Unveiling the Collaborative Strategy and Synergistic Effects of Pd/V2O5‐fAC towards Glycerol Electrooxidation

A series of Pd nanoparticles supported on V2O5 immobilized on functionalized carbon, % Pd (1, 3, and 5) and % V2O5 (10, 20, and 30), were prepared by sodium borohydride‐assisted microwave polyol synthesis for glycerol oxidation reaction (GlyOR) in an alkaline medium. Electrocatalysts loading, temperature, V2O5 immobilization, and their synergistic effect on the electrocatalytic performance are systematically studied. The electrocatalysts′ morphology and electronic properties were investigated using X‐ray diffraction, Fourier transform infrared spectroscopy, scanning electron microscopy, Transmission electron microscopy, and X‐ray photoelectron spectroscopy. A significantly improved GlyOR is observed with increased V2O5 content and Pd percentage. The 5 %Pd/30 %V2O5‐fAC showed the highest mass activity of 2157.3 mA . mgPd−1, a more negative onset potential of 0.62 VRHE, versus the commercial equivalent, and possessed high stability and durability. The increase in electrocatalytic activity is attributed to the effective immobilization of V2O5 on fAC efficient synergism between Pd and V2O5, strong metal support interaction (SMSI), and great exposure of the electroactive sites. The results herein contribute significantly to the understanding of the physicochemical and electrochemical effects of metal oxide immobilization, microwave irradiation, % Pd/% Metal oxide optimization, and SMSI on metal oxide‐carbon hybrid electrocatalysts for GlyOR, opening new avenues for fabricating high‐performance direct alkaline glycerol fuel cells.


Introduction
Fuel cells (FCs) are highly appealing regarding new energy conversion and environmental protection.Of the fuel cell technologies, direct alcohol fuel cells are acknowledged as a technology applicable to portable and mobile devices.Monohydric, e. g., methanol, ethanol, etc., and polyhydric alcohols, e. g., glycerol and ethylene glycol, etc., have high energy densities, high boiling points, low volatilities, and are easy to store.In the green chemistry drive, bio-alcohols are rendered the best alternative fuels.Glycerol is the most promising of the fuels because of good theoretical energy density (5.96 kW h L À 1 ), fuel efficiency (95.1 %), and a high number of extractable electrons (14 e À ).The mechanisms of the electrooxidation of glycerol have been presented in many review papers. [1,2]It has been postulated that the complete glycerol oxidation in alkaline media forming carbonates produces 14 e À as per equation 1 below: [3] CH 2 OH-CHOH-CH 2 OH þ 20 OH À !
While glycerol has appealing properties, direct glycerol fuel cells operating in alkaline media are demerited by complex electrooxidation reactions and the associated intermediates.The intermediates are due to incomplete electro-oxidation to CO 2À 3 (carbonates in alkaline media) [4] , resulting from a strong CÀ C bond that is difficult to cleave.The primary connotation of the intermediates is that they bind strongly and ultimately block the electrocatalyst's active sites, which inherently causes sluggish anodic reaction kinetics.Thus, there has been a race toward attaining an electrocatalyst that enhances and boosts poison tolerance and overall reaction kinetics.
Electrocatalyst engineering is governed mainly by cost, electroactivity, and stability.For cost alleviation, Pt-free catalysts have been employed.Palladium-based electrocatalysts are used as anode materials because they are more active in alkaline electrolytes and can be alloyed with 3d transition metals.Cost reduction is achieved by reducing the loadings of the precious metal aided by better dispersion on a suitable support material, e. g., carbon nanotubes, carbon nanofibers, carbon nanodots, graphene, etc. [5] However, there is a poisoning effect emanating from strongly adsorbed carbonaceous intermediates and a very high potential needed for hydroxyl formation on the Pd-based electrocatalysts surfaces that are necessary for further oxidation of the intermediates.Research has pointed to the introduction of oxophilic metals being capable of introducing more oxygencontaining species for the oxidation of poisoning intermediates. [6]The interaction of the oxophilic metal with Pd facilitates Pd electronic structure tuning, improvement of Pd charge transfer kinetics, synergistic effect, and bi-functional mechanism.Therefore, these oxophilic metals are termed promoters in the electrocatalytic reaction.
