Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/nanoenergy

Nano Energy (2013) 2, 677–687

2211-2855/$ - see frohttp://dx.doi.org/1

nCorresponding aufax: +48 223 433 333

nnCorresponding aE-mail addresses

maiyalagan@gmail.c

RAPID COMMUNICATION

One-pot synthesis of chain-like palladiumnanocubes and their enhancedelectrocatalytic activity for fuel-cellapplications

Palanisamy Kannana,n, Thandavarayan Maiyalaganb,nn,Marcin Opalloa

aInstitute of Physical Chemistry, Polish Academy of Sciences, 44/52 ul. Kasprzaka, 01-224 Warszawa, PolandbMaterials Science and Engineering Program, The University of Texas at Austin, Austin, TX, USA

Received 7 May 2013; received in revised form 10 August 2013; accepted 11 August 2013Available online 23 August 2013

KEYWORDSChain-likenanoparticles;Cubic shape;Formic acid;Methanol and ethanol

nt matter & 20130.1016/j.nanoen.2

thor. Tel.: +48 22.uthor. Tel.: +1 51: ktpkannan@gmaiom (T. Maiyalagan

AbstractA simple approach has been developed for the synthesis of anisotropic cubic chain-like Pd nano-structures in an aqueous medium. The cubic chain-like Pd nanostructures show better performancetoward oxidation of formic acid and methanol. The cubic chain-like Pd nanostructures show �11.5times more activity on the basis of an equivalent noble metal mass for the formic acid and methanolthan the spherical shaped Pd nanoparticles and commercial Pd/C catalysts. Further, the superiorelectrocatalytic performance of the present anisotropic cubic chain-like Pd nanostructures towardthe oxidation of alcohols makes them excellent candidates as high performance multipurposecatalysts for direct formic acid fuel cells (DFAFC), direct methanol fuel cells (DMFC), direct ethanolfuel cells (DEFC), respectively.& 2013 Elsevier Ltd. All rights reserved.

Introduction

Controlling the morphology of noble metal nanocrystals hasbeen one of the most recent and attractive research topics

Elsevier Ltd. All rights reserved.013.08.001

3 433 375;

2 772 3084l.com (P. Kannan),).

for the past few decades, especially in the field of catalysis,because the catalytic activity and stability of catalystsare strongly correlated with the size and shape of the nano-crystals in a variety of chemical reactions [1–3]. It has beendemonstrated that the intrinsic properties of nanostruc-tured material can be dramatically enhanced by shape andstructural variation [4–6]. The one dimensional metal nanos-tructures like nanowires, nanobelts, and nanotubes haveattracted significant interest due to their exotic technologicalapplications [7–10]. The surface roughness and inherent high

P. Kannan et al.678

index facets of anisotropic metal nanoparticles stimulateimpressive applications in electro-catalysis, sensors, SERSand many more [11–15]. At present, Pt is still the mostcommonly used electrocatalytic material for fuel cells interms of both activity and stability. However, the highcost of Pt is one of the most important barriers thatlimits the large-scale commercialization of fuel cells [16].Therefore, economical and effective alternative catalysts arerequired.

Among the metal based catalysts studied to date, Palla-dium (Pd) is considerably less expensive than Pt, and Pdhas been focusing on recent research as an attractive andalternative catalysts nanomaterial for a wide range of applica-tions in catalysis, hydrogen storage and biosensing, reduction ofautomobile pollutants and so forth [17–20]. It has been foundthat upon appropriate modification of their surface atomicstructure, Pd-based nanomaterials can become promising elec-trocatalysts by simultaneously decreasing the material cost andenhancing the performance [20,21]. Recently, Pd nanoparticleshave been accepted as a better alternative material forpolymer electrolyte membrane fuel cells towards catalyticoxidation of formic acid as well as oxygen reduction in aproton-exchange membrane (PEM) fuel cell [22,23]. Further-more, Pd nanoparticles have been chosen as the best alter-native material for solving many industrial process problemsand demand the exploration of a new synthetic approach forproduction of well-defined shapes and structures. Attemptshave been made to develop anisotropic Pd nanostructuresmainly using surfactants and polymers or high temperaturereactions in organic medium [24–31]. Although, these methodswere successful in producing well-defined Pd nanostructures,the strong adsorption of these protecting agents resulted in adecrease of catalytic activity [22]. Nevertheless the aqueousmedium based synthesis of metal nanoparticles is more promis-ing from an environmental standpoint, adding an advantageover the use of toxic organic solvents [32–34]. Moreover theseprotocols are limited to producing nanoparticles with a smoothsurface morphology. However, it has been observed that theinteresting nanoarchitectures with high surface roughness andsurface steps can contribute to the increased accessibility ofreactant species and are more attractive for enhancing cataly-tic application [35]. Therefore, it is highly desirable to explorea facile approach to produce nanocatalysts with high surfaceareas and high degree of structural anisotropy for improvedperformance in technological applications.

Herein, we explore a new approach for the synthesis ofchain-like Pd (branches in a tree) nanostructures (PdCNs)with a cubic morphology in an aqueous medium. Herewe use the neurotransmitter 5-hydroxytryptamine (HT) toaddress the environment-friendly and green perspectives. Anotorious problem in the shape-controlled synthesis ofmetal nanoparticles is the shape heterogeneity of thesynthesized product. This report clearly demonstrates theformation of uniform PdCNs with a unique morphology. Tothe best of our knowledge, this is the first report thatdescribes rapid eco-friendly synthesis of cubic chain-like Pdnanostructures without any template, polymer, surfactantor without any heat treatment procedures. The cyclicvoltammetry and chronoamperometry are employed toinvestigate the morpho-dependent electrocatalytic activityof these chain-like cubic Pd nanostructures toward theoxidations of formic acid, methanol and ethanol.

