Is Pd2(DBA)3 a Feasible Precursor for the Synthesis of Pd Nanoparticles?

D. N. Leonard and S. Franzen*Department of Physics and Astronomy, Appalachian State UniVersity, Boone, North Carolina 28608, Department ofChemistry, North Carolina State UniVersity, Raleigh, North Carolina 27695, and W.M. Keck Center for RNAMediated EVolutionary Materials Synthesis, North Carolina State UniVersity, Raleigh, North Carolina 27695

ReceiVed: April 14, 2009; ReVised Manuscript ReceiVed: May 23, 2009

The organometallic reagent tris(dibenzylideneacetone) dipalladium(0) [Pd2(DBA)3] has been used as a reagentfor synthesis of hexagonal microstructures ranging from 0.5-10 µm. We have shown elsewhere that Pd2(DBA)3

spontaneously forms hexagonal and cubic microstructures. The results obtained herein confirm thatmicrostructures formed by spontaneous aggregation of Pd2(DBA)3 produce the same morphology andcarbonaceous chemical composition as those produced with a process that has been called RNA-mediationby pyridyl-modified RNA (pmRNA). Synthesis products from both methods of preparation were analyzedusing a combination of scanning transmission electron microscopy (STEM), electron energy loss spectroscopy,energy dispersive X-ray spectroscopy, and selected area electron diffraction. Indexing of electron diffractionpatterns from hexagonal molecular crystals and metallic Pd nanoparticles combined with STEM Z-contrastimaging and X-ray dot maps were used to positively differentiate the two materials. On the basis of theseresults, we conclude that the observed microparticles with hexagonal morphology are composed of Pd2(DBA)3.Surfactants such as pmRNA sequence Pd017 or Triton X-100 stabilize suspensions of spontaneously formedcrystalline aggregates of Pd2(DBA)3 in aqueous mixtures of tetrahydrofuran and H2O, but they are not requiredfor their formation. Crystalline Pd metal nanoparticle composites that range from 50 to 200 nm in diameter,that are observed in both spontaneous and RNA-mediated preparations, are an impurity that arises during thesynthesis of Pd2(DBA)3 by chemical reduction. There is no evidence for a competing decomposition processthat forms crystalline Pd metal nanoparticles on the time scale of the experiments.

Introduction

Tris(dibenzylideneacetone) dipalladium(0), Pd2(DBA)3, is arelatively stable complex of zerovalent Pd, in which each Pdcenter is coordinated by three η2-alkene ligands.1 The decom-position of Pd2(DBA)3 to Pd nanoparticles in the solid stateoccurs on the time scale of months and is quite slow comparedto other zerovalent Pd solids such as Pd(PPh3)4. However, thesolution chemistry of Pd2(DBA)3 is complex due to ligandexchange reactions including oxidative addition leading tocatalytic cycles during which Pd changes its redox state.Colloidal Pd often forms as a byproduct of redox cycling. Onthe other hand, in the absence of ligands that undergo oxidativeaddition in solution, Pd2(DBA)3 crystallizes in CH2Cl2, CHCl3,benzene, and toluene to yield crystals of Pd2(DBA)3(solvent).2-4

Thus, there is a solvent- and ligand-dependent competitionbetween formation of metallic Pd nanoparticles and molecularcrystals of Pd2(DBA)3.

The coordination chemistry of Pd(II) and Pd(0) has beenextensively studied to understand the role played by bothoxidation states in Pd catalysis.5 Studies of the displacementreactions of Pd2(DBA)3 revealed that it is susceptible tooxidative addition of propenyl chloride,2,6,7 which is useful forits application as a heterogeneous catalyst. However, given thatreductive elimination does not return the Pd atom to the originalPd2(DBA)3 adduct, the catalytic species is the Pd atom itselfrather than Pd2(DBA)3. The Pd(0) species that forms during thecycle can aggregate to form colloidal particles known as Pdblack.8 Formation of Pd black is not rapid compared to catalysis

under most conditions, and it is dependent on the ligationstrength of the Pd(0) coordination sphere. Pd is stabilized bysoft ligands such as PPh3 and the η2-coordinated alkenes of DBAas well as many π-systems including quinones, alkynes,cyclopentadiene, etc.9-11 Nitrogenous ligands such as bidentateligands bipyridine and o-phenanthroline12-15 and pyridine haveplayed a more limited role in the coordination chemistry ofPd(0). Primary amines such as hexadecyl amine have hadapplication as a stabilizer in nanoparticle synthesis followinghydrogenation reactions.16 Systematic study of these ligandsleads to the concept that Pd acts as the donor (nucleophile) inthese adducts and the π*-acceptor orbitals of the ligand(electrophile) are key to stability of the complex. The “inverseelectron demand”14 of Pd(0) leads to a trend in ligation strengththat is the reverse of that normally observed where the ligandacts as the nucleophile and the metal is the electrophile. In thistype of bonding Pd(0) acts as an electron donor to π* orbitalsof the coordinating ligand. Unlike more conventional π-back-bonding, there is little stabilization arising from σ donation toPd(0). Pyridine is a poor π* acceptor, which may explain whyno homoleptic pyridine adduct of Pd(0) has been reported. Theoccurrence of pyridine in Pd(0) coordination chemistry is rare.17

Pyridine is known to play a role in certain catalytic reactionsthat may be dependent upon ligation to Pd(II) adducts.18,19

Although nitrogenous ligands are capable of stabilizing Pdnanoparticles, they do not displace η2-coordinating ligands suchas DBA unless the olefin is hydrogenated.16 An understandingof ligand coordination strength to Pd is of key importance forcatalyst design. Not only must the relative ligation strength ofsubstrate and product be tuned for catalytic turnover, but the

* To whom correspondence should be addressed. E-mail: stefan_franzen@ncsu.edu.

J. Phys. Chem. C 2009, 113, 12706–1271412706

10.1021/jp903417r CCC: $40.75 2009 American Chemical SocietyPublished on Web 06/18/2009

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

stability in solution must be considered since formation ofcolloidal Pd removes catalytic Pd from solution.

