Synthesis and Electrochemistry of 6 nm Ferrocenated Indium−TinOxide NanoparticlesJoseph J. P. Roberts, Kim T. Vuong, and Royce W. Murray*

Department of Chemistry, University of North Carolina, Chapel Hill, North Carolina 27599, United States

*S Supporting Information

ABSTRACT: Indium−tin oxide (ITO) nanoparticles, 6.1 ± 0.8 nm in diameter, were synthesized using a hot injection method.After reaction with 3-aminopropyldimethylethoxysilane to replace the initial oleylamine and oleic acid capping ligands, theaminated nanoparticles were rendered electroactive by functionalization with ferrocenoyl chloride. The nanoparticle colorchanged from blue-green to light brown, and the nanoparticles became more soluble in polar solvents, notably acetonitrile. Thenanoparticle diffusion coefficient (D = 1.0 × 10−6 cm2/s) and effective ferrocene concentration (C = 0.60 mM) in acetonitrilesolutions were determined using ratios of DC and D1/2C data measured by microdisk voltammetry and chronoamperometry. TheD result compares favorably to an Einstein−Stokes estimate (2.1 × 10−6 cm2/s), assuming an 8 nm hydrodynamic diameter inacetonitrile (6 nm for the ITO core plus 2 nm for the ligand shell). The ferrocene concentration result is lower than anticipated(ca. 1.60 mM) based on a potentiometric titration of the ferrocene sites with Cu(II) in acetonitrile. Cyclic voltammetric dataindicate tendency of the ferrocenated nanoparticles to adsorb on the Pt working electrode.

■ INTRODUCTION

Being both optically transmissive and electrically conductive,the n-type semiconductor indium−tin oxide (ITO) hassubstantial industrial utility in flat panel display technology,1

solar cell windows,2 gas sensing devices,3 and transparentelectrodes.4 Methods for generating bulk ITO materials tend torequire specialized instrumentation and are subject to variabilityof produced materials. Preparation of ITO in nanoparticleforms and spreading as films followed by sintering provides aless expensive route while retaining desired conductive andoptical characteristics.5 Modern synthetic methods are able toproduce crystalline ITO nanoparticles with well-definedmorphologies and narrow size distributions.6 One synthesisinvolves decomposition of indium and tin precursors(commonly as acetates or acetylacetonate complexes) in ahigh boiling point solvent (at ca. 300 °C) in the presence offatty amine and carboxylic acid surfactants.7 Nanoparticles areformed in which tin atoms are substituted for indium sites inthe cubic bixbyte structure of In2O3. Thusly prepared ITOnanoparticles can be formed as films by dip-coating or spin-coating, followed by sintering. The films can be tuned bycontrolling the capping surfactant and its concentration insolution.8

This paper reports the synthesis of small (6 nm diameter)indium tin oxide (ITO) nanoparticles capped with a redoxactive species (ferrocene). Rather than ITO film preparation,

the goal is to create ITO nanoparticles having ferrocenatedsurfaces (abbreviated as FcITO) and to examine theirelectrochemical behavior as diffusing nanoparticle species,including assessing the nanoparticles’ ferrocene surface cover-age. Very small nanoparticles are desired, in the interest ofsolution solubility. ITO surfaces, like those of many metaloxides, are known to react with chloro- and alkoxysilanes.9

Here, reaction with an excess of the monoalkoxysilane 3-aminopropyldimethylethoxysilane (APDMES) is calculated todisplace the original capping surfactants from the surfaces ofprepared ITO nanoparticles. The amine-terminated silanizednanoparticle surface provides a platform for an amide-couplingreaction with ferrocenoyl chloride. The study provides a furtherexample10 of redox-labeled nanoparticle electrochemicalbehavior, of assessment of the surface coverage of redox labels,and is part of an ongoing exploration of electroactivenanoparticle materials11 as components of electrochemicalcharge storage systems.

