Reactivity of [Au25(SCH2CH2Ph)18]1- Nanoparticles with Metal Ions†

Jai-Pil Choi,‡ Christina A. Fields-Zinna, Rebecca L. Stiles,§ Ramjee Balasubramanian,|

Alicia D. Douglas,⊥ Matthew C. Crowe,# and Royce W. Murray*Kenan Laboratories of Chemistry, UniVersity of North Carolina, Chapel Hill, North Carolina 27599-3290

ReceiVed: October 22, 2009; ReVised Manuscript ReceiVed: December 17, 2009

We report reactivity of the gold nanoparticle [TOA+] [Au25(SC2Ph)18]1- (TOA+ ) tetraoctylammonium;SC2Ph ) phenylethanethiolate ) L; [Au25(SC2Ph)18]1- ) Au25L18

1-) with Ag+, Cu2+, and Pb2+ ions. Titrationof solutions of Au25L18

1- in CH2Cl2 with one and two equivalents of Ag+ produces changes in absorbancespectra with isosbestic points, and a titration curve break at 1:1 mol ratio, indicating a stoichiometric interaction.Similar effects are seen with Cu2+ and Pb2+ additions, but the break occurs at 0.5:1 mol ratio metal/nanoparticle.Changes in Au25L18

1- absorbance and fluorescence spectra are qualitatively similar to those accompanyingoxidation of the Au25L18

1- nanoparticle anion, but the spectra of the stoichiometric products differ slightlyaccording to the metal ion. Addition of higher excess of Ag+ or Cu2+ causes loss of characteristic[Au25(SC2Ph)18]1- UV-vis spectral fine structure and apparent irreversible refining into larger nanoparticles.Voltammetric currents for nanoparticle 0/-1 and +1/0 redox waves are depressed by Ag+ addition. Electrosprayionization mass spectra of products of addition of up to two equivalents Ag+ show prominent peaks for[Au25(SC2Ph)18]1- but also peaks corresponding to bimetal nanoparticles [Au24Ag(SC2Ph)18]2+,[Au23Ag2(SC2Ph)18]2+, and [Au22Ag3(SC2Ph)18]2+. We propose a redox model of reaction of the Au25L18

1-

nanoparticle with metal ions, in which the Au25L181- nanoparticle acts as a reductant toward the metal ion,

forming Au25M(SC2Ph)18 adducts that become oxidatively dissociated in the mass spectral cationizationenvironment to yield the bimetals observed.

For over a decade, metal nanoparticles capped by thiolateligands (or monolayer-protected clusters, MPCs) have been ofinterest, having possible application to catalysis,1 chemical andbiological sensors,2 optoelectronics,3 and surface-enhancedRaman scattering (SERS),4 among others. Because physico-chemical properties of very small metal MPCs can vary withtheir size, protecting ligands, and core-metal compositions, onecan envision controlling nanoparticle properties by adoptingspecific reaction routes that select for desired sizes, protectingligands, or core-metal compositions. For example, the originalthiolate ligands of monolayer-protected gold nanoparticles canbe exchanged with different target thiolate ligands withoutchanging the nanoparticle core size.5 It is possible to prepareapparently alloyed bimetallic nanoparticles from monometallicones composed of less noble metals (e.g., Ag, Cu, and Pdclusters) by galvanic exchange reactions6 with salts of morenoble metal thiolate complexes (such as AuI-SR). To realizethe full scope of possible transformations and uses, it isimportant to further understand and elucidate the reactivitiesof metal MPCs.

In the present work, we report that solutions containing metalions less noble than gold, react with gold MPCssspecifically

in CH2Cl2 solutions of the MPC nanoparticle [Au25(SC2Ph)18]1-

(abbr. Au25L181-, SC2Ph ) phenylethanethiolate ) L). The

composition and detailed structure of this nanoparticle, an anionin its native state, are known from mass spectrometry7,8

crystallographic9 measurements and are supported by theory.10

(The Au25L181- structure is shown in the Supporting Informa-

tion, Figure S-1.) Here, we report that incremental additionsof Ag+ to CH2Cl2 solutions of Au25L18

1- MPCs induce, over1:1 and 1:2 stoichiometric Ag+/nanoparticle ratios, changesin optical absorbance spectra. Isosbestic points are seenduring the spectral changes; these imply quantitative conver-sions of the initial Au25L18

1- to other species after one andthen two equivalents of Ag+ have been added. Similar effectsare seen when adding Cu2+ or Pb2+ ions. Addition of largerproportions of Ag+ or Cu2+ (but not Pb2+) ions causes lesscontrolled changes and ultimately evokes changes in nano-particle size. We have examined this unanticipated chemistryby a range of measurements including photoluminescence(PL), electrochemistry, and electrospray ionization massspectrometry (ESI-MS).

Theoretical approaches to structures of small thiolate-protected Au nanoparticles have enjoyed recent successes,anticipating11 novel aspects of their thiolate ligand shells,supporting detailed crystallographic structures of Au102 and Au25

nanoparticles,9,11b,12 and predicting ligand shell structure forAu144.13 The above metal ion/Au25L18

1- reactivities are particu-larly interesting in light of recent detection of Au24PdL18

nanoparticles14 and ensuing theoretical studies on a variety ofanalogous bimetals.15 The replacement of a single Au site withanother metal is capable of evoking significant electronicchanges that additionally differ according to whether thereplacement occurs at different Au sites (center, core surface,and semiring).

