Synthesis and characterization of Pd-Ni nanoalloy electrocatalysts for oxygen reduction reaction in...

KFPEOS

mtpohtchHcc

ny[ctAa

Electrochimica Acta 55 (2010) 1756–1765

Contents lists available at ScienceDirect

Electrochimica Acta

journa l homepage: www.e lsev ier .com/ locate /e lec tac ta

ynthesis and characterization of Pd-Ni nanoalloy electrocatalysts for oxygeneduction reaction in fuel cells

uan Zhao, Arindam Sarkar, Arumugam Manthiram ∗

lectrochemical Energy Laboratory and Materials Science and Engineering Program, The University of Texas at Austin, Austin 78712, TX, USA

r t i c l e i n f o

rticle history:eceived 15 July 2009eceived in revised form 21 October 2009ccepted 25 October 2009vailable online 1 November 2009

eywords:

a b s t r a c t

Carbon-supported Pd-Ni nanoalloy electrocatalysts with different Pd/Ni atomic ratios have been synthe-sized by a modified polyol method, followed by heat treatment in a reducing atmosphere at 500–900 ◦C.The samples have been characterized by X-ray diffraction (XRD), energy dispersive spectroscopy (EDS),transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), cyclic voltammetry(CV), rotating disk electrode (RDE) measurements, and single-cell proton exchange membrane fuel cell(PEMFC) tests for oxygen reduction reaction (ORR). XRD and TEM data reveal an increase in the degree of

uel cellsalladium-based nanoalloyslectrocatalystsxygen reduction reactionynthesis

alloying and particle size with increasing heat-treatment temperature. XPS data indicate surface segrega-tion with Pd enrichment on the surface of Pd80Ni20 after heat treatment at ≥500 ◦C, suggesting possiblelattice strains in the outermost layers. Electrochemical data based on CV, RDE, and single-cell PEMFCmeasurement show that Pd80Ni20 heated at 500 ◦C has the highest mass catalytic activity for ORR amongthe Pd-Ni samples investigated, with stability and catalytic activity significantly higher than that foundwith Pd. With a lower cost, the Pd-Ni catalysts exhibit higher tolerance to methanol than Pt, offering an

. Introduction

Proton exchange membrane fuel cells (PEMFC) and directethanol fuel cells (DMFC) are gaining considerable interest due

o their high energy density and relatively low operating tem-eratures [1–4]. However, several challenges such as the sluggishxygen reduction and methanol oxidation reactions as well as theigh cost of Pt catalysts impede their commercialization [5,6]. Inhis regard, extensive efforts have been focused on Pt-based alloyatalysts, and optimum alloy catalysts have been found to offerigher catalytic activity than Pt while lowering the cost [7–16].owever, the dissolution of non-noble metal components and theonsequent performance loss during cell operation remain a con-ern [17–19].

With an aim to lower the cost and find platinum-free alter-atives, Pd-based catalysts have drawn much attention in recentears. Also, Pd is known to be inactive for methanol oxidation20–29], offering high tolerance to methanol poisoning as a cathode

∗ Corresponding author. Tel.: +1 512 471 1791; fax: +1 512 471 7681.E-mail address: rmanth@mail.utexas.edu (A. Manthiram).

involve either formation of Pd(II), Pd(IV), and Pd(VI) surface oxidesor Pd electrodissolution. To overcome these problems, efforts havebeen made to alloy Pd with other elements such as Ti, Fe, Co, Au,Mo, and W [20–26,35–40]. Especially, Pd–Co and Pd–Fe alloy cata-lysts have been found to exhibit high catalytic activity for ORR andgood stability [24,26,28]. It has been suggested that alloying Pd withother metals having smaller atomic size such as V, Cr, Fe, and Co isparticularly effective in enhancing the catalytic activity [41,42]. Niis another possible choice to alloy with Pd and enhance its catalyticactivity. Lee et al. [29] reported that Pd-Ni thin film alloys fabri-cated by an rf sputtering method exhibit higher catalytic activitythan Pd, although lower than that of Pt. Ramos-Sanchez preparedthe Pd-Ni alloys by a metallic reductive precipitation [43], but thecatalytic activity of the sample was much lower than that of thethin film sample fabricated by the sputtering method. However, asystematic investigation of the effect of Ni on alloying with Pd isstill lacking.

We present here the synthesis of carbon-supported nanos-tructured Pd-Ni catalysts with various atomic ratios by amodified polyol reduction process, followed by heat treatment atvarious temperatures. The synthesized Pd-Ni alloys are character-

J. Zhao et al. / Electrochimica Ac

5sRNerh(5icTawitaals

cathode. The membrane-electrode assemblies (MEAs) were fab-

ig. 1. X-ray diffraction patterns of Pd and Pd80Ni20 after heat treatment at variousemperatures. The dotted line refers to the expected position of the (1 1 1) reflectionf Pd.