■ EXPERIMENTAL SECTIONChemicals. Indium(III) acetate (In(ac)3; 99.99%) and tin(II)

(acetate) (Sn(ac)2; 99.99%) were obtained from Alfa Aeser,

Received: November 7, 2012Revised: December 13, 2012Published: December 26, 2012

Article

pubs.acs.org/Langmuir

© 2012 American Chemical Society 474 dx.doi.org/10.1021/la304443k | Langmuir 2013, 29, 474−479

oleylamine (C9H18C9H17NH2; 80−90%), oleic acid (C9H18C8H15COOH; 97%), octadecene (C16H33CH2; 90%), ferrocene-carboxylic acid (97%), oxalyl chloride (98%), imidazole (>99%),pyridine (>99.0%), triethylamine (>99%), and copper(II) perchloratehexahydrate (98%) were from Sigma-Aldrich, and tetrabutylammo-nium perchlorate (Bu4NClO4) was from Fisher Scientific. Acetonitrile(CH3CN), absolute ethanol, hexanes, dichloromethane (DCM),petroleum ether, toluene, and chloroform obtained from FisherScientific were dried over 4 Å molecular sieves. The functionalizedsilane, 3-aminopropyldimethylethoxysilane (APTMES), was obtainedfrom Gelest, Inc. (Morrisville, PA) and stored refrigerated. Use of amonoalkoxy reagent was calculated to avoid polymeric siloxane layerformation.Instruments. The prepared ITO and FcITO nanoparticles were

imaged by transmission electron microscopy (TEM; JEOL 2010F andHitachi 9500). Electrochemistry in 0.1 M Bu4NClO4/acetonitrilesolutions was performed in a standard three-electrode cell (Pt WE area= 0.02 cm2; Pt microdisk radius = 5 μm; Ag/AgCl RE; Pt wire CE)using a CH Instruments CHI660a. Centrifugation was done with anEppendorf 5810 centrifuge with a fixed-angle rotator at 3000−4000rpm for 10 min. A CAL 9500P programmable process controller wasused to provide accurate temperature control during nanoparticlesynthesis. XPS data were taken on a Kratos Axis Ultra DLD systemwith monochromatic Al Kα X-ray source. High-resolution scans weretaken at pass energy = 20 eV, and the spectral energy axis was alignedat the C 1s peak at 284.6 eV. EDS analysis was performed usingOxford INCA EnergyTEM 250 TEM microanalysis system attached tothe JEOL 2010F.Synthesis of 6 nm Indium Tin Oxide Nanoparticles. The

procedure was modified from Sun et al.6c In(acac)3 (0.20 mmol),Sn(acac)2Cl2 (0.02 mmol), and 0.60 mL of oleic acid (1.9 mmol) wereadded to octadecene (5 mL) in a 50 mL three-neck round-bottomflask equipped with a magnetic stir bar, thermocouple, and condenser,closing the third port with a septum. Using a 100 mL heating mantlepacked with sand to promote even heating, the vessel was evacuated

and heated 1 h at 120 °C with vigorous stirring. The temperature wasthen rapidly increased to 295 °C at which point 0.80 mL of oleylamine(2.4 mmol) in 0.2 mL of octadecene was injected via syringe. Thesolution color changed from bright orange to murky green, signalingformation of ITO nanoparticles. The reaction was refluxed at 320 °Cfor an hour and allowed to cool to room temperature. The cloudyblue-green suspension was transferred to a 50 mL centrifugation tubeand 40 mL of absolute ethanol was added, precipitating thenanoparticles, followed by centrifugation at 3000 rpm for 10 min.This process was repeated three times. The nanoparticles were thentaken up into 3 mL of either hexanes or chloroform and stored atroom temperature. The solution ranged from dark blue to green, andthe nanoparticles produced had average diameter 6.1 ± 0.8 nm. Figure1 shows TEM images.

Silane-Capped ITO Nanoparticles. A 1 mL hexane solution ofnanoparticles was rotovapped to dryness; the nanoparticles wereweighed and redissolved in 4 mL of toluene in a scintillation vial.APTMES was added in 10-fold excess (relative to estimated totalnanoparticle surface area), which was ∼0.03 g per 0.1 g ofnanoparticles. 0.05 g (0.75 mmol) of imidazole was added as catalyst,and the solution, which became a turbid blue, was stirred vigorously atroom temperature for 1 h. The solution was transferred to a 15 mLcentrifuge tube and centrifuged at 4000 rpm for 10 min. The aminatednanoparticles were washed twice more with fresh toluene and thendispersed in 3 mL of toluene for storage.