† Part of the “Protected Metallic Clusters, Quantum Wells, and Metal-Nanocrystal Molecules Symposium” special issue.

* To whom correspondence should be addressed. E-mail: rwm@email.unc.edu.

‡ Present address: Dept. Chem., California State Univ. - Fresno, M/SSB70, Fresno, CA.

§ Present address: Air Protection Technologies, Nalco, Naperville, IL60563.| Present address: Dept. Chem. Biochem., Old Dominion Univ., Norfolk,

VA 23529.⊥ Present address: CEM Corp., Charlotte, NC.# Present address: Dept. Chem. Univ. Washington, Seattle, WA.

J. Phys. Chem. C 2010, 114, 15890–1589615890

10.1021/jp9101114 2010 American Chemical SocietyPublished on Web 02/24/2010

Experimental Section

Chemicals. Hydrogen tetrachloroaurate(III) trihydrate(HAuCl4 ·3H2O) was prepared by a literature method.16 Phe-nylethanethiol (PhCH2CH2SH, 98%), sodium borohydride(NaBH4, 99%), tetra-n-octylammonium bromide (Oct4NBr,98%), tetra-n-butylammonium perchlorate (Bu4NClO4, g 99.0%),cerium(IV) sulfate (Ce(SO4)2), silver nitrate (AgNO3, g99.0%),cupric chloride (CuCl2, 97%), and lead(II) perchlorate trihydrate(Pb(ClO4)2 ·3H2O, 98%) were obtained from Aldrich or Fluka(Milwaukee, WI). Toluene (ACS-certified grade), tetrahydro-furan (THF, HPLC grade), methanol (optima grade), acetonitrile(CH3CN, optima grade), and dichloromethane (CH2Cl2, optimagrade) were purchased from Fisher Scientific (Suwanee, GA).Ethanol (absolute - 200 proof) was from Aaper Alcohol andChemical Company (Shelbyville, KY). All chemicals were usedwithout further purification. Deionized water (>18MΩ) wasprepared with a Millipore Nanopure water purification system.

Synthesis of Au25L181- MPCs. Au25L18

1- MPCs weresynthesized by the previously reported method, in a paper17 mis-identifying them as Au38. Briefly, AuCl4

- was reacted with3-fold excess PhC2SH in toluene to form polymer-like Au(I)-phenylethanethiolates, which were then reduced by 10-foldexcess (against AuCl4

-) aqueous BH4- at 0 °C for 24 h. After

completion of the reduction reaction, the toluene phase wasseparated, and the toluene removed under reduced pressure atroom temperature. The nanoparticle product was extracted withCH3CN (which preferentially dissolves Au25L18

1-; Au25L180 is

insoluble), the nanoparticles then being precipitated by addingCH3OH or CH3CH2OH and re-extracting and reprecipitatingwith CH3CN and CH3OH, respectively, until desired purity ofAu25L18

- was obtained. The final product was inspected byTEM, NMR, UV-vis absorption, and voltammetry. We willterm the as-prepared Au25L18

- anion, the “native state”.Procedures for Metal Ion Titrations. Using a Shimadzu

UV-1601 spectrophotometer (Shimadzu, Columbia, MD) tocollect UV-vis spectra, µL volumes (10-µL microsyringe;Hamilton, Reno, NV) of 10 mM of metal ion (Ag+, Cu2+, andPb2+) solutions in CH3CN (AgNO3, CuCl2, and Pb(ClO4)2 ·3H2Oare not soluble in CH2Cl2) were incrementally added to 3 mLof Au25L18

1- MPC solution in CH2Cl2 in the quartz cuvette.The same titration procedures were employed for photolumi-nescence measurements, taken with a modified ISA FluorologFL321 spectrometer (HORIBA Jovin Yvon Inc., Edison, NJ)equipped with a 450 W xenon source and Hamamatsu R928PMT (visible wavelengths) and InGaAs (near IR wavelengths,connected via a T channel) detectors. No optical polarizationwas used or measured. NIR photoluminescence spectra weretaken using a long-pass filter (cutoff wavelength of 650 nm)placed between the cuvette and the NIR detector to avoid thedetection of higher order excitation peaks (secondary harmonicpeaks caused by a grating monochromator). Measurements weretaken immediately after titration. (Titrations performed inCH3CN alone caused precipitation, hence the use of CH2Cl2

solvent.)Voltammetry. A CHI-660C electrochemical analyzer/

workstation (CH instruments, Inc., Austin, TX) was used tomeasure cyclic voltammetry, in a three-electrode cell containingPt disk working (3 mm dia.), Pt mesh (1 cm × 1 cm) auxiliary,and Ag/AgCl reference (10 mM AgNO3 in CH3CN) electrodes.

