Electrocatalytic oxidation of ethanol at Pd, Pt, Pd/Pt and Pt/Pd nano particles supported on poly 1,8-diaminonaphthalene film in alkaline medium

An ethanol oxidation reaction (EOR) in alkaline medium was carried out at palladium (Pd) or platinum (Pt) nanoparticles/poly 1,8-diaminonaphthalene (p1,8-DAN) composite catalyst electrodes. Pd and Pt were incorporated onto a p1,8-DAN/GC electrode by a cyclic voltammetry (CV) strategy. The obtained Pd/p1,8-DAN/GC, Pt/p1,8-DAN/GC, Pt/Pd/p1,8-DAN/GC and Pd/Pt/p1,8-DAN/GC modified electrodes were characterized by scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX) and cyclic voltammetry (CV) techniques. Electrode surface areas (ESAs) of the obtained catalysts were calculated by carbon monoxide (CO) adsorption using differential electrochemical mass spectroscopy (DEMS). The electrocatalytic oxidation of ethanol (EtOH) at the catalyst electrodes was considered in 0.5 M NaOH solutions by CV and chronoamperometric techniques. The catalyst electrodes significantly enhanced the catalytic efficiency for EOR compared to a bare glassy carbon (GC) electrode. Bimetallic catalyst electrodes demonstrate improved catalytic activity, superior durability and higher tolerance to (CO) poison generated in the development of EOR compared with Pd/p1,8-DAN and Pt/p1,8-DAN catalysts, giving priority to Pt/Pd/p1,8-DAN/GC electrodes. Viability parameters, such as NaOH and EtOH concentrations, scan rate and upper potential limits, were examined and analyzed. This study suggests that the prepared catalysts have pronounced potential applications in direct EOR in fuel cells.


Introduction
In recent decades, direct oxidation fuel cells have pulled in extraordinary consideration since they offer a productive and green innovation for energy conversion. Nevertheless, the utilization of hydrogen is restricted by issues of fabrication, purication, storage and distribution. Alkaline direct oxidation fuel cells engaging on several liquid fuels promise to be an environmentally friendly energy technology, mainly because of the enhanced performance as a result of fast electrochemical kinetics on both the anode and cathode. [1][2][3][4][5][6][7] Among several liquid fuels, ethanol (EtOH) has been documented to be the most appropriate fuel as it is a sustainable and carbon-neutral transport fuel. 8 Insignicant poisoning effects in alkaline solutions were also detected. 6 In electrochemical oxidation, the electrode material is a fundamental factor, and a highly effective electrocatalyst is required. Conducting polymer matrixes with excellent stability, a porous nature and high active surface area (like poly 1,8-diaminonaphthalene) are attractive and favorable supports for incorporation of the catalyst nanoparticles (NPs). [9][10][11] The high available surface area, the synergistic impact between NPs and the conducting polymer and the improved tolerance against poisoning by adsorbed carbon monoxide (CO ads ), which is formed during the EtOH electrooxidation, enhance the effectiveness of the electrocatalyst and reduce the surface poisoning. The supported structure prevents agglomeration and affords a high degree of distribution of MNPs during the electrodeposition for liquid fuel oxidation. It could be considered as a redox mediator between the electrode and the electroactive reactant by electron-proton exchange. Composites of conducting polymers with NPs offer a rapid electron transfer pathway through the polymer layer during the electrochemical process. Therefore, metal electrodeposition on the conducting polymers can provide a low-cost and appropriate approach for electrode modication. [12][13][14] Pt has been broadly used as the anode catalyst for liquid fuel oxidation due to its excellent electrocatalytic capability. [15][16][17] Disregarding the critical changes made up to now in the performance of direct oxidation fuel cells, various technical obstacles to their marketable application have continued, including the low performance of the anodes for the oxidation process and the high cost of noble metal platinum-based (Ptbased) catalysts. [18][19][20] Diverse methodologies have been utilized to help the commercial application of direct oxidation fuel cells: for example, the incorporation of another metal into Pt, controlling the catalyst morphology, and choosing appropriate support materials. 11,21,22 The incorporation of a second metallic element into Pt has been recognized to help the development of liquid fuel oxidation kinetics considerably. 9,11,12,23 Referring to the dual-function and fundamental mechanism, a second metal can provide oxygen containing species at lower potentials and reduce the bond strength of Pt-CO ads , and then support the oxidation removal of CO to CO 2 . 21 Palladium (Pd) has been used as an alternative material to Pt, mainly in fuel cells, due to its reasonably abundant source and proper electrocatalytic activity. 6,21,24 There are few reports on utilizing conducting polymers as dispersion media for Pt and Pd NPs. In this paper, we report a simple strategy for the preparation of a poly 1,8diaminonaphthalene/glassy carbon modied electrode as an effective support to anchor single and bimetallic Pd and Pt NPs. 25 This material offers a high potential for an electrocatalyst support for an EtOH oxidation reaction (EOR).