There has been an interest in Pd nanoparticle catalysts as analternative to homogeneous catalysis by Pd complexes.20-25

Aside from potential for novel reactivity, heterogeneous catalysisby Pd has the potential to mitigate the competition betweenredox cycling of catalytic Pd complexes and formation ofnoncatalytic Pd black. This strategy turns the colloidal foe ofcatalysis into a friend by harnessing well-defined nanoparticlestructures as catalytic sites. Heterogeneous Pd nanoparticlecatalysts have been found to have activity in hydrogenationreactions and, more recently, in C-C bond formation.20,21 Thesynthesis of Pd nanoparticles26 and nanoparticle clusters, alsocalled giant nanoparticles,27 has been demonstrated starting fromPd(II) compounds such as Pd(OAc)2, Pd(acac)2, and Na2PdCl4.The discovery Pd particles of well-defined stoichiometry28 hasled to an interest in the synthesis of particles of controlled size.29

Cubo-octahedral and icosahedral Pd particles are formed thatappear to follow the relationship of Chini’s series where thenumber of atoms around a central Pd atom, N ) 1/3(10m3 +15m2

+11m+ 3) ) 13, 55, 147, 309, 561, etc., for m ) 1, 2, 3, 4, 5,etc., defines successive coordination spheres.27-31 Shape controlhas been introduced into the synthesis of Pt and Pd nanoparticlesby control of surfactant and reducing conditions starting fromPd(II) and Pt(II) precursors.24,32-34 In all cases the particles areformed by chemical reduction. When Pd2(DBA)3 is used asstarting material, H2 gas is used to hydrogenate the olefinicbonds of the DBA ligand in order to release Pd(0).16 Pdnanoparticle clusters of 50-200 nm diameter, observed inpreparations of Pd2(DBA)3, appear to be impurities present fromthe synthesis of Pd2(DBA)3,35,36 which is accomplished byreduction of Pd(II) complexes in the presence of excess DBA.1

Spontaneous formation of Pd nanoparticles from Pd2(DBA)3

in the absence of a reducting agent appears to be sufficientlyslow that heating or other input of energy is required even tomake composite materials. Nanoparticle synthesis methods usingPd2(DBA)3 as the precursor including sonochemical decomposi-tion37 and thermal annealing38 have been reported. Theseconditions for nanoparticle formation involve elevated temper-ature as a necessary step to break the η2-bonds that stabilizezerovalent Pd in Pd2(DBA)3. However, neither of these methodsis sufficient to create pure Pd nanoparticles, but rather theyproduce carbon fibers that contain Pd nanoparticles. The strongaffinity of Pd(0) for unsaturated carbon bonds was recognizedby Ito et al. in studies of olefinic groups that were capable ofdisplacing DBA ligands.9 This reactivity, noted above as theinverse electron demand of Pd, apparently extends to carbonnanotubes, which present a novel templating agent for thecreation of Pd wires capable of hydrogen sorption based onreaction with Pd2(DBA)3.39 Given these considerations, thereport of pyridyl-modified ribonucleic acid (pmRNA) mediationof Pd-Pd bond formation at room temperature is surprising.40-42

When Pd2(DBA)3 is added to a solvent such as tetrahydro-furan (THF) or THF/H2O mixtures, micrometer-sized structuresform spontaneously.35,36 These structures have hexagonal plate,rod, and star morphologies that may arise from the nearlytrigonal symmetry of molecular Pd2(DBA)3 crystals obtainedin THF. The triclinic unit cells of Pd2(DBA)3 crystals in solventssuch as CHCl3 and CH2Cl2 contain a cocrystallized solventmolecule.2,3 The lattice parameters are sufficiently close to atrigonal unit cell that these cocrystallized solvent molecules maybe responsible for a distortion of the lattice. While there is noX-ray crystal structure of Pd2(DBA)3 in THF, the hexagonalselected area electron diffraction (SAED) spot pattern observed

from Pd2(DBA)3 hexagonal microcrystals at short times (i.e.,prior to beam damage) is consistent with a trigonal unit cellthat contains only Pd2(DBA)3 and no solvent molecule.35 Hence,the observed hexagonal-shaped organic crystal could be apseudohexagonal habit of the trigonal Pd2(DBA)3 unit cell. Thepseudohexagonal form of microcrystalline Pd2(DBA)3 formedin THF is difficult to study by electron diffraction due to thefact that these hexagonal organic crystals are only transientlystable under a 200 kV electron beam.35 After less than oneminute in the electron beam, the sample begins to losecrystallinity due to the formation of small (2-4 nm) Pdnanoparticles throughout the microcrystal. The formation of Pdnanoparticles35 inside a larger structure provide an indicationthat the hexagonal microplatelets are composed of Pd2(DBA)3

rather than metallic Pd.

Pd2(DBA)3 microcrystals of both cubic and hexagonalmorphology35,36 are remarkably similar to the forms observedfor metallic Pd nanocrystals, but the length scale of thePd2(DBA)3 microcrystals is at least 10 times larger in all casesand can be up to 1000 times larger in specific cases.26,28 Giventhe morphological similarity but difference in length scale, it isof interest to compare Pd nanocrystals to microcrystals withthe hexagonal and cubic structures which were reported to bemediated by pmRNA in a sequence-dependent manner usingPd2(DBA)3 as the starting material.41 The hexagonal Pd nano-crystals observed on the 10-100 nm length scale24,26,43,44 beara stunning similarity to the morphology of Pd2(DBA)3 micro-crystals on the 0.5-10 µm length scale. Cubic Pd nanoparticleshave also been reported.28,45 These are also observed on a lengthscale of 2-4 nm,46 which is ∼500 times smaller the micrometer-sized cubic particles reported as products pmRNA-mediation.41

Larger Pd nanoparticles have been observed, but these aremultiply twinned clusters of Pd561 particles, which is the mostcommonly observed type of cuboctahedral particle.28 Themorphological similarities of different shapes, hexagons, andcubes on different length scales suggests that the chemicalcomposition and structure of the micrometer-sized Pd2(DBA)3

particles should be carefully verified. In previous studies, weestablished that the micrometer-sized Pd2(DBA)3 particles ofhexagonal and cubic morphology are essentially identical tothose reported as the products of pmRNA mediation.36 In thefollowing, we study particles produced from Pd2(DBA)3 withand without pmRNA to compare two hypotheses that have beenadvanced to explain the observation of micrometer-sizedparticles derived from Pd2(DBA)3 in THF/H2O suspensions.

The hypothesis of pmRNA-mediated metal-metal bondformation40-42 states that pmRNA strips away at least one ofthe DBA ligands and presents the face of a Pd crystal that growsin a controlled morphology. The inspiration for this type ofcontrolled growth can be found in biomineralization of allkinds.47 This type of process is observed in the interaction ofnucleic acids during the growth of CdS quantum dots48-51 andAg nanoclusters.52 In the structures studied here,35,36,40-42,53 apyridyl modification is carried on every uridine of the RNAsequence Pd017 as a 5-(4-pyridylmethyl) carboxamide group.The Pd017 sequence contains 12 uridines in its variable region.40

Thus, the metalation and templating mechanism would neces-sarily involve binding interactions of the pyridinyl groups withPd(0) atoms from Pd2(DBA)3. The proposed mechanism isunusual since all reported syntheses of Pd nanoparticles usereduction of Pd(II),20-26,28,32,33 and the formation of Pd blackfrom Pd(0) in organic reaction mixtures has not led to theisolation of nanoparticles. Moreover, the pyridine side chain in