Electrospray Ionization Mass Spectrometry. Products ofmixing Au25L18

1- MPCs with Ag+ were studied by electrospraymass spectrometry (ESI-MS). A total of 150 µg of[TOA+][Au25L18

-] dissolved in ∼400 µL of CH2Cl2 and µLvolumes of a AgNO3 solution in acetonitrile were added to attain

1:1, 1.5:1, and 2:1 mol ratio Ag+/Au25L181-. After sonicating

this mixture for a few seconds (the reactions are rapid, accordingto UV-vis observations), drying quickly under vacuum (ap-proximately 1 min), the product was examined by ESI-MSwithin the hour by taking the sample up into a suitable ESIsolventseither 70:30 methanol:toluene or 70:30 methanol:CH2Cl2 containing 2.6 mg of Cs(acetate) per mg of nanoparticle.The experimental plan assumed that the product of interactionof Au25L18

1- and Ag+ in methylene chloride might have limitedstability. In some cases, the dried sample was first washed withmethanol to remove any free AgNO3 present. The time periodof exchange for mass spectrometry experiments is longer thanthat for spectral experiments.

The overall behavior suggested that the bimetal species arefragile, and little effort was expended on isolating Au24AgL18

nanoparticles. (This has been successfully accomplished forAu24PdL18 MPCs14a but the isolation was quite difficult.) Also,due to general similarities in properties, mass spectra of samplestitrated with Cu2+ and Pb2+ were not sought.

The instruments were Bruker BioTOF II and MicromassQuattro II mass spectrometers, with spectra were taken inpositive mode. For reference, as previously established,7a,b acomparison ESI-TOF-MS spectrum of Au25L18

1- to which noAg+ has been added, was electrosprayed in 70:30 CH3OH:CH2Cl2 solution, giving 2+ peaks assignable toCsxAu25(SC2Ph)18

2+ where x ) 1, 2, and 3 with relativeintensities x ) 1 > 2 > 3. (see Supporting Information FigureS-7). The peak for x ) 0 was very small.

MALDI Mass Spectrometry. Matrix-assisted laser desorp-tion/ionization mass spectra of Ag+ titrated Au25L18

1- werecollected on a Voyager DE Biospectrometry Workstation(Perspective Biosystems, Inc., Framingham, MA) in the linearmode using a nitrogen laser (337 nm). trans-2-[3-(4-Tert-butylphenyl)-2-methyl-2-propenylidene] malononitrile (DCTB)was used as the MALDI matrix.7c Spectra were obtained bothin negative and positive ion mode using an acceleration voltageof 25 kV and a delay time of 350 ns, and at serially reducedlaser pulse intensities (laser fluence) to search for minimalnanoparticle fragmentation at 10% above threshold observation.While the data shown are in negative mode, positive modespectra are (at low laser pulse intensity) substantially the sameas reported previously.7c

Results

Absorbance Spectra Changes Caused by Added Ag+.UV-vis spectra of Au25L18

1- are substantially unaffected bythe choice of solvent or alkali metal or tetra-alkylammoniumelectrolyte cosolutes. In contrast, incremental additions of upto 2:1 mol ratio of Ag+/Au25L18

1- (Ag+ as AgNO3) to 15.3 µMsolutions of Au25L18

1- in CH2Cl2 causes rapid spectral changesfrom the initial12,14,15,17 spectral fine structure (SupportingInformation, Figure S-2 shows a wider wavelength range).Figure 1A shows that the addition of up to 1:1 mol ratio ofAg+/Au25L18

1- (a) suppresses the shoulder at ca. 790 nm, (b)shifts the 682 nm band to lower wavelength, (c) “fills in” thespectral valley at 605 nm, and (d) causes roughly linear changesin absorbance (Figure 1A, inset). The spectral changes in Figures1A are accompanied by appearance of isosbestic (see *, acommon crossing) points at 438, 670, and 865 nm. Continuedaddition of Ag+, up to 2:1 Ag+/Au25L18

1- mol ratio (Figure 1B),produces further changes in spectral shape and evokes a newisosbestic point at 628 nm.

Further additions of Ag+ do not produce more isosbesticfeatures. The spectrum has lost most of its original17-19 fine

Reactivity of [Au25(SCH2CH2Ph)18]1- Nanoparticles J. Phys. Chem. C, Vol. 114, No. 38, 2010 15891

structure and remains more or less featureless. TEM measure-ments on products isolated from the solution after addition of10:1 mol ratios of Ag+/Au25L18

1- reveal a general increase innanoparticle core size, to ca. 1.3 ( 0.8 nm. Consistent with anincrease in size,20 the organic fraction as measured by thermo-gravimetric analysis (36.3%; ideal is 37.4% for the (Oct4)N+

salt of Au25L181-) had decreased to 27.7% (Figure S-3). XPS

of isolated (unfractionated) material revealed a Ag 3d peak aswell as Au 4f (Figure S-4). These products were not furthercharacterized as it seemed evident that reaction with >2:1 molratio Ag+/Au25L18

1- leads to instability and loss of identity asan Au25L18

1- nanoparticle. Further work focused on the lowermole ratio range of Ag+ additions.

While the isosbestic crossings were not always as tightlybunched as seen in Figure 1, they were always readily noticed.An isosbestic point classically signals a stoichiometric conver-sion of one substance to another. This is particularly clear inthe “titration curve” plot of 605 nm absorbance against moleratio in Figure 1B, inset, which shows a clear break at 1:1stoichiometry. Adding equivalent quantities of KNO3 causes nospectral change, while identical changes are evoked by addinga different silver salt (AgClO4). Qualitatively, the absorbanceincrease at 605 nm and decrease at 790 nm in Figure 1Aresemble the spectral effects18a,b of electrochemical18a or chemi-

cal oxidation of Au25L181- to its neutral form (see Figure S-5).