Instruments
SEM images were recorded using QUANTA FEG 250 equipped with an energy dispersive ray spectrometer (EDS). Electrochemical measurements were recorded using a potentiostat Model BASi EPSILON. All electrochemical experiments were achieved using an electrochemical cell, with a conventional three-electrode system. A GC electrode (3.0 mm diameter) was used as the working electrode and Pt wire as an auxiliary one. All reported potentials were recorded with respect to an Ag/AgCl reference electrode. Electrochemical surface area (ESA) measurements were carried out using an EG&G potentiostat (model 273A) in combination with LabVIEW soware (National Instruments GmbH, Munich, Germany) for recording cyclic voltammograms. Carbon dioxide (CO 2 ) was detected by differential electrochemical mass spectroscopy (DEMS) using a quadrupole mass spectrometer (Balzer QMG-422) with dual thin layer ow through a cell in which a hydrophobic Teon membrane forms the interface between the electrolyte and the vacuum.  supporting electrolyte. Aer formation of the CO monolayer, the solution was replaced by a pure 0.5 M supporting electrolyte under potential control (E ¼ 0.06 V) in order to maintain a solution free of CO. The faradic and ionic currents were recorded during the positive potential sweep at a scan rate of 0.01 V s À1 and a ow rate of 5 mL s À1 .

Electrochemical performances and surface morphology of the catalysts
Cyclic voltammograms (CVs) of both Pd/p1,8-DAN/GC and Pt/ p1,8-DAN/GC catalyst electrodes in 0.5 M NaOH in the potential range of À0.9 V to 0.3 V were recorded at scan rate of 0.05 V s À1 , as shown in Fig. 1. In the forward scan, oxidation of Pd or Pt forming an oxide layer was observed in a potential range from 0.0 V to 0.3 V. In the backward sweep, the oxide stripping peaks for Pd/p1,8-DAN/GC and Pt/p1,8-DAN/GC electrodes were revealed at À0.53 V and À0.15 V, respectively. 24,27,28 To date, a lot of bimetallic materials with synergistically improved activities have been considered based on Pt in combination with its neighboring transition metals. Among these materials, a Pt/Pd catalyst is more stable than other bimetallic catalysts at high potentials. 29,30 The electrochemical behaviors of bimetallic Pd/Pt/p1,8-DAN/GC and Pt/Pd/p1,8-DAN/GC catalysts were investigated in 0.5 M NaOH in the potential range of À0.9 V to 0.3 V at a scan rate of 0.05 V s À1 , as exhibited in Fig. 2. Both catalysts demonstrated characteristic anodic peaks for metal oxide formation in the forward sweep, whereas the cathodic peaks at À0.35 V and À0.5 V were due to the oxide stripping of the Pd/Pt/p1,8-DAN/GC and Pt/Pd/p1,8-DAN/GC electrodes, respectively.
ESAs of the prepared catalysts were computed by recording faradic and ionic currents for the oxidation of CO ads to CO 2 by CV and DEMS procedures (gure not shown). Anodic oxidation peaks appearing at 0.9 V for Pd/p1,8-DAN/GC, and at 0.7 V for the other three catalysts were ascribed to the combined impact of oxidation of CO ads and the partial surface oxidation of Pt or Pd to metal oxide (PtO or PdO). Cathodic reduction peaks shown in the reverse scan were credited to metal oxide reductions. Charges associated with the metal oxide reductions were subtracted from the anodic charges and the charge corresponding to CO ads oxidation. [31][32][33] The ionic signal for m/z ¼ 44 in 0.5 M H 2 SO 4 at a scan rate of 0.01 V s À1 at a ow rate of 5 mL s À1 was utilized for ESA calculations according to the following equation: where F is the Faraday constant, G M is the surface concentration of the CO ads monolayer (assuming a G M of 1.45 nmol cm À2 corresponding to 280 mC cm À2 ) and K* is the calibration constant measured by CO stripping on polycrystalline Pt at a scan rate of 10 mV s À1 . To compare the activities of different catalysts in terms of protable efficiency, the current is generally normalized by the mass of loaded metal. Despite the fact that the mass-current density characterizes the economic efficiency of a catalyst, this does not consider the surface area of active metal sites. Electrochemical active surface area (ECSA) is an essential parameter that explains the number of electrochemical active sites with reference to the mass of noble metal 34,35 as follows: where Q is the coulombic charge of the metal oxide reduction peak; s is the proportionality constant that correlates charge with area (0.405 mC cm À2 ) and l is the electrocatalyst loading (g m À2 ). The metal loading and ECSA of the modied electrodes were computed and are listed in Table 1. Pt/Pd/p1,8-DAN/GC showed more active reaction centers than the other electrocatalysts, an important parameter for their electrocatalytic activity due to its higher ECSA. Generally, the addition of the second metal nano particles to obtain the bimetallic catalysts increased the ECSA values. The surface topographies of the obtained catalysts were evaluated using SEM, as shown in Fig. 3A-D, which displays signicant differences in the surface structures of the four catalysts.
Electrodeposited Pd (shiny particles) and Pt (light-grey) particle sizes are in the range of 88.4-96.3 nm and 89.3 nm to 168.2 nm, respectively, as shown in Fig. 3A and B.     electrodes, two well-characterized oxidation peaks in the forward and reverse scans (E pf and E pb , respectively) 6 appeared at À0.20 V and À0.44 V, À0.20 V and À0.35 V, À0.12 V and À0.24 V, and À0.06 V and À0.22 V, respectively (Fig. 4A-D). These oxidation peaks could be ascribed to EtOH  (2) The current density and ECSA were different due to the surface area and the charge of the catalyst electrodes.