There is a generality that the OH À is responsible for the first deprotonation step leading to the formation of alkoxides.Also, the products formed depend on Pd's potential, electronic structure, and morphology.It has been reported that at low potentials, glyceraldehyde and glyceric acid are predominant, whilst at higher potentials, hydroxypyruvic acid and dicarboxylates are selectively produced, and products from CÀ C bond cleavage. [1]Drawing a conclusion, based on literature, electrocatalysts dispersed carbonaceous and a metal oxide (MO) matrix influences the electrocatalytic activity via strong metal support interaction, SMSI.In some cases, the oxide is not involved in the electrocatalytic activity promotion of Pd, the oxide matrix ensures the development of a high surface area electrocatalyst.A variety of oxide matrices have been reported as CeO 2 , [7] ZrO 2 , [8,9] TiO 2 , [10,11] WO 3, [12,13] and CuO, [14,15] to aid the improvement of the electrooxidation of activity and to enhance the poisoning resistance.Therefore, including the MO promotes the Pd d-character, which is essential in forming chemisorption bonds, improving electrocatalytic performances, and promoting the scission of CÀ C bonds. [16,17]anostructured vanadium oxides have drawn intensive exploration for energy conversion owing to their great interactions with molecules or ions, superior electrocatalytic activity, and/or high electron-electron correlations. [18]V 2 O 5 is a strong oxidant mainly used as a cathode in lithium-ion batteries. [19]The V 2 O 5 is a multivalent redox-sensitive transition metal, V 2 + to V 5 + .This transition in redox probes enables the creation of oxygen moieties via a redox reaction.For the VÀ O systems, the V 2 O 5 (3d 0 , V 5 + ) is thermodynamically stable and the most saturated oxide, with an orthorhombic unit cell (space group Pmmn).The vacant d-orbitals of vanadium have the LUMO character acting as Lewis acid sites, whilst the lone pairs of electrons of bridging oxygen atoms possess the HOMO character behaving as Lewis basic sites.V 2 O 5 , an n-type semiconductor with a layered structure, permits the formation of intercalation compounds, making it versatile in scientific and engineering applications.Based on our literature search, there are very few reports of V 2 O 5 in fuel cells.More recently, T. Maiyalagan et al. [20] reported using Pt nanoparticles, NPs supported on V 2 O 5 -C to support high methanol electrooxidation and stability in alkaline and acidic medium.In Pt/V 2 O 5 -C study, the improved activity was reported to be from synergistic effects of Pt and V 2 O 5 , leaving out an explanation about the contribution of electronic effect, active site exposure, and efficient active site synergism as there was a solid-state mixture of V 2 O 5 and the carbon support.In another study, Li et al. [21] reported the synthesis of Pd-V 2 O 5 /C and the electrooxidation of methanol in alkaline electrolytes.The Pd-V 2 O 5 /C was reported to have high electrocatalytic performances and anti-poisoning ability.In this study, efficient synergism was said to be from the "high valence state V and Pd NPs", but the explanations and origins of the electrocatalytic activity, the relative contributions to high electrocatalytic activity, stability, and high poisoning tolerance resulting from the interaction between carbon, V 2 O 5 , and Pd are still at infantry stages.In our study, V 2 O 5 was used because of its (i) Oxidationreduction properties (has reversible redox probes that enhance effective oxidative reactions), (ii) tuneable electronic structure for improved electron transfer and efficient charge transfer between electrode-electrolyte interface, (iii) efficient synergy with Pd promoting SMSI, hence improved electronic coupling interaction for ultimate excellent glycerol electrooxidation and (iv) V 2 O 5 is a common oxide hence industrial manufacturing systems are already there hence setup and manufacturing cost will be reduced.
The synthesis methods of electrocatalysts are fundamental in controlling structural and surface morphology, particle size, and dispersion, which are all key to electrocatalytic performances.The microwave-assisted synthesis methods are still in the infantry, although they have been reported to enhance electrocatalytic activity despite oblique around the causes of such performance.The selection of the microwave technique, as reported by Manthiram et al. [22] is a simple, fast, and clean synthesis method.The microwave method also enhances the electrocatalytic activity towards the electroreduction of oxygen and alcohol electrooxidation.
This work utilized the NaBH 4 -ethylene glycol facile microwave-induced reduction method to synthesize Pd/V 2 O 5 -fAC.The AC was made from seaweed for value addition to this menace weed, thereby protecting the aquatic life.To the best of our knowledge, no research group has used (1) NaBH 4ethylene glycol facile microwave-induced reduction method for the synthesis of Pd/V 2 O 5 -fAC, (2) used V 2 O 5 -fAC as a double support system, (3) simultaneous dual optimization of Pd % and V 2 O 5 % (% Pd less than the commercial 10 % to minimize cost) (4) use of Pd/V 2 O 5 -fAC for electrooxidation of glycerol as a fuel for fuel cell application.We used a double support system constituting V 2 O 5 -fAC as the supporting platform for Pd nanoparticles.The V 2 O 5 -fAC platform provided room for a strong metal support interaction as induced by microwave irradiation.We proved that the functional groups and redox probes are significant in the physicochemical properties of the Pd/V 2 O 5 -fAC that instigate electrocatalytic oxidation of glycerol.The subjection of Pd/V 2 O 5 -fAC into a microwave environment tunes the physicochemical properties by enhancing the crystallinity, surface area, and the reaction of oxygen vacancies.The property enhancement collectively improves the electrocatalytic activity toward glycerol electrooxidation.

Results and discussions
The synthesis of V 2 O 5 nanocrystals is proposed as follows: The NH 4 VO 3 has a polymeric anion structure, which reduces its solubility.Thus, the NH 4 VO 3 must be warmed, where it undergoes dissociation to form VO 2þ (Eqs. 2 and 3).
