Abstract

The Pt-Pd/C electrocatalyst was synthesized on graphite substrate by the electrochemical codeposition technique. The physicochemical characterization of the catalyst was done by SEM, XRD, and EDX. The electrochemical characterization of the Pt-Pd/C catalyst for methanol electro-oxidation was studied over a range of NaOH and methanol concentrations using cyclic voltammetry, quasisteady-state polarization, chronoamperometry, and electrochemical impedance spectroscopy. The activity of methanol oxidation increased with pH due to better OH species coverage on the electrode surface. At methanol concentration (>1.0 M), there is no change in the oxidation peak current density because of excess methanol at the electrode surface and/or depletion of OH at the electrode surface. The Pt-Pd/C catalyst shows good stability and the low value of Tafel slope and charge transfer resistance. The enhanced electrocatalytic activity of the electrodes is ascribed to the synergistic effect of higher electrochemical surface area, preferred OH adsorption, and ad-atom contribution on the alloyed surface.

1. Introduction

The direct alcohol fuel cell (DAFC) has now emerged as one of the prospective power sources, since liquid alcohol fuels have several merits over gaseous fuel such as high energy density and, availability of gasoline infrastructure, and these can be used directly without the necessity of reforming. Among the different fuel candidates, methanol has been considered as one of the most appropriate fuel for the DAFCs because of its low molecular weight, simplest structure, and very high energy density (6.1 kWh kg−1). Methanol can be generated from a number of different sources like natural gas, oil, coal, or biomass, since direct methanol fuel cell (DMFC) can operate by feeding methanol directly to the reactor without using a reformer and the system is compact and suitable for portable and mobile power generation [1, 2]. However, it is difficult to carry out the complete oxidation of methanol in acid with Pt alone as the electrocatalyst. The electro-oxidation of methanol in acid media has been studied extensively, and several parallel reactions are reported to limit the electropotentiality of methanol in acid media [3, 4]. Formaldehyde, formic acid and CO are considered as reaction intermediates in acidic media. The formation of surface adsorbed intermediate species could act as poison for subsequent methanol adsorption and oxidation. The CO is strongly adsorbed and linearly bonded, which leads to self-poisoning of Pt electrocatalyst. Further, adsorption of methanol on poisoned Pt cannot occur, and methanol oxidation drops to a minimum rate [57]. Another problem associated with DMFC in acid media is the methanol crossover from anode through the polymeric electrolyte membrane (PEM) poisoning cathode, which results in depolarizing reaction on the cathode and considerable potential loss even under open-circuit conditions [8]. The rate of methanol permeation is particularly high due to electro-osmotic drag associated with the proton conduction mechanism.

Platinum, mostly used as an electrocatalyst, has a low activity for methanol oxidation in acid solution due to its inability to adsorb suitable oxygen-containing species in potential region of methanol adsorption. To overcome this problem, a second and some time a third metal have been advocated as cocatalyst. Transition metal that display one or several redox couples containing hydrous species at potential close to that of hydrogen or methanol oxidation are mainly utilized for the purpose [9, 10]. Superior catalytic activities have been reported for methanol oxidation on Pt-based alloys in acidic media, such as Pt-Ru, Pt-Sn, Pt-Os, Pt-Ru-Os, and Pt-Ru-Ir [1116]. Although incorporation of second and third metal enhances the catalytic activity to some extent, further improvement of the reaction kinetics of anode is essential to make it suitable for commercial viability of DMFCs. To date, although, the majority of the work related to DMFC has been conducted in acidic media, few studies of methanol oxidation in alkaline solutions have been reported [4, 1722]. These studies have demonstrated higher activity for methanol oxidation as compared to that in acid media. Interest in the alkaline methanol fuel cell has arisen because of its better oxygen reduction kinetics in alkaline condition than in acidic environment, simplicity, low cost, and comparable efficiency compared to other types of fuel cells [23]. Beside these, alkaline fuel cell provides benefits like higher efficiency and wider selection of possible electrocatalysts. The generation of H+ ions in the anodic reaction is an obvious reason for the favored kinetics in alkaline solution. The electrocatalysts used in the alkaline fuel cell are Pt, Ni [24, 25], Fe(III) [26], Al [26], and Pt-Ru [27]. The application of alkaline electrolytes to the DMFC could lead to a reduction in catalyst loading and allow the use of less expensive nonprecious metal catalysts. Despite the fact that the efficiency of DMFCs can be improved by well-balanced coadsorption of methanol and OH anions at low potential in alkaline solution, the effective use of alkali media has come to realization only in the last few years. Consequently, the kinetics and mechanistic study of methanol oxidation in alkaline solution at different Pt-based catalysts may provide valuable information for the development of new and powerful catalysts.

The overall oxidation reaction equation for the methanol in alkaline medium is supposed to beCH3OH+6OHCO2+5H2O+6e.(1) Complex process of methanol electro-oxidation involving 6-electron transfer and several intermediate organic species are produced. Methanol oxidation in alkaline medium proceeds through following paths:

Formate and CO2 (as carbonate) are the main soluble reaction products, and CO has been detected as the main poisoning species in alkaline medium [2830]. The (CO)ads species formation follows pathway (B) or (C) and is responsible for current decay during methanol oxidation. The selectivity of product formation largely depends on the potential as well as temperature. The first step of the oxidation process is the dissociative adsorption of methanol followed by subsequent dehydrogenation, water activation, noble metal oxide formation at higher potentials, its reduction, and finally desorption or oxidation of the organic species by reactivation.