Electrocatalytic activities of the catalysts toward EOR
(3) E pf and E pb varied with the metal type of the supported catalyst, giving lower values at Pt/Pd/p1,8-DAN to reach À0.06 and À0.22 V, respectively, as presented in Table 2.
(4) In the EOR mechanism at Pt and Pd catalysts it is considered that the onset potential (E onset ) is related to the breaking of C-H bonds and the consequent removal of intermediates such as CO ad by oxidation with adsorbed OH À (OH ad À ) provided by Pd-OH and/or Pt-OH sites. 27 The ratio of the current density value of the forward oxidation peak to that of the backward oxidation peak was utilized to determine the tolerance of the prepared catalysts to accumulated carbonaceous intermediates generated during EOR on the electrode surface. 34 A higher tolerance ratio demonstrates greater efficiency of EOR during the forward scan and less accumulation of carbonaceous residues on the electrode surface. 27 Tolerance ratios for the four studied catalysts were computed and are collected in Table 2. It was discovered that the tolerance order was Pt/Pd/p1,8-DAN > Pd/Pt/p1,8-DAN > Pt/p1,8-DAN > Pd/p1,8-DAN. The improved catalytic activity of bimetallic over pure metals is generally ascribed to several effects. It is considered that the alloying component tends to leach out under electrochemical conditions and results in a surface rich in noble metal. 45 Consequently, it produces an additional active surface compared to monometallic alone. The alloy has also changed the geometric ligand (e.g. diminishes the Pt-Pt bond distance) or electronic impact (e.g. increase in Pt delectron vacancies), so one of the components modies the electronic properties of the other to yield a more active catalytic surface. [46][47][48] These results demonstrated that the Pt/Pd/p1,8-DAN catalyst has the highest electrocatalytic activity (the lowest E onset ) and the most elevated tolerance ratio. This could be ascribed to the synergetic impacts of Pt/Pd NPs for the breaking of the C-C bond of EtOH. Additionally, the presence of p1,8-DAN as a highly conducting polymer is an effective support for the catalyst. Therefore, the presence of a support plays an important role in the activity of Pd and Pt NPs towards the EOR, suggesting that Pt/Pd/p1,8-DAN is a good electrocatalyst for EOR in an alkaline medium. 26