The VO 2þ binds to EDTA to form a vanadium (II) ethylenediaminetetraacetate ( VO ½ ðEDTAÞ� 2À ) complex (Eq.4): This vanadium (II) ethylenediaminetetraacetate complex will decompose on calcination at 600 °C giving vanadium pentoxide and other by-products, as shown in Equation 5 below: The double support system ensured the embedding and immobilization of V 2 O 5 on Fac (Eq.6): X represents 10, 20, and 30 %. Reduction of Pd 2 + to Pd 0 , with PdCl 2 as the precursor (Eq.7):

Physicochemical characterization
The phase and crystal structures of carbon (C), vanadium oxide (V 2 O 5 ), and Pd/V 2 O 5 -C arrays were determined by XRD analysis, Figure 1 (a).The appearance of (001) and (100) crystal planes at 2q values of 20°and 21°confirms the formation of α-V 2 O 5 crystals on V 2 O 5 and Pd/V 2 O 5 -C arrays alluding to the orthorhombic crystal system. [25,26]The diffraction peak located at a 2 q value of 26°alludes to the (002) carbon plane and the Pd/ V 2 O 5 -C array.Other five diffraction peaks located at 2q values of 40°, 47°, 68°, 82°, and 86°are characteristic features of the face center cubic (FCC) crystalline Pd and correspond to the (111), ( 200), ( 220), (311) and ( 222) miller planes respectively on the Pd/V 2 O 5 -C series. [27]In the case of the 1 %Pd/V 2 O 5 -C series, the diffraction peaks corresponding to the Pd nanoparticles have reduced intensity due to small size and small percentage loading.Apart from the reduced intensity of the Pd nanoparticles, no impurities were detected in all the Pd/V 2 O 5 -C arrays.The sharp diffraction peaks for all the miller planes allude to the nanomaterials' highly crystallized nature.These results indicate that the Pd nanoparticles deposited on the V 2 O 5 -C hybrid support system have single-face structures.The calculated interplanar spacing, d, according to Bragg's law (Equation 8) for (002), was approximated to be between 3.36 to 3.40 A ∘ for the Pd/V 2 O 5 -C series, which was slightly higher than that of the pristine carbon (3.35A

∘
).That feature indicated the presence of COOH, C=O, and OH functional groups at the edge and in between the activated carbon layers to a higher degree on the Pd/V 2 O 5 -C array, facilitating the anchoring of the palladium and metal oxide on the carbon sheet.The average crystal sizes calculated were according to the Scherrer equation (Equation 9), and the lattice parameters (calculated from Equation 10) for all nanohybrid materials alluded to high crystallinity for all the composites.One thing to note is that the average crystal size increases at a high percentage of palladium on the Pd/V 2 O 5 -C array due to a plausible synergy between vanadium oxide and palladium which might be linked to high conductivity.The peaks of 3 %Pd/30 %V 2 O 5 -C are broader, meaning that this particular sample is amorphous.The Pd peaks are slightly wider, indicating much smaller crystal size with the 3 % Pd at 10, 20, and 30 % V 2 O 5 , [28] the same phenomenon is seen for 5 %Pd samples.As seen in Table 1, the Interplanar spacing space for 3 %Pd/30 %V2O5-C is 2.75 A ∘ bigger than others, this increase is probably due to Pd intercalation into the V 2 O 5 -C lattice.The Pd intercalation can induce surface restructuring, leading to rearrangement of atoms within the hybrid.The bonding information and corresponding functional groups on activated carbon (C), vanadium oxide (V 2 O 5 ), and Pd/ V 2 O 5 -C array were determined by FTIR analysis in the range of 450-4000 cm À 1 , Figure 1 (b).In the pristine V 2 O 5 , the 451-598 cm À 1 peaks correspond to the symmetric and asymmetric stretching vibrations of the triply coordinated oxygen bonds to the metal.The V=O terminal oxygen and bridge oxygen stretching vibration on the pristine V 2 O 5 is located around 829 cm À 1 and 1006 cm À 1 , respectively.The vibration peak around 1006 cm À 1 splits into two components in the case of α-V 2 O 5 . [29]This g-V 2 O 5 splitting is absent on the pristine V 2 O 5 and all the Pd/V 2 O 5 -C array, which alludes to the formation of the orthorhombic α-V 2 O 5 . [30]For the Pd/V 2 O 5 -C arrays, there is a shifting of the peaks located at 598, and 1006 cm À 1 to 609, and 1014 cm À 1 , which is ascribed to the incorporation of the Pd nanoparticles that facilitate the stretching vibrations of the α-V 2 O 5 on those nanocomposites.Some prominent peaks correspond to various functional groups for the functionalized activated carbon on the Pd/V 2 O 5 -C array.The peak around 1015 cm À 1 is a