Hydroxide ions get adsorbed on the electrode surface and initiate the formation of surface hydroxide and oxide species. The associated electrochemical reaction occurring at the electrode-electrolyte interface is as follows [22, 27]: M+OHPt(OH)ads+e,(2)M+6OH[]M(OH)62sol+4e,(3)MOH+MHMO+H2O.(4) A low level of hydrous oxide may be formed at the metal sites and surface defects due to the repetitive formation and reduction of thin compact layer via the above reactions. This hydrous oxide supplies the active oxygen atom to oxidize the organic species. The OH ions required for the equilibrium are mainly supplied by the solution OH ion at higher potential. Pd incorporation into the Pt matrix, the hydrogen adsorption, and desorption characteristics is significantly improved at much lower potential than that on the Pt alone. The hydrogen desorption particularly at the Pd sites plays important role in the anodic potential region in alkali medium and activates the catalyst surface for deprotonation. The surface hydroxides are more readily formed in the presence of Pd in the catalyst matrix. Thus, the poisoning effect of COad is efficiently reduced due to the synergic effect of Pt and Pd in the matrix.

Dissociative chemisorption of methanol molecules on anode surface proceeds through the following steps:M+CH3OHsolMCH3OHads,(5)MCH3OHads+OHMCH3Oads+H2O+e,(6)MCH3Oads+OHMCH2Oads+H2O+e,(7)MCH2Oads+OHM(CHO)ads+H2O+e,(8)M(CHO)ads+OHM(CO)ads+H2O+e,(9)M(CO)ads+OHM(COOH)ads+e,(10)M(COOH)ads+OHM+CO2+H2O+e,(11) where M stands for Pt-Pd.

In the presence of adequate alkali, the oxidation of M–(CHO)ads and M–(CO)ads may proceed directly through the steps (12) and (13) to the ultimate production of CO2M(CHO)ads+3OHM+CO2+2H2O+3e,(12)M(CO)ads+2OH(onM)M+CO2+H2O+2e.(13) The strength of the bonding of (CHO)ads on the surface probably determines the entire rate of the reaction. In general, the chemisorbed bonding of (CHO)ad on Pt group of metals in alkali is weak such that further oxidation takes place without much difficulty, that is, without irreversibly blocking the electrode’s active sites. The several intermediate organic species have been identified by FTIR spectroscopic study which revealed that the CO species are found to be linearly bonded to Pt sites (Pt–C=O) and bridge bonded to Pd sites (Pd2C=O) [27]. Complete removal of CO species may not be possible at low potential, as they strongly bind to the electrode surface. However, at higher potential, highly active oxygen atoms again become available and the poisonous PtCO or Pd2CO species are oxidized via the reaction (10) or (13).

The reduction of oxygen takes place at cathode and proceeds through following steps: 32O2+3H2O+6e6OH,(14)CO2+2OHCO32+H2O.(15) Therefore, the overall reaction is as follows:CH3OH+8OHCO32+6H2O+6e.(16)

The disadvantage of alkaline solution is its progressive carbonation with CO2 (15), whereby carbonate and bicarbonate are produced, which lower the concentration of the electrolytes affecting the cell performance [26, 31]. The performance of the AFC can be improved by the electrolyte recirculation, and continuous CO2 removal may prevent significant carbonation [27] with the development of a membrane-free laminar flow fuel cell (LFFC) and alkaline-anion exchange membrane (AAEM) for application in AFC which have shown improved performance and very little methanol permeability from anode to cathode chamber [3234].

In fuel cells, the high loading of expensive Pt on carbon has severely limited their use making DAFCs still prohibitive from broad commercialization. One of the ways for cutting down the Pt consumption in DAFCs is to reduce the metal loading. Other ways deal with the development of an anode catalyst which is cheaper, efficient, and more abundant than Pt [35]. In this context, our aim is to reduce the Pt loading without compromising the efficiency of the catalyst, and Pd is of our interest, because it is less costly and at least 50 times more abundant in the earth than Pt [36]. It has been used as an excellent electrocatalyst for oxidizing small organic molecules including methanol [37]. Pd has the power to reduce protons, store, and release hydrogen and is further used to remove adsorbed CO formed from the methanol electro-oxidation, thus abating the poisoning effect [38, 39]. It has been reported that the presence of Pd assists the oxidation of C1 alcohol to CO2 by reaction of CO with hydrogen occluded in the Pd lattice at significantly more negative potential [39]. The release of hydrogen by Pd may thus provide a viable route for lowering the surface concentration of adsorbed CO, permitting the continual oxidation of organic molecules at the Pd surface. Pd is just above Pt in the same group of the periodic table. Therefore, they share some common features and can be coupled with each other. Moreover, Pd can form stronger bonds with other metals other than itself [40]. Therefore, a combination of Pt and Pd that changes the electronic character in thin coating of catalysts layer deposit may be helpful in reducing poisoning by oxidizing CO to CO2. It is also reported that electro-oxidation of methanol on Pd-modified Pt proceeds through the “direct pathway”, and the suppression of CO formation on Pd-modified Pt was observed [41]. The atomic radius of Pt is similar to that of Pd (0.138 and 0.137 nm, resp.) so that the substitution of Pt with Pd atoms expands the lattice only slightly and Pt-Pd alloy is found to bear the f.c.c. structure. Despite possessing interesting electrocatalytic properties, Pd has been less extensively studied for fuel cell applications compared to Pt.