These results imply that a stoichiometric oxidation-like processis invoked by 1:1 Ag+ additions.

Photoluminescence Spectra Changes Caused by AddedAg+. Au25L18

1- MPCs luminesce at near-infrared (NIR)wavelengths18,19 when excited anywhere in the UV-vis range,and like absorbance spectra, the emission is sensitive tonanoparticle oxidation. The spectral sensitivity is illustrated inFigure S-6 for a CH2Cl2 Au25L18

1- solution oxidized by aqueousCe(IV) to Au25

0 and to Au25+1.17 Oxidation to Au25

0 shifts theemission from 1070 to 950 nm. The oxidation states wereidentified by the solution electrochemical rest potentials fol-lowing reaction with Ce(IV), comparing them to the formalpotentials of the Au25

0/-1 and Au25+1/0 redox couples and invoking

the Nernst equation.21-23

Figure 2 shows that incrementally adding Ag+ to Au25L181-

solutions in CH2Cl2 (up to ∼1:1 mol ratio) causes a shift inemission to the same wavelength at which the oxidized (Au25

0)nanoparticle emits, and a linear decrease in (integrated) emissionintensity. Above a 1:1 mol ratio, there is little further changein the emission energy (Figure S-7) but the (integrated) emissionintensity now increases, again linearly, up to ∼3:1 mol ratio.These results are qualitatively consistent with the absorbancespectral results described above; addition of Ag+ initiates astoichiometric, oxidation-like process.

Changes in Voltammetry Caused by Added Ag+. Cyclicvoltammetry (CV) of Au25L18

1- MPCs in 0.1 M Bu4N+PF6-/

CH2Cl2 (Figure 3a, upper left) shows two well-formed, revers-ible one electron waves at - 0.30 and - 0.02 V vs. Ag/Ag+,corresponding to the Au25

0/-1 and Au25+1/0 redox couples,

respectively. (This voltammetry was reported previously,18a,24

but at the time the nanoparticle was mis-identified as Au38. Themis-identity was corrected7 based on high resolution electrospraymass spectrometry observations). The Au25L18

1- nanoparticlehas a molecule-like 1.3 eV HOMO-LUMO energy gap.18a Thedoublet of one electron oxidations, with formal potentials spacedby several hundred mV, is classically typical18a,24-27 for reactionof a molecular species having a two-electron occupied highestmolecular orbital (HOMO). The spectral results were supportedby a subsequent report.18b

Upon incremental additions of 0.5 µmol of Ag+ (Figure3B-H) to the 1.5 µmol of Au25L18

1- (0.3 mM solution), thecurrents for both oxidation peaks (iP1 and iP2) gradually decrease,and at >3:1 mol ratio (Figure 3H) are difficult to see. Thevoltammetry in Figure 3 is consistent with the absorbance and

Figure 1. UV-vis absorption spectra of 0 to 2:1 mol ratio of Ag+/Au25. 46 nmol of Au25

- (15.3 µM in CH2Cl2) with Ag+ added as 10mM AgNO3 in CH3CN: Panel a: 0 to 1:1 Ag+/Au25 mole ratio; Curves1-9: 0, 6, 12, 18, 24, 30, 36, 42, and 48 nmol added Ag+; Panel b: 1:1to 2:1 Ag+/Au25 mole ratio; Curves i to vi: 54, 60, 72, 84, 96, and 102nmol added Ag+. Dotted line is no added Ag+. Insets: Panel a,absorbance at 605, 682, and 790 nm vs mole added Ag+; Panel b:absorbance at 605 nm vs mole ratio Ag+/Au25. Isosbestic points (*) at438, 670, and 865 nm.

Figure 2. Plot of (integrated) photoluminescence (PL) intensity vsnmole Ag+ added to 32.4 nmol Au25L18

- in CH2Cl2. Inset is PL spectrafor 0 to 30 nmole Ag+. (See Supporting Information, Figure S-6, forPL spectra for 30-72 nmole Ag+.)

15892 J. Phys. Chem. C, Vol. 114, No. 38, 2010 Choi et al.

fluorescence observations (above) in signaling a reaction thatproduces a changed nanoparticle. It is notable in Figure 3 thatas higher equivalents of Ag+ are added (Panels E-H), featurelesscathodic currents (above baseline) appear at negative potentials,as if the bulk solution now contains a readily reducibleconstituent. This effect is consistent with the apparent oxidationeffect implied by the spectral results above, but also could reflectonset of nanoparticle destruction.28 It is also notable that, whiledecreasing in size, the formal potentials of the Au25L18

1- wavesremain unchanged (Table 1). Given the known sensitivity24 offormal potentials to the ligands of this nanoparticle, it seemsunlikely that the metal ion reaction involves dissociation ofligands to yield, for example, AgL complexes.