. Experimental

.1. Synthesis

Approximately 50 mg of carbon-supported Pd100−xNix (0 ≤ x ≤0) electrocatalysts with 20 wt% metal loading were synthe-ized by a modified polyol reduction process as described below.equired amounts of 0.5 M (NH4)2PdCl4 (Alfa Aesar) and 0.5 Mi(CH3COO)2·6H2O (Acros Organics) were dissolved in 25 mL ofthylene glycol (Fisher Scientific), followed by an addition ofequired amount of Vulcan XC-72R carbon black (Cabot Corp.) toave 20 wt% metal loading and 12.5 mg of poly(vinylpyrrolidone)PVP, MW = 40,000, MP Biomedicals, LLC) under constant stirring.mL of 0.1 M NaOH (Fisher Scientific) was then added dropwise

nto the solution, and the mixture was refluxed for 10 h at ∼198 ◦C,ooled to room temperature, filtered, washed, and dried overnight.he resulting black powders obtained are denoted hereafter ass-synthesized Pd-Ni catalysts. The as-synthesized Pd-Ni catalystsere subsequently heat-treated at 500, 700, and 900 ◦C in a flow-

ng 10% H2–90% Ar atmosphere for 2 h and then cooled to roomemperature. For a comparison, 50 mg of carbon-supported Pt cat-lyst with 20 wt% metal loading was also synthesized by refluxingrequired amount of H2PtCl6 (Strem Chemicals) in 25 mL of ethy-

ene glycol. The carbon-supported Pt catalyst was not, however,ubjected to any post heat treatment.

.2. Structural and microstructural characterizations

Phase identification was performed by X-ray diffraction withu K� radiation (� = 0.154 nm). Bulk compositional analysis wasarried out by energy dispersive spectroscopy with a JEOL-JSM5610

able 1haracterization data from XRD and EDS of Pd80Ni20 before and after heat treatment at v

Sample Pd:Ni ratio from EDS/% Lattice parameterfrom XRD/nm

As-synthesized 83.7:16.3 0.3913500 ◦C 83.4:16.6 0.3841700 ◦C 82.9:17.1 0.3835900 ◦C 81.1:18.9 0.3833

a The details are described elsewhere [40].b Obtained using the (1 1 1) reflection; the numbers in parentheses give the standard d

ta 55 (2010) 1756–1765 1757

SEM having an Oxford Instruments EDS attachment. Particle sizeand distributions were studied with a JEOL 2010F TEM operatingat 200 keV. Surface compositions were assessed by XPS with Al K�radiation.

2.3. Electrochemical characterization

2.3.1. Cyclic voltammetry and linear polarization measurementsThe electrochemical measurements were performed with an

Autolab potentiostat PGSTAT302N and a standard three-electrodecell at room temperature, employing a saturated calomel electrode(SCE) as the reference electrode and a platinum mesh as the counterelectrode. All potentials are, however, reported here against thenormal hydrogen electrode (NHE). A glassy carbon (GC) disk elec-trode (5 mm), polished to a mirror-like finish with a 0.05 �malumina suspension before each experiment, was used as the sub-strate for the catalyst layer. Typically, the catalyst layer as theworking electrode was prepared by dispersing 2 mg of the carbon-supported catalyst in a mixture consisting of 1 mL of isopropylalcohol and 1 mL of 0.15 wt% Nafion solution, ultrasonicating untila homogeneous ink is formed, applying 20 �L of the ink onto theglassy carbon electrode to obtain a catalyst loading of 20.4 �g cm−2,and drying under an infrared lamp. The cyclic voltammetry exper-iments were conducted in N2 purged 0.5 M H2SO4 between 0 and1.1 V at a scan rate of 50 mV s−1. The potential cycling to cleanand activate the electrode surface was carried out until repro-ducible voltammograms were obtained (ca. 50 cycles), and the CVsobtained after a steady state has been reached are presented inthe paper. Although the as-synthesized sample did not reach asteady state after 50 cycles, the same protocol was used for allthe samples for the sake of comparison. For the RDE experiments[44], the electrolyte (0.5 M H2SO4) was saturated with oxygen,and the polarization curves were recorded at 1600 rpm from 0 to1.0 V at a scan rate of 5 mV s−1. Methanol tolerance of the catalystwas assessed by scanning the potential between 0 and 1.1 V in N2purged 0.5 M H2SO4 + 1 M methanol solution at 20 mV s−1. Accel-erated durability tests were performed at room temperature in N2purged 0.5 M H2SO4 solution by applying cyclic potential sweepsbetween 0.6 and 1.1 V at a sweep rate of 50 mV s−1 for a givennumber of cycles.