Synthesis of Ferrocenoyl Chloride and of Ferrocene-Functionalized ITO Nanoparticles. 0.23 g of ferrocenecarboxylicacid (1.00 mmol) was suspended in 4 mL of petroleum ether in a 25mL round-bottom flask. Two drops of pyridine were added followedby slow addition of 100 μL of oxalyl chloride (1.20 mmol); thesolution bubbled and gassed upon this addition. The reaction solutionwas stirred under Ar for 1.5 h, and the petroleum ether and excessoxalyl chloride removed by pumping under vacuum (ca. 2 h). The red-orange solid was collected in CH2Cl2 and placed into a 15 mLcentrifuge tube and centrifuged at 4000 rpm for 10 min to separate the

Scheme 1. The As-Synthesized ITO Nanoparticles Are Covalently Reacted with APDMES To Displace the Initial Surfactantsand To Aminate the Nanoparticle Surfaces; Ferrocenoyl Chloride Then Reacts with the Primary Amine

Figure 1. TEM Images of ITO nanoparticles (6.1 ± 0.8 nm); panel A and B scale bars are 50 and 10 nm, respectively. The inset shows a histogramof the nanoparticle size dispersion.

Langmuir Article

dx.doi.org/10.1021/la304443k | Langmuir 2013, 29, 474−479475

unreacted FcCOOH from the FcCOOCl. The bright red productsolution was decanted and immediately reacted with silane-cappedITO nanoparticles.The 3 mL solution of silane-capped ITO nanoparticles in toluene

plus 50 μL of triethylamine (to aid activation of the acid chloride) wasreacted for 1 h. in a scintillation vial with 100 μL of the ferrocenoylchloride solution and then transferred to a 15 mL centrifuge tube andcentrifuged at 4000 rpm for 10 min. The blue-brown precipitate waswashed twice more with 5 mL of toluene, then dissolved in 3 mL ofacetonitrile, and stored at room temperature.

■ RESULTS AND DISCUSSIONITO Synthesis. The ITO nanoparticle synthesis was

modified from Sun et al.6c Rather than heating the reactionmixture with all reagents present, the oleylamine, whichactivates the nanoparticle formation, was hot injected at 25°C below the solvent boiling point. This modification, alongwith maintaining accurate temperature control, substantiallyimproved reproducibility of synthetic nanoparticle batches. Thecolor of the as-synthesized ITO nanoparticle solutions rangefrom dark blue to vivid green, after ca. 24 h shifting towardblue, or teal, and is blue after the silanization step. It issuggested that the color changes reflect the slow equilibrationof SnIV sites with interstitial oxygen. The ITO nanoparticles areby TEM (Figure 1) reasonably monodisperse, with littleaggregation. The nanoparticle diameter, from TEM images ofmultiple synthetic batches, averages 6.1 ± 0.8 nm.Replacement of Surfactants with Silane Ligands and

Ferrocenation. Aiming at gaining control of the ITOnanoparticle surface chemistry, the original capping ligandson the synthesized ITO nanoparticles were displaced bysilanization of the ITO surface. Silanization of macroscopicITO surfaces is a known reaction12 and can be anticipated forITO nanoparticle surfaces. The goal was to form an aminated,functionalized nanoparticle surface for reaction with ferrocenoylchloride. (Electroactive ligands have been previously attachedto Au and SiO2 nanoparticles10,13 by this laboratory.) Amonodentate alkoxysilane reagent was selected to avoidformation of silane polymer on the nanoparticles. After someexperimentation, the silanization reaction was carried out atroom temperature for ∼1 h, with little flocculation beingnoticed by TEM imaging. (Test silanizations for longer periods(12−24 h) and at elevated temperatures (∼80 °C) tended toprovoke nanoparticle aggregation.)The silanized ITO nanoparticles would usually remain