Pd2(DBA)3: Precursor for Pd Nanoparticles? J. Phys. Chem. C, Vol. 113, No. 29, 2009 12707

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

pmRNA is an unusual choice as a chemical mediator becauseof its weak coordination to Pd(0).12-15,54

The alternative hypothesis is that the hexagonal particles areformed spontaneously as microcrystals of Pd2(DBA)3 in THF/H2O solutions and that pmRNA plays the role of a surfactantwhen it is present. Stabilization of particles of Pd2(DBA)3 inaqueous solution has precedent using the detergent Triton X-100as a stabilizer.55 Thus, pmRNA may act to slow precipitation,thereby providing metastable suspensions of Pd2(DBA)3 undersome conditions. The use of surface-attached pmRNA has alsobeen reported, and particles of both cubic and hexagonalmorphology on surfaces have been observed using TEM.42

Atomic force microscopy (AFM) has also provided evidenceon the nanometer scale for hexagonal particles on etchedsurfaces that present a grid of surface attached pmRNAmolecules.53 However, control experiments reveal that hexagonaland cubic particles are formed on surfaces even withoutpmRNA.35,36 Our previous reports35,36 have established theidentity of Pd2(DBA)3 particles that spontaneously formed inTHF/H2O solutions and on surfaces. By use of transmissionelectron microscopy/selected area energy diffraction (TEM/SAED), scanning tunneling electron microscopy/electron energyloss spectroscopy (STEM/EELS), and scanning electron mi-croscopy/electron diffraction spectrometry (SEM/EDS) methods,we determined the morphology, chemical composition, andstructure of the hexagonal Pd2(DBA)3 microcrystals (0.5-10µm) and smaller Pd impurities that are clusters of 2-4 nmnanoparticles with a total diameter of 50-200 nm. Herein, wepresent further supporting evidence for the hypothesis thatmediation by pmRNA Pd017 does not significantly alter thecomposition of the spontaneously formed Pd2(DBA)3 microc-rystals either in solution or on surfaces. The present studyrepresents the first direct comparison of microparticles producedwith and without added pmRNA Pd017.

Methods

Sample Preparation. The Pd2(DBA)3 reagent powder(STREM, Lot#A4805106) was used as received and unpurified.14.65 mg were dissolved in 2 mL of THF (from Fischer, distilledprior to use) to produce 8 mM solutions of Pd2(DBA)3. Thissolution was added to variable amounts of water and THF tocreate 400 µM Pd2(DBA)3 solutions and resulted in suspensionsof Pd2(DBA)3 in binary solutions with ranging from 30 vol %THF/H2O, to 100 vol % THF. When pmRNA was dissolved inthe suspensions, diethylpyrocarbonate (DEPC) water was used.These solutions were then cast and allowed to dry in ambientatmosphere onto either indium tin oxide (ITO, Delta Technolo-

gies), lacey carbon TEM grids (Ted Pella Part No. 01883),Formvar-coated TEM grids (Ted Pella Part No. 01820), or bareAu. More in-depth sample preparation and observations of thesolution turbidity and precipitate formation after addingPd2(DBA)3 was previously reported.35

RNA Pd017 was made by in vitro transcription from PCR-generated or synthesized DNA templates as described previ-ously.41 For production of modified RNA, 5-(4-pyridylmethyl)carboxamide UTP, analyzed by liquid chromatography-tandemmass spectrometry,56 was substituted for unmodified UTP instandard transcription reactions. RNA concentrations wereestimated by absorbance at 260 nm using calculated extinctioncoefficients.

Surface attachment of RNA to ITO was carried out bymodification of the ITO surface based on the following protocol,which is illustrated in Figure 1. ITO slides were first hydroxy-lated, by immersion in a 10 mM NaOH solution overnight at60 °C (2 mL/2 slides; NaOH solution was made using enzyme-free water). Then, they were rinsed with enzyme-free water,N2 dried, and UVO cleaned for 5 min. The slides were immer-sed in a 1% solution of 3-amino-propyltrimethoxysilane (3-APS)in anhydrous toluene (total volume: 2 mL/2 slides) as shownin Figure 1A. The reaction was carried out at 60 °C, in air, for30 min. The slides were then rinsed with toluene and absoluteethanol and cured overnight at 120 °C. The heterobifunctionalcross-linking agent SSMCC (4-(N-maleimidomethyl)cyclohex-ane-1-carboxylic 3-sulfo-n-hydroxysuccinimide ester) was coupledto the amino groups presented on the surface using 2 mg/mLof SSMCC in 0.05 M KP buffer at pH 7.2 by soaking for 1 hat ambient temperature (Figure 1B). The maleimide-activatedslides were then rinsed and immersed in the solution (1 µMpmRNA; 0.05 M KP buffer; pH 7.2) for 3 h (Figure 1C).Subsequently the slides were rinsed thoroughly, first with bufferand then with water, and dried in N2. To create a surface-attached pmRNA, 5′-poly(ethylene)glycyol-thioester-linked-pmRNA was added to the maleimide-coated ITO surface (Figure1D). The Michael addition is the coupling reaction that bindspmRNA to ITO (Figure 1E). The preparation of 5′-poly(ethyl-ene)glycyol-thioester-linked-pmRNA shown in Figure 1F hasbeen discussed elsewhere.42 The surfaces were hydrated with athin layer of water, treated with 50 vol % THF/H2O to 100%vol THF solution, and allowed to evaporate to dryness followingthe previous protocol.42

SEM and EDS. A Hitachi S-3200N, equipped with aRobinson backscattered electron detector (BSE) (for atomicnumber contrast imaging) and Oxford Link ISIS EDS (forquantitative chemical analysis and X-ray dot maps), was

Figure 1. (A) Depiction of the steps to functionalize an ITO surface. Following hydroxylation 3-APS is added. Reaction of 3-APS presents anamide group on the surface. The SSMCC linker is then added and creates an amide bond to create SSMCC-ITO. The hydroxysuccinimide groupof the NHS-ester is an excellent leaving group. (B) The 5′-thioester-labeled-pmRNA is added to the solution. A Michael addition takes placecreating a covalent bond between the pmRNA and the ITO surface (pmRNA-ITO). (C) The pyridyl-modified uridine is shown. The5′-poly(ethylene)glycol thioester linker is also shown.

12708 J. Phys. Chem. C, Vol. 113, No. 29, 2009 Leonard and Franzen

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

employed in this study. The operating voltage for imaging andchemical analysis was 20 kV. To eliminate sample charging,the variable pressure of the sample chamber was set to 60 mbar.Digital image acquisition of BSE micrographs and quantitativechemical analysis was accomplished by using the Oxford LinkISIS software.