2.3.2. Single-cell evaluationFor single-cell measurements in PEMFC, the electrodes con-

sisted of gas-diffusion and catalyst layers. The catalyst layer wasprepared by spraying a mixture consisting of required amount ofthe carbon-supported catalyst, isopropyl alcohol, water, and 40 wt%Nafion onto the top of a commercial gas diffusion layer (BASF)(2.5 cm × 2.5 cm), followed by drying in air at 60 ◦C for 1 h. Themetal loading was kept at 0.4 mg cm−2 both for the anode and the

ricated by uniaxially hot-pressing the anode and cathode onto aNafion 112 membrane (DuPont) at 130 ◦C for 2 min. The perfor-mances of the MEAs with commercial Pt (Alpha Aesar, 20 wt%metal in carbon) as the anode and the Pd-Ni alloy catalysts as the

arious temperatures.

Degree of alloying: atomicpercentage of Ni in the alloya

Crystallite sizefrom XRDb/nm

0 4.6 (0.1)13.3 6.0 (0.2)14.9 14.9 (0.5)15.4 35 (0.4)

eviation.

1758 J. Zhao et al. / Electrochimica Acta 55 (2010) 1756–1765

Fig. 2. TEM images and particle size distributions of (a) as-synthesized Pd80Ni20 and Pd80Ni20 after heat treatment at (b) 500 ◦C and (c) 700 ◦C.

J. Zhao et al. / Electrochimica Acta 55 (2010) 1756–1765 1759

Fig. 3. XPS profiles and fitting results of Pd80Ni20 before and after heat treatmentat various temperatures: (a) Pd 3d core levels showing Pd [- - -] and PdOy (0 ≤ y ≤ 2)[dc

c5t(tb1

atrtcapHipTce

. . .], and (b) Ni 2p core levels. Solid lines refer to the experimental curves andotted lines refer to the fitted curves. C 1s binding energy (284.5 eV) was used foralibration.

athode were evaluated at 60 ◦C with a single-cell fixture havingcm2 active area and humidified hydrogen and oxygen gas reac-

ants at, respectively, 14 psi (96.53 kPa), 0.15 L min−1 and 16 psi110.32 kPa), 0.2 L min−1. IR corrections were made by correctinghe cell voltage with the measured ohmic resistance of the fuel celly an inbuilt AC impedance analyzer operating at a frequency ofkHz.

. Results and discussion

.1. Structural and compositional characterization

Fig. 1 compares the XRD patterns of the carbon-supported Pdnd Pd80Ni20 samples before and after heat treatment at variousemperatures in the reducing atmosphere. All the samples exhibiteflections characteristic of a fcc structure, and the reflections shifto higher angles with increasing heat-treatment temperature, indi-ating lattice contraction caused by a substitution of smaller Ni forlarger Pd. The degree of shifting increases with increasing tem-erature, suggesting an increasing degree of alloying of Pd with Ni.owever, no reflections corresponding to Ni or NiOx could be seen

n the as-prepared sample or in the samples annealed at lower tem-eratures due to the poor cystallinity or low concentration of Ni.he lattice parameter values, degree of alloying, and the averagerystallite size values obtained from the XRD data using Scherrerquation for the Pd80Ni20 sample before and after heat treating at

Fig. 4. Cyclic voltammograms of Pd80Ni20 in 0.5 M H2SO4 solution at a sweep rateof 50 mV s−1: (a) as-synthesized sample, (b) after heat treatment at 500 ◦C, and (c)after heat treatment at 700 ◦C.

various temperatures are listed in Table 1. The calculation of degreeof alloying is described elsewhere [41]. Since no clear shift in thepositions of the reflections of the as-prepared Pd80Ni20 samplescompared to those of Pd is seen, the degree of alloying (at.% Ni in

Fig. 5. (a) Reproducible cyclic voltammograms (in 0.5 M H2SO4 at a sweep rate of5 −1

phsts7tdd

Pvipwia0drsp

Fig. 6. (a) Hydrodynamic polarization (RDE) curves of Pd80Ni20 before and afterheat treatment at various temperatures in O2-saturated 0.5 M H2SO4 with a rotationspeed of 1600 rpm and (b) mass activity and (c) specific activity measured at 0.7 Vfor Pd80Ni20 before and after heat treatment at various temperatures. Mass and

0 mV s ) (b) and the specific ECSA of Pd80Ni20 before and after heat treatment atarious temperatures.

uggests that not all the Ni atoms get into the Pd lattice, indicatinghe presence of Ni or/and NiOx and incomplete alloying of Ni withd.

Fig. 2 compares the TEM photographs and the correspondingarticle size distributions of the Pd80Ni20 samples before and aftereating at various temperatures. As seen, the nanoparticles showpherical morphology with a uniform dispersion on carbon. Whilehe as-synthesized and 500 ◦C samples have a narrow particleize distribution centered around, respectively, 5.6 and 6.7 nm, the00 ◦C sample shows a broader distribution with a much larger par-icle size of 11.8 nm. The particle size values obtained from the TEMata are consistent with the crystallite size obtained from the XRData.