dispersed in toluene for about a week (in some cases someprecipitation then commenced). Variability of solubilityprobably reflects incomplete displacement of the originalsurfactant. The nanoparticles could be isolated from tolueneby centrifugation. Nanoparticle surface composition wasmonitored with XPS (Figure 2). Postsilanization, indium 3dpeaks are seen at 444.5 and 451.5 eV and tin 3d peaks at 486and 494.5 eV. Before silanization, these peaks are masked byscattering of photoelectrons by the nanoparticle capping ligands(Figure S-1) which produces the large C 1s peak.Postsilanization, the C 1s peak is greatly diminished andindium and tin peaks can be seen, as are low-intensity Si 2p, Si2s, and N 1s peaks.Following silanization, a measured mass of the aminated ITO

nanoparticles is reacted with ferrocenoyl chloride. In thisprocedure and in subsequent washing steps, considerable lossof product occurs in the centrifugal isolation. At least two washsteps were necessary to remove unattached ferrocene reagent,whose presence would interfere in the subsequent electro-

chemistry studies. TEM imaging shows that the average corediameters of the ferrocenated ITO nanoparticles (abbreviatedFcITO) are substantially unchanged from the as-synthesizedITO nanoparticles (Figure 1). (The lower atomic weightferrocene substituents are not expected to contributesignificantly to the image diameter.) Assuming completenanoparticle surface coverage by the amine silane reagent, itscomplete coupling with ferrocenoyl chloride, and a ferrocenesurface footprint of 1 nm2, the typical 0.006 g mass of isolatedFcITO nanoparticle product in a 3 mL acetonitrile solutionwould ideally correspond to an overall ca. 5 mM ferroceneconcentration.The actual reactive ferrocene concentration in 1 mL of the

ITO nanoparticle solution was assessed by a potentiometrictitration (Figure 3) with Cu2+, which in acetonitrile is an

effective ferrocene oxidant.14 During the titration, the nano-particle solution color changes from light blue-brown to yellow.This coloration is due to the combination of the ITOnanoparticles and the ferrocene attached to the surface.Multiple trials give indicate an overall ferrocene concentrationof 1.3 ± 0.2 mM. Based on the model coverage calculationabove, the actual ferrocene coverage of the ITO surface is about25%.The ferrocenated (FcITO) nanoparticles were stably soluble

in acetonitrile, which was chosen for the electrochemicalexperiments. (Before ferrocenation, the ITO nanoparticles are

Figure 2. XPS spectrum of ITO nanoparticles after silanization.

Figure 3. Potentiometric titration of a 1 mL CH3CN solution offerrocenated ITO nanoparticles with 1.18 mM Cu(ClO4)2 CH3CNsolution. The equivalence point is marked with a diamond; Cu2+

titration volume is 1.15 mL.

Langmuir Article

dx.doi.org/10.1021/la304443k | Langmuir 2013, 29, 474−479476

insoluble in acetonitrile.) The presence of iron on the FcITOnanoparticles was confirmed by XPS (Figure S-2) and EDX(Figure S-3). The observed N/Fe peak ratio (with normal-ization for elemental sensitivity) was ca. 1:0.5, differing fromthe ideal 1:1 for the ferrocene amide grouping. This result isconsistent with the above indication of nonexhaustive couplingbetween the amine sites and the ferrocenoyl chloride.Voltammetry of Ferrocenated ITO Nanoparticles.

Voltammetry of the FcITO nanoparticles was carried out in0.1 M Bu4NClO4/acetonitrile solutions. The ferrocene wavewas readily seen in cyclic voltammetry (Figure 4) although the

currents were modest due to the low concentration and slowednanoparticle diffusion rates. Measurement of the nanoparticlediffusion coefficient (D) was set as a target in quantifying theelectrochemical assessment, using cyclic voltammetry, chro-noamperometry, and microdisk electrode voltammetry.The diffusion coefficient expected was estimated using the

no-slip version of the Stokes−Einstein equation, which is

πη=D

kTr6 (1)

where r is nanoparticle hydrodynamic radius and η isacetonitrile viscosity (0.343 cP). Based on the 6.1 nm TEMnanoparticle core diameter and plus a 2 nm increment for theferrocene sites, this equation predicts a FcITO nanoparticlediffusion coefficient of D = 2.1 × 10−6 cm2/s. Previousmeasurements of diffusion coefficients of organothiolate-coatedAu nanoparticles (1−5 nm diameter), using the classical Taylordispersion method,15 show that (a) eq 1 can predictnanoparticle diffusion coefficients to within a factor of ca. 2and (b) that the organothiolate monolayer on the nanoparticlesmay be partially free-draining.Cyclic voltammetry of the amide-connected ferrocene on the