characteristic feature of the = CÀ O stretching.Peaks around 1410 and 1564 cm À 1 were attributed to the stretching vibration of the CÀ C bonds attached to the À C=C, which indicates the graphitic nature of the activated carbon (C) structure.The 1640 and 1720 cm À 1 peaks correspond to the stretching vibration of the sp 2 hybridized carbon (C=C) and the C=O of the carboxylic group, respectively.Finally, the peak around 3448 cm À 1 was ascribed to the stretching vibration of the OÀ H group of the phenolic or alcoholic group on the activated carbon support. [31]The FTIR also showed absorption peaks around 2297 cm À 1 for the 5 %Pd/V 2 O 5 -C series due to CO 2 sorption at the surface of those nanocomposites. [32]aman spectrum was introduced to further confirm the phase of V 2 O 5 , functionalized activated carbon (fAC), and the Pd/V 2 O 5 -C array, Figure S1(a-c) and Figure 1(c).From the spectra, intense vibrational peaks at 143 and 195 cm À 1 corresponded to the skeleton bending vibration (B 3 g) and the stretching vibration of the OÀ VÀ O bonds (Ag) on V 2 O 5 and the Pd/V 2 O 5 -C series respectively. [33]The peaks at 283 and 404 cm À 1 were assigned to the V=O bonds of the metal oxide on V 2 O 5 and the Pd/V 2 O 5 -C array.The peak located at 481 cm À 1 on V 2 O 5 and the Pd/V 2 O 5 -C series corresponds to bending vibrations of the double coordinated oxygen bonds (OÀ VÀ O).In contrast, the peak located at 527 cm À 1 for V 2 O 5 , 1 %Pd/30 %V 2 O 5 -C, 3 %Pd/ 30 %V 2 O 5 -C, and 5 %Pd/30 %V 2 O 5 -C is a characteristic feature of the triply coordinated oxygen bonds (V 3 À O).The intense peak around 994 cm À 1 on V 2 O 5 was assigned to the stretching vibration of the shortest vanadium oxygen bond and was not visible on the Pd/V 2 O 5 -C series due to the carbon support. [34]For fAC and the Pd/V 2 O 5 -C array, there are two characteristic peaks located at 1357 and 1599 cm À 1 , corresponding to the lattice defects (D band) and the stretching vibration of the carbon support's C=C bond (G band).These results are consistent with the XRD and FTIR data which alludes to the existence of palladium decorated on the α-V 2 O 5 and fAC as a hybrid support system.The intensity ratio (I D /I G ) (Table 2), which is used to 3 %Pd/ 30 %V 2 O 5 -C has a more sequential graphene-like structure due to the high I D /I G ratio.The increase in disorder is beneficial in electrocatalysis as it exposes electrocatalytic active sites and promotes edge catalysis.The absence of second-order peaks in fAC qualifies it to be indeed AC.
Morphological features of the as-fabricated electrocatalysts were analyzed through SEM, Figure S2 and 2. As seen from images in Figure S2, the fAC shows intercalated porous carbon sheets.The fAC has a rough, coarse, and irregular surface (from the acid functinalization) covered with very small agglomerated carbon particles. [35,36]V 2 O 5 exhibits irregular micro rods; clearly, the rod arrays form the rod-like morphology exhibiting well-defined boundaries.These defined edge boundaries are important in anchoring Pd.In Figure 2(a-i), there is a large amount of neighboring, interconnected, and immobilized V 2 O 5 on the fAC that constructs numerous pathways for rapid electrolyte diffusion, which is necessary for exposure of active sites, strong metal support interaction (SMSI), and excellent conductivity.The fAC acted as a physical scaffold where the V 2 O 5 subunits were tightly linked, embedded, and immobilized.Interestingly, the synthesized electrocatalysts still exhibit a porous nature; this confirms the suitability and versatility of the microwave irradiation synthesis method.As from the comparison of 10 % V 2 O 5 (1 %!3 %) Figure 2  (a, d, g), 20 % V 2 O 5 (1 %!3 %) Figure 2 (b, e, h), and 30 % V 2 O 5 (1 %!3 %) Figure 2 (c, f, i) we see that there is an increased and strong contact between the V 2 O 5 and C. The contact increased with an increase in Pd content for every % V 2 O 5 .Since the C is constant throughout and V 2 O 5 is fixed at a given % with varying Pd, this trend can be explained bringing to light that Pd content helps in making interconnected electrocatalyst system.The interconnection helps reduce the interparticle distance, which is favourable in electrocatalysis on the electrode-electrolyte interface.As can be seen in Figure 2 (g, h, i) when % Pd is 5 %, the are interconnected particles with clear diffusion pathways and some pores, meaning that all the electrolyte can reach exposed electrocatalytic sites.