Mass spectrometry of Ag+/MPC Reaction Products. Moredirect characterization of reaction products of the above addi-tions of Ag+ to Au25L18

1- was pursued by mass spectrometry.We have previously demonstrated7b electrospray ionization massspectrometry (ESI-MS) of Au25L18

1-, analyzing it as Cs(acetate)solutions. Figure S-8 illustrates similarly obtained spectra ofAu25 adducts. Spectra were also detected for the oxidized

nanoparticle Au25L182+ without Cs+ in the solution, the nano-

particle apparently becoming oxidized as part of the ESI process.ESI-MS experiments were carried out on nanoparticle samples

to which 1, 1.5, or 2 equivalents of Ag+ had been added and,assuming possible instabilities, examined as quickly as possibleby ESI-MS. (In the UV-vis experiments, spectra were obtainedwithin minutes following metal ion additions, whereas the massspectrometry measurements entailed longer period of time asindicated in Experimental.) Results are shown in Figures 4 and5. Figure 4 shows the ESI-QQQ-MS 2+ ion spectrum obtainedfrom a 1:1 mol ratio of Ag+/MPC mixture, which after mixingwas immediately dried and redissolved in methanol:toluene:Cs(acetate) solution. The three peaks to the right, labeled (0,0),(1,0), and (2,0) correspond to adducts with 0, 1, and 2 Cs+ ionsin which the Au25 nanoparticle core bears 2+, 1+, and 0charges, becoming oxidized7b in the electrospray environment.The small peak at the left labeled (0,1) is assignable to a bimetalnanoparticle Au24Ag(SC2Ph)2+ in which Au has been replacedby a single Ag with no change in the total number of ligands.

Figure 5A,B shows similar experiments but with 1.5 and 2added equivalents of Ag+, respectively. The presence of Cs+

adducts appears to be suppressed by addition of larger amountsof Ag+, Figure 5 displays peaks assignable as CsXAu(25-Y)-AgY(SC2Ph)18sions that contain no Cs+ with y ) 3, 2, 1, and0 (especially panel B), and that contain (Panel A) one Cs+ withy ) 2, 1, and 0 and two Cs+ with y ) 1 and 0. These resultsagain show replacement of Au by Ag (as opposed to Ag+ simplyreplacing the ionizing Cs+, without loss of Au). To assesswhether residual AgNO3 in the dried reaction mixture (beforeanalysis, see above) might have any impact on the ESI-MSobservations, the dried reaction mixture was washed withmethanol prior to dissolving in methanol:toluene or methanol:CH2Cl2 for ESI-MS. For 1.5 added equivalents of Ag+, the

Figure 3. Cyclic voltammetry of 0.3 mM Au25L18- (1.5 µmole in 5 mL CH2Cl2; 0.1 M Bu4NPF6) with addition of Ag+: Curves a to h are 0, 0.5,

1.0, 1.5, 2.0, 3.0, 4.0, 5.0 µmole Ag+. Three mm diameter Pt working, Pt counter, Ag/Ag+ (10 mM AgNO3/CH3CN) reference electrodes. Potentialscan rate 100 mV/s.

TABLE 1: Electrochemical Dataa for Addition of Ag+ to aCH2Cl2 Solution Containing 1.5 µmol Au25(SC2Ph)18

-

µmol Ag+ Ep#1°′ (V)b Ep#2°′ (V)b - ip#1 (µA) - ip#2 (µA) ip#2/ip#1

0.0 -0.30 -0.02 0.47 0.46 0.980.5 -0.31 -0.02 0.36 0.40 1.111.0 -0.31 -0.01 0.31 0.35 1.131.5 -0.32 -0.01 0.19 0.24 1.212.0 -0.32 -0.02 0.15 0.21 1.403.0 -0.32 -0.02 0.13 0.20 1.544.0 -0.32 -0.02 0.05 0.12 2.405.0

a CV at 100 mV/s. b E°′ vs Ag/Ag+ (10 mM AgNO3 in CH3CN).

Reactivity of [Au25(SCH2CH2Ph)18]1- Nanoparticles J. Phys. Chem. C, Vol. 114, No. 38, 2010 15893

spectra showed the same peaks as seen in Figure 5A, except inmethanol:CH2Cl2 solvent the relative intensities were changed(Figure S-9). These tests produced spectra at good resolution,as shown in Figure 5A inset for the (0,1) peak that is well-fitted by an isotopic simulation for the bimetal nanoparticleAu24Ag(SC2Ph)18

2+. Attempts at MS/MS spectra were notsuccessful.

A secondary concern in these mass spectrometry experimentswas whether methanol might have caused some Au25L18

-

oxidation (seen sometimes in practice). As a control, MALDIwas performed on a Au25 sample to which 1.5 equivalents ofAgNO3 was added, afterward drying and removing any excessAg+ by washing the solid with water. The resulting spectrum,shown in Figure 6, showed both Au25L18

+ and Au24AgL18+,

demonstrating that observation of the bimetal nanoparticle isindependent of both the ionization process and the use ofmethanol.

These mass spectrometry observations are significant byshowing, unequivocally at good spectral resolution, that bimetal

ions exist in which Au has been replaced with equal numbersof Ag, with no change in ligand count. The observations arefurther significant, but puzzling, in that the most intense peaksare those in which no such replacement has taken place, e.g.,the original Au25(SC2Ph)18 nanoparticle. It is understood thatESI peak intensities do not necessarily parallel the populationdistribution, but the observation of unmodified Au25 seeminglycontradicts the spectral observations of Figure 1, where isos-bestic points indicate a quantitative reaction of the originalAu25L18

1- nanoparticle, in CH2Cl2 solution, with added Ag+,leading to anticipation that the original Au25L18

- compositionmight be absent. The Discussion Section addresses this apparentdisconnect with a model of reversible adduct formation.