Experimental section

Materials

PdCl2 and 5-hydroxytryptamine were obtained from Sigma-Aldrich Chemical Company. The phosphate buffer solution(PBS; pH=7.2) was prepared using Na2HPO4 and NaH2PO4.Double distilled water was used to prepare the solutions inthis investigation. All other chemicals used in this investigationwere of analytical grade. All the solutions were prepared withdeionised water (18 mΩ) obtained from Millipore system.

Synthesis of cubic chain-like Pd nanostructures

Glasswares used for synthesis were well cleaned with freshlyprepared aqua regia (3:1 HCl and HNO3), then rinsedthoroughly with water and dried prior to use (Caution! aquaregia is a powerful oxidizing agent and it should be handledwith extreme care). In a typical synthesis, 10 ml aqueoussolution of PdCl2 (0.1 mM) was taken in a beaker and then0.1 ml of 5-hydroxytryptamine (5-HT) (10 mM) was rapidlyinjected into the solution, allowed for 30 min in static state.The resulting nanocolloid was stored at 4 1C for further use.

Characterization

The morphologies of the cubic chain-like, dendrites andfractal Pd nanostructures were characterized by a fieldemission-scanning electron microscopy (FE-SEM JEOL JSM-6301F). The specimens were prepared by dropping 3 μl ofcolloidal solution onto silicon wafer substrate. The crystal-lographic information of the prepared Pd chain-like nano-structures were studied by the powder X-ray diffractiontechnique (XRD, Shimadzu XRD-6000, Ni filtered CuKα(λ=1.54 Å) radiation operating at 30 kV/40 mA).

Electrochemical measurements

Electrochemical measurements were performed using twocompartment three-electrode cell with a glassy carbonworking electrode (geometric area=0.07 cm2), a Pt wireauxiliary electrode and Ag/AgCl (3 M KCl) as referenceelectrode. Cyclic voltammograms were recorded using acomputer controlled CHI643B electrochemical analyzer. Allthe electrochemical experiments are carried out in an argonatmosphere. The 3 μl of as-synthesized chain-like nano-structures are dispersed over glassy carbon electrode withnafion and dried prior to electrochemical experiments.

Results and discussion

The cubic chain-like Pd nanostructures have been charac-terized to establish their nanostructure and mechanism ofshape evolution. Figure 1 shows the FE-SEM images obtainedfor the nanoparticle synthesized using 5-HT. The FE-SEMimage shows that Pd nanoparticles were cubic morphologywith chain like-structures (branches in a tree; Figure 1a–d).The cubic chain-like Pd nanocubes have an average sizebetween 140� 160 nm and 140� 210 nm. Furthermore, weexamine the stability of cubic chain like-structures PdCNs

Figure 1 FE-SEM images of (a) simple branched chain-like cubic Pd nanostructures using 5-HT (0.1 mM), (b) a portion of image “a”,(c) a portion of image “b” and (d) a portion of image “c”.

679One-pot synthesis and enhanced electrocatalytic activity of chain-like palladium nanocubes

prepared at a conc. of 0.1 mM 5-HT, the FE-SEM images weretaken after 1 and 5 h time interval and presented inFigure 2a–c. From the FE-SEM measurements the density,length and number of cubic branches per nanochains werealmost same over the growth period up to 5 h (Figure 2a and b).These time-dependent features can be ascribed either to theaggregation of nanoparticles or the growth of anisotropicnanostructures. The morphology of cubic chain-like Pd nano-structures remained the same even after 5 h, implying that theobserved feature is not due to the aggregation of nanoparticles.The chain-like Pd nanocubes grow up to the size of �80 mmwith more than 25 branches in 5 h (Figure 2d). The observedresults suggest that the cubic chain-like morphology is highlystable by varying the time intervals (both 30 min, 1 and 5 h). Tothe best of our knowledge, this is the first report that describesrapid eco-friendly synthesis of cubic chain-like Pd nanostruc-tures (branches in a tree) at room temperature without anyheat treatment, template, polymer, or surfactant. The over-view of the mechanism for formation of such a unique shapecan be outlined in Scheme 1 as: (i) in the initial stage ofreaction, Pd2+ ions are reduced by 5-HT to form nano-sizedsmall Pd cubic shaped nanoparticles (around 5–10 min), (ii) ahomogeneous nucleation of these cubic nanoparticles formssmall, primary Pd cubic nano-chains by gradual assembly ofthese particles that grows into a one-dimensional intercon-nected nanostructure and a complete growth of cubic chain-like nanostructures was observed after 30 min of reaction(iii) these primary particles undergo nucleation/assemble duecourse over time into a 1-D nanostructure and reconstruction ofthese small nanocubes into cubic dendritic nanowires [36,37].We propose that, 5-HT molecule (stabilising/reducing agent)plays a vital role on selective binding on the surface of

1-D nanostructure and reconstruct the tiny nanocubes intodendritic-like nanochains. The growth of nanostructured parti-cles is highly perceptive to the concentration of 5-HT andprecursor, Pd2+ (Figure 3). Here we demonstrate that cubicchain-like Pd nanostructures, cubic small dendrites of Pdnanostructures, cubic big-dendrites of Pd nanostructures andfractal Pd nanostructures could be selectively prepared bymanipulating and controlling the concentration of 5-HT andPd2+ (Figures 4, 5 and Figure S1). It should be noted here thatthis procedure establishes a universal protocol for synthesis ofmonodisperse Pd nanostructures with a variety of shapeswithout adopting different strategies, reaction conditions, orinjection of foreign reagents. Minor change in the concentra-tions of stabilizer and a fixed concentration of precursor orvice versa results in the formation of different morphologies ofthe nanoparticles assembly (Figure 3).