The work in this paper entails a comprehensive study on the physicochemical as well as electrochemical characterization of the synthesized Pt-Pd/C electrode towards oxidation of methanol in NaOH media. For physicochemical characterization, different techniques like scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX), and X-ray diffraction (XRD) study were employed. Electrochemical techniques like voltammetry, chronoamperometry, and electrochemical impedance spectroscopy (EIS) were used to derive the electrocatalytic behavior of the electrodes towards methanol oxidation.

2. Experimental

2.1. Materials

Methanol (MeOH) was purified by using standard procedure [42] and stored over a Type 3A molecular sieve beads. The other reagents such as NaOH and HCl used in this work were of GR grade purity (Merck) and H2PtCl6. 6H2O and PdCl2 were obtained from Arora Matthey Ltd. The graphite block (saw cut finish grade) used as the catalyst substrate was procured from Graphite India Limited. All the solutions were freshly prepared with Milli-Q water.

2.2. Preparation of Pt/C, Pd/C, and Pt-Pd/C Electrocatalysts

Electrochemical deposition of Pt and Pd or codeposition of Pt and Pd were made on coupons of graphite samples with surface area of 0.65 cm2. Electrodeposition of Pt and Pd were carried out using 0.05 M H2PtCl6·6H2O and 0.05 M PdCl2 solutions, respectively. Both H2PtCl6·6H2O and PdCl2 solutions were prepared in 2.0 M HCl and Milli-Q water. For electrochemical codeposition of Pt and Pd, an equimolar mixture of 0.05 M H2PtCl6·6H2O and 0.05 M PdCl2 was used. For each case, electrodeposition was performed at room temperature under galvanostatic control by applying 5 mA cm−2 current with the help of computer controlled PG Stat (AUTOLAB 30) for 10 minutes duration. Catalyst loading for each electrode was maintained at 0.5 mg cm−2.

2.3. Physicochemical Characterization of the Electrocatalysts

Bulk composition of the Pt-Pd/C electrocatalyst was determined by energy dispersive X-ray spectroscopy (EDX), while scanning electron microscopy (SEM) was employed to reveal the surface morphology. All these measurements were done with a JSM-6700F FESEM at an accelerating potential of 5 kV. In order to obtain information about the surface and bulk structure of the catalyst, X-ray diffraction (XRD) study was carried out with the help of Philips PW 1140 parallel beam X-ray diffractometer with Bragg-Bretano focusing geometry and monochromatic Cu Kα radiation (λ = 1.54 Å).

2.4. Electrochemical Measurements

Electrochemical measurements like cyclic voltammetry and chronoamperometry were carried out in the standard three-electrode cell at different temperature with the help of computer controlled AUTOLAB 30 Potentiostat/Galvanostat from Ecochemie B.V., The Netherlands. The working solutions were deaerated by purging with nitrogen (XL grade). Electrochemical impedance spectroscopy studies were conducted using the same AUTOLAB 30 PGSTAT with a frequency response analyzer (FRA) module. EIS was performed with amplitude of 5 mV for frequencies ranging from 40 kHz to 100 mHz. Each scan contained about 100 data points (20 data points per decade). The impedance spectra were fitted to an equivalent circuit model using a nonlinear fitting program. A mercury-mercuric oxide (MMO) reference electrode and a platinum foil (2 × 1 cm2) counter electrode were used for all electrochemical experiments. The experiments were performed at different concentrations of NaOH. Since the potential of the MMO reference electrode depends on the activity of OH, the potential correction was made by𝐸=0.098𝑅𝑇𝐹alnOHa1/2H2O,(17) where R is the universal gas constant; T is the temperature; F is the Faraday; aOH and aH2O are activities of OH and H2O, respectively. All the potentials in this paper are referenced to the standard hydrogen electrode (SHE) at cell operating temperature, and the current is normalized to the geometrical surface area.

3. Results and Discussion

3.1. Structural Characterization of Pt-Pd/C Electrocatalyst

The morphologies of the catalyst surface were studied using scanning electron microscopy. Pt-Pd particles show better dispersion than Pt on the graphite substrate shown in Figure 1. As platinum and palladium were coelectrodeposited on the graphite surface, both the metals were grown together on the same sphere making an assembly of ad-atoms in the form of agglomerated particles. The size of the particles does not show much variation and were found to remain within the range of (80–100 nm).

The Pt and Pd nanoparticles coelectrodeposited on graphite surface exhibited an XRD pattern of a typical face-centeredcubic-lattice structure as shown in Figure 2. The strong diffraction peaks at the Bragg angles of 40.08°, 46.67°, 68.02°, 82.08°, and 86.45° correspond to the (111), (200), (220), (311), and (222) facets of Pt-Pd crystal. Alloying of Pt and Pd does not change the diffraction pattern. With the incorporation of Pd into the fcc structure of Pt, the diffraction peaks were shifted to higher values of 2θ, which is indicative of contraction of lattice. No characteristic peaks of Pt or Pd oxides were detected. The (111) peak was used to calculate the particle size of the Pt-Pd crystal according to the Debye-Scherrer equation. The average particle size in the Pt-Pd/C catalyst matrix was found to be 6.5 nm. However, the morphology indicated agglomeration of smaller particles throughout the matrix. The bulk compositions of electrocatalysts were investigated using energy dispersive X-ray spectroscopy (EDX), and atomic ratio of Pt to Pd was found to be 1 : 2.