Low-Voltage STEM. A Hitachi S5500 low-voltage STEM(Plesanton, CA) equipped with a cold cathode field emissionelectron source was capable of 0.4 nm resolution in secon-dary electron mode when operated at 30 kV. This in-lens field-emission STEM was equipped with secondary and backscatterelectron detectors with the ability to mix both signals and captureimages in a magnification range of 100-2,000,000×. Themicroscope was also equipped with a bright field and anadjustable dark field detector allowing a range of collectionangles to be captured for diffraction or Z-contrast imaging.Chemical analysis in the microscope was performed with anOxford INCA (Oxfordshire, UK) energy dispersive X-rayspectrometer attached to the LV-STEM sample chamber. TheX-ray dot maps and spectrum reports were generated using theINCA software.

TEM and SAED. Phase contrast imaging with the HRTEMand EELS analysis was performed with a JEOL 2010Ftransmission electron microscope operated at 200 kV (λ ) 2.51pm) equipped with a high-resolution pole piece (URP2F) usedto capture lattice fringe micrographs.57 Column alignment wasperformed with the first condenser lens in TEM probe mode,spot size set at 5 (∼4.48 V) and the condenser mini lens set atR ) 3 (∼3.93 V). Conventional brightfield (BF) images werecaptured with a Gatan Imaging Filter (GIF) coupled to themicroscope using Digital Micrograph 3.3.1. Digital imageacquisition parameters included a scan rate of 0.5 s and a finalimage resolution of 1024 × 1024 pixels. All HRTEM imageswere acquired at the instruments optimum contrast transferfunction (termed Scherzer focus).58 Obtaining this imagingcondition by utilizing the live FFT of the real image ensuredtheoretical requirements for micrograph contrast interpretationcould be applied.

SAED was used to document the long and short-range atomicordering of the features of interest. The resulting SAED patternswere acquired at camera lengths of 60, 80, and 100 cm andthen recorded on Kodak Electron image film (SO - 163). Toindex the resulting polycrystalline SAED ring pattern, thecamera constant of the JEOL 2010F was calibrated at cameralength set to 60 cm according to standard procedures.

Electron beam analysis with fast electrons can cause tempo-rary or permanent changes to occur in the sampled volume dueto both elastic and inelastic scattering of incident electrons.Inelastic scattering can cause ionization damage, changes tointernal structure and surface hydrocarbon contamination. Incrystalline materials of atomic numbers greater than ∼40, theenergy required to “knock” an atom from its position in thelattice was well above the incident electron energy used in thisstudy. But for carbon-based materials (except diamond), multipleinelastic scattering events lead to irradiation artifacts above 140keV.59 Low-voltage (i.e., 30 kV) STEM-minimized imagingartifacts created by electron beam damage by allowing observa-tion and chemical analysis of the Pd2(DBA)3 nanostructures.The combined utilization of 10 kV and 20 kV FESEM/EDS,30 kV LV-STEM/EDS, 80 kV and 200 kV TEM/STEM/EELSimproved flexibility for employing the correct “tool” for theanalysis needed to identify various organic and inorganicstructures on the nanoscale.

Results

The formation of hexagonal microcrystals was compared withand without surface-attached pmRNA (see Methods). Initially,we compared the appearance of the hexagonal microplateletsby optical microscopy (OM). Then, HRTEM and SEM methodswere employed to determine the composition of the hexagonalmicroplatelets when pmRNA is present. These results arecompared to previous results showing the morphologicalsimilarity to microcrystals of Pd2(DBA)3. In all cases, beamdamage alters the crystals, which is consistent with theirassignment as organic rather than metallic microcrystals. Allof the methods agree that the composition of the pmRNA-mediated particles is the same as the spontaneously formedparticles.

Hexagonal Microstructures Formed on Surfaces with andwithout pmRNA. By use of OM we compare the morphologyof hexagonal platelets prepared in pmRNA-free solutions andin the presence of surface-attached pyridyl-modified-RNA Pd017using the methods shown in Figure 1. Figure 2 showscomparison of hexagonal platelets, observed by OM, formedon two different material surfaces with and without pmRNApresent. Figure 2A shows an ITO substrate prepared with surfaceattached pmRNA Pd017 as described in the Methods section.41,42

Hexagons, with lateral dimensions on the micrometer scale, wereobserved on the pmRNA modified surface after Pd2(DBA)3 wasadded in a THF solution. Figure 2B shows the results of anexperiment in which the linker molecule SSMCC was attachedto the surface. Figure 2C shows relatively small hexagonsformed on an ITO surface with H2O added. Figure 2D showshexagonal platelets formed on a bare hydrophobic gold substrate,without surface attached pmRNA Pd017. The platelets in Figure2D have the same morphology and same average size as thoseplatelets prepared in the presence of pmRNA, as observed inFigure 2A.

Structure and Composition of pmRNA-Mediated Pd2-(DBA)3 Microstructures. TEM grids with microstructures createdby incubation of Pd2(DBA)3 with pmRNA P017 were preparedaccording to the solution protocols described elsewhere.40-42

Morphology, elemental composition, and crystallographic ori-entation of the hexagons of several samples were studied in a

Figure 2. OM images of formation of hexagonal microplatelets onvarious surfaces. (A) Deposition of a ∼100% THF solution of 200µM Pd2(DBA)3 with only a small layer of water covering the pmRNAPd017 surface-modified ITO slide. (B) Deposition of a ∼100% THFsolution of 200 µM Pd2(DBA)3 on an ITO slide coated with the SSMCClinker. (C) Deposition of a ∼100% THF solution of 200 µM Pd2(DBA)3

on a hydrated bare ITO slide. (D) Deposition on a gold substrate surfacecoated with no RNA.

Pd2(DBA)3: Precursor for Pd Nanoparticles? J. Phys. Chem. C, Vol. 113, No. 29, 2009 12709

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

200 kV TEM with EELS and SAED. Figure 3 shows oneexample of a bright field (BF) micrograph and correspondingdiffraction pattern (inset) acquired to determine the atomicordering of the hexagonal sample. Lack of thickness fringes inthe BF micrograph and diffuse rings in the diffraction patternsfrom the morphologically hexagonal material provide evidencethat the particles were neither single crystal nor polycrystallinePd metal. The type of diffuse SAED shown in Figure 3 isconsistent with a predominantly carbon composition as observedin samples prepared without pmRNA Pd017.35 In general,diffuse patterns of this type arise from beam damage.59 We havefurther shown that extreme beam damage can even lead to theformation of small (<5 nm) Pd nanoparticles inside thehexagonal microcrystals.