Recently, Willner and co-workers have reported the catae-cholamine neurotransmitters mediated growth of Au nanopar-ticles and their quantification by optical method [38]. Thecatacholamine neurotransmitters and their metabolites inducethe growth of spherical shape nanoparticles, and they do notinduce the formation of any anisotropic nanoparticles [38]. Inthe present investigation, we observed cubic Pd nanochains,cubic nanostructures of big-dendrites, and fractal nanostruc-tures by the indoleamine neurotransmitter. The formation ofchain-like nanostructured particles is attributed to the oxida-tion of 5-HT by PdCl2. As shown in Scheme 2, reduction ofPdCl2 occurs due to the transfer of electrons from HT to themetal ion, resulting in the formation of Pd0. The metallic Pdthen undergoes nucleation and growth with time to form cubicchain-like Pd nanostructures, cubic nanostructures of den-drites and fractal nanostructures. It is well documented in the

Figure 2 FE-SEM images of simple branched chain-like cubic Pd nanostructures using 5-HT (0.1 mM) after 1 h (a), 5 h (b) and closeview of image “b”. The image “c” represents the length of the branches in a sample. The image “d” represents maximum length ofchain-like nanocubic particles.

Nucleation

Self-assembly

Reconstruction

5-HT

Pd cubes

Primary cubic Pd chain-like nanostructure

Scheme 1 Schematic presentation showing the formation/growth process of cubic chain-like Pd nanostructures anddendritic nanochains.

P. Kannan et al.680

literature that the oxidation of 5-hydroxyindole generates afree radical [39–43]. The radical generated by the oxidation ofHT can undergo a chemical reaction in aqueous solution toyield different products. The reaction of radical strongly

depends on the experimental condition. The hydroxylateddimers (5,5'-Dihydroxy-4,4'-bitryptamine (DHB)) and trypta-mine-4,5-dione (TAD) (Scheme 2) are the major products inneutral and acidic pH. These oxidation products are expectedto adsorb on the surface of the nanoparticles (vide infra) bydifferent orientations, resulted that different cubic chain-likePd nanostructures. In Scheme 2, “a” and “b” representing thereversible redox reaction, these oxidized products wereexpected to adsorb on the surface of the nanoparticlesthrough amino groups (–NH2), resulted the Pd nanoparticles.It is expected that the hydrogen bonding between above twocompounds with PdNPs (cellulose type hydrogen bonding),resulting that chain-like nanostructures.

We suggest that the balance act between the concentra-tion of precursor and stabilizer determine their shapeevolution process. The effect of Pd2+ concentration canbe speculated as at high concentration, Pd2+ ions arefloating around to seed the growth of the Pd nanocubesand speed up the nucleation to form compact large nano-cubes. On the other hand at lower concentration the Pd2+

ions are diluted enough to decrease the seeding and smallcubic chain-like nanostructures were formed. Those smallnanocubes undergo controlled nucleation to form thedendritic morphology. When the concentration of 5-HTis increased, Pd nanostructures with a big dendritic andfractal shaped morphology were obtained. This may be dueto the high concentration of 5-HT (reducing/stabilisingagents) which not only expedites the reduction of Pd2+ to

Pd (0.1 mM) in 10 mL

+

5-Hydroxytryptamine (HT)

HT (0.1 mM)

HT (0.3 mM)

HT (0.5 mM)

HT (0.75 mM)

Figure 3 FE-SEM measurements of different morphologies of Pd nanoparticles induced at different concentrations of 5-HT andPd(II).

Figure 4 FE-SEM images of 5-HT-induced (0.3 mM) formation of cubic chain-like small dentries of Pd nanostructures.

681One-pot synthesis and enhanced electrocatalytic activity of chain-like palladium nanocubes

Figure 5 FE-SEM images of 5-HT-induced (0.5 mM) formation of cubic chain-like big-dendrites of Pd nanostructures.

P. Kannan et al.682

Pd0, but also controls the nucleation process. Our examina-tion concludes that the optimal metal precursor concentra-tion for the preparation of Pd nanoparticles with uniformdistribution of nanocubes was of 0.5 mM. When the con-centration goes up to 0.75 mM, the chain-like cubic nano-particles were aggregated. The sufficient numbers of 5-HTmolecules present around the reaction direct the nuclea-tion/assembly of the reduced primary Pd cubic nanoparti-cles in different arrays or unique direction to produce cubicchain-like Pd dendritic nanostructures, nanostructures ofsmall, big-dendrites, and multiple chain-like fractal shapes.So it can be speculated that the concentration of 5-HT notonly stabilizes the nano-sized particles but its structureplays a vital role in controlling their nucleation or growthprocess in a universal pattern to produce cubic chain-like Pdnanostructures, cubic chain-like dendritic or fractal shapes.It is well established that using surfactants or polymers withdifferent functional groups having different binding strate-gies such as acids and amines regulate the morphology ofnanopararticles [44]. The formation of different shapes andmorphology is controlled by the faceting tendency of thestabilizer and the growth kinetics to crystallographic planes[6,29,30]. We anticipate here that the structure and func-tional groups of reducing/stabilizing agents play a vital rolein shape and morphology evolution of nanoparticles.Our efforts are yet underway to establish the structure–function relationship of the stabilizer for shape evolution ofPd nanoparticles using the bi-products of 5-HT (DHB andTAD). We propose a reasonable explanation towards thebalancing act between the concentration of precursor andstabilizer for shape evolution of Pd nanostructures inSchemes 1 and 2.