Figures S3 and 3(a-i) show TEM images of the synthesized electrocatalysts.Figure S3 (a) shows the fAC amorphous sheets, and (b) the V 2 O 5 shows the interconnected rods.A closer look at the rods shows they are connected side by side, clearly showing the connection boundaries.Such a connection enhances electron movement.In Figure 3(a-i), the catalysts clearly show a homogeneous distribution of Pd nano-pods on the surface of the double support system.It can be seen that ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi     [37,38] The shift in the C 1s can be due to adventitious carbon, conformational changes (as also visible in XRD), and charge transfer as V 2 O 5 can induce the carbon surface affecting the electron density.Pd 3d core level spectra are depicted in Figure 4c(i, ii, iii) and the relative percentage area of the spectral weight of this core level in Tables 3-5 [39,40] As shown in Figure 4, the electronic interaction induced by the high valence state of V results in the profiles of the Pd/V 2 O 5 -C array being different from one another and having varying chemical shifts.Higher palladium oxide formation also results from V's high valence state, which promotes palladium oxidation.The shift in the Pd core level BE can also be due to the electronic coupling interaction with different V oxidation states, as can be seen in Table S4 the 20 % sample has the lowest V 5 + of 2 %.The core binding energy levels for V 2p of the 5 %Pd10 % V 2 O 5 -C, 5 %Pd20 %V 2 O 5 -C, and 5 %Pd30 %V 2 O 5 -C samples are shown in Figure S4.The V 2p 3/2 spectra of the 5 %PdV 2 O 5 -C array can be deconvoluted into two peaks at binding energies between 514.73-518.51eV and 517.05-520.83eV correspond-ing to the V 4 + and V 5 + states in the vanadium oxide respectively.As described in Table S4, the V species on the carbon support mainly exist as V 4 + due to the reduction of V 5 + by carbon species and NH 3 through the decomposition of the precursor used.The highly dispersed VO x species mainly exist as  mixed V 4 + and V 5 + on the activated carbon hybrid nanomaterial.A strong electronic effect is indicated by the redox between palladium and the high valence state V. Increased À OH adsorption ability and other oxygen-containing species are also promoted by the presence of palladium oxide and high valence state V. [41,42] O 1s core level spectra depicted in Figure 4d(i, ii, iii) and the relative % area of the spectral weight of this core level in Tables S5 can be clearly observed. [43,44]This is consistent with the C 1s and V 2p3/2 spectra results.We postulate that the shift of the O 1s for the 20 % sample Figure 4 (c)(ii) can be due to the higher % of OH relative to samples which might form or interfere with PdÀ V or PdÀ OÀ V in small quantities or as a result of oxygen loss/reduction of V 2 O 5 .The O 1s can also be due to the hybrid easily reacting with oxygen species.

Cyclic voltammetry in alkaline medium
Electrochemical experiments in oxygen-free solutions utilizing very pure electrolytes and clean electrochemical cells are widely established as effective surface characterization methods.Fig- ure 5(a) shows the electrocatalysts' cyclic voltammograms (CVs) in an N 2 0.5 M NaOH solution at 20 mVs À 1 .The CVs show the signature characteristic peaks for Pd containing electrocatalysts in alkaline: (i) Hydrogen adsorption/desorption redox processes around 0.01 to 0.55 (V Vs.RHE), as represented (Eq.11): (ii) at 0.72 (V Vs.RHE) PdO formation occurs, partially overlapping with H des .This process is initiated by the chemisorption of OH À as shown below (Eqs.12-14): (iii) and reduction of palladium oxide (PdO) to Pd 0 around 0.65 to 0.75 (V Vs.RHE) window.The reduction from PdO to Pd (equation 15) is represented as follows: There was a negative shift in potential for the PdO peak, indicating some alteration of the Pd electronic structure with the introduction of V 2 O 5 .The magnitude of the potential shift increased with an increase in the percentage of Pd and V 2 O 5 .The MO shift is very complex because the PdO reduction peak shift can also indicate that the electronic structure and another probability being an indication that the amount of PdO on the electrode was changed.It is important to note that there are however some broad peaks around 0.6 and 0.84 V Vs.RHE are attributed to the oxidation and reduction of V [45,46] which are different from the PdO formation around 0.72-0.8(VVs.RHE).These V redox probes also explain the PdO peak shift.
The electrochemically active surface area (ECSA) (Eq.16) for the electrocatalysts was determined from the PdO reduction peak based on literature [47,48] as depicted in Figure 5 (a): where Q denotes the integrated Coulombic charge for the reduction of PdO (in mC), S is a proportionality constant (0.424 mC cm À 2 ), and I is the loading in μg.It is important to note that this procedure considers the formation of a PdO monolayer when the potential is swept until E max .However, the influence of V 2 O 5 on PdO formation is assumed.To date there is no universally established protocol to estimate electrochemical area from PdO.
The ECSA follows the trend: 1 % to 5 % Pd with an increase in V 2 O 5 content (10 %!30 %) in (Table 1).The high ECSA value entails the introduction of the V 2 O 5 promoter and improved conductivity from fAC. Figure 5(b) shows cyclic voltammetry curves for the electrooxidation of glycerol on 1 %Pd/30 %V 2 O 5 -fAC, 3 %Pd/30 %V 2 O 5 -fAC, and 5 %Pd/30 %V 2 O 5 -fAC.The electrocatalytic performance in the CV curve can be evaluated based on the following parameters: (1) Onset potential (E onset ) and ( 2) forward peak current (I f ).The forward oxidation peak signifies freshly chemisorbed glycerol species, and the backward peak in the negative scan resembles further oxidation of the glycerol molecules and its intermediates on the recovered active surface. [45,49]Linking our research to the research by Zhao et al. [50] the most probable explanation could be that glycerol adsorption can occur at low potentials and its intermediates be oxidized on the forward peak, freshly adsorbed glycerol and its intermediates can also be oxidized during negative going scan.There is an increase in the forward peak current density with an increase in Pd content from 1 % to 5 %, corresponding to the increase in V 2 O 5 Figure S5 (a), (b), and (c).As the % of Pd increased, the electrocatalytic active site for glycerol oxidation increased, corroborated by the increased surface for anchoring on V 2 O 5 .Thus, for the realized performance, we assume there was an enhanced strong support metal