3.2. Comparison of Background Behavior for the Pt/C, Pd/C, and Pt-Pd/C Electrodes in Alkaline Medium

Figure 3 shows the cyclic voltammogram of Pd/C, Pt/C, and Pt-Pd/C electrodes in deoxygenated 0.5 M NaOH solution. Three peaks can be observed for Pd/C, which correspond to three electrochemical processes occurring on the surface of Pd/C. Peak a, which appears at −360 mV, can be attributed to the adsorption of OH [43, 44]Pd+OHPdOHads+e.(18) Peak b, which emerges above ca. −30 mV, can be attributed to the formation of Pd oxide layerPdOHads+OHPdO+H2O+e.(19) Peak c, at ca. −266 mV, can be attributed to the corresponding reduction of Pd oxide to Pd during the cathodic sweepPdO+H2O+ePd+2OH.(20) In contrast, the CV of Pt/C shows three potential peaks during the positive-going sweep as shown in the inset of the Figure 3. Peak a′ centering at −568 mV is due to the oxidation of the adsorbed hydrogen [45]PtHads+OHPt+H2O+e.(21) Peak b′, which emerges above −350 mV, can be attributed to the adsorption of OHPt+OHPtOHads+e.(22) Peak c′, can be attributed to the formation of Pt oxide layer which emerges above ca. 190 mVPtOHads+OHPtO+H2O+e.(23) The adsorption of OH starts at the more negative potential on Pd/C electrode than on the Pt/C electrode and overlaps the hydrogen desorption peak.

Peak d′, centering at −215 mV, can be attributed to the reduction of the Pt(II) oxide during the cathodic sweep such asPtO+H2O+ePt+2OH.(24) A small hydrogen adsorption peak, e/, appears at −625 mV with the Pt/C electrode, which is not clearly visible for Pd/C electrode. At the Pd/C surface, hydrogen adsorption and absorption occur simultaneously. The hydrogen adsorption occurs on the Pt surface as follows: Pt+H2O+ePtHads+OH.(25) The feature of Pt-Pd/C catalyst demonstrated both the characteristics of Pd/C and Pt/C and the difference of the potentiodynamic curve for Pt-Pd/C catalyst indicates the formation of an alloy of these metals. The X-ray analysis also supports this fact. For the Pt-Pd/C alloyed electrode, the hydrogen desorption as well as OHadsorption peaks overlaps and appears as a single peak, a′′, at −340 mV. The OH adsorption clearly dominates in this region and starts around −730 mV. The presence of Pd facilitates the adsorption of OH and suppress the hydrogen desorption. The board peak in the potential region −730 and −170 mV can be attributed to the hydrogen desorption and OH adsorptionMHads+OHM+H2O+e,(26)M+OHMOHads+e,(27) M stand for Pt-Pd.

The double layer region for Pt-Pd/C is more compressed, and oxide formation starts at lower potential ca. −165 mV. The appearance of single oxide formation as well as reduction peak is in accordance with alloyed behavior of the Pt-Pd. The Pt-Pd/C electrodes deliver oxide reduction peak, c′′, at ca. −395 mV. The oxide formation and reduction reactions on the Pt-Pd/C surface can be represented byMOHads+OHMO+H2O+e,(28)MO+H2O+eM+2OH.(29) The hydrogen adsorption peak, d′′, appears at around −760 mV, and the corresponding reaction is as follows:M+H2O+eMHads+OH.(30) The advantage of Pt-Pd/C catalyst in alkaline medium is clearly visible from the remarkable negative shift of hydrogen adsorption/desorption (Had/Hdes) and oxide formation/reduction peak as compared to Pd/C and Pt/C. The CVs (Figure 3) were used for the estimation of the electrochemically active surface area (ECSA) of the synthesized catalysts in alkaline medium. The high charge corresponding to the hydrogen adsorption/desorption and oxide reduction peak on Pt-Pd/C catalyst is indicative of very high ECSA compared to Pd/C and Pt/C [46]. The ECSA of the Pd/C electrodes was measured by determining the coulombic charge (Q) for the reduction of palladium oxide and using the relation𝑄ECSA=,𝑆L(31) where “𝑆” is the proportionality constant used to relate charge with area and “l” is the catalyst loading in “g”. The charge required for the reduction of PdO monolayer is assumed as 405 mC cm−2 [47].

The ECSA of the Pt/C catalyst was determined by measuring the charge collected in the hydrogen adsorption/desorption region (𝑄𝐻) after double-layer correction and assuming a value of (𝑄ref = 210 μC cm−2 ) the charge needed for oxidation of a single layer of hydrogen on a smooth Pt surface [48]. Then, the specific ECSA was calculated based on the following relation:𝑄ECSA=𝐻𝑄ref𝑙,(32) where “l” is the Pt loading in “g”.

For Pt-Pd/C catalyst the ECSA was measured by determining the coulombic charge (Q) corresponding to the oxide reduction peak. However, it is not possible to measure ECSA value accurately by using this method, because the reduction peaks may be ascribed to the reduction of the oxides of Pd and Pt formed on the surface of the Pt-Pd/C catalyst during the positive scan. [49]. The charge required for the reduction of PdO (qPdO-red) and PtO (qPdO-red) monolayer were assumed as 405 and 420 μC cm−2, respectively [48, 50]. For the Pt-Pd/C catalyst having Pt to Pd ratio (1 : 2), the mean value of charge required for oxide reduction was calculated as 410 μC cm−2 and used for the calculation of ECSA of the alloyed catalyst. The high charge corresponding to the hydrogen adsorption and oxygen reduction peak particularly on Pt-Pd/C catalyst which is indicative of very high electroactive surface area (ECSA) compared to Pt/C and Pd/C [46, 51].