The X-ray diffraction (XRD) patterns were recorded toconfirm the lattice facets of cubic chain-like Pd nanostruc-tures as shown in Figure 6. Peaks corresponding to Pd (111),(200), (220) and (311) were obtained. The relative peakintensities were compared using the peak area of (111) as areference (JCPDS card number: 41-1021). The ratio of therelative peak intensity of (200) with respect to (111) isfound to be 0.52 versus 0.6 of the standard value. However,the ratio of the relative peak intensities of high index planes(220) and (311) are higher than the standard values prev-iously reported (0.73 versus 0.42) and (0.71 versus 0.55),respectively [17,45,46]. This observation reveals that thecubic chain-like Pd nanostructures were abundant in highindex facets. Further, XRD patterns for cubic small dendritesPd nanostructures, cubic chain-like big-dendrites Pd nano-structures and fractal nanostructures of Pd nanostructureswere recorded to confirm their lattice facets (Figure 6 andFigure S2). For instance, the ratio of the relative peakintensities of high index planes (220) and (311) is 0.61 versus0.42 and 0.65 versus 0.55 respectively for cubic smalldendrites Pd nanostructures, which is lower than the valuesobtained for cubic chain-like Pd nanostructures (0.73 versus0.42) and (0.71 versus 0.55). Moreover, the formations of Pdnanoparticles were also investigated in the presence of5-hydroxyindoleacetic acid and N-aceltyserotonin, that arefunctionally correlated with respect to 5-HT. In similarconditions, both these structurally different moleculesinduce the formation of Pd nanostructures but the mor-phology differs (Figure S3). We anticipate here that thestructure and functional groups of reducing/stabilizingagents affect the shape and morphological evolution ofnanoparticles.

‘a’ and ‘b’ represent the reversible redox reaction.

NH2

HN

OH

5-hydroxytryptamine

Pd2+

Pd0

-H+

NH2

HN

O*

N

N

ONH2NH2 N

H

HO

NH2HO

OH

N

HN

HONH2NH2

HO NH

NH2O

O

2e-

2H+2e-

2H+

tryptamine-4,5-dione (TAD)

5,5'-Dihydroxy-4,4'-bitryptamine (DHB)

a b

N

N

ONHNH

NH

HN PdNPsN

N

O

HNNH

HN

NHPdNPs

OH

OH

OH

OH

OH

OH

Scheme 2 Mechanism for 5-HT induced formation of Pd nanostructures with different branches with cubic and fractal shapes.

20 402 Theta (Degree)

Inte

nsity

(a.u

.)

60 80

(111)

(200) (220) (311)

20 402 Theta (Degree)

Inte

nsity

(a.u

.)

60 80

(111)

(200)

(220) (311)

Figure 6 XRD pattern of as synthesized cubic chain-like Pd nanostructures (a) and cubic chain-like small dendrite of Pdnanostructures (b).

683One-pot synthesis and enhanced electrocatalytic activity of chain-like palladium nanocubes

Pd is well-known to absorb massive quantities of hydro-gen in bulk to form hydrides and therefore has a greattechnological potential as a hydrogen storage material [47].We have carried out cyclic voltammetric studies in acidicmedium to examine the affinity of hydrogen toward allcubic chain-like Pd nanostructures. Figure 7 shows thevoltammetric behavior of the cubic chain-like Pd nano-structures modified electrode in 0.5 M HClO4 at a scan rate

of 50 mVs�1. The cyclic voltammogram (CVs) obtained forcubic chain-like Pd nanostructures in 0.5 M HClO4 showssharp distinguished anodic peaks in between �0.25 and–0.15 V. This broad peak is due to the oxidation of both theabsorbed and adsorbed hydrogen on the cubic chain-like Pdnanostructures and has been earlier reported for bulk Pdelectrodes [48]. However, the CV for the cubic chain-like Pdnanostructures coated disk electrode shows two distinct

–0.5 0 0.5 1

a

d

Potential (V) vs. Ag/AgCl

Cur

rent

Den

sity

(mA

/cm

2 )

0

0.4

0.4

0.8

0.8

0.5

dCur

rent

Den

sity

(mA

/cm

2 )

Potential (V) vs. Ag/AgCl

0.05

0

a

Figure 7 (A) CVs of fractal Pd nanostructures (a), cubic Pd big-dendrites (b), cubic chain-like small dendrites of Pd nanostructures(c) and cubic chain-like Pd branched nanostructures in N2-saturated 0.5 M HClO4 solution at a scan rate of 50 mV s�1.

P. Kannan et al.684

voltammetric peaks, one of which is a small shoulder peakin the negative potential region during the anodic sweep[48–51]. The first peak at �0.23 V corresponds to the oxida-tion of the adsorbed hydrogen (Had), while the subsequentpeak at �0.18 V is due to the oxidation of the absorbedhydrogen (Hab) on Pd surface [49]. The cathodic peak duringthe reverse scan is due to both absorption and adsorption ofhydrogen. It is known from the literature that the two well-resolved peaks for the oxidation of adsorbed and absorbedhydrogen in Pd are manifested exclusively in the case of Pdnanoparticles. This behavior is attributed to the largernumber of surface sites available for adsorption on thenanoparticle's surface [48–51]. The large anodic peak corre-sponding to the desorption of hydrogen in the case of cubicchain-like Pd nanostructures modified electrode shows itspotential as an excellent hydrogen storage material. Theelectroactive surface area of Pd has been measured fromthe palladium oxide stripping analysis [48]. The chargeconsumed during the reduction of Pd oxides has beenestimated by integrating the area under the reductionwave. The electric charge of the hydrogen desorption ofcubic chain-like Pd nanochains, small cubic nanostructuresof dendrites, big-dendrites and fractal nanostructures weremeasured to be 132.4, 366.2, 509.4, and 784.2 μC (a–d)respectively. The electrochemically accessible area of cubicchain-like Pd nanostructures was calculated to be 1.476 cm2

using the reported value of 424 μC cm�2. This is a very largereal surface area for the other chain-like Pd nanostructuressuch as cubic nanostructures of small dendrites, big-den-drites, and fractal nanostructures are 0.875 0.554, and0.253 cm�2, respectively. This large surface area arises due tocubic chain-like Pd nanostructures in which the Pd nanocubeswere highly dispersed making a branches-on a tree type ofnanostructure. Since the Pd nanoparticles were directly mod-ified on the electrode surface, the available effective activecenters were also higher, which makes them an ideal electro-catalytic nanomaterial.