Experiments With Other Metal Ions. Experiments werealso conducted in which small volumes of Cu2+ and Pb2+

solutions (concentrated, in CH3CN) were added to Au25L181-

solutions in CH2Cl2. These experiments were not as extensiveas for Ag+ but the results obtained were largely parallelsapparentstoichiometric reactions as reflected by changes in absorbancespectra that are shown in Figures S-10-S-13. Specifically, forboth metal ions, the Au25L18

1- shoulder at ca. 790 nm becomesdepressed and the spectral valley at 605 nm becomes “filledin”. A plot of absorbance against Cu2+/MPC exhibits a clearbreak at 0.5:1 mol ratio; Pb2+ does also but much less clearly.In both cases, isosbestic crossing points are seen up to 0.5:1mol ratio. At higher mole ratios of Cu2+, isosbestic behavior isnot seen; the spectral absorbance becomes generally lowered,and it is apparent that decomposition is occurring. Isolation ofproduct after 10 equivalents of Cu2+ were added produced largernanoparticles (1.6 ( 0.5 nm in diameter). For Pb2+, no furtherspectral change was seen. Photoluminescence spectra, voltam-metry, and mass spectrometry experiments were not conductedfor these metal ions. The absorbance spectral changes for bothmetal ions, like that for Ag+, imply an oxidation-like process,and unlike Ag+, a 0.5:1 reaction mole ratio.

Additionally, a careful comparison of the longer wavelengthchanges evoked for the three metal ions shows that they arenot exactly identical; the spectra for 1:1 Ag+/MPC mole ratioand 0.5:1 mol ratios for Cu2+ and Pb2+ are all somewhatdifferent (Figure S-14, the middle spectra are Figure 1a.) Thisimplies that the “oxidized” nanoparticles produced, while similarin broad form, are in their details, unique to the metal ionreaction that produced them.

Figure 4. Electrospray ionization (ESI) mass spectra of Au25L18- with

1 equivalent of added Ag+. The analyzed solution was 70:30 methanol:toluene containing 2.6 mg of Cs(acetate) per mg nanoparticle. Numbersin parentheses correspond to x and y of formula CsXAu(25-Y)AgY-(SC2Ph)18. [Au25(SC2Ph)18]2+ is the prominent peak; at left is a smallpeak for [Au24Ag(SC2Ph)18]2+.

Figure 5. Electrospray ionization (ESI) - mass spectra of Au25L18- with 1.5 (panel A) and 2 (panel B) equivalents of added Ag+. The analyzed

solution was 70:30 methanol:toluene containing 2.6 mg Cs(acetate) per mg nanoparticle. Numbers in parentheses correspond to x and y of formulaCsXAu(25-Y)AgY(SC2Ph)18. Panel A inset is high resolution of (0,1) peak of Panel A with simulation (red line) for [Au24Ag(SC2Ph)18]2+. This peakwas obtained in a different run from Panel A and sample had been washed with methanol to remove any free AgNO3.

15894 J. Phys. Chem. C, Vol. 114, No. 38, 2010 Choi et al.

Discussion

We note above that the spectral results imply a 1:1 stoichio-metric, oxidation-like reaction between Ag+ and the Au25L18

1-

MPC, and the mass spectral results show that Au24AgL18

nanoparticles are produced. However, the mass spectra showrelatively larger peaks for the precursor Au25 nanoparticle, inapparent contradiction by implying a substantial remnant ofunreacted Au25 nanoparticles.

We rationalize the above results with a model that permitsthe metal ion reaction to be reVersible. We propose that theAu25L18

1- reaction with metal ions is a redox process, in whichthe Au25L18

1- nanoparticle acts as a reductant toward the metalion, in the case of Ag+ evoking the formation of an adduct

where the Ag0 added to the nanoparticle occupies a former Ausite and/or is coordinated to the Au13 core. The other metalsbehave similarly but act as two electron oxidantsshence their0.5:1 stoichiometry. The thermodynamics of such redox pro-cesses can be justified by relative nanoparticle and metal ion(Ag+ and Cu2+) redox potentials (crudely, since E°′ in CH2Cl2

and water will differ29), and invoking ideas of underpotentialdeposition30 in binding a Cu0 or Pb0 atom to a Au13 nanoparticlecore.

This model rationalizes the stoichiometric reaction behaviorexpressed in the absorbance spectral data. It also accounts forthe oxidation-like changes in the absorbance spectra, and isconsistent with the differing absorbance spectra of oxidized-like nanoparticles (Figure S-14) produced by metal ion addition;the [Au25ML18]0 species produced would differ with the identityof the metal “M”. The original well-defined voltammetricdoublet is lost when the adduct is formed.

The adducts are not very stable and the nanoparticles arereformed into larger ones by further reactions with added metalions. Importantly, adduct formation is reversed in the cation-ization processes of the ESI and MALDI mass spectra experi-ments. That both experiments are effectively mild oxidizing

environments is evident in the production of Au25L182+ (Figures

4 and 5) and Au25L181+ (Figure 6). Ag rather than Au loss is

apparently favored in oxidative adduct dissociation, producingmore intense mass spectral peaks at the original Au25L18 massthan Au24AgL18. The latter is, however, produced in sufficientabundance to be observed. When the mole ratio Ag+/nanopar-ticle is >1:1, higher adducts are formed by reactions like eq 1,producing Au23Ag2L18.