FcITO nanoparticles shows a one-electron chemically rever-sible wave with formal potential ca. 0.46 mV vs. Ag/AgCl(aq)(Figure 4). The Randles−Sevcik equation for linear sweepvoltammetry16 is

= ×i n AD Cv2.69 10p5 3/2 1/2 1/2

(2)

where n is the number of electrons delivered (per ferrocene), D(cm2/s) is nanoparticle diffusion coefficient, and C (mol/cm3)is the concentration of reactive ferrocene. Figure 5 shows that

the peak currents do not follow the expected v1/2 potential scanrate dependency but show a nonzero intercept and curveupward in a manner suggesting adsorption of the ferrocenatednanoparticles on the electrode. Note the lessening of ΔEpeak atthe fastest scan rate in Figure 4, which is consistent with thisconclusion. We therefore elected not to employ linear sweepcurrent results in estimates of the nanoparticle diffusioncoefficient.Currents from potential step chronoamperometry and from

microelectrode voltammetry should be relatively little affectedby a modest extent of nanoparticle adsorption. These twoexperiments yield nanoparticle diffusion coefficient andconcentration, without assumptions, by combining theirD1/2C and DC results, as shown next. Microdisk voltammetry,

seen in Figure 6, yields the product DC from the relation forsteady-state limiting current (ilim) and C

=i nFDCr4lim 0 (3)

where r0 is microdisk radius (cm). Solving for DC gives a valueof 2.1 × 10−13 mol/(s cm). Currents for ferrocenes on anyadsorbed nanoparticles would be minimized by the slowpotential scan rate.

Figure 4. Background subtracted cyclic voltammetry of FcITO inacetonitrile (0.1 M Bu4NClO4) at indicated potential scan rates.Electrode area = 0.020 cm2.

Figure 5. Plot of ferrocene oxidation peak current (ipeak) vs square rootof potential scan rate for solution of FcITO in acetonitrile. Thecurvature and nonzero intercept are interpreted as indicatingnanoparticle adsorption.

Figure 6. Voltammetry of a solution of FcITO nanoparticles inacetonitrile (0.1 M Bu4NClO4) using a 5 μm radius disk micro-electrode; potential scan rate 1 mV/s. ilim = 1.28 × 10−11 A.

Langmuir Article

dx.doi.org/10.1021/la304443k | Langmuir 2013, 29, 474−479477

In potential step chronoamperometry, currents for ferrocenes

on any adsorbed nanoparticles should contribute only at short

electrolysis times. Currents at longer times should follow a

normal Cottrell relationship, which is seen in the result of

Figure 7 where current is measured out to 6 s. The Cottrellequation is

π= ⎜ ⎟⎛

⎝⎞⎠i t nFAC

Dt

( )1/2

(4)

where t is time. From this experiment, D1/2C = 1.2 × 10−10

mol/(s1/2 cm2). Taking the ratio of the two parametersproduces D and C without assumptions and yields D = 1.0 ×10−6 cm2/s and ferrocene concentration C = 0.60 mM. We seethat the diffusion coefficient determined by this method iswithin a factor of 2 of the Einstein−Stokes equation prediction,and the determined ferrocene concentration is more than 2-foldsmaller than the titration-predicted 1.3 mM.The preceding results from the ferrocene titration and

combined microelectrode−chronoamperometry indicate thatferrocene sites coupled to the ITO nanoparticles are accessibleto a small, diffusing oxidizing reagent (e.g., Cu2+) but not fullyaccessible during nanoparticle diffusion-collision with theworking electrode (a large reagent). This conclusion is testedin Figure 8, where increments of 1 mM ferrocene carboxylic

