Synthesis, Characterization and Electrocatalytic study of Pd supported on CeO2 doped on N, S-rGO towards Hydrogen and Oxygen Evolution

The sustainable production of hydrogen and oxygen through electrolysis of water requires the development of an ecient electrocatalyst. Herein, we synthesized Pd nanoparticles dispersed on CeO 2 /N, S-rGO by hydrothermal method followed by chemical reduction of Pd nanoparticles. Electrochemical measurements towards hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) show a high electrocatalytic activity of the catalyst. Among the synthesized electrocatalysts Pd/CeO 2 /N, S-rGO exhibits lower overpotential (75 mV and 240 mV) at 10 mAcm − 2 and lower Tafel slope value (44 mV dec − 1 and 42 mV dec − 1 ) for HER and OER, respectively. The chronoamperometric and linear sweep voltammetry (LSV) of the electrocatalyst shows a negligible decrease in the current density for twelve hours and a minor change in the polarization curve after 10,000 cycles, respectively. The high electrocatalytic activity and superior stability of the synthesized electrocatalyst could be attributed to the synergetic effect between Pd nanoparticles and CeO 2 /N, S-rGO support. This work demonstrates a facile way to develop effective and stable electrocatalysts by exploiting the Pd/Metal oxide interface.


Introduction
The electrochemical water splitting in acidic polymer electrolyte water electrolyzers (PEWEs) has many advantages over the alkaline system. This includes higher kinetics of the cathodic hydrogen evaluation reaction, high electrolyte conductivity, and higher pressure above 50 bar [1].
Hydrogen has been a promising clean energy source for the future [2]. The production of hydrogen through water splitting sustainably by using an effective electrocatalyst is attracting more attention [3,4].
Pt and Pt-based electrocatalyst are mostly used for hydrogen production [5]. However, the low natural abundance of Pt hinders large-scale applications. These limitations forced the research community to nd a non-Pt or less Pt content electrocatalyst [6,7] or Pt alloys with other metals such as Pd [8,9], Ni [10,11], etc. Pd and Pd-based electrocatalysts attract more attention because Pd has a similar property with Pt, relatively more abundant and lower cost compared to Pt. It is also less poisonous during processing [12]. Moreover, Pd has very close adsorption energy and exchange current density to Pt towards HER, as Norskov et al. [13] calculated.
Pd and Pd-based electrocatalyst are mostly used in catalytic reactions related to hydrogen [14,15].
However, Pd and hydrogen's strong binding energy in Pd-H bonding makes the Pd electrocatalyst inactive towards hydrogen evolution reaction. Many strategies have been employed to alter hydrogen binding energy on the Pd surface [16]. This includes Pd's combination with heteroatomic metallic elements that could affect the lattice strain [17] and the dispersion of Pd on high surface area support that could weaken the bond between Pd and Hydrogen [18,19]. This report synthesized Pd/CeO 2 /N, S-rGO, and Pd-Ni/CeO 2 /N, S-rGO by hydrothermal, and wet chemical reduction method followed by characterization of the composite and its application in electrochemical activities towards hydrogen evolution reaction and oxygen evolution reaction. Among the synthesized electrocatalysts, Pd/CeO 2 /N, S-rGO demonstrated the highest electrocatalytic activity than Pd-Ni/CeO 2 /N, S-rGO, and CeO 2 /N, S-rGO.

Reagents
All chemicals used for this analysis were analytical grade and utilized without further puri cation. Graphene oxide was synthesized from graphite powder by the modi ed Hummer method as our previous report [44]. N, S-rGO was synthesized by the hydrothermal method as our previous report [44]. Brie y, G.O. (70 mg) was added to de-ionized water (70 ml), followed by sonication for an hour. Thiourea (1 g) was added to the above mixture as a source of both N and S, followed by sonication for 30 minutes. The mixture was transferred into a 100 ml Te on lined stainless steel autoclave for the hydrothermal reaction at 180 o C for 12 h. Finally, the product was collected by washing with de-ionized water and ethanol, followed by drying at 60 0 C for 12 h.