interaction, synergy, and exposure of active site resulting from the immobilization of V 2 O 5 on the activated carbon.Similar findings were reported by Li et al. [51] where they attributed the high electrocatalytic activity to the synergistic effect between high valence state V and Pd NPs.The 5 %Pd/30 %V 2 O 5 -fAC showed the highest forward peak current and the most negative E onset = 0.54 (V Vs.RHE) (80 mV negative shift from Pd/C) followed by 3 %Pd/30 % V 2 O 5 -fAC, 1 %Pd/30 %V 2 O 5 -fAC and Pd/C with E onset = 0.56, 0.59 and 0.62 (V Vs.RHE) respectively.The negative shift indicates favorable glycerol electrooxidation kinetics on 5 %Pd/30 %V 2 O 5 -fAC.Generally, the E onset for Gly-OR decreased with an increase in the oxide content, as was observed by Liu et al. [52] with an increase in NiCo 2 O 4 .The comparison of mass activity, MA, and  Pd and SA = 149.8mA cm À 2 which are 1.02, 1.2 (MA) times and 1.7, 5.7 (SA) times more than 1 % and 3 % Pd electrocatalysts, respectively.It is important to note that there was not much difference in electrocatalytic activity from 3 %Pd/30 %V 2 O 5 -fAC to 5 %Pd/30 %V 2 O 5 -fAC, a similar trend was reported by Li et al., [21] thus high % of V 2 O 5 is not beneficial to overall activity.Based on the LSV profiles in Overly, the observed performances explained above are shown in Figure 5 (e).In this figure, the brown arrow shows the increase in activity based on the optimization of Pd and V 2 O 5 , also ECSA relationship is shown.Thus, the increase in the Pd and V 2 O 5 resulted in a corresponding increase in electrocatalytic activity and ECSA.The realized trend is due to the collaborative effect of fAC and V 2 O 5 towards improved Pd GlyOR electro- catalytic activity, SMSI between the immobilized V 2 O 5 and Pd, and the resulting synergy thereof with alters the Pd electronic structure.Figure 5(e) also show that at 3 and 5 % the hybrid support has little contribution to Pd electroactivity properties.
The observed faster electrocatalytic kinetics were further verified by assessing the solution resistance (R s ) and charge transfer resistance, (R ct ) respectively, as derived from the electrochemical impedance spectrum data, Figure 6 (a).With Nyquist plot fitting to probe the electrocatalyst-electrolyte interfacial properties, the R ct increased as follows: 5 %Pd/30 % V 2 O 5 -fAC (Ω) < 3 %Pd/30 %V 2 O 5 -fAC (Ω) < 1 %Pd/30 %V 2 O 5 -fAC (Ω).Thus, the 5 %Pd/30 %V 2 O 5 -fAC had the smallest R ct hence the greater charge transfer ability.The same trend is shown in Figure S10 (a-c), and Table 6.There was a general decrease in R ct with an increase in % Pd (1 %!3 %) and %V 2 O 5 (10 %!30 %).The high R ct at low a lower % are due to reduced Pd-V 2 O 5 synergism and low promotional effect, as seen with lower electrocatalytic activity in Figure S5 (a).The high R ct compromised activity at low % results from insufficient Pd-promoted active sites for Gly-OR.
The robustness of an electrocatalyst is fundamental for the long-term electrocatalytic performance towards Gly-OR.Thus, long-term stability and durability were acquired using potentiostatic chronoamperometry, CA, and cyclic voltammetry cycles, CV, respectively.As shown in Figure S11 (a-c), within the first 1000 s, there was a sharp current decay, probably from chemisorbed species, after which the CA attained a pseudosteady state.Figure 6 (b) shows that 5 %Pd/30 %V 2 O 5 -fAC exhibited excellent stability, retaining 42.90 % relative to the commercial, retaining 3.14 %.The general current retention was as followed the order (1 %!5 %) with (10 %!30 %) increase for (Pd % and % V 2 O 5 ).Of greater importance, The long term poisoning rate from these adsorbed species, s, was calculated from the linear decay current density portion from CA using the following relationship (Eq.17): Where dI dt � � t>5000s is the slope of the linear portion of CA, t is time, and I 0 is the back extrapolated current density at the point of linear CA portion.The calculated poisoning rates are presented in Table 6, showing a decrease in the poisoning rate with increase in %Pd(1 %!5 %) with % V 2 O 5 (10 %!30 %).This is explained by an increase in SMSI, and effective immobilization of the V 2 O 5 of AC, which aids in the reduction of the overall electrocatalyst corrosion.The s corresponds with the CV I f values and the CV cyclic studies.Figure S12 shows TEM images after the 10000 s chronoamperometry stability test; there are not many morphological changes for the Pd NPs, but they still maintain the pod-like morphology.The NPs showed little agglomeration after the CA test, showing greater electrocatalyst stability.Interestingly to note, is the steady rise of the CA curve for the most stable 5 %Pd/30 %V 2 O 5 -fAC.This slight rise is due to the removal of passive species and the reactivation of the electrocatalytic active sites, that is self-regeneration of the active sites.The self-regeneration means the availability of more active sites, hence increased migration of the electrolyte species to the available sites (creation of a concentration gradient favouring the electrocatalyst surface).To confirm the above CA discussed findings, CV cycle studies compared the electrocatalyst deactivation rate and durability Figure S13 (a-c).In a study by Hameed et al. [53] explained the improved activity to be emanating from the MO ability to generate OH ads that enhances CO oxidation enhancing the overall stability, we perceive that the generated OH ads plays a fundamental role in Pd active site protection.The current density, I, was normalized by the initial current (I 0 ). [54]The ( I I 0 ) ratios are shown in Table 6, the 5 %Pd/30 %V 2 O 5 -fAC a ( I I 0 ) of 0.96, which is high than that of Pd/C (0.64) and relatively all the electrocatalysts signifying superior cyclic stability.The ( I I 0 ) is corroborated with the current loss.As can be seen from Figure 3 (c), the rate of current loss followed the order 5 %Pd/30 %V 2 O 5 -fAC < 3 %Pd/30 %V 2 O 5 -fAC < 1 %Pd/30 %V 2 O 5 -fAC.Thus, the electrocatalysts gained high durability as the % of V 2 O 5 increased.The realized superior cyclic stability, low rate of deactivation, and high durability are based on (i) immobilization of V 2 O 5 on fAC enables strong metal-support interaction, SMSI, (greater synergy), enhancing Pd NPs anchoring, which offers structural stability and prevents agglomeration during cycling, (ii) utilizing fAC matrix as a substrate improves the electrocatalyst's stability and conductivity further while preventing the aggregation of Pd NPs, and (iii) resistance to corrosion at high % V 2 O 5 .