Finally, recent experimental14 and theoretical work15 hasshown that the replacement of a Au site in Au25 with anothermetal results in electronic and optical disturbances. The calcula-tions include15a,b incorporation of Ag into a Au nanoparticle,[AgAu24L18]1-. According to the above interpretation of metalion reactions with [Au25L18]1-, however, the optical spectrareported here are instead for adducts and so cannot be readilycompared with results of the theoretical efforts. The massspectral data do confirm, on the other hand, that bimetalnanoparticles can be formed.

Acknowledgment. This work was supported in part by theNational Science Foundation. Equal contributions were madeby coauthors J.P.C. and C.A.F.Z.

Supporting Information Available: UV-vis and PL spectrawith different metals added, thermogravimetric data, X-rayphotoelectron spectra, ESI-TOF mass spectra, titration plots forCu2+ and Pb2+. This material is available free of charge via theInternet at http://pubs.acs.org.

References and Notes

(1) (a) Schmid, G. Chem. ReV. 1992, 92, 1709. (b) Zhong, C. J.; Maye,M. M. AdV. Mater. 2001, 13, 1507.

(2) (a) Bergers, R.; Delamarche, E.; Lang, H. P.; Gerber, C.; Gimze-wski, J. K.; Meyer, E.; Guntherodt, H.-J. Science 1997, 276, 2021. (b)Dermody, D. L.; Lee, Y.; Kim, T.; Crooks, R. M. Langmuir 1999, 15, 8435.(c) Ostuni, E.; Yan, L.; Whitesides, G. M. Colloids Surf. B 1999, 15, 3. (d)Cooper, M. A. Nat. ReV. Drug DiscoV. 2002, 1, 515.

(3) (a) Tanabe, K. Mater. Lett. 2007, 61, 4573. (b) Zheng, J.; Nicovich,P. R.; Dickson, R. M. Annu. ReV. Phys. Chem. 2007, 58, 409. (c) Kamat,P. V. J. Phys. Chem. B 2002, 106, 7729.

(4) (a) Willets, K. A.; Van Duyne, R. P. Annu. ReV. Phys. Chem. 2007,58, 267. (b) Moskovits, M. J. Raman Spectrosc. 2005, 36, 485. (c) VanDuyne, R. P. In Chemical and Biochemical Applications of Laser; Moore,C. B., Ed.; Academic Press: New York, 1979; Vol. 4, pp 101-185.

(5) (a) Hostetler, M. J.; Green, S. J.; Stokes, J. J.; Murray, R. W. J. Am.Chem. Soc. 1996, 118, 4212–4213. (b) Ingram, R. S.; Hostetler, M. J.;Murray, R. W. J. Am. Chem. Soc. 1997, 119, 9175. (c) Templeton, A. C.;Wuelfing, W. P.; Murray, R. W. Acc. Chem. Res. 2000, 1, 27–36. (d)Warner, M. G.; Reed, S. M.; Hutchison, J. E. Chem. Mater. 2000, 12, 3316.(e) Yang, Y.; Chen, S. Nano Lett. 2003, 3, 75. (f) Choi, J.-P.; Coble, M. M.;Branham, M. R.; DeSimone, J. M.; Murray, R. W. J. Phys. Chem. C 2007,111, 3778.

(6) (a) Shon, Y.-S.; Dawson, G. B.; Porter, M.; Murray, R. W.Langmuir 2002, 18, 3880. (b) Huang, T.; Murray, R. W. J. Phys. Chem. B2003, 107, 7434.

(7) (a) Tracy, J. B.; Kalyuzhny, G.; Crowe, M. C.; Balasubramanian,R.; Choi, J.-P.; Murray, R. W. J. Am. Chem. Soc. 2007, 129, 6706. (b)Tracy, J. B.; Crowe, M. C.; Parker, J. F.; Hampe, O.; Fields-Zinna, C. A.;Dass, A.; Murray, R. W. J. Am. Chem. Soc. 2007, 129, 16209. (c) Dass,A.; Stevenson, A.; Dubay, G. R.; Tracy, J. B.; Murray, R. W. J. Am. Chem.Soc. 2008, 130, 5940.

(8) (a) Negishi, Y.; Nobusada, K.; Tsukuda, T. J. Am. Chem. Soc. 2005,127, 5261. (b) Shichibu, Y.; Negishi, Y.; Tsukuda, T.; Teranishi, T. 2005,127, 13464. (c) Negishi, Y.; Chaki, N. K.; Shichibu, Y.; Whetten, R. L.;Tsukuda, T. J. Am. Chem. Soc. 2007, 129, 11322.

(9) Heaven, M. W.; Dass, A.; White, P. S.; Holt, K. M.; Murray, R. W.J. Am. Chem. Soc. 2008, 130, 3754.

(10) Akola, J.; Walter, M.; Whetten, R. L.; Hakkinen, H.; Gronbeck,H. J. Am. Chem. Soc. 2008, 130, 3756–3757.

(11) (a) Hakkinen, H.; Walter, M.; Groenbeck, H. J. Phys. Chem. B2006, 110, 9927–9931. (b) Walter, M.; Akola, J.; Lopez-Acevedo, O.;Jadzinsky, P. D.; Calero, G.; Ackerson, C. J.; Whetten, R. L.; Gronbeck,H.; Hakkinen, H. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 9157–9162.