Inspired by the attractive cubic chain-like structures andunique morphology of Pd nanostructures, we further provedthese as nanoelectrocatalyst. Very recently Mohanty et al.observed a dramatic catalytic performance of dendritic noblemetal nanoparticles towards the Suzuki–Miyaura and Heckcoupling reaction [17]. Here we use methanol, ethanol andformic acid as model molecules for studying the

electrocatalytic performance of cubic chain-like Pd nanos-tructures. The electrochemical oxidation of methanol andformic acid has attracted much attention due to theirpotential energy related applications for direct methanolfuel cells (DMFC) and direct formic acid fuel cells (DFAFC),respectively. It is generally accepted that noble Pt metal isthe best catalyst for the formic acid oxidation but it sufferspoisoning due to a strong adsorbed CO species generated asreaction intermediates [52]. A noble Pd-based catalyst wasexplored to be efficient and possesses superior performancesfor fuel cell applications because it is highly resistant to COpoisoning [22]. We have examined the electrocatalyticactivity ofas-synthesized chain-like cubic Pd nanostructures towardsformic acid and methanol oxidation (Figure 8). Well-definedvoltammograms with prominent peaks in the forward andreverse scans were obtained, respectively. The oxidationcurrent has been normalized to the electroactive surfacearea of the cubic chain-like Pd nanostructures. This surfacearea was determined from the coulombic charge reduction ofPd oxide (vide supra), according to the reported value of424 mC cm�2. As shown, the as-synthesized cubic chain-likePd nanostructures of different branched morphologies exhibitdifferent electrocatalytic activities.

The oxidation of formic acid on Pt involves a dual pathmechanism, a dehydrogenation path to the direct formationof CO2 and a dissociative adsorption to form poisoning COspecies (a dehydration path) [53–55]. It can be seen that theformic acid electro-oxidation on these present cubic chain-like Pd nanostructures shows two anodic peaks in theforward scan. The onset potential and peak potential oncubic chain-like Pd nanostructures are �0.15 and 0.13 V,respectively (Figure 8A, curve b). The first small anodicpeak is attributed to the direct oxidation of formic acid toCO2 on the remaining sites unblocked by intermediatespecies. At higher potentials, the current further increasesand reaches another peak. The second peak is related to theoxidation of adsorbed intermediate species (CO), whichreleases the free surface sites for the subsequent directoxidation of formic acid. In the reverse scan, the currentreaches a high peak located at around 0.02 V, which ismuch higher than those in the positive scan, due to the bulkHCOOH oxidation and the absence of CO poison. However,with the potential scanning to more negative values, the

-0.3 -0.2 -0.1 0 0.1 0.2 0.3

Potential (V) vs. Ag/AgCl

0.0

2.0

10.0

4.0

6.0

8.0

abcdef

Cur

rent

Den

sity

(mA

/cm

2 )

Cur

rent

Den

sity

(mA

/cm

2 )

-0.3 -0.2 -0.1 0 0.1 0.2 0.3

Potential (V) vs. Ag/AgCl

0.0

1.0

5.0

2.03.0

4.0

bc

ef

6.0a

d

-0.1

Figure 8 Cyclic voltammograms show the oxidation of (A) formic acid (0.25 M) in 0.5 M HClO4 and (B) methanol (0.25 M) in0.1 M KOH at (a) bare GC, (b) cubic chain-like Pd nanochains, (c) cubic chain-like Pd nanochains in the absence of both formic acidand methanol, (d) cubic small dendrites of Pd nanostructures, (e) cubic big-dendrites Pd nanostructures and (f) fractal Pdnanostructures modified electrodes. Scan rate: 50 mV s�1.

8.0

6.0

4.0

2.0

01.0

3.0

5.0

7.0

0 2000 4000 6000

Cur

rent

Den

sity

(mA

/cm

2 )

Time (Sec)

a

bc

d

Figure 9 Polarization current vs. time plots of branched cubicchain-like Pd nanostructures (a), cubic small dendrites of Pdnanostructures (b), cubic big-dendrites of Pd nanostructures(c) and fractal Pd nanostructures (d) modified electrodes measuredin formic acid (0.25 M) in 0.5 M HClO4 solution.