In contrast to our findings, a recent report of a SAEDpattern,40 reproduced in Figure 4, is indicative of a polycrys-talline metal lattice with a face-centered cubic unit cell.However, no accompanying TEM image was presented to permitidentification of the structure that gave rise to the SAED pattern.Therefore, it is not possible to determine the morphology ofthe structure that gave rise to the diffraction pattern in Figure4. Moreover, this SAED pattern does not agree with that ofFigure 5, which shows the only other published SAED evidenceby these authors in support of the assignment of the structureof the hexagonal particles.60 The spot pattern in Figure 5B hasa reported lattice constant of 10.9 Å.60 The accepted value forface-centered cubic crystalline palladium metal from X-raydiffraction card files is 3.89 Å. Small regions of a polycrystalline

sample such as demonstrated by the lattice fringes of Figure5A would not produce a spot pattern, since spot patterns areindicative of a single crystal. However, the reported beamdamage60 agrees with our report of beam damage35 of organiccrystals of Pd2(DBA)3. The beam damage is also consistent withthe observation of diffuse scattering (vide supra) since theelectron beam would cause a breakdown of the initially formedcrystalline Pd2(DBA)3.

The hexagons often form in clusters when deposited fromsolution. The pmRNA Pd017-incubated clusters were furtheranalyzed with 200 kV STEM Z-contrast imaging, EELS, andEDS. The secondary electron (SE) and Z-contrast (compositiondependent contrast) STEM micrographs of parts A and D ofFigure 6 show clusters of hexagons with edge lengths on themicrometer scale and are an example of representative featuresfound during STEM analysis of multiple regions on the TEMgrid. At higher magnification, the light contrast of the hexagonalfeature in the Z-contrast micrograph of Figure 6C is consistentwith the carbon-rich composition of the hexagonal molecularcrystals.35,36 Core-loss EELS spectra from the hexagonal particleswere analyzed and the hexagons were found to be composedof an average of 92.5 ((12.9) atom % C and 6.40 ((0.80) atom% Pd. These data confirm that the composition of hexagonalparticles produced using pmRNA Pd017 using the reportedmethods41,42 are mostly carbon as observed previously forhexagonal particles produced with pmRNA Pd017.

The SAED and EELS data collected from the pmRNA Pd017-incubated samples indicate that the hexagonal microstructuresare not crystalline Pd metal. The only evidence of Pd metalwas the relatively sparse ∼200 nm nanoparticle clusters thatare a spontaneous breakdown product of Pd2(DBA)3, as shownin both the Z-contrast and diffraction contrast micrographs ofparts E and F of Figure 6, respectively.35,36

Structure and Composition of Spontaneously FormedPd2(DBA)3 Microstructures. The spontaneous formation ofhexagonal microstructures was also observed in binary solutionsof THF/H2O without pmRNA incubation.35,36 A low-magnifica-tion SEM backscattered electron (BSE) micrograph of a TEMgrid (corner of Cu grid square top left) prepared with featurescreated in a 50 vol % THF/H2O binary mixture is shown inFigure 7A. Three micrometer-sized clusters of hexagons areseen; two are suspended by the lacey carbon film, and onecluster is located on the Cu grid bar. Figure 7B is a highermagnification BSE micrograph of a hexagonal cluster locatedon a different area of the TEM grid. The contrast of the hexagonsdenotes a low Z material, when compared to the contrast of the

Figure 3. BF TEM micrograph showing the morphology of hexagonalmicrostructures resulting from synthesis using pmRNA Pd017 incuba-tion with Pd2(DBA)3.

Figure 4. Reproduction of SAED data presented in Eletter attachedto Ref. 40. The original figure caption is “Electron diffraction fromhexagonal particles synthesized in the presence of the cycle 8 of theRNA pool of sequences.”

Figure 5. High-resolution TEM image obtained from a singlehexagonal palladium containing particle. Localized areas of observedlattice fringes are marked in boxes. Accelerating voltage, 200 kV;magnification 1,500,000×. (B) Electron diffraction obtained from asingle hexagonal palladium containing particle. Accelerating voltage,80 kV; camera length, 60 cm. This figure is a reproduction of Figure 2.3(p. 52) of ref 60.

12710 J. Phys. Chem. C, Vol. 113, No. 29, 2009 Leonard and Franzen

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

Cu grid, and indicates these hexagons were not composed ofhigh atomic number material like Pd. Quantitative EDS analysisof 5 representative hexagons was performed and the resultingaverage atomic percentage of carbon and palladium was >90 at% and <10 at %, respectively. Indexed diffraction patterns aswell as microanalysis results on these spontaneously formedhexagonal microstructures have been presented elsewhere.35,36

These data confirm that the chemical composition of thehexagons is the same within experimental error, whether theyare incubated with pmRNA Pd017 or in the absence of pmRNA.

A Critical Examination of the Assignment of CubicParticles as Metallic Pd. Cubic-shaped particles have beenobserved to form spontaneously from Pd2(DBA)3

36 and in the

presence of the pmRNA sequence known as Pd034.41,42 Thetwo SAED patterns that have been presented as evidence forassignment of these cubic particles as metallic Pd are presentedas an overlay of SAED patterns shown in Figure 8.60 The overlayshows a significant discrepancy between the spot and ringpattern in parts A and B of Figure 8, obtained from the centerand edge of a cube-shaped particle, respectively.60 The cameralength (L) is usually calibrated with a metal standard. Diffractionpattern indexing typically includes calculating the relativisticcorrected electron wavelength (λ) (λ ) 0.0418 Å @ 80 kV andλ ) 0.0251 Å @ 200 kV), measuring distances (R) from the(000) spot to the (hkl) reflections, and finding angles betweenthe lattice planes. Since Rd ) Lλ, the accuracy of L, λ, and Ris critical for determining the lattice spacing (d) of atomic planesin the material. For a cubic metal, d spacings can be used tocalculate the lattice parameter (a) using the relationship a )d�(h2 + k2 + l2), where h, k, and l are the Miller indices.However, there was no reported calibration of L, calculation ofλ, or indexed values of R, which complicates interpretation ofthe d spacings of the cube-shaped particles.60 Therefore, tounderstand the significance of the data, we proceeded byassuming that L ) 40 cm as reported in the figure caption.60

While this uncalibrated value would not be appropriate for anaccurate determination of the lattice constant, it is sufficient forthe purpose of comparison of the d spacings in the spot andring pattern. Since both SAED patterns were acquired with thesame L and λ, the lattice plane reflections should match exactlyif they originated from the same Pd lattice. However, the indexed(111) plane reflections of the spot pattern (red boxes) exhibit a∼20% difference in d spacing when compared to the (111)reflection of the ring pattern. Moreover, the (200) ring and (220)spot reflections were incorrectly indexed to have the same dspacing. Such mismatches between the reported ring and spotdiffraction data indicate these patterns cannot have originatedfrom the same material. The lack of representative HRTEMmicrographs and EDS data further hampers the assignment ofthe cube-shaped particles.