The value of ECSA for Pt-Pd/C was found to be far high (51.4 m2 g−1), about 15 and 5.8 times higher than the value obtained for Pt/C (3.4 m2 g−1) and Pd/C (8.8 m2 g−1). The high ECSA of Pt-Pd/C catalyst could be attributed to the smaller particle size and more uniform size dispersion.

3.3. Comparison of Methanol Oxidation on Pd/C, Pt/C, and Pt-Pd/C Electrodes in NaOH Solution

The chemical nature and the structure of the electrode material play a key role in the adsorption and electro-oxidation of aliphatic alcohols. The chemisorption on the catalyst surface is one of the prime factors affecting the electrocatalytic activity. This is illustrated by the electro-oxidation of methanol under potentiodynamic conditions in alkaline medium in the voltammograms corresponding to Pd/C, Pt/C, and Pt-Pd/C electrodes. Figure 4 shows the cyclic voltammograms (CV) of methanol oxidation recorded in the solution containing 1.0 M methanol and 0.5 M NaOH on different electrodes at a scan rate of 50 mV s−1. The Pt-Pd/C catalyst shows lower onset potential for methanol oxidation and a large anodic peak current density (130 mA cm−2) compared to Pt/C (100 mA cm−2) and Pd/C (40 mA cm−2). The lower onset potential (magnified in the inset of Figure 4) indicates that methanol oxidation commences at lower potential on Pt-Pd/C (−360 mV) than Pt/C (−345 mV) and Pd/C (−350 mV), respectively. Thus, Pt-Pd/C electrode demonstrates a much higher catalytic activity towards the methanol oxidation, probably due to the fact that alloying of Pd with Pt is capable of controlling the poisoning effect and selected as the working electrode in course of studying the electro-oxidation kinetics of methanol.

The inset of the Figure 4 illustrates the methanol oxidation under potentiodynamic conditions on Pt-Pd/C in the solution containing 0.5 M NaOH and 1.0 M MeOH which has been superimposed with the basic voltammogram for the alkaline solution only. A detailed cyclic voltammetric study on the Pt-Pd/C electrode shows that in blank NaOH solution, a visible hydrogen adsorption/desorption peak of current density 3 mA cm−2 appeared at potential around −400 mV. The broad second oxidation peak at potential ~200 mV was attributed to the formation of metal oxide [22, 52]. A methanol oxidation peak was clearly observed in the CV of Pt-Pd/C electrode in the presence of 1.0 M methanol at potential ~700 mV, and the oxide formation peak was suppressed by this large current of methanol oxidation. The electro-oxidation of methanol was characterized by two well-defined current peaks on the forward and reverse scans. In the forward scan, the oxidation peak corresponds to the oxidation of freshly chemisorbed species coming from the alcohol adsorption. The reverse scan peak is primarily associated with the removal of carbonaceous species formed during the oxidation of freshly chemisorbed species in the forward scan [53, 54]. The magnitude of the peak current on the forward scan indicates the electrocatalytic activity of the Pt-Pd/C electrode for oxidation reaction of methanol. The initial small current density peak (5 mA cm−2) in the hydrogen desorption region in the presence of methanol is attributed to the combined effect of hydrogen desorption and the dissociative adsorption of methanol in this potential region. A small oxide reduction peak was observed in the reverse scan in the absence of methanol, while the same was not prominent in presence of methanol. The presence of unoxidized organic residue suppresses the oxide reduction peak during reverse scan.