CeO 2 / N, S-rGO Synthesis
CeO 2 / N, S-rGO was synthesized by following the procedure reported by G.Tengyang et al. [45]. Brie y, Ce(NO 3 ) 3 .6H 2 O (10.1 g) dissolved in 250 ml of de-ionized water and N, S-rGO (4 g) was added to the above solution, followed by sonication of the whole solution for 30 minutes. The solution's pH was adjusted to neutral value using 0.1 M KOH, followed by stirring the solution for 2 h. Finally, the product was collected by ltration and washed with de-ionized water until the pH becomes neutral and dried at 60 o C. The powder was further heated in a tube furnace at 250 o C for two hours.
Pd/CeO 2 / N, S-rGO and Pd-Ni/CeO 2 / N, S-rGO Synthesis Palladium nanoparticle anchored on CeO 2 / N, S-rGO, was synthesized using the method reported by G.Tengyang et al. [45]. Accordingly, CeO 2 / N, S-rGO (500 mg) was dissolved in 100 ml of de-ionized water followed by vigorous stirring (for one hour) and sonication (for 30 minutes). K2PdCl4 (0.17 g) solution in 20 ml of de-ionized water was added to the above mixture under stirring and stirred for an additional one hour. The solution's pH was adjusted by adding 2 ml of 2.5 M KOH solution dropwise, followed by 15 ml of Ethanol (1 ml /min). The whole solution was heated for one hour at 80 o C. Finally, the product was collected by cooling, ltering, and washing until the pH becomes neutral and dried at 60 o C. The Pd-Ni/CeO 2 / N, S-rGO composite was synthesized by co-reduction of the metal salts.
Physicochemical characterization X-ray diffraction (XRD) was used to investigate the crystalline structure of the crystals using Shimadzu powder XRD-600 with Cu kα radiation. Horiba-Jobin Raman spectra (model: -LabRAM HR Evolution) with 633 nm Ar Laser source was used to evaluate the prepared sample's defect structure. HRTEM image was taken with JEOL JEM 2100(200kv) with a LaB6 electron gun (manufactured in Japan). The X-ray Photoelectron Spectroscopy (XPS model PHI 5000 Versa Probe III) and XPSPEAK4.1 software were used to analyze the synthesized sample's surface chemistry. A eld emission scanning electron microscope (FESEM) (FEI, Quanta 200) was used to study the surface morphology.

Electrode fabrication
The glassy carbon electrode was polished with different particle size alumina polishing slurry. The geometric area of the electrode was 0.0706 cm 2 . The Pd loading amount on the glassy carbon was 0.14 mg/cm 2 . The catalyst ink was prepared by dissolving 5 mg of electrocatalyst sample in water/ethanol (1:1) and 10 µl 5% Na on solution (used as a binder). The suspension was ultrasonicated for 30 minutes.
The ink (5 µl) dropped cast on the polished glassy carbon.

Electrochemical measurement
All electrochemical measurements were carried out using a potentiostat (Biologic SP-300 using software EC-Lab V11.10) with a three-electrode con guration. The electrocatalyst modi ed glassy carbon electrode was used as a working electrode, Ag/AgCl (saturated KCl) as a reference electrode. A graphite electrode was used as a counter electrode in N 2 saturated 0.5 M H 2 SO 4 for hydrogen evolution and in 0.1M KOH for oxygen evolution.
The electrocatalytic performance evaluated using linear sweep voltammetry (LSV) measurement due to the ohmic resistance (iR) the current measured may not indicate the intrinsic behavior, so iR correction was done using LSV measurement. Stability is an essential criterion for practical applications. Commonly there are two ways to study the stability of electrocatalyst. One way is measuring the change in current with time (i.e., the I-t curve). In this case, the current was set at a current density of 10 mA cm -2 for twelve hours. The second method was by conducting a cycling experiment for 10,000 cycles at 25 o C in 0.5 M H 2 SO 4 and 0.1 M NaOH, for HER and OER, respectively. The electrochemical impedance spectroscopy (EIS) was used to evaluate the solution's charge transfer resistance, which was conducted at a frequency range of 100 kHz to 0.01 HZ at open circuit potential with perturbation 5 mV. All the potential mentioned in this report converted to RHE by using