Figure 6. (A) STEM SE micrograph showing a representative exampleof clustered particles of hexagonal morphology, which resulted fromsynthesis using pmRNA Pd017 incubation with Pd2(DBA)3. (B) Ahigher magnification STEM SE micrograph of the hexagonal featureat the tip of the arrow in A. (C) The contrast of this STEM Z-contrastmicrograph, of the exact region of part B, shows low contrast whichresults from the carbonaceous (low Z) composition of the hexagon.(D) A lower magnification STEM Z-contrast micrographs of anotherhexagonal cluster shows a bright contrast region, confirmed to be Pdrich by EDS analysis. (E) High-magnification BF STEM micrographshowing the morphology of the Pd-rich region indicated by the arrowin part D. The contrast of this micrograph shows smaller grains withinthe nanoparticle diffract the electrons differently. (F) This STEMZ-contrast micrograph of the exact region in part E also shows thenanoparticle is composed of yet smaller nanoparticles which haveagglomerated. The low contrast feature in the bottom right of the imageis the edge of a carbonaceous hexagon to which the nanoparticle waseither attached or embedded.

Figure 7. (A and B) BSE SEM micrographs of hexagonal micro-structures formed spontaneously without RNA in a 50 vol% THF/H20binary solution containing Pd2(DBA)3. These hexagonal clusters arerepresentative of the features sampled for quantitative EDS analysis.The hexagonal shape and carbon composition of these particles matchesthe morphology and composition of samples created with RNA.

Figure 8. Electron diffraction obtained from a single cubic palladiumcontaining particle (A) near the edge and (B) in the center of the particle.Accelerating voltage, 200 kV; camera length, 40 cm. This figure is anoverlay of the Figure 2.5 (p. 54) of ref 60.

Pd2(DBA)3: Precursor for Pd Nanoparticles? J. Phys. Chem. C, Vol. 113, No. 29, 2009 12711

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

Structure and Composition of the Decomposition Productof Pd2(DBA)3. We have studied the polycrystalline Pd nano-particles observed in various solution preparations of Pd2(DBA)3.We have noted that these 50-200 nm Pd nanoparticles can beobserved in the reagent powder as purchased from STREM.35

In all of the samples that were prepared without pmRNA, usingPd2(DBA)3 as a reagent crystalline Pd nanoparticles were foundto coexist with the hexagonal molecular crystals.36 X-ray dotmaps were used to observe the occurrence of Pd metal on aTEM grid prepared with 8 mM Pd2(DBA)3/THF solution. Figure9A shows a BSE image of an area on the grid which containeda hexagonal particle (white arrow). The X-ray dot map of PdLR X-rays (Figure 9B) proves the presence of Pd metal in the100% THF/Pd2(DBA)3 solution. However, no X-ray diffractionsignal was detected from the hexagon feature (white arrow).On the other hand, the C KR X-ray map does show thehexagonal feature (white arrow) was composed mostly ofcarbonaceous material.

The morphology of the Pd nanoparticles shown in the BFTEM micrograph of Figure 10A is markedly different from thefaceted hexagonal structures shown in Figures 6, 7, and 9. Theatomic ordering of the Pd metal nanostructures analyzed withSAED and indexed diffraction patterns, like the one shown inFigure 10B, proved the metal-metal bond length was 0.275nm, which matches the published value for Pd metal.61 TheHRTEM BF micrographs of parts C and D of Figure 10 showlattice fringes from a 50-200 nm Pd metal discoid that iscomposed of smaller nanoparticles clustered together. Thepresence of these lattice fringes provides more evidence ofthe polycrystalline nature of the Pd nanoparticle impurities. Thecommon occurrence of Pd metal nanoparticles in samplesincubated with and without pmRNA indicated that the metallicnanoparticles were present a priori and impurities from synthesisor degradation during storage of the reagent Pd2(DBA)3 wasthe source.

Conclusions

There are two competing aggregation processes that occurin solutions of Pd2(DBA)3. Hexagonal microcrystals ofPd2(DBA)3 form spontaneously in THF/H2O and on surfacesprovided the THF volume fraction is greater than 50%.35,36 Asthe H2O percentage increases in THF/H2O binary mixture,Pd2(DBA)3 is destabilized and it forms aggregates that some-times have a tendency to cluster. In this work we observedclusters of hexagons (Figures 6 and 7) with similar carboncomposition and dimensions to those obtained in solutionsamples incubated with pmRNA Pd017. In all cases studied,i.e., even in the presence of pmRNA, the composition ofhexagons of aggregated Pd2(DBA)3 was determined to be >90

atom % carbon by both EELS and EDS. Under appropriategrowth conditions, Pd2(DBA)3 will form molecular crystals witha pseudohexagonal habit in this solvent system, but othercrystalline habits are also observed to form spontaneouslydepending on the solvent conditions.36 A second much slowerprocess leads to formation of Pd nanoparticles from thedecomposition of Pd2(DBA)3, in an electron. This observationof such nanoparticles should not be confused with the 50-200nm Pd nanoparticles present in the Pd2(DBA)3 reagent powder.These impurities are byproducts of Pd2(DBA)3 synthesis byreduction of PdCl2 in the presence of DBA. Although nano-particle formation inherently competes with formation ofPd2(DBA)3 under reducing conditions, our observations suggestthat the 50-200-nm Pd nanoparticles do not form rapidly fromPd2(DBA)3 itself relative to the time scale of the experimentalprotocols (minutes to tens of minutes). This observation agreeswith the thermal stability of Pd2(DBA)3 and with its knowncoordination chemistry.

A major conclusion of the present work is that the spontane-ous aggregation of Pd2(DBA)3 is the major product observedin the mixtures of THF/H2O studied,35,36,40-42 even in thepresence of surfactants such as pmRNA. We have attempted tocompare the two methods using conditions that are as close aspossible to those reported in the pmRNA studies. However, ithas not been possible to reproduce the reported solutionconditions of 5% THF/H2O using pmRNA-free solvent. Solubil-ity of Pd2(DBA)3 is an issue as acknowledged in a recenterratum, which states that ∼50% THF, rather than 5% THF,was used in experiments designed to deposit hexagonalPd2(DBA)3 microcrystals on pmRNA.62 Here we have alsocompared Pd2(DBA)3 microcrystals made on surfaces usingthose protocols. In the surface studies, there was no significantdifference in morphology of samples formed spontaneously andthose with added pmRNA.

Figure 9. (A) BSE image from a TEM grid prepared with a dropletof 8 mM solution of THF/Pd2(DBA)3 was analyzed using EDS X-raydot maps. (B) The Pd LR X-ray map positively identifies the presenceof Pd metal in the solution. (C) The C KR X-ray dot map shows thehexagonal feature observed in part A., not in part B, is composed ofmostly carbon.