3.4. Effect of Methanol Concentration on Methanol Oxidation

Cyclic voltammograms for the methanol oxidation on Pt-Pd/C electrode in 0.5 M NaOH are displayed in Figures 5(a) and 5(b) for different concentration of methanol. For low concentration of methanol (0.1 M), the oxide formation and their reduction peaks were observed along with the feeble hydrogen adsorption/desorption peaks still visible, implying that methanol does not prevent completely the hydrogen adsorption. The formation of metal oxide gives rise to nearly inhibition of methanol oxidation, and a decreasing trend in current was observed [55]. This is also because of the deactivation of the electrode surface by poisoning species and the shortage of methanol close to the electrode surface due to the oxidation. But as the electrode surface was reactivated by reduction of metal oxide in the reverse scan, the renewed oxidation became prominent. The magnitude of hydrogen adsorption/desorption were reduced as the concentration of methanol increases, implying that methanol is adsorbed preferentially on the electrode surface at this potential [55, 56]. A second anodic oxidation peak was observed at methanol concentration as high as 2.0 and 3.0 M. This reflects the possibility of oxidation of organic residue produced at high concentration of methanol. The increase in the concentration of methanol results in an increase in anodic peak current in the forward scan up to the concentration 1.0 M, and beyond that concentration, the increase in anodic peak current becomes negligible. This behavior may be explained by a shift from a diffusion-controlled reaction at low methanol concentration to reaction, which is inhibited by adsorbed reaction intermediates at higher concentrations and insufficient availability of OH in the solution [57]. Figure 5(c) demonstrates a plateau in the plot of the peak current density versus log [MeOH] at concentration greater than approximately 1.0 M. The figure reveals that methanol oxidation peak current increases with increasing methanol concentration up to 1.0 M, after which they reach almost a constant value independent of the concentration. Thus, a concentration value of 1.0 M represents a critical concentration after which the adsorption of oxidation products at the electrode surface causes the hindrance to further oxidation. At higher concentration of methanol, diffusion control reaction can no longer occur, and the reaction is inhibited by adsorbed reaction intermediates. Diffusion control reaction with respect to methanol occurs when OH ions are available in excess and chemisorbed methanol species is insufficient [57]. Thus, at lower methanol concentration, when methanol/OH ratio ≤1, the peak current in the voltammograms was controlled by the diffusion transport of methanol due to excess availability of OH. At higher methanol concentrations, methanol/OH ratio >1, the peak current in voltammograms were controlled by the diffusion transport of OH ions because of excess production of reaction intermediates and insufficient availability of OH ions compared to methanol. However, maximum peak current was observed with equimolar proportion of OH and methanol in solution. It is clearly observed that the anodic peak current with increasing methanol concentration levels off at concentration higher than 1.0 M. This effect may be assumed due to the saturation of active sites on the surface of the electrode. In accordance with this result, the optimum concentration of methanol may be considered to be 1.0 M to achieve a reasonably high current density of oxidation. An increase in reverse oxidation peak current and a positive shift of both forward and reverse oxidation peak potential was encountered. This phenomenon can be explained by the increase in concentration of the unoxidized organic residue with the increase in the concentration of methanol and require higher potential to oxidize this strongly adsorbed residue. Figure 5(d) represents the dependence of anodic peak potential (𝐸𝑝) of methanol oxidation on the bulk concentration of methanol. The relation between methanol concentration and the oxidation peak potential may be attributed to the IR drop due to high oxidation current at higher concentrations [25]. However, the overall process is not a case of single kinetic control but rather is more complicated with a series of competing reactions. The broadening of methanol oxidation peak with the increase of methanol concentration is due to the accumulation of methanol molecules on the catalyst surface that require higher potential to oxidize though oxidation starts more or less at the same potential. The diminishing trend of the oxide formation as well as the reduction peak is due to higher concentration of methanol that suppresses both the peaks. The inset of Figure 5(b) shows the variation of onset potential with the concentration of methanol. A small negative shift of the onset potential was observed with the increase in methanol concentration and is attributed to the increased rate of methanol adsorption at the catalyst surface with adequate supply of methanol molecules. This behavior is also reflected in Figure 6, which shows a nonlinear behavior in the potentiostatic polarization plot at high anodic potential.

The reaction order for methanol in the methanol oxidation reaction was determined by plotting log i versus log [MeOH] up to methanol concentration of 1.0 M at a selected low potential region of −260 mV to −60 mV as presented in Figure 7. The order with respect to methanol was derived from the relation Rate𝑖=𝑘𝐶𝑛,log(𝑖)=log𝑘+𝑛log𝐶,(33) where, “i” is the current density, k is the reaction rate constant, C is the bulk concentration of methanol, and “n” is the reaction order. A slope of 0.45 was obtained, which was independent of the potential. A reaction order close to 0.5 is also reported in alkaline solution on platinized Ti mesh electrode which suggested that the adsorption of methanol and intermediates on Pt-Pd/C electrode followed a Temkin-type isotherm in the Tafel region [25, 57]. Higher concentration of methanol was avoided, because at higher concentration, the presence of other adsorbed intermediate species makes the oxidation of methanol more complicated. Partial oxidation of CH3OH leads to the formation of (HCO)ads as the intermediate by CH3OH+3OH𝑘1HCOad+3H2O+3e(34) and the coverage of intermediate HCOads can be expressed in terms of methanol concentration as 𝜃HCO=𝑘1𝐶0.45CH3OH.(35)

3.5. Effect of NaOH Concentration on Methanol Oxidation

The influence of the hydroxide concentration on the oxidation of methanol, in alkaline medium on Pt-Pd/C electrode, was studied by recording CVs in solution containing 1.0 M methanol and x M NaOH, with x ranging between 0.1 and 2.0 M. Typical CVs at a scan rate of 50 mV s−1 are shown in Figure 8. The current density increases rapidly with OH concentration at the same methanol concentration. The magnitude of reverse oxidation peak current density is much less than the anodic peak current density. This is suggestive of the trend towards complete oxidation of methanol at higher concentration of OH when only a small amount of methanolic residue may be left on the electrode surface. The reaction proceeds further on the surfaces covered by reversible OHad species, showing relatively faster kinetics. The current maxima are attained at the potentials (𝐸𝑝), where the reaction kinetics is optimized by the balance between the rate of methanol dehydrogenation and the rate of oxidation of dehydrogenated products with OHads species [58]. The balance is rapidly disturbed immediately after reaching the maximum reaction rate, probably due to the fast transition from the reversible to the irreversible state of oxygenated species. The lowering of onset potential for methanol oxidation was observed with the increase in the concentration of NaOH and is shown in the inset of Figure 8. This behavior also suggests that the kinetics of methanol oxidation was enhanced by the availability of OH ions in solution in ample quantity and higher OH ion coverage on the electrode surface [57]. It was found that 0.5–1.0 M NaOH appears to contribute to the effective electrocatalytic performance of Pt-Pd/C electrode as an optimal pH condition. Tafel plots for the methanol oxidation reaction for a range of NaOH concentration are given in Figure 9. At concentration greater than 0.5 M NaOH, the Tafel slopes were identical (125 mV dec−1) and suggest that the same reaction mechanism occurs throughout this pH range. At very low concentration of NaOH (0.1 M), the Tafel slope increased to 266 mV dec−1, showing poor reactivity at lower pH condition. Figure 10 shows the dependence of log i versus log [NaOH] at several fixed potentials in the linear region of the Tafel plots of Pt-Pd/C catalysts. The reaction order for OH, obtained from the slope of the log i versus log [OH] plot, was 0.7 and was independent of the potential within the Tafel potential range. However, the plot of log 𝑖𝑝 versus log [OH] gave slope close to 1.0. This behavior was also observed with methanol oxidation on Pt and platinised Ti mesh electrode in alkaline medium [57, 58]. The fractional reaction order for OH in Tafel region implies that adsorption of OH on the electrode follows Temkin type isotherms. The energy of adsorption at any adsorption site depends on whether or not its neighbors are already occupied. It is assumed that the rate-determining step involves adsorption of OHad as well as that of the intermediate HCOad [58]HCOad+OHad𝑘HCOOH.(36) The fractional coverage can be expressed as OHOHad+e,𝜃OH=𝑘1𝐶0.5OH𝑒𝛼𝐹𝜂/𝑅𝑇,(37) where 𝑘1 is rate constant, α is transfer coefficient and η is the over potential. The different reaction order at the peak current density and in the Tafel region suggests different reaction process occurs in these two potential regions. Thus, the OH adsorption on the electrode is dependent on the potential.