Conclusions
This work studied the behaviour of a Pd NPs supported on a hybrid support system constituting of V 2 O 5 immobilised on functionalised activated carbon for the electrocatalytic oxidation glycerol in alkaline media.We demonstrated that metal oxide immobilisation plays a vital role in physicochemical and electrochemical properties of the optimised Pd/V 2 O 5 -fAC.offering highly dispersed pod like Pd structures over the entire hybrid support surface and the internal network of both supports.The lower onset potential, electrocatalytic activity, poison tolerance, durability, and stability followed the order: Pd increase (1 % to 5 %) with V 2 O 5 increase (10 %!30 %).The high electrochemical performance is attributed to exposure of the electrocatalytically active sites, high electrochemical surface area, intimate Pd-V 2 O 5 , strong metal support interaction, and efficient synergy (V induced electronic effect and ligand effects resulting in altered Pd d-band).We realized that as V 2 O 5 approached 20 % at a 5 % Pd loading the was not much difference with the 30 % at the same loading.Whilst, at low V 2 O 5 there is less synergy and SMSI hence the reported low electrocatalytic activity.The research has paved the way for future optimisation of these Pd-based electrocatalysts, encompassing nanocarbon platforms, metal sulphides, phosphides, selenides etc., other PGMs e. g., Pt, for enhanced selective electrooxidation alcohols in DAFCs.

Experimental Section Chemicals
Ammonium metavanadate (NH 4 VO 3 ) and ethylene diamine tetra acetic acid (EDTA) were purchased from Merck and used as received.Seaweed (macroalgae-Chlorophyta) from Harare River.Double distilled water, DH 2 O from our in-house purification system.Sodium borohydride, sodium hydroxide pellets (NaOH), Sodium tetrahydridoborate NaBH 4 , Ethylene glycol, Nafion solution, palladium chloride, PdCl 2 , commercial Pd/C electrocatalyst, glycerol, and Isopropanol these materials were bought from Sigma-Aldrich and without further purification without purification.

Synthesis and Chemical Activation of Activated Carbon
The AC was prepared by a step process: pre-carbonization and chemical activation.Seaweed (SW), collected from the Harare River in Zimbabwe, was washed with double distilled water to remove impurities.Firstly, the SW was dried in an oven at 100 °C for 36 h to drive out moisture content.The dried SW was heated at 450 °C at a ramp rate of 2 °C min À 1 and let to cool naturally.The sample was labelled pre-carbonized seaweed PC-SW.Secondly, chemical activation was done by mixing 400 g of PC-SW with 200 mL of 6 M NaOH, a ratio of 2 : 1.The PC-SW and NaOH mixture was homogenously stirred at 90 °C for 6 h.The resulting slurry was subjected to microwave for 6 min and then centrifuged at 8.000 rpm to remove excess NaOH.The slurry was dried at 100 °C for 36 h.Afterward, the sample was heated at 900 °C at a ramp rate of 2 °C min À 1 under an Ar atmosphere.The sample was allowed to cool to room temperature and was washed with double distilled water until a neutral pH was obtained.The washed activated carbon was finally dried at 100 °C for 24 h and stored in a desiccator. [23]

Functionalization of Activated Carbon
The activated carbon was functionalized using wet -air chemistry method.Functionalization was performed by treating 5 g of activated carbon with 100 mL of 6 M HNO 3 with O 2 bubbling and 400 rpm at 90 °C for 8 h.The sample, now termed functionalized activated carbon, fAC was washed with copious amounts of water until a neutral pH was reached.The fAC was then finally dried at 100 °C.