Figure 6. MALDI-MS spectrum of Au25L18- with 1.5 equivalents of

added Ag+; sample had been washed with water to remove any freeAg+. Spectrum shows predominantly Au25L18

+ but also the bimetalAu24AgL18

+, showing that presence of bimetals is innate to adding Ag+

and not to the mass spectrometry ionization process or solvent system.

Au25L18- + Ag+ f [Au25AgL18]

0 (1)

Reactivity of [Au25(SCH2CH2Ph)18]1- Nanoparticles J. Phys. Chem. C, Vol. 114, No. 38, 2010 15895

(12) Jadzinsky, P. D.; Calero, G.; Ackerson, C. J.; Bushnell, D. A.;Kornberg, R. D. Science 2007, 318, 430–433.

(13) Lopez-Acevedo, O.; Akola, J.; Whetten, R. L.; Gronbeck, H.;Hakkinen, H. J. Phys. Chem. C 2009, 113, 5035–5038.

(14) (a) Fields-Zinna, C.; Crowe, M. C.; Dass, A.; Weaver, J. E. F.;Murray, R. W. Langmuir 2009, 25, 7704–7710. (b) Murray, R. W. Internat.Sympos. Monolayer-Protected Clusters, Univ. Jyvaskyla: Finland, 2008.

(15) (a) Katarzyna, A.; Kacprzak, K. A.; Lehtovaara, L.; Akola, J.;Lopez-Acevedo, O.; Hakkinen, H. Phys. Chem. Chem. Phys. 2009, 11,7123–7129. (b) Jiang, D.; Dai, S. Inorg. Chem. 2009, 48, 2720–2722. (c)Walter, M.; Moseler, M. J. Phys. Chem. C 2009, 113, 15834–15837.

(16) (a) In Handbook of PreparatiVe Inorganic Chemistry; Brauer, G.,Ed.; Academic Press: New York, 1965; p 1054. (b) Block, B. P. Inorg.Syn. 1953, 4, 14.

(17) Donkers, R. L.; Lee, D.; Murray, R. W. Langmuir 2004, 20, 1945.(18) (a) Lee, D.; Donkers, R. L.; Wang, G.; Harper, A. S.; Murray, R. W.

J. Am. Chem. Soc. 2004, 126, 6193. (b) Zhu, M. Z.; Eckenhoff, W. T.;Pintauer, T.; Jin, R. C. J. Phys. Chem. C 2008, 112, 14221–14224.

(19) Wang, G.; Guo, R.; Kalyuzhny, G.; Choi, J.-P.; Murray, R. W. J.Phys. Chem. B 2006, 110, 20282.

(20) Hostetler, M. J.; Wingate, J. E.; Zhong, C.-J.; Harris, J. E.; Vachet,R. W.; Clark, M. R.; Londono, J. D.; Green, S. J.; Stokes, J. J.; Wignall,G. D.; Glish, G. L.; Porter, M. D.; Evans, N. D.; Murray, R. W. Langmuir1998, 14, 17.

(21) Choi, J.-P.; Murray, R. W. J. Am. Chem. Soc. 2006, 128, 10496.

(22) Wuelfing, W. P.; Green, S. J.; Pietron, J. J.; Cliffel, D. E.; Murray,R. W. J. Am. Chem. Soc. 2000, 122, 11465.

(23) Pietron, J. J.; Hicks, J. F.; Murray, R. W. J. Am. Chem. Soc. 1999,121, 5565.

(24) Chen, S.; Ingram, R. S.; Hostetler, M. J.; Pietron, J. J.; Murray,R. W.; Schaaff, T. G.; Khoury, J. T.; Alvarez, M. M.; Whetten, R. L. Science1998, 280, 2098.

(25) Guo, R.; Murray, R. W. J. Am. Chem. Soc. 2005, 127, 12140.(26) Antonello, S.; Holm, A. H.; Instuli, E.; Maran, F. J. Am. Chem.

Soc. 2007, 129, 9836.(27) Lee, D.; Donkers, R. L.; DeSimone, J. M.; Murray, R. W. J. Am.

Chem. Soc. 2003, 125, 1182.(28) The reduction currents in Figure 3, Panels E-H are not caused by

adventitious deposition of Ag0 on the working electrode. No change incurrents at negative potentials was observed in a control CV experimenton a 1 mM Ag+ solution in CH2Cl2.

(29) Redox potentials vs. SHE (the comparisons are approximate,between aqueoussthe metal ionssand organic solvents): Ag+/0(aq), E° )0.799V; [Au25L18]0/1- (in CH2Cl2, similar in CH3CN), E°′ ) 0.16V;Cu2+/0(aq), E° ) 0.34V; Pb2+/0(aq), E° )-0.13V.

(30) (a) Hernandez, J.; Solla-Gullon, J.; Herrero, E.; Feliu, J. M.; Aldaz,A. J. Nanosci. Nanotechnol. 2009, 9, 2256–2273. (b) Herrero, E.; Buller,L. J.; Abruna, H. D. Chem. ReV. 2001, 101, 1897–1930.

JP9101114

15896 J. Phys. Chem. C, Vol. 114, No. 38, 2010 Choi et al.