3.6. Current-Time (Chronoamperometric) Behavior for Methanol Electro-Oxidation
3.6.1. Effect of Methanol Concentration

Current-time transient curve were recorded over a period of time in order to characterize the stability of methanol oxidation reaction on the Pt-Pd/C catalyst in alkaline media allowing building up reaction intermediates or products over the catalyst surface. Figure 11 shows current-time response on Pt-Pd/C at −160 mV for methanol oxidation in 0.5 M NaOH at different concentration of methanol for 1000 seconds. For all concentrations, the methanol oxidation current gradually decay with time due to the formation of subsequent intermediate species such as (CH3OH)ad, CHOad, and COad during the methanol oxidation. Long-term poisoning rate (δ) can be calculated by measuring the linear decay of the current at times greater than 500 seconds from Figure 11 [14]𝛿=100𝑖0×𝑑𝑖𝑑𝑡𝑡>500sec(%persec),(38) where (di/dt)t>500 sec is the slope of the linear portion of the current decay and 𝑖0 is the current at the start of polarization back extrapolated from the linear current decay. The calculated δ values are given in the Table 1. It is evident from the Table that the long-term poisoning rates increase with the increase in concentration of methanol although the initial methanol oxidation current is relatively high for higher methanol concentration. A steady value of current was observed up to 0.5 M methanol, and at higher concentration of methanol, the value of current decreases even after 1000 seconds of measurement. With the accumulation of adsorbed intermediate species at higher concentration methanol, the current decreases with time. At this stage, some amount of CO2 production is expected, which may increase the carbonate concentration in the anolyte resulting in the concentration polarization [4]. At lower concentration of methanol, the formation of CO2 as well as adsorbed intermediates is negligible resulting in steady current over a long period of time. The overall methanol oxidation current on Pt-Pd/C electrode is maintained at higher level up to methanol concentration 2.0 M. beyond this concentration at 3.0 M methanol; there is a clear tradeoff for the oxidation current due to the combined effect of electrode poisoning and concentration polarization. The results are in agreement with the result of the cyclic voltammetry. In all cases, initial decay in methanol oxidation current is related to the formation of metal oxide [59]. The rate of current decay is very low at longer times, indicating the prevailing pseudosteady state kinetic conditions [60]. In chronoamperometric experiments, the current does not reach a stationary behavior even after several hours. Electrode deactivation is caused mainly due to the oxidation of metal surface. The other factor causing the decay of current is apparently the blockage of surface by some organic residue, which is slowly formed and can only be oxidized at high anodic potentials to the ultimate product [61].

3.6.2. Effect of NaOH Concentration

The stability of methanol oxidation on Pt-Pd/C electrode at a fixed concentration of 1.0 M methanol with various concentration of NaOH was investigated by chronoamperometry at a potential of −160 mV, as shown in Figure 12. An increase in methanol oxidation current was observed with the increase in NaOH concentration, and results are in agreement with the results of cyclic voltammetric experiments (Figure 8). The long-term poisoning rate (δ) for methanol oxidation with various concentration of NaOH was calculated by the Equation (38), and the values are given in the Table 1. At low concentration of NaOH (0.1 M), the methanol oxidation current is comparatively less although the rate of poisoning is the least. This is due to the presence of fewer number of OH ions at the electrode surface, compared to methanol molecules in large number. At the concentration of NaOH as high as 3.0 M, rapid decay of the oxidation current was observed (Figure 12) which is also evident from the high rate of poisoning (Table 1) of the electrode surface.