Fig. 3 compares the Pd 3d and Ni 2p core level XPS data ofd80Ni20 before and after heating at 500 and 700 ◦C. The decon-oluted peaks of Pd 3d5/2 and Pd 3d3/2 core levels in Fig. 3(a)ndicate that both Pd and PdOy exist on the surface of all the sam-les due to the exposure of the samples to ambient air [45,46]. Also,ith increasing heat-treatment temperature or degree of alloy-

ng, the ratio of Pd:PdOy decreases gradually from 1.42 for thes-synthesized Pd80Ni20 sample to 1.13 for the 500 ◦C sample and.78 for the 700 ◦C sample. Moreover, the Pd 3d5/2 binding energy

specific activities are given as kinetic current densities (jk) normalized in referenceto, respectively, the loading amount and ECSA of the metal.

2p3/2 and Ni 2p1/2. The main peaks centered at 855.8 and 873.5 eVand the satellite peaks centered at 861.7 and 880.2 eV indicate thatthe Ni on the surface exists as an oxidized Ni species like Ni(OH)2or NiO [47–48]. The Pd:Ni ratio determined from the XPS peak

J. Zhao et al. / Electrochimica Ac

as7itpds

hissrdct[tcl

Pd, Pd Ni Pd Ni Pd Ni Pd Ni and Pd Ni after heat

ig. 7. (a) Reproducible cyclic voltammograms (in 0.5 M H2SO4 at a sweep rate of0 mV s−1) of Pd-Ni alloys after heat treatment at 500 ◦C and their comparison withhat of home-made Pt and (b) the specific ECSA of Pd-Ni alloys after heat treatmentt 500 ◦C.

re found with an increased Pd /Ni ratio. Also, the peaks corre-ponding to metallic Ni and Ni oxide vanish on heat treating at00 ◦C. The observations suggest that in addition to the increas-

ng degree of alloying with increasing heat-treatment temperature,he surface may become richer in Pd. As the typical mean freeath for photoelectrons in alloys is in the range of 1–2 nm, XPSata are representative of several layers beneath the outermosturface.

.2. Electrochemical characterization

Fig. 4 shows the CV plots of the Pd80Ni20 sample before and aftereat treatment at 500 and 700 ◦C. As seen, the peak current density

ncreases initially with cycle number, which could be attributed tourface roughening and removal of contaminants from the sampleurface. Although the as-synthesized sample shows a large oxideeduction peak during the early cycles, the peak current densityecrease substantially after 50 cycles, indicating instability. Thisould be related to the sintering of small particles and forma-ion and dissolution of oxides (or hydroxides) of both Pd and Ni19,34,50,51]. Interestingly, after heat treatment at 500 and 700 ◦C,he peak current density remains the same after the initial surface

Fig. 5 compares the stable CV plots obtained after the initialurface cleaning and the specific electrochemical active surface

ta 55 (2010) 1756–1765 1761

area (ECSA) per unit weight of metal in Pd80Ni20 before and afterheat treatment at various temperatures. Since the electrochemi-cal active surface area of Pd-alloy samples could not be obtainedfrom the hydrogen desorption region due to the dissolution ofhydrogen into the bulk of palladium, the charge correspondingto surface oxide reduction region was used [41] here to comparethe ECSA. The charge transferred during surface oxide reductioncan be evaluated by measuring the peak area and dividing it bythe scan rate. The charge thus evaluated was further normalizedwith respect to the mass of the metal in the sample. It is noticedthat the potential where the formation of surface oxide adsorptionand the reduction of surface oxides begin does not change muchwith increasing heat-treatment temperature. It is found that boththe peak current density corresponding to the cathodic peak andthe specific ECSA increase with increasing heat-treatment tem-perature, reach a maximum at 500 ◦C, and then decrease. Thesmall ECSA for the as-synthesized samples is attributed to thesintering of smaller particles on potential cycling and the dissolu-tion of surface oxides in acid as mentioned earlier. The decreasein ECSA with heat-treatment temperature above 500 ◦C is dueto the obvious increase in particle size. Also, the peak poten-tial corresponding to the oxide reduction peak shifts by 25 mVto the left on going from the as-synthesized to the 500 ◦C sam-ple.

The ORR measurements were performed in O2-saturated 0.5 MH2SO4 solution with a glassy carbon rotating disk electrode, andthe polarization curves for ORR on the Pd80Ni20 samples before andafter heat treating at various temperatures are shown in Fig. 6(a).For all of the catalysts studies here, a mixed kinetic-diffusion con-trol occurs in the whole potential region. The kinetic current wascalculated from the ORR polarization curve using mass-transportcorrection and normalized to the amount of metal loading in orderto compare the mass activities of different catalysts (Fig. 6(b)). Asseen, the mass activity increases as we go from the as-synthesizedsample to the 500 ◦C heat treated sample, and then decreases ongoing from 500 to 900 ◦C. For a better understanding of the observeddifferences in the ORR activity, we normalized the kinetic currentagainst the ECSA for each catalyst as shown in Fig. 6(c). The specificactivity (i.e., kinetic current per unit surface area of the catalystin mA/mC) is found to jump on going from the as-synthesized tothe 500 ◦C sample, indicating an accelerated ORR kinetics on theelectrocatalyst surface.