685One-pot synthesis and enhanced electrocatalytic activity of chain-like palladium nanocubes

surface was again poisoned by CO, resulting in the decrease ofthe oxidation current [56–59]. The oxidation of formic acid atcubic chain-like Pd nanostructures is higher than those oncubic chain-like small dendrites of Pd nanostructures (�0.10and 0.16 V), cubic big-dendrites Pd nanostructures (�0.10 and0.16 V) and fractal Pd nanostructures (�0.07 and 0.18 V),indicating that the formic acid is easier to oxidize on cubicchain-like Pd nanostructures. The peak current density forcubic chain-like Pd nanostructures is �760 mA/cm2, which ismuch higher than that for cubic small dendrites of Pdnanostructures (�480 mA), cubic big-dendrites Pd nanostruc-tures (�460 mA) and fractal Pd nanostructures (�250 mA).This should be resulted from its highest ECSA. The specificsurface activities of these catalysts could be obtained fromthe normalized peak current density with ECSA to furtherinvestigate the possibility of the enhancement effect of cubicbranched nanostructures on the intrinsic electrocatalyticactivity, which is 10.8, 6.8, 6.5 and 3.5 Am�2 for cubicchain-like Pd nanostructures, cubic small dendrites of Pdnanostructures, cubic big-dendrites Pd nanostructures andfractal Pd nanostructures, respectively. The low onset andoxidative peak potential value (less positive potential), muchhigher specific surface activities of the cubic chain-like Pdnanostructures catalysts towards formic acid oxidation may beattributed to the presence of branched (branches in a tree)morphology. On the other hand, the present Pd nanostructuresalso efficiently catalyze the oxidations of methanol (Figure 8B)and ethanol (Figure S4). We summarized the activity of Pdnanostructures in terms of oxidation potential and generatedcatalytic current-density and these data are provided in(Table S1). Further, we compared the electrocatalytic per-formance of the present branched cubic chain-like nanopar-ticles to commercial Pd/C and spherical PdNPs toward theoxidation of formic acid and methanol (Figures S5 and S6).The current densities were also normalized to the electro-active surface area, which were measured from the electriccharge of hydrogen adsorption/desorption on Pd surfaces.Such normalization allowed the current density to applydirectly for comparison of the catalytic activities of thedifferent catalysts. The cubic chain-like Pd nanostructuresshow Z11.5 (comparison for both cases) times more activityon the basis of an equivalent noble metal mass for the formicacid and methanol than the spherical shaped PdNPs and

commercial Pd/C catalysts. This may be attributed to thehigh degree of structural anisotropy of cubic chain-like Pdnanostructures and branched tree-like morphology contribut-ing to their superior electrocatalytic activities. The richbranched sharp edges and corner atoms of tree-like struc-tures were highly preferred for improving the catalyticactivity [60]. The long-term stability of these catalysts wasevaluated by the chronoamperometry method and also thecurrent decay, which is associated with the poisoning ofintermediate species on the nanoparticles determining thefate of its practical application. The chronoamperometryprofile is sufficiently stable for cubic chain-like Pd nanos-tructures and cubic nanostructures of dendrites morpholo-gies. The chronoamperometric stabilities of cubic chain-likePd nanostructures, cubic nanostructures of small dendrites,cubic big-dendrites and fractal nanostructures were investi-gated using towards formic acid oxidation and as shown inFigure 9. The polarization current for the formic acidoxidation reaction on these three supported Pd catalystsshows a significant decay initially and reaches a stable valueafter polarization at 0.20 V for �400–600 s. The current decayfor the formic acid oxidation reaction indicates the slowdeactivation of the nanostructued Pd-based electrocatalysts

P. Kannan et al.686

by the slow adsorption of CO or CO-like intermediates [61–63].This is supported by the detection of the adsorption of COspecies on Pd catalysts during formic acid oxidation bysurface-enhanced infrared absorption spectroscopy [64]. How-ever, the stable mass specific current for formic acid oxidationreaction on cubic chain-like Pd nanostructures is �540 mA,significantly higher than cubic nanostructures of small den-drites (�170 mA) and cubic big-dendrites (�90 mA). Thisindicates that cubic chain-like Pd nanostructures catalystspossess much better stability against the poisoning byadsorbed CO or CO-like intermediate species. The stabilityof these catalysts was also studied by the cyclic voltammetricmethod. The first peak currents at forward scan are recordedas a function of the cycle numbers, as shown in Figure S7.Using the peak current of the 10th cycle as the baseline, thepeak current after 100 cycles on cubic chain-like Pd nano-structures is reduced by 29.6%, which is significantly lowerthan the reduction in activity on cubic nanostructures of smalldendrites (38.5%) and cubic big-dendrites (57.2%) nanocata-lysts, further indicating that cubic chain-like Pd nanostruc-tures exhibit the highest catalytic stability and is in goodagreement with the results of chronoamperometry curves.Therefore, the as-synthesized cubic chain-like Pd nanostruc-tures were highly resistant to poisoning due to intermediatesand were favorable for practical uses. We could also evidentlyargue that the presence of cubic chain-like Pd nanostructuressignificantly enhances the electrocalytic activity and dimini-shes the poisoning of the Pd catalysts for the highest stability.We anticipate that the cubic chain-like Pd nanostructures(Figure 1) expose a larger number of active sites for theadsorption of active oxygen atoms, which readily oxidizes thereaction intermediate species, protecting it from surfacepoisoning [65,66]. No change in the morphology of branchedcubic chain-like Pd nanostructures was observed after severalvoltammetric cycles (Figure S8) and hence acclaims its robustnature. Our efforts are underway to modify these cubic chain-like Pd nanostructures on the graphene, carbon support andcompare their enhanced electrocatalytic activity with thecommercially available Pd/C.

Conclusions

In summary, we have explored a facile approach to tailorthe shape and tree-like branched morphology of Pd nano-structures in aqueous medium. We propose that, 5-HTmolecule (stabilising/reducing agent) plays a vital role onselective binding on the surface of 1-D nanostructure andreconstruct the tiny nanocubes into dendritic-like cubicnanochains. The as-synthesized chain-like, small and big-dendrites, and fractal Pd nanostructures showed cubicshape and morphology dependent electrocatalytic activitytowards oxidation of formic acid, methanol and ethanol forDFAFC, DMFC and DEFC applications.