Parameters affecting EOR
In order to achieve better electrocatalytic EOR, different parameters like NaOH and EtOH concentrations, scan rate, and upper potential limits were examined.
3.3.1. Effect of NaOH concentration. Fig. 5A-D indicate distinctive CVs for the four catalysts in the presence of 0.5 M EtOH and different NaOH concentrations. The information obtained demonstrated that the current densities of the forward and backward peaks increased, E pf together with E onset values shied to more negative values and the tolerance ratio diminished with increasing NaOH concentration, while E pb had a positive shi with a linear dependence and a correlation coefficient of 0.98 (Fig. 5 inset). It was concluded that increasing OH À ion concentration facilitated EOR due the removal of the adsorbed intermediates and the high coverage of catalysts with OH À . 27 3  This journal is © The Royal Society of Chemistry 2018 using different sweep rates from 0.025 to 0.3 V s À1 (Fig. 7A-D). The current densities of the forward peak (J pf ) increased linearly with rising scan rate (n) (gure not shown), while E pf and E pb shi to a positive potential with a linear relationship between E pf and log(n), as shown in Fig. 7 (inset), 27 which suggests that the electrooxidation of EtOH is an irreversible process.
For each catalyst, the diffusion coefficient (D) was calculated according to the following Randles-Sevcik equation. 50,51 I p ¼ 2.69 Â 10 5 n 3/2 AD 1/2 n 1/2 C where I p is the peak current in ampere, n is the number of electrons transferred in the rate-determining step, A is the electrode surface area, n is the scan rate in V s À1 , C is the solution concentration in mole cm À3 and D is the diffusion coefficient in cm 2 s À1 , which is found by plotting the relation between the square root of the scan rate and the current density. Diffusion coefficients for Pd/p1, 8 In the forward sweep, J pf and E pf remain unaffected, while in the backward scan, values of J pb , E pb , reduction peak current densities (J pr ) and reduction peak potentials (E pr ) were changed. This could be explained by the fact that at lower potential, metal oxides had not been signicantly created and so their impact on EOR in the backward scan was minor. Increasing the positive potential limits accelerates metal oxide formation leading to an increase in J pr . On the other hand, E pb had a continuous positive shi and J pb improved. 27

Chronoamperometry
The impact of poisoning the Pt and Pd NPs surfaces was analyzed using the chronoamperometry (CA) technique. A chronoamperometric study was carried out to better understand the electrocatalytic performance and stability of Pd/ p1,8-DAN, Pt/p1,8-DAN, Pd/Pt/p1,8-DAN and Pt/Pd/p1,8-DAN catalysts towards EOR (gure not shown) where the potential was held at À0.02 V. By applying potential to each electrode, a steady decrease in current was observed within the initial couple of minutes for all catalysts, followed by the establishment of an almost steady current at longer times. The decrease in current with time could be ascribed to the intermediate poisoning species accumulated during the development of oxidation. The catalytic stability was found to follow the order Pt/Pd/p1,8-DAN > Pd/Pt/p1,8-DAN > Pt/p1,8-DAN > Pd/p1,8-DAN. The results suggested that the Pt/Pd/p1,8-DAN catalyst electrode shows higher catalytic activity and stability towards EOR, proving a superior tolerance to the carbonaceous intermediates generated during the oxidation process, as represented in Table 1. 40,52,53

Conclusion
In this work, we have presented a support of a modied p1,8-DAN/GC electrode with Pt, Pd, Pt/Pd and Pd/Pt NPs for EOR. The performances of the resulting catalyst electrodes were considerably enhanced in terms of increasing the catalytic current and depressing the onset potential of EtOH oxidation. DEMS, CV and CA techniques were used to characterize the electrochemical behaviors of the prepared catalysts. Moreover, CA analysis was useful to prove the electrocatalytic performance, catalytic activity and stability of the prepared catalysts toward EOR. Experimental results revealed that the Pt/Pd/p1,8-DAN catalyst demonstrated considerably improved electrocatalytic activity and tolerance to CO poisoning (J Pf /J Pb ¼ 2.6). Additionally, it revealed a superior electrochemically available surface area and faster charge transfer rate at the electrode/ electrolyte interface than the other catalysts, making it a smart anode for the manufacture of an EtOH fuel cell.

Conflicts of interest
There are no conicts to declare.