Recognizing that heat treatment at 500 ◦C gives the maximummass catalytic activity for all the samples, Fig. 7(a) compares the CVplots of the carbon-supported Pd-Ni alloy catalysts with differentNi contents after heat treatment at 500 ◦C with that of home-madePt. As seen, the onset potential of Pd-Ni for the formation of sur-face oxide adsorption and the reduction of surface oxides doesnot vary much with composition. However, the onset potentialis about 50 mV lower compared to that for Pt. All the Pd-Ni cat-alysts show large peak area for hydrogen adsorption comparedto Pt due to dissolution of adsorbed hydrogen into bulk Pd, andthe decreased intensity of hydrogen desorption peak with increas-ing Ni content may suggest the restrained dissolution of hydrogeninto bulk Pd-Ni due to the alloying of Pd with Ni. Moreover, thecathodic peak current density and the specific ECSA (Fig. 7(b))increase with Ni incorporation, reach a maximum at 20 at.% Ni, andthen decrease with increasing Ni content. Additionally, the peakpotential for Pd80Ni20 shifts by 55 mV to the left compared to thatfor Pd.

Fig. 8(a) compares the hydrodynamic polarization curves of

treatment at 500 ◦C with that of home-made Pt. All the Pd-Ni cata-lysts are under mixed control of kinetics and mass transfer, and theORR onset potential of ∼0.85 V observed for Pd80Ni20 is similar tothat of Pt and 0.1 V higher than that of Pd. The mass activity and the

F with at tivity5 at tre

sbBNeb5atPflP

cmgcrFtaeotae

ig. 8. (a) Hydrodynamic polarization (RDE) curves (in O2-saturated 0.5 M H2SO4

reatment at 500 ◦C and their comparison with that of home-made Pt, (b) mass ac00 ◦C, and (d) comparison of the mass activities at various potentials of Pd after he

pecific activity of Pd-Ni catalysts heated at 500 ◦C were calculatedy the method used in Fig. 6(b) and are shown in Figs. 8(b) and (c).oth of them are found to have a “volcano” relationship with thei content with a maximum at 20 at.% Ni. To emphasize the differ-nces between the best Pd-Ni catalyst and Pd and to know how farehind is the best one from Pt, we compare the mass activity of the00 ◦C heat treated Pd80Ni20 with those of 500 ◦C heat treated Pdnd home-made Pt at various potentials in Fig. 8(d). It is evidenthat the ORR activity of Pd80Ni20 is significantly higher than that ofd, and around 36% of that of the home-made Pt at 0.8 V. The dif-erence between Pd80Ni20 and Pt may be related the significantlyarger particle size of Pd80Ni20 (6.7 ± 1.1 nm) compared to that oft (2.6 ± 0.7 nm).

The measured PEMFC potential–current (V–I) polarizationurves of Pd heated at 500 ◦C, Pd80Ni20 heated at 500 ◦C, and home-ade Pt are compared in Fig. 9(a). The V–I curves are mainly

overned by the surface reaction kinetics up to the geometricurrent densities of 0.2 A/cm2. At larger current densities, ohmicesistance losses as well as mass transfer effects become significant.ig. 9(a) and (b) compare the mass activity for oxygen reduction ofhe three catalysts. The logarithmic current density scale allows forn accurate relative activity ranking. At 0.7 and 0.75 V, the Pd80Ni20xhibits two to three times higher mass activity than Pd and halff that of Pt. The results obtained in single cell are consistent withhose obtained in acid electrolyte in this study, but the differencemong the catalysts in single cell is lower than those found in acidlectrolyte.

Oxygen reduction reaction in acidic media may take place

rotation speed of 1600 rpm) of Pd-Ni alloys with different Ni contents after heatand (c) specific activity measured at 0.7 V for Pd-Ni alloys after heat treatment atatment at 500 ◦C, Pd80Ni20 after heat treatment at 500 ◦C, and home-made Pt.

from our hydrodynamic polarization curves (Fig. 10) at variousrotation rates for the 500 ◦C heat treated Pd80Ni20 clearly illus-trate that the oxygen reduction pathway in this bimetallic alloysystem follows a four-electron transfer (n = 4). The number ofelectrons transferred per oxygen molecule on the catalyst is cal-culated from the Koutecky–Levich equation I−1 = I−1

k+ I−1

1 = I−1k

+(0.62nFAD2/3C�−1/6ω1/2)

−1and the measured slope, where n is the

total number of electrons transferred, F is the Faraday constant(96485 C mol−1), A is the area of the electrode (0.1963 cm2), D isthe diffusion coefficient of oxygen in cm2 s−1, C is the concentra-tion of oxygen in mol cm−3, � is the kinematic viscosity in cm2 s−1,and ω is the rotation rate in rad s−1. The values of D, C, and � weretaken to be respectively, 18 × 10−6 cm2 s−1, 1.13 × 10−6 mol cm−3,and 8.93 × 10−3 cm2 s−1 [54,55].