Figure 10. (A) HRTEM BF image showing several Pd metalnanoparticles with a diameter of 50-200 nm. (B) The SAD pattern,acquired from the same area as imaged in part A was indexed andused to calculate lattice spacings of atomic Pd metal. (C) HRTEM BFmicrograph of a crystalline Pd metal nanoparticle. (D) Lattice fringesshowing smaller ∼5 nm diameter Pd nanoparticles are contained inthe larger Pd metal agglomerate.

12712 J. Phys. Chem. C, Vol. 113, No. 29, 2009 Leonard and Franzen

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

We have addressed the metal-metal bonding hypothesis bydirect comparison and found that the hexagonal microcrystalsare not metallic Pd. Our findings are corroborated by the erratumto Ref. 53, which states that the authors could “not be certain”that the particles were metallic Pd.62 The data presented hereresolve the uncertainty. As determined by EDS, EELS, andSAED, the hexagonal nanoparticles are not metallic Pd in anyof the solution or surface preparation methods studied. On thebasis of the data above and on the erratum, the ring pattern inFigure 4 is most readily explained as a SAED pattern of one ofthe Pd impurities found in the reagent powder as evidenced bycomparison with the Pd impurity particle in Figure 10A thatgives rise to a ring pattern in Figure 10B, which matches thatof Figure 4.

The reproducible pseudohexagonal form of Pd2(DBA)3

resulting from evaporation of THF/H2O solutions providesa method for deposition of a Pd-containing material on asurface with a controlled geometry. Although no essentialrole can be identified for pmRNA Pd017 in the formation ofthe hexagonal Pd2(DBA)3 particles, it could act as a detergentthat would stabilize the formation of a growing aggregate ina suspension. The role of wetting and surface structureappears to play a significant role in determining the size anddistribution of Pd2(DBA)3 hexagonal microcrystals. If thegrowth is slowed, Pd2(DBA)3 aggregates will remain longerin solution, consistent with observations that Pd2(DBA)3

suspensions in THF/H2O mixtures appear more soluble whenpmRNA is present. Since more uniform microcrystal forma-tion is observed without pmRNA, we conclude that thehexagonal particles can be prepared with greater reproduc-ibility without the complication of the pmRNA surfactant.

Acknowledgment. The authors would like to thank Drs.Gugliotti, Feldheim, and Eaton, formerly of the W.M. KeckCenter for Evolutionary Materials Synthesis, for guidance inthe synthesis of pmRNA Pd017 and preparation of samplesusing Pd2(DBA)3. We thank Dr. Marta Cerruti formerly of theW.M. Keck Center for preparation of pmRNA Pd017 in solutionand surface studies. We also thank Hitachi High TechnologyAmericas Inc.’s electron microscopy division for use of aHD2300 equipped with a Gatan, Inc. Enfina EELS spectrometerand Dr. Russell of Appalachian State University for donatingaccess to a state-of-the-art low kV STEM/EDS for nanomaterialanalysis.

References and Notes

(1) Takahashi, Y.; Ito, T.; Sakai, S.; Ishii, Y. J. Chem. Soc., Chem.Commun. 1970, 1065–1066.

(2) Ukai, T.; Kawazura, H.; Ishii, Y.; Bonnet, J. J.; Ibers, J. A. J.Organomet. Chem. 1974, 65, 253–266.

(3) Pierpont, C. G.; Mazza, M. C. Inorg. Chem. 1974, 13, 1891–1895.(4) Mazza, M. C.; Pierpont, C. G. J. Chem. Soc., Chem. Commun. 1973,

207–208.(5) Trost, B. M. Tetrahedron 1977, 33, 2615–2649.(6) Ito, T.; Ishii, Y.; Takahashi, Y. J. Chem. Soc., Chem. Commun.

1972, 629–630.(7) Trost, B. M.; Brieden, W.; Baringhaus, K. H. Angew. Chem., Int.

Ed. Eng. 1992, 31, 1335–1336.(8) Steinhoff, B. A.; Stahl, S. S. J. Am. Chem. Soc. 2006, 128, 4348–

4355.(9) Ito, T.; Hasegawa, S.; Takahash, Y.; Ishii, Y. J. Organomet. Chem.

1974, 73, 401–409.(10) Ishii, Y. Ann. N.Y. Acad. Sci. 1974, 239, 114–128.(11) Kawazura, H.; Tanaka, H.; Yamada, K. I.; Takahashi, T.; Ishii, Y.

Bull. Chem. Soc. Jpn. 1978, 51, 3466–3470.(12) Elsevier, C. J. Coord. Chem. ReV. 1999, 1999, 809–822.(13) Milani, B.; Anzilutti, A.; Vicentini, L.; Santi, A. S. O.; Zangrando,

E.; Geremia, S.; Mestroni, G. Organometallics 1997, 16, 5064–5075.

(14) Stahl, S. S.; Thorman, J. L.; de Silva, N.; Guzei, I. A.; Clark, R. W.J. Am. Chem. Soc. 2003, 125, 12–13.

(15) Stahl, S. S.; Thorman, J. L.; Nelson, R. C.; Kozee, M. A. J. Am.Chem. Soc. 2001, 123, 7188–7189.

(16) Ramirez, E.; Jansat, S.; Philippot, K.; Lecante, P.; Gomez, M.;Masdeu-Bulto, A. M.; Chaudret, B. J. Organomet. Chem. 2004, 689, 4601–4610.

(17) Kluwer, A. M.; Elsevier, C. J.; Buhl, M.; Lutz, M.; Spek, A. L.Angew. Chem., Int. Ed. 2003, 42, 3501–3504.

(18) Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. J. Org. Chem. 1999,64, 6750–6755.

(19) Steinhoff, B. A.; Guzei, I. A.; Stahl, S. S. J. Am. Chem. Soc. 2004,126, 11268–11278.

(20) Beller, M.; Fischer, H.; Kuhlein, K.; Reisinger, C. P.; Herrmann,W. A. J. Organomet. Chem. 1996, 520, 257–259.

(21) Reetz, M. T.; Lohmer, G. Chem. Commun. 1996, 1921–1922.(22) Bei, X. H.; Turner, H. W.; Weinberg, W. H.; Guram, A. S.; Petersen,

J. L. J. Org. Chem. 1999, 64, 6797–6803.(23) Li, Y.; Boone, E.; El-Sayed, M. A. Langmuir 2002, 18, 4921–

4925.(24) Redjala, T.; Apostolecu, G.; Beaunier, P.; Mostafavi, M.; Etche-

berry, A.; Uzio, D.; Thomazeau, C.; Remita, H. New J. Chem. 2008, 32,1403–1408.

(25) Narayanan, R.; Tabor, C.; El-Sayed, M. A. Top. Catal. 2008, 48,60–74.

(26) Xia, Y.; Xiong, Y. J.; Lim, B.; Skrabalak, S. E. Angew. Chem.,Int. Ed. 2009, 48, 60–103.