acid solution are added to a 3 mL solution of ferrocenated ITOnanoparticles. The choice of ferrocene carboxylic acid wasbased on its formal potential, which is only slightly morepositive than that of the nanoparticle-coupled ferrocene amidegroup; the leading edge of the ferrocene carboxylic acid wavecould mediate reaction of accessible nanoparticle-coupledferrocene amide groups. Figure 8 shows that addition of asmall volume of FcCOOH solution immediately enhances theferrocene amide oxidation wave by roughly a factor of 2.Further additions produce a small additional enhancement atthe ferrocene amide oxidation potential. This behavior confirmsthe hypothesis that the titration-detected ferrocene amides arereactive toward a small molecule oxidant.The incomplete electrochemical reactivity of the ferrocene

amide functionalities on the ITO nanoparticles stands incontrast to results for ferrocenated 13 nm diameter SiO2nanoparticles,10 where Cu2+ titration analysis indicated 590ferrocenes had been attached (ca. a complete monolayer).Model calculations indicated that rotational diffusion wasadequately fast (τROT ∼ 0.25 μs) to (sequentially) bring all ofthese ferrocene sites into proximity with the electrode. Sincethe FcITO nanoparticles have a smaller diameter, rotationaldiffusion should in the present ideally not be a factor lesseningthe reactivity of all of the ITO-attached ferrocene sites, so someother factors require consideration. One is that the tendency ofFcITO nanoparticles to adsorb precludes an entirely freerotational diffusion at the electrode interface. A second andequally likely issue is the lack of complete accessibility(mentioned above) owing to the roughness or porosity ofthe ITO nanoparticle surfaces. This latter factor couldpotentially be overcome using a longer chain linker betweenthe nanoparticle and ferrocene.

■ ASSOCIATED CONTENT*S Supporting InformationFigures S-1 to S-3. This material is available free of charge viathe Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail rwm@email.unc.edu.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis research was supported in part by the Office of NavalResearch and the National Science Foundation. We gratefullyacknowledge TEM and EDX measurements performed by theAnalytical and Nanofabrication Laboratory of the UNCInstitute for Advanced Materials.

■ REFERENCES(1) Shen, Z. P. E. Three-Color, Tunable, Organic Light-EmittingDevices. Science 1997, 276 (5321), 2009.(2) Cho, Y.-J.; Ahn, T. K.; Song, H.; Kim, K. S.; Lee, C. Y.; Seo, W.S.; Lee, K.; Kim, S. K.; Kim, D.; Park, J. T. Unusually HighPerformance Photovoltaic Cell Based on a [60]Fullerene MetalCluster−Porphyrin Dyad SAM on an ITO Electrode. J. Am. Chem. Soc.2005, 127 (8), 2380−2381.(3) Neri, G.; Bonavita, A.; Micali, G.; Rizzo, G.; Pinna, N.;Niederberger, M.; Ba, J. Effect of the chemical composition on thesensing properties of In2O3−SnO2 nanoparticles synthesized by a non-aqueous method. Sens. Actuators, B 2008, 130 (1), 222−230.

Figure 7. Cottrell plot of ferrocenated ITO in acetonitrile. Thepotential step was from 0.45 to 0.85 V vs Ag/AgCl. The plot is forcedthrough the origin as is required for linear diffusion control. Ptelectrode area is 0.020 cm2.

Figure 8. Cyclic voltammetry of ferrocenated ITO solution following90 μL additions of 1 mM FcCOOH with a scan rate of 100 mV/s.Additions of 0, 90, 360, 630, and 900 μL are shown.