Acknowledgments

Palanisamy Kannan and Marcin Opallo thank NanOtechnologyBiomaterials and aLternative Energy Source for ERA Integration[FP7-REGPOT-CT-2011-285949-NOBLESSE] Project from Eur-opean Union.

Appendix A. Supplementary material

Supplementary data associated with this article can be foundin the online version at http://dx.doi.org/10.1016/j.nanoen.2013.08.001.

References

[1] D. Kim, Y.W. Lee, S.B. Lee, S.W. Han, Angewandte ChemieInternational Edition 51 (2012) 159–163.

[2] Y. Xia, Y. Xiong, B. Lim, S.E. Skrabalak, Angewandte ChemieInternational Edition 48 (2009) 60–103.

[3] M. Grzelczak, J. Perez-Juste, P. Mulvaney, L.M. Liz-Marzan,Chemical Society Reviews 37 (2008) 1783–1791.

[4] R. Jin, Y. Charles Cao, E. Hao, G.S. Metraux, G.C. Schatz,C.A. Mirkin, Nature 425 (2003) 487–490.

[5] C. Burda, X. Chen, R. Narayanan, M.A. El-Sayed, ChemicalReviews 105 (2005) 1025–1102.

[6] C.J. Murphy, T.K. Sau, A.M. Gole, C.J. Orendorff, J. Gao,L. Gou, S.E. Hunyadi, T. Li, The Journal of Physical Chemistry B109 (2005) 13857–13870.

[7] X. Teng, W.-Q. Han, W. Ku, M. Hücker, Angewandte ChemieInternational Edition 47 (2008) 2055–2058.

[8] K.K. Caswell, C.M. Bender, C.J. Murphy, Nano Letters 3 (2003)667–669.

[9] S. Guo, S. Dong, E. Wang, Chemical Communications 46 (2010)1869–1871.

[10] N. Tian, Z.-Y. Zhou, S.-G. Sun, Chemical Communications0 (2009) 1502–1504.

[11] S. Cheong, J.D. Watt, R.D. Tilley, Nanoscale 2 (2010) 2045–2053.[12] Y. Wook Lee, M. Kim, S. Woo Han, Chemical Communications

46 (2010) 1535–1537.[13] B. Hu, K. Ding, T. Wu, X. Zhou, H. Fan, T. Jiang, Q. Wang,

B. Han, Chemical Communications 46 (2010) 8552–8554.[14] B. Lim, M. Jiang, P.H.C. Camargo, E.C. Cho, J. Tao, X. Lu,

Y. Zhu, Y. Xia, Science 324 (2009) 1302–1305.[15] M.J. Mulvihill, X.Y. Ling, J. Henzie, P. Yang, Journal of the

American Chemical Society 132 (2009) 268–274.[16] R. Bashyam, P. Zelenay, Nature 443 (2006) 63–66.[17] A. Mohanty, N. Garg, R. Jin, Angewandte Chemie International

Edition 49 (2010) 4962–4966.[18] M. Cao, J. Lin, H. Yang, R. Cao, Chemical Communications 46

(2010) 5088–5090.[19] C. Langhammer, I. Zorić, B. Kasemo, B.M. Clemens, Nano

Letters 7 (2007) 3122–3127.[20] N. Tian, Z.-Y. Zhou, N.-F. Yu, L.-Y. Wang, S.-G. Sun, Journal of

the American Chemical Society 132 (2010) 7580–7581.[21] L. Xiao, L. Zhuang, Y. Liu, J. Lu, H.C.D. Abruna, Journal of the

American Chemical Society 131 (2008) 602–608.[22] V. Mazumder, S. Sun, Journal of the American Chemical

Society 131 (2009) 4588–4589.[23] C. Bianchini, P.K. Shen, Chemical Reviews 109 (2009) 4183–4206.[24] X. Huang, N. Zheng, Journal of the American Chemical Society

131 (2009) 4602–4603.[25] Y.-H. Chen, H.-H. Hung, M.H. Huang, Journal of the American

Chemical Society 131 (2009) 9114–9121.[26] J. Watt, S. Cheong, M.F. Toney, B. Ingham, J. Cookson,

P.T. Bishop, R.D. Tilley, ACS Nano 4 (2009) 396–402.[27] V.P. Ananikov, N.V. Orlov, I.P. Beletskaya, V.N. Khrustalev,

M.Y. Antipin, T.V. Timofeeva, Journal of the American Chemi-cal Society 129 (2007) 7252–7253.

[28] Y. Sun, L. Zhang, H. Zhou, Y. Zhu, E. Sutter, Y. Ji,M.H. Rafailovich, J.C. Sokolov, Chemistry of Materials 19(2007) 2065–2070.

[29] Y. Xiong, Y. Xia, Advanced Materials 19 (2007) 3385–3391.

687One-pot synthesis and enhanced electrocatalytic activity of chain-like palladium nanocubes

[30] J. Watt, N. Young, S. Haigh, A. Kirkland, R.D. Tilley, AdvancedMaterials 21 (2009) 2288–2293.

[31] S.-W. Kim, J. Park, Y. Jang, Y. Chung, S. Hwang, T. Hyeon,Y.W. Kim, Nano Letters 3 (2003) 1289–1291.

[32] B.K. Jena, S.C. Sahu, B. Satpati, R.K. Sahu, D. Behera,S. Mohanty, Chemical Communications 47 (2011) 3796–3798.