(27) Oleshko, V. P. J. Mol. Catal. A 2006, 249, 4–12.(28) Vargaftik, M. N.; Zagorodnikov, V. P.; Stolyarov, I. P.; Moiseev,

I. I.; Likholobov, V. A.; Kochubey, D. I.; Chuvilin, A. L.; Zaikovsky, V. I.;Zamaraev, K. I.; Timofeeva, G. I. J. Chem. Soc., Chem. Commun. 1985,937–939.

(29) Schmid, G.; Harms, M.; Malm, J. O.; Bovin, J. O.; Vanruitenbeck,J.; Zandbergen, H. W.; Fu, W. T. J. Am. Chem. Soc. 1993, 115, 2046–2048.

(30) Chini, P. Gazz. Chim. Ital. 1979, 109, 225–240.(31) Stolyarov, I. P.; Gaugash, Y. V.; Kryukova, G. N.; Kochubei, D. I.;

Vargaftik, M. N.; Moiseev, I. I. Russ. Chem. Bull. 2004, 53, 1194–1199.(32) Wu, L.; Li, Z. W.; Zhang, F.; He, Y. M.; Fan, Q. H. AdV. Synth.

Catal. 2008, 350, 846–862.(33) Ren, J. T.; Tilley, R. D. J. Am. Chem. Soc. 2007, 129, 3287–3291.(34) Lim, B.; Xiong, Y. J.; Xia, Y. N. Angew. Chem., Int. Ed. 2007,

46, 9279–9282.(35) Franzen, S.; Cerruti, M.; Leonard, D. N.; Duscher, G. J. Am. Chem.

Soc. 2007, 129, 15340–15346.(36) Leonard, D. N.; Cerruti, M.; Duscher, G.; Franzen, S. Langmuir

2008, 24, 7803–7809.(37) Dhas, N. A.; Cohen, H.; Gedanken, A. J. Phys. Chem. B 1997,

101, 6834–6838.(38) Liu, C. J.; Wu, T. W.; Hsu, L. S.; Su, C. J.; Wang, C. C. Carbon

2004, 42, 2635–2640.(39) Anson, A.; Lafuente, E.; Urriolabeitia, E.; Navarro, R.; Benito,

A. M.; Maser, W. K.; Martinez, M. T. J. Alloys Compd. 2007, 436, 294–297.

(40) Gugliotti, L. A.; Feldheim, D. L.; Eaton, B. E. Science 2004, 304,850–852.

(41) Gugliotti, L. A.; Feldheim, D. L.; Eaton, B. E. J. Am. Chem. Soc.2005, 127, 17814–17818.

(42) Liu, D. G.; Gugliotti, L. A.; Wu, T.; Dolska, M.; Tkachenko, A. G.;Shipton, M. K.; Eaton, B. E.; Feldheim, D. L. Langmuir 2006, 22, 5862–5866.

(43) Xiong, Y. J.; McLellan, J. M.; Chen, J. Y.; Yin, Y. D.; Li, Z. Y.;Xia, Y. N. J. Am. Chem. Soc. 2005, 127, 17118–17127.

(44) Xiong, Y. J.; Washio, I.; Chen, J. Y.; Cai, H. G.; Li, Z. Y.; Xia,Y. N. Langmuir 2006, 22, 8563–8570.

(45) Xiong, Y. J.; Wiley, B.; Chen, J. Y.; Li, Z. Y.; Yin, Y. D.; Xia,Y. N. Angew. Chem., Int. Ed. 2005, 44, 7913–7917.

(46) Volkov, V. V.; Vantendeloo, G.; Vargaftik, M. N.; Moiseev, I. I.J. Cryst. Growth 1993, 132, 359–363.

(47) Crookes-Goodson, W. J.; Slocik, J. M.; Naik, R. R. Chem. Soc.ReV. 2008, 37, 2403–2412.

(48) Berti, L.; Burley, G. A. Nat. Nanotechnol. 2008, 3, 81–87.(49) Kumar, A.; Kumar, V. J. Phys. Chem. C 2008, 112, 3633–3640.(50) Dooley, C. J.; Rouge, J.; Ma, N.; Invernale, M.; Kelley, S. O. J.

Mater. Chem. 2007, 17, 1687–1691.(51) Green, M.; Smith-Boyle, D.; Harries, J.; Taylor, R. Chem. Commun.

2005, 4830–4832.(52) Petty, J. T.; Zheng, J.; Hud, N. V.; Dickson, R. M. J. Am. Chem.

Soc. 2004, 126, 5207–5212.(53) Chung, S. W.; Presley, A. D.; Elhadj, S.; Hok, S.; Hah, S. S.;

Chernov, A. A.; Francis, M. B.; Eaton, B. E.; Feldheim, D. L.; De Yoreo,J. J. Scanning 2008, 30, 159–171.

Pd2(DBA)3: Precursor for Pd Nanoparticles? J. Phys. Chem. C, Vol. 113, No. 29, 2009 12713

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r

(54) Milani, B.; Vicentini, L.; Sommazzi, A.; Garbassi, F.; Chiarparin,E.; Zangrando, E.; Mestroni, G. J. Chem. Soc., Dalton Trans. 1996, 3139–3144.

(55) Rodman, D. L.; Carrington, N. A.; Xue, Z. L. Talanta 2006, 70,426–431.

(56) Gagnon, K. T.; Ju, S.-Y.; Goshe, M. B.; Maxwell, E. S.; Franzen,S. Nucleic Acids Res. 2009, 37, 3074–3082.

(57) Fraundorf, P.; Qin, W. T.; Moeck, P.; Mandell, E. J. Appl. Phys.2005, 98.

(58) Scherzer, O. J. Appl. Phys. 1949, 20, 20–29.(59) Egerton, R. F.; Li, P.; Malac, M. Micron 2004, 2004, 399–409.

(60) Gugliotti, L. RNA-Mediated Synthesis of Particles from Organo-metallic Palladium and Platinum Precursors, PhD Thesis; North CarolinaState University: Raleigh, NC, 2006.

(61) Sutton, L. E. Table of Interatomic Distances and Configuration inMolecules and Ions; Chemical Society: London, UK, 1965.

(62) Chung, S. W.; Presely, A. D.; Elhadj, S.; Hok, S.; Hah, S. S.;Chernov, A. A.; Francis, M. B.; Eaton, B. E.; Feldheim, D. L.; De Yoreo,J. J. Scanning 2008, 30, 474–474.

JP903417R

12714 J. Phys. Chem. C, Vol. 113, No. 29, 2009 Leonard and Franzen

Dow

nloa

ded

by N

OR

TH

CA

RO

LIN

A S

TA

TE

UN

IV o

n Ju

ly 3

1, 2

009

Publ

ishe

d on

Jun

e 18

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

jp90

3417

r