3.7. Electrochemical Impedance Spectroscopic (EIS) Study
3.7.1. Effect of Potential on the Impedance Response

EIS was used to investigate the overall kinetics of methanol oxidation at different potentials. The technique enabled to dissect the various impedance parameters for the charge transfer reaction occurring across the electrode solution interface. Nyquist plots for methanol oxidation for 1.0 M methanol in 0.5 M NaOH solution at 25°C are shown in Figure 13. The effective charge transfer resistance (𝑅ct) was used to analyze the electrode kinetics of the reaction process. In the low potential region (E < −360 mV), shown in Figure 13(a), the Nyquist plot is characterized by a capacitive feature, as expected for hydrogen adsorption/desorption and double layer charging/discharging phenomena [57]. However, at moderate potential region that is, the Tafel region for the present case (−360 < E <−110 mV), there is a changeover from capacitive behavior to resistive behavior, as shown in Figure 13(b). At these potentials, the Nyquist plot resembles a semicircle, which can be assigned to kinetically controlled reaction. The small arc in the high frequency region may be associated with the chemisorption and dehydrogenation of the methanol molecule at the initial stage of the oxidation process. A charge transfer resistance (𝑅ct) for the kinetically controlled reactions may be represented by the diameter of the semicircle in the medium frequency and is related to the charge transfer reaction kinetics according to𝑅ct=RTnFi0,𝑖𝑜=𝑛𝐹𝐴𝑘𝑜𝐶𝑜(1𝛼)𝐶𝛼𝑅,(39) where R: molar gas constant (J mol−1 k−1); T: temperature (K); n: number of electrons transferred; F: Faraday constant (C); i0: exchange current (A); A: reaction area (cm2); 𝑘0: standard heterogeneous rate constant (cm sec−1); C0, CR: bulk concentration of oxidant and reductant species (mol L−1); α: transfer coefficient. The charge transfer resistance is closely related to potential within the Tafel region. The values of charge transfer resistance at different potentials are given in the Table 2, from where it is observed that 𝑅ct decreases with increasing potential, indicating faster reaction kinetics at higher potentials. The depression in the arcs with potential may result from a high degree of heterogeneity at the electrode surface, and in such a case, the double-layer capacitance may be expressed taking into consideration the constant phase element (CPE) [4, 57].

Impedance can be written as 𝑍CPE=1𝑇,(𝑗𝜔)(40) where T is a constant in F cm−2 s ø−1j = (−1) ω is the angular frequency, and Ø is related to the depression angle α according to α = (1 − Ø) 90°. At potentials greater than −160 mV, Figure 13(c), a straight line appeared on the Nyquist plot at low frequencies. This can be attributed to the Warburg impedance (𝑍𝑤) associated with a diffusion control process [57]. Potential independent nature of the charge transfer resistance beyond −60 mV (Table 2) also support the diffusion control nature of the reaction. In addition to the Warburg impedance, two arcs at higher frequencies are associated with the adsorption (high frequency) and kinetic (medium frequency) process. On the basis of the impedance behavior, an equivalent circuit is suggested in Figure 14. For curve fitting calculations, impedance spectra for methanol oxidation at 140 mV (Figure 13(c)) was considered. The calculated impedance parameters are shown in Table 3. The low level of OH perhaps generates unfavorable impedance parameters even at higher potential values.

3.7.2. Effect of Methanol Concentration on the Impedance Response for Methanol Oxidation

The effect of the methanol concentration on the impedance response in 0.5 M NaOH is shown by the Nyquist plots, recorded at −160 mV in Figure 15. It is observed from Table 1 that at lower concentration range of methanol, the charge transfer resistance decreased with increasing methanol concentration. This behavior suggests that the reaction kinetics were enhanced by the adsorption of methanol on the electrode surface. However, too high concentrations of methanol do not favor the reaction kinetics, and the catalyst may be subject to poisoning effect in this potential and concentration range. The high values of charge transfer resistance for all concentration of methanol suggest slow reaction kinetics. These results are in agreement with those obtained from voltammetric studies. For low methanol concentration and at this potential, a feeble inductive feature was observed at low-frequency region, indicating the possibility of oxidation of organic residues including CO present at the electrode surface [28].

3.7.3. Effect of NaOH Concentration on the Impedance Response for Methanol Oxidation

Figure 16 shows Nyquist plots for methanol oxidation at −160 mV for a set of NaOH concentrations. The values of the charge transfer resistance (𝑅ct) and solution resistance (𝑅𝑠) are given in the Table 1. From Figure 16 and Table 1, it is clear that the charge transfer resistance and solution resistance decreased with increase of NaOH concentration. This indicates that the reaction becomes facile with increasing OH concentration, which also supports the voltammetric results.

4. Conclusion

The electrochemically synthesized Pt-Pd/C catalyst shows remarkably high activity towards methanol oxidation in alkaline media both in terms of oxidation peak current density and the onset potential. The peak current density of methanol oxidation on Pt-Pd/C catalyst increases with increasing the methanol concentration. For methanol concentration higher than 1.0 M, no appreciable change in the oxidation peak current density was observed. The results indicate that the reaction of the electro-oxidation of methanol is activation controlled process. A linear increase in methanol oxidation current density was observed with the increase in NaOH concentrations, suggesting better electrode kinetics at higher concentration of NaOH. The optimum methanol/OH concentration ratio is 1 :  2. The kinetics of the methanol oxidation reaction on the Pt-Pd/C electrode was determined by the surface coverage of OH species on the electrode surface. A reaction order of 0.45 and 0.70 were obtained for methanol and NaOH, respectively.

The Pt-Pd/C catalyst showed good stability over a moderately extended period of time. The low value of Tafel slope and charge transfer resistance also suggests that the electrode could be promising for application in DMFCs in alkaline medium.

Acknowledgment

The authors gratefully acknowledge the financial support of the Defense Research and Development Organization (DRDO), New Delhi, India.