[33] Q. Yuan, X. Wang, Nanoscale 2 (2010) 2328–2335.[34] B. Lim, M. Jiang, J. Tao, P.H.C. Camargo, Y. Zhu, Y. Xia,

Advanced Functional Materials 19 (2009) 189–200.[35] L. Wang, Y. Yamauchi, Journal of the American Chemical

Society 131 (2009) 9152–9153.[36] F. Viñes, F. Illas, K.M. Neyman, Angewandte Chemie Interna-

tional Edition 46 (2007) 7094–7097.[37] H. Kura, T. Ogawa, Journal of Applied Physics 107 (2010)

074310–074315.[38] R. Baron, M. Zayats, I. Willner, Analytical Chemistry 77 (2005)

1566–1571.[39] E. Perez-Reyes, R.P. Mason, Journal of Biological Chemistry 256

(1981) 2427–2432.[40] M.Z. Wrona, G. Dryhurst, Bioorganic Chemistry 18 (1990)

291–317.[41] C.R. Raj, S. Chakraborty, Biosensors and Bioelectronics 22

(2006) 700–706.[42] R.N. Goyal, A. Sangal, Journal of Electroanalytical Chemistry

578 (2005) 185–192.[43] M.Z. Wrona, G. Dryhurst, The Journal of Organic Chemistry 52

(1987) 2817–2825.[44] E. Ramirez, S. Jansat, K. Philippot, P. Lecante, M. Gomez,

A.M. Masdeu-Bultó, B. Chaudret, Journal of OrganometallicChemistry 689 (2004) 4601–4610.

[45] Y. Sun, Y. Xia, Science 298 (2002) 2176–2179.[46] B.D. Cullity, Elements of X-ray Diffraction, 3rd Edition,

Prentice-Hall, Upper Saddle River, NJ, 2001.[47] J.C. Barton, J.A.S. Green, F.A. Lewis, Nature 197 (1963)

1293–1294.[48] W. Pan, X. Zhang, H. Ma, J. Zhang, The Journal of Physical

Chemistry C 112 (2008) 2456–2461.[49] H.-P. Liang, N.S. Lawrence, T.G.J. Jones, C.E. Banks, C. Ducati,

Journal of the American Chemical Society 129 (2007)6068–6069.

[50] Y. Gimeno, A. Hernández Creus, S. González, R.C. Salvarezza,A.J. Arvia, Chemistry of Materials 13 (2001) 1857–1864.

[51] C. Batchelor-McAuley, C.E. Banks, A.O. Simm, T.G.J. Jones,R.G. Compton, ChemPhysChem 7 (2006) 1081–1085.

[52] S. Zhang, Y. Shao, G. Yin, Y. Lin, Angewandte Chemie Interna-tional Edition 49 (2010) 2211–2214.

[53] A. Capon, R. Parson, Journal of Electroanalytical Chemistryand Interfacial Electrochemistry 44 (1973) 1–7.

[54] A. Capon, R. Parsons, Journal of Electroanalytical Chemistryand Interfacial Electrochemistry 45 (1973) 205–231.

[55] M.E. Vela, R.O. Lezna, N.R. De Tacconi, A.J. Arvia, B. Beden,F. Hahn, C. Lamy, Journal of Electroanalytical Chemistry 323(1992) 289–302.

[56] Y. Li, Y. Jiang, M. Chen, H. Liao, R. Huang, Z. Zhou, N. Tian,S. Chen, S. Sun, Chemical Communications 48 (2012) 9531–9533.

[57] H. Zhao, C. Yu, H. You, S. Yang, Y. Guo, B. Ding, X. Song,Journal of Materials Chemistry 22 (2012) 4780–4789.

[58] H. Yang, L. Dai, D. Xu, J. Fang, S. Zou, Electrochimica Acta 55(2010) 8000–8004.

[59] T. Jin, S. Guo, J.-l. Zuo, S. Sun, Nanoscale 5 (2013) 160–163.[60] L. Wang, H. Wang, Y. Nemoto, Y. Yamauchi, Chemistry of

Materials 22 (2010) 2835–2841.[61] Z. Cui, C. Gong, C.X. Guo, C.M. Li, Journal of Materials

Chemistry A 1 (2013) 1179–1184.[62] X. Yu, P.G. Pickup, Electrochemistry Communications 11 (2009)

2012–2014.[63] X. Yu, P.G. Pickup, Journal of Power Sources, 187, 493–499.[64] H. Miyake, T. Okada, G. Samjeske, M. Osawa, Physical

Chemistry Chemical Physics 10 (2008) 3662–3669.[65] X. Zhang, W. Lu, J. Da, H. Wang, D. Zhao, P.A. Webley, Chemical

Communications 0 (2009) 195–197.[66] R. Manoharan, J. Prabhuram, Journal of Power Sources 96

(2001) 220–225.

Palanisamy Kannan is currently workingas Research Fellow in Institute of PhysicalChemistry, Polish Academy of Sciences,Warsaw, Poland. He received his Ph.D. in2010, Gandhigram Rural University andcompleted the Post-Doctoral Program inSchool of Chemical and Biomedical Engi-neering, Nanyang Technological University,Singapore. His research interests are mainlyon the development of novel nanomaterials

for electrochemical biosensor and fuel cell applications.

Thandavarayan Maiyalagan is currentlyworking as Research Fellow in Texas Materi-als Institute, The University of Texas atAustin, United States. He received his Ph.D.in 2007, IIT Madras and completed the Post-Doctoral Programs at New Castle University,United Kingdom and Nanyang TechnologicalUniversity, Singapore. His current researchinterest includes development of nanomater-ials for energy conversion and storage

devices, electro-catalysis, fuel cells and biosensors.

Marcin Opallo is a Professor of Departmentof Electrode Process, Institute of PhysicalChemistry, Polish Academy of Sciences,Warsaw, Poland. He received his Ph.D. in1987 completed the Post-Doctoral Programsat Tohoku University and University ofCalifornia, Davis, United States. He haspublished more than 100 papers. Currently,his research focuses on developing newmaterials for energy conversion and sensing.