The atomic adsorption pathway involves the dissociativeadsorption of O2 (reaction (1)) and subsequent electroreductionand removal of the adsorbed oxygen atoms (reaction (2)):

2 M + O2 → 2MO (1)

2MO + 4H+ + 4e → 2 M + 2H2O (2)

AE = E(M–Oads) − E(M) − 12 E(O2) (3)

where the adsorption energy (AE) is defined as the change in thepotential energy E of the system (substrate + adsorbate) on adsorp-tion (Eq. (1)). Lu’s group has reported that the AE of Oads is adescriptor of the specific activity of Pd-alloy catalysts towards ORRbased on the atomic adsorption pathway [41]. They found that

J. Zhao et al. / Electrochimica Acta 55 (2010) 1756–1765 1763

Fig. 9. Comparison of (a) the current-voltage characteristics, (b) the correspondingTafel plots, and (c) mass activity at various potentials of Pd after heat treatmenta ◦ ◦

Oartiiibtlsirs

Fig. 10. (a) Hydrodynamic polarization curves of Pd80Ni20 after heat treatment at500 ◦C at various rotation rates and (b) the Koutecky–Levich plots of Pd80Ni20 afterheat treatment at 500 ◦C and home-mad Pt (n refers to number of electrons trans-

t 500 C, Pd80Ni20 after heat treatment at 500 C, and home-made Pt in single-cellroton-exchange membrane fuel cell (PEMFC) at 60 ◦C.

RR kinetics [56]. The Pd-Ni samples synthesized by our methodfter heat treatment at above 500 ◦C have a Pd-rich surface asevealed by the XPS data, suggesting the presence of possible lat-ice strain. At low degree of alloying, the surface concentration of Nis negligible and the lattice-strain effect dominates. With increas-ng heat-treatment temperature or Ni content, more Ni atoms getnto the Pd lattice, resulting in higher lattice strain, weaker M–Oadsonding, and higher specific activity. At high degrees of alloying,he surface concentration of Ni could increase and the surface-igand effect may become dominating, resulting in a decrease in

ferred).

imum at 20 at.% Ni because both the specific activity and the ECSAare maximal at that composition.

It has been demonstrated that palladium-based catalysts exhibithigh tolerance to methanol oxidation and are stable in the pres-ence of methanol [23,25,37–40]. Fig. 11 compares the CV plots of500 ◦C heat treated Pd80Ni20 with that of home-made Pt in the pres-ence and absence of methanol. As seen, the CV plot of Pd80Ni20remain unchanged on going from 0.5 M H2SO4 solution to 0.5 MH2SO4 + 1 M methanol solution in contrast to that of Pt, demonstrat-ing the high tolerance of Pd80Ni20 to methanol and its potential asa cathode catalyst in a DMFC.

We also performed accelerated durability tests by applyingcyclic potential sweeps between 0.6 and 1.1 Vat 50 mV/s in N2-purged 0.5 M H2SO4 solutions at room temperature. The CVmeasurements showed a loss of 20% in ECSA for the 500 ◦C heattreated Pd80Ni20 after 200 cycles and 60% after 400 cycles (Fig. 12),suggesting the less durability of Pd80Ni20 for long-term operation.This may be related to the incomplete alloying of Pd with Ni dueto the low alloying temperature used during the formation of thealloy (see Table 1). Also, even well-alloyed Ni may dissolve or leachout of the surface since thermodynamically less stable base-metalsare more unstable under potential in acidic electrolytes. From apractical point of view, the improvement of durability is critical for

1764 J. Zhao et al. / Electrochimica Ac

Fig. 11. Comparisons of the cyclic voltammogram of (a) Pd80Ni20 after heat treat-ment at 500 ◦C and (b) home-made Pt in 0.5 M H2SO4 and in 0.5 M H2SO4 + 1 Mmethanol solutions at a sweep rate of 20 mV s−1 at room temperature.

[[[[[[40] W. Wang, D. Zheng, C. Dua, Z. Zoua, X. Zhang, B. Xia, H. Yang, D.L. Akins, J. Power

ig. 12. Cyclic voltammograms of 500 ◦C heat treated Pd80Ni20 before and afterccelerated durability test.

. Conclusions

Carbon-supported Pd-Ni nanoalloy electrocatalysts with con-rolled particle size and distribution have been synthesized by a

odified polyol reduction method, followed by heat treatment at

ta 55 (2010) 1756–1765

elevated temperatures in 10% H2–90% Ar atmosphere. The Pd80Ni20sample with a Pd:Ni atomic ratio of 4:1 after heat treatment at500 ◦C is found to exhibit the highest ECSA and mass catalytic activ-ity among the Pd-Ni samples investigated. The enhanced activity ofPd80Ni20 compared to that of Pd is attributed to Pd enrichmentin the surface and the consequent lattice-strain effects and thedecrease in the surface oxide adsorption energy. With a lower costand a much higher tolerance to methanol compared to those of thePt catalyst, the Pd-Ni alloys are attractive as cathode catalysts indirect methanol fuel cells.

Acknowledgements

Financial support by the National Science Foundation grantCBET-0651929 and Welch Foundation grant F-1254 is gratefullyacknowledged.

References

[1] B.D. Mcnicol, D.A.J. Rand, K.R. Williams, J. Power Sources 83 (1999) 15.[2] E. Antolini, J.R.C. Salgado, E.R. Gonzalez, J. Electroanal. Chem. 580 (2005) 45.[3] H. Yang, W. Vogel, C. Lamy, N. Alonso-Vante, J. Phys. Chem. B 108 (2004) 11024.[4] V.S. Murthi, R.C. Urian, S. Mukerjee, J. Phys. Chem. B 108 (2004) 11011.[5] T.D. Jarvi, S. Stuve, E.M. Sriramulu, J. Phys. Chem. B 101 (1997) 3646.[6] T.R. Ralph, M.P. Hogarth, Platinum Met. Rev. 46 (2002) 3.[7] L. Xiong, A.M. Kannan, A. Manthiram, Electrochem. Commun. 4 (2002) 898.[8] L. Xiong, A. Manthiram, J. Mater. Chem. 14 (2004) 1454.[9] L. Xiong, A. Manthiram, Electrochim. Acta 50 (2005) 2323.10] S. Mukerjee, S. Srinivasan, M.P. Soriaga, J. McBreen, J. Phys. Chem. 99 (1995)

4577.11] V. Jalan, E.J. Taylor, J. Electrochem. Soc. 130 (1983) 2299.12] V. Stamenkovic, B.S. Mun, K.J. Mayrhofer, P.N. Ross, N.M. Markovic, J. Rossmeisl,

J. Greeley, J.K. Nørskov, Angew. Chem. Int. Ed 45 (2006) 2897.13] J. Greeley, J.K. Norskov, M. Mavrikakis, Annu. Rev. Phys. Chem. 53 (2002) 319.14] N.M. Markovic, T.J. Schmidt, V. Stamenkovic, P.N. Ross, Fuel Cells 1 (2001) 105.15] V. Stamenkovic, B.S. Mun, K.J. Mayrhofer, P.N. Ross, N.M. Markovic, J. Am. Chem.

Soc. 128 (2006) 8813.16] V. Stamenkovic, B. Fowler, B.S. Mun, G. Wang, P.N. Ross, C.A. Lucas, N.M.

Markovic, Science 315 (2007) 493.17] A. Bonakdarpour, J. Wenzel, D.A. Stevens, S. Sheng, T.L. Monchesky, R. Lobel,

R.T. Atanasoski, A.K. Schmoeckel, G.D. Vernstorm, M.K. Debe, J.R. Dahn, J. Elec-trochem. Soc. 151 (2005) A61.

18] T. Toda, H. Igarashi, H. Uchida, M. Watanabe, J. Electrochem. Soc. 146 (1999)3750.

19] H.R. Colon-Mercado, H. Kim, B.N. Popov, Electrochem. Commun. 6 (2004) 795.20] O. Savadogo, K. Lee, K. Oishi, S. Mitsushima, N. Kamiya, K.I. Ota, Electrochem.

Commun. 6 (2004) 105.21] J.L. Fernandez, D.A. Walsh, A.J. Bard, J. Am. Chem. Soc. 127 (2005) 357.22] J.L. Fernandez, V. Raghuveer, A. Manthiram, A.J. Bard, J. Am. Chem. Soc. 127

(2005) 13100.23] V. Raghuveer, A. Manthiram, A.J. Bard, J. Phys. Chem. B 109 (2005) 22909.24] V. Raghuveer, P.J. Ferreira, A. Manthiram, Electrochem. Commun. 8 (2006) 807.25] W.E. Mustain, K. Kepler, J. Prakash, Electrochem. Commun. 8 (2006) 406.26] M.-H. Shao, K. Sasaki, R.R. Adzic, J. Am. Chem. Soc. 128 (2006) 3526.27] M.R. Tarasevich, A.E. Chalykh, V.A. Bogdanovskaya, L.N. Kuznetsova, N.A.

Kapustina, B.N. Efremov, M.R. Ehrenburg, L.A. Reznikova, Electrochim. Acta 21(2006) 4455.

28] T. Huang, P. Liu, J. Zhang, R.R. Adzic, The 207th Electrochemistry Society Meet-ing, vol. 501, 2006, p. 1588.

29] K. Lee, O. Savadogo, A. Ishihara, S. Mitsushima, N. Kamiya, K. Ota, J. Electrochem.Soc. 153 (2006) A20.

30] P. Pattabiraman, Appl. Catal. A 153 (1997) 9.31] M.T. Giacomi, M. Balasubramanian, S. Khalid, J. McBreen, J. Electrochem. Soc.

50 (2003) A593.32] J. Solla-Gullón, V. Montiel, A. Aldaz, J. Clavilier, Electrochem. Commun. 4 (2002)

716.33] X. Li, I.-M. Hsing, Electrochim. Acta 51 (2006) 3477.34] M. Grden, M. Łukaszewski, G. Jerkiewicz, A. Czerwinski, Electrochim. Acta 53

(2008) 7583.35] W.E. Mustain, J. Prakash, J. Power Sources 170 (2007) 28.36] L. Zhang, L. Kunchan, J. Zhang, Electrochim. Acta 52 (2007) 3088.37] H. Liu, A. Manthiram, Electrochem. Commun. 10 (2008) 740.38] A. Sarkar, A.V. Murugan, A. Manthiram, J. Phys. Chem. C 112 (2008) 12037.39] A. Sarkar, A.V. Murugan, A. Manthiram, J. Mater. Chem. 19 (2009) 159.

Sources 167 (2007) 243.41] Y. Suo, L. Zhuang, J. Lu, Angew. Chem. Int. Ed. 46 (2007) 2862.42] Y.-X. Wang, P.B. Balbuena, J. Phys. Chem. B 109 (2005) 18902.43] G. Ramos-Sánchez, H. Yee-Madeira, O. Solorza-Feria, Int. J. Hydrogen Energy 33

(2008) 3596.

ica Ac

J. Zhao et al. / Electrochim

44] T.J. Schmidt, H.A. Gasteiger, G.D. Stab, P.M. Urban, D.M. KoIb, R.J. Behm, J. Elec-trochem. Soc. 145 (1998) 2354.

45] K.S. Kim, A.F. Gossmann, N. Winograd, Anal. Chem. 46 (1974) 197.

4919.48] J. Zang, S. Yang, Q. Xue, J. Mater. Res. 15 (2000) 541.49] S.M. Yoon, J.Y. Choi, Y.K. Lee, E.S. Lee, S.K. Kim, Mater. Res. Soc. Symp. Proc. 848

(2005), pp. FF3.35.1.

ta 55 (2010) 1756–1765 1765

50] K. Juodkazis, J. Juodkazyt, B. Sebeka, G. Stalnionis, A. Lukinskas, Russ. J. Elec-trochem. 39 (2003) 954.

51] L.D. Burke, J.K. Casey, J. Electrochem. Soc. 140 (1993) 1284.

tions, John Wiley, New York, 2001.55] S. Gottesfeld, I.D. Raistrick, S. Srinivasan, J. Electrochem. Soc. 134 (1987) 1455.56] M. Mavrikakis, B. Hammer, Nørskov. Phys. Rev. Lett. 81 (1998) 2819.

Synthesis and characterization of Pd-Ni nanoalloy electrocatalysts for oxygen reduction reaction in...

Documents

Transcript of Synthesis and characterization of Pd-Ni nanoalloy electrocatalysts for oxygen reduction reaction in...

THE PLATINIFEROUS Au-Pd BELT OF MINAS GERAIS, BRAZIL, AND GENESIS OF ITS BOTRYOIDAL Pt-Pd AGGREGATES

Ang Dakung Away ni San Miguel ug Ni Lucifer (1928)

NI\S/\ - CORE

"Tu n'es ni chaud ni froid" : Fondane lecteur de l'Apocalypse

Xavier Guix - Ni me Explico, ni me Entiendes

Growth of NiO ultrathin films on Pd(1 0 0) by post-oxidation of Ni films: the effect of pre-adsorbed oxygen

Cristical raw material-free catalysts and electrocatalysts

Ni fiction ni réalité

Preparation of Supported Pd Catalysts: From the Pd Precursor Solution to the Deposited Pd 2+ Phase

Ni filosofía sin ciencia, ni ciencia sin filosofía

Talambuhay ni Andres Bonifacio

PD 723 - City of Chicago

daftar isi - PD. Pasar Jaya

NI te cases, ni te embarques

“Ni säger att ni älskar era barn framför allt annat, samtidigt stjäl ...

03 PD part6of8

Proj ni leclec

NOROX PD-40 - United Initiators

65 Pd Pd JAC 1998

DNB PD-CET January 2014 Secondary Course