Synthesis of vanadium pentoxide, V 2 O 5
Briefly, 1 g of the weak alkaline ammonium salt of NH 4 VO 3 and 1 mmol of ethylene diamine tetra acetic acid, EDTA were mixed in a borosilicate beaker containing 30 mL double distilled water under stirring for 30 min at 50 °C until a clear solution was obtained.The borosilicate beaker containing the above mixture was subjected to domestic microwave irradiation (50 Hz at 1500 W) for 6 min continuously until the solution was evaporated until dry, obtaining a powder.The powder obtained from microwave irradiation was calcined in a tube furnace at 600 °C in the air for 3 h to get pure V 2 O 5 nanoparticles. [24]nthesis of fAC-V 2 O 5 ratios The fAC-V 2 O 5 composite of different ratios, for example, fAC-10 % V 2 O 5 , was prepared by mixing different amounts of fAC and V 2 O 5 .Typically, 5 g of fAC was dispersed in 20 mL of ethylene glycol using sonication, solution A. The V 2 O 5 was dispersed in distilled water, DH 2 O : H 2 O 2 (2 : 1), solution B. Solution B was added dropwise to solution A under stirring, forming solution C. Solution C was subjected to microwave irradiation for 10 min continuously (left in solution phase).All other ratios were prepared following the same route.

Synthesis of Pd/fAC-V 2 O 5
Pd nanoparticles supported on fAC-V 2 O 5 nanocomposite were prepared through the NaBH 4 -ethylene glycol facile microwaveinduced reduction palladium.For the preparation of 5 % Pd loaded fAC-V 2 O 5 , 1.000 mg of fAC-10 % V 2 O 5 was added to 40 mL of ethylene glycol under stirring, solution D. 82,1 mg of PdCl 2 , was dissolved in 40 mL of DH 2 O, solution E. Solution E was added dropwise to solution D under stirring, simultaneously with NaBH 4 , forming solution F. The pH of solution F was controlled by adding 2 M NaOH solution.The pH-controlled solution F was subjected to microwave irradiation 4 times for 4 mins with 60 s pauses.Solution F was the 10 times using a mixture of ethanol/DH 2 O (1 : 2) and thrice with DH 2 O. Finally, the sample was dried in an oven at 100 °C for 24 h, labeled 1 %Pd/fAC-10 % V 2 O 5 .The other catalysts were synthesized following the same route.

Physicochemical characterization
The crystalline nature of samples was conducted at room temperature on an X'Pert PRO PANalytical diffractometer (CuKα, λ = 1.5406Å), Powder X-ray diffraction (pXRD) over the 2θ range of 10-90°.FTIR spectra were obtained from a Perkin Elmer FTIR spectrophotometer.Raman was conducted on a WITec Alpha300R Confocal Raman spectrometer equipped with grating used (600 grooves per mm), laser wavelength (532 nm), using a 50X objective and a 100 μm fibre which also acted as the entrance slit to the spectrometer.The microstructure images were obtained by scanning transmission electron microscopy (STEM, JEOL JEM-ARM200F, 200 kV).Pd Concentrations were measured using Inductively coupled plasma-optical emission spectrometry (ICP-OES) instrument (Perkin Elmer, Optima 7000, USA).X-ray photoelectron spectroscopy (XPS) measurements were acquired at room temperature in an ultra-high vacuum (UHV) chamber equipped
), where (h, k, l are the Miller indices of the crystallographic planes).
1.59  the V 2 O 5 shows the interconnected rods are embedded and immobilized in fAC, as seen by the dark portions on the TEM images.The introduction of V 2 O 5 immobilized on fAC offered anchoring sites Pd and enhanced and controlled nucleation of Pd NPs.It is important to note that there is an interesting trend between the Pd NPs sizes and the % V 2 O 5 .For 1!5 % Pd at 10 % V 2 O 5 , 20 % V 2 O 5 , and 30 % V 2 O 5 are approximately ~6 nm, ~7 nm, and ~8 nm.The notable trend shows a general increase of ~1 nm in Pd NPs for every 10 % increase in V 2 O 5 .The surface composition and chemical states of the species in the PdV 2 O 5 -C series were determined using X-ray photoelectron spectroscopy (XPS).
specific activity, SA, are shown in Figure 5 (c), 5 %Pd/30 %V 2 O 5 -fAC had MA = 2157.3mA mg À 1 Figure S(6-9) (a-c) and their linear fittings in Figure S(6-8) (d), the E a of Gly-OR at the synthesized electrocatalysts, based on the Arrhenius equation showed to decrease with increasing Pd % and V 2 O 5 are content, Table 6.To be specific, as per Figure 5(d), the E a 1 %Pd/30 %V 2 O 5 -fAC, 3 %Pd/30 %V 2 O 5 -fAC, and 5 %Pd/30 %V 2 O 5 -fAC were determined to be 31.86,22.63, and 19.53 kJmol À 1 , respectively, suggesting that the charge transfer process of 5 %Pd/30 %V 2 O 5 -fAC is faster.It is worthwhile to note that Pd/C has a lower E a than 1 %Pd/30 %V 2 O 5 -fAC and 3 %Pd/30 %V 2 O 5 -fAC with lower activity due to less exposed electrocatalytic sides and Pd not promoted by any MO.

Figure 6 .
Figure 6.(a) Nyquist plots, (b) Chