33) Fritz, K. P.; Scholes, G. D. J. Phys. Chem. B 2003, 107, 10141.(34) Lin, X. M.; Wang, G. M.; Sorensen, C. M.; Klabunde, K. J. J.

Phys. Chem. B 1999, 103, 5488.(35) Zubarev, E. R.; Xu, J.; Sayyad, A.; Gibson, J. D. J. Am. Chem.

Soc. 2006, 128, 4958.(36) Gibson, J. D.; Khanal, B. P.; Zubarev, E. R. J. Am. Chem. Soc.

2007, 129, 11653.(37) Link, S.; Wang, Z. L.; El-Sayed, M. A. J. Phys. Chem. B 1999,

103, 3529.(38) Storhoff, J. J.; Mirkin, C. A. Chem. ReV. 1999, 99, 1849.(39) Moskovits, M.; Vlckova, B. J. Phys. Chem. B 2005, 109, 14755.(40) Kreibig, U.; Genzel, L. Surf. Sci. 1985, 156, 678.

(41) Link, S.; El-Sayed, M. A. J. Phys. Chem. B 1999, 103, 4212.(42) Alvarez, M. M.; Khoury, J. T.; Schaaff, T. G.; Shafigullin, M. N.;

Vezmar, I.; Whetten, R. L. J. Phys. Chem. B 1997, 101, 3706.(43) Li, J.; Gauza, S.; Wu, S.-T. J. Appl. Phys. 2004, 96, 19.(44) Johnson, P. B.; Christy, R. W. Phys. ReV. B 1972, 6, 4370.(45) Berciaud, S.; Cognet, L.; Tamarat, P.; Lounis, B. Nano Lett. 2005,

5, 515.(46) Novo, C.; Gomez, D.; Perez-Juste, J.; Zhang, Z.; Petrova, H.;

Reismann, M.; Mulvaney, P.; Hartland, G. V. Phys. Chem. Chem. Phys.2006, 8, 3540.

(47) Boichuk, V.; Kucheev, S.; Parka, J.; Reshetnyak, V.; Reznikov,Y.; Shiyanovskaya, I.; Singer, K. D.; Slussarenko, S. J. Appl. Phys. 2001,90, 5963.

JP907923V

Plasmonic Nanoparticles-Liquid Crystal Composites J. Phys. Chem. C, Vol. 114, No. 16, 2010 7257