Result And Discussion
The XRD pattern was used to analyze the crystalline nature of the synthesized sample. Accordingly, the diffraction peak of the electrocatalyst displayed in Fig. 1(a)  The Raman spectroscopy was used to analyze the defect in the synthesized sample in a non-distractive way. The presence of more defects in the Nano composite structure helps form a more active site for electrochemical reactions.
The two band in the Raman spectrum are D-band, which is related to out of plane vibration of SP2 bonded carbon associated with a structural defect, and G-band due to in-plane vibrations of SP 2 bonded carbon. Moreover, the ID/IG ratio indicates the degree of defect in the synthesized sample. The Raman analysis result of the electrocatalyst displayed in Fig.1   The X-ray Photoelectron Spectroscopy (XPS) measurement was done to study the composite elements' chemical state. The XPS of the best performing electrocatalyst displayed in Fig and Pd 3d 5/2 level of Pd 0 , respectively [46,47]. The observed increase in Pd's binding energy in Pd/CeO 2 /N, S-rGO, could be due to the strong interaction between Pd and CeO 2 , which resulted from electron transfer between Pd and CeO 2 [48]. The Peak at 337.0 eV and 342.2 eV corresponds to palladium (II). The higher intensity of the former peak indicates the higher dispersion of palladium nanoparticles.
The XPS spectrum of Ce3d is displayed in Fig.2 (b). The highest peaks at 905.3 eV and 886.1 eV is associated with Ce (Ce 4+ ). The satellite peaks relatively lower binding energies are related to trivalent Ce (Ce 3+ ). The peak at 284.7 of C 1s shown in Fig.2(c) corresponds to graphitic carbon due to distortion in the sp 3 carbon structure. The peak at 289.4 eV corresponds to the oxygen-containing carbon functional group (O-C=O) [49]. The XPS spectrum of S 2p is displayed in Fig.2 (d) with a peak at 162.3 eV, and 163.03 eV corresponds to S 2p 3/2 and S 2p 5/2. The N 2p XPS spectrum shown in Fig.2(e) shows major peaks around 398.6 eV and 399.0 eV associated with nitrogen bonded to Sp2-hybridized carbon (C-C=N-C) [50,51]. The XPS analysis con rms the successful incorporation of N and S.
The HRTEM and FESEM analysis was conducted to study the morphology and microstructure of the best performing electrocatalyst Pd/CeO 2 /N, S-rGO. The result shows uniform dispersion of Pd N.P. on CeO 2 /N, S-rGO support with fringe calculated lattice spacing (0.21 nm) that corresponds to Pd (111), and a fringe of CeO 2 with lattice spacing (0.32 nm) corresponds to CeO 2 (111) plane. The result suggests the nanocomposite was formed by transferring an electron from Pd NP to CeO 2 by reverse spillover effect [52], which agreed with the XRD and XPS results.

Electrochemical activity measurement
The electrochemical HER activity was investigated in 0.5 M H 2 SO 4 . Fig.5 shows the iR-corrected polarization curve (LSV) of the synthesized electrocatalyst. The CeO 2 /N, S-rGO exhibit very weak activity with an onset potential of 261 mV and overpotential of 197 mV at 10 mA cm -2 . Pd/CeO 2 /N, S-rGO demonstrates the highest HER activity with the onset potential of 45 mV, overpotential of 55 mV at 10 mA cm-2, resulting from the strong interaction between Pd and CeO 2 . Comparatively, Pd-Ni/CeO 2 /N, S-rGO exhibited lower HER activity than Pd/CeO 2 /N, S-rGO with an onset potential of 75 mV, overpotential of 105 mV at 10 mA cm -2 under the same condition. This could be related to the availability of less active sites for an electrocatalytic reaction.
The Tafel slope value derived from the polarization curve indicates the rate of increase in current against overpotential. The increase in current density with a small change in overpotential signi es faster electrocatalyst kinetics and is also used to estimate the hydrogen desorption mechanism from the electrocatalyst's surface. Accordingly, Pd/CeO 2 /N, S-rGO reveals 49 mV dc -1 , Pd-Ni/CeO 2 /N, S-rGO 64 mV dc -1 and CeO 2 /N, S-rGO 101 mV dc -1 . Therefore, the hydrogen evolution follows the Volmer-Heyrovsky mechanism in which the desorption of hydrogen was the rate-determining step.
The EIS was measured in 0.5 M H 2 SO 4 with a frequency range from 100 kHz to 0.01 HZ and displayed in Moreover, the chronoamperometric stability study shows good stability for 12 h, and EIS measurement was conducted in 0.1 M KOH with a frequency range from 100 kHz to 0.01 HZ and amplitude 5 mV. The small semicircle demonstrated by Pd/CeO 2 /N, S-rGO, shows a faster electron transfer rate. PdNiMo lm 85 110 227 [54] Co@Pd/N,S-rGO 94 58 54 [43] Pd16-CoCNTs -120 79 [55] Pd/MoS 2 /CB 78 57 [56] Pd/g-C 3 N 4 105 69 [56] Conclusion In this study, we synthesize Pd/CeO 2 /N, S-rGO, Pd-Ni/CeO 2 /N, S-rGO and CeO 2 /N, S-rGO successfully by hydrothermal, followed by chemical reduction. The Pd/CeO 2 /N, S-rGO, demonstrates a higher electrocatalytic activity than Pd-Ni/CeO 2 /N, S-rGO, and CeO 2 /N, S-rGO. The observed decrease in electrocatalytic activity after incorporating Ni in Pd/CeO2/N, S-rGO, could be due to decreased active cite for electrochemical reactions. The strong interaction between Pd NP and CeO 2 con rmed by HRTEM, XRD, and XPS analysis leads to higher electrocatalytic activity. Furthermore, CeO2 and N's coexistence, S-rGO support was bene cial for anchoring and stabilizing the metal nanoparticles. It is also evident interaction in the nanocomposite metal with the support). This study could help investigate the further improvement of the catalytic activity of non-Pt electrocatalyst through Pd/Metal oxide interface.