Langmuir Article

dx.doi.org/10.1021/la304443k | Langmuir 2013, 29, 474−479478

(4) Shi, Y.; Slaterbeck, A. F.; Seliskar, C. J.; Heineman, W. R.Spectroelectrochemical Sensing Based on Multimode SelectivitySimultaneously Achievable in a Single Device. 1. Demonstration ofConcept with Ferricyanide. Anal. Chem. 1997, 69 (18), 3679−3686.(5) Reindl, A.; Mahajeri, M.; Hanft, J.; Peukert, W. The Influence ofDispersing and Stabilizing of Indium Tin Oxide Nanoparticles uponthe Characteristic Properties of Thin Films. Thin Solid Films 2009, 517(5), 1624−1629.(6) (a) Choi, S.-I.; Nam, K. M.; Park, B. K.; Seo, W. S.; Park, J. T.Preparation and Optical Properties of Colloidal, Monodisperse, andHighly Crystalline ITO Nanoparticles. Chem. Mater. 2008, 20 (8),2609−2611. (b) Gilstrap, R. A.; Capozzi, C. J.; Carson, C. G.;Gerhardt, R. A.; Summers, C. J. Synthesis of a NonagglomeratedIndium Tin Oxide Nanoparticle Dispersion. Adv. Mater. 2008, 20 (21),4163−4166. (c) Sun, Z.; He, J.; Kumbhar, A.; Fang, J. NonaqueousSynthesis and Photoluminescence of ITO Nanoparticles. Langmuir2010, 26 (6), 4246−4250.(7) Zhang, Z.; Lu, M.; Xu, H.; Chin, W.-S. Shape-ControlledSynthesis of Zinc Oxide: A Simple Method for the Preparation ofMetal Oxide Nanocrystals in Non-aqueous Medium. Chem.Eur. J.2007, 13 (2), 632−638.(8) Mahajeri, M.; Voigt, M.; Klupp Taylor, R. N.; Reindl, A.; Peukert,W. Evaluation of the Film Formation and the Charge TransportMechanism of Indium Tin Oxide Nanoparticle Films. Thin Solid Films2010, 518 (12), 3373−3381.(9) Neouze, M.-A.; Schubert, U. Surface Modification andFunctionalization of Metal and Metal Oxide Nanoparticles by OrganicLigands. Monatsh. Chem. 2008, 139 (3), 183−195.(10) Beasley, C. A.; Murray, R. W. Voltammetry and Redox ChargeStorage Capacity of Ferrocene-Functionalized Silica Nanoparticles.Langmuir 2009, 25 (17), 10370−10375.(11) Chow, K.-F.; Sardar, R.; Sassin, M. B.; Wallace, J. M.; Feldberg,S. W.; Rolison, D. R.; Long, J. W.; Murray, R. W. 3D-AddressableRedox: Modifying Porous Carbon Electrodes with Ferrocenated 2 nmGold Nanoparticles. J. Phys. Chem. C 2012, 116 (16), 9283−9289.(12) Haensch, C.; Hoeppener, S.; Schubert, U. S. ChemicalModification of Self-Assembled Silane Based Monolayers by SurfaceReactions. Chem. Soc. Rev. 2010, 39, 2323−2334.(13) Stiles, R. L.; Balasubramanian, R.; Feldberg, S. W.; Murray, R.W. Anion-Induced Adsorption of Ferrocenated Nanoparticles. J. Am.Chem. Soc. 2008, 130 (6), 1856−1865.(14) (a) Quirk, P. F.; Kratochvil, B. Determination of FerroceneDerivatives by Oxidation with Copper(II) in Acetonitrile. Anal. Chem.1970, 42 (4), 535−536. (b) Schmidt, S. P.; Basolo, F.; Trogler, W. C.Reactions between Dimanganese, Dirhenium, and Manganese−Rhenium Decacarbonyl and Oxidants. Inorg. Chim. Acta 1987, 131(2), 181−189.(15) Wuelfing, W. P.; Templeton, A. C.; Hicks, J. F.; Murray, R. W.Taylor Dispersion Measurements of Monolayer Protected Clusters: APhysicochemical Determination of Nanoparticle Size. Anal. Chem.1999, 71 (18), 4069−4074.(16) Faulkner, L. R.; Bard, A. J. Electrochemical Methods:Fundamentals and Applications, 2nd ed.; John Wiley & Sons, Inc.:Phoenix, AZ, 2001.

Langmuir Article

dx.doi.org/10.1021/la304443k | Langmuir 2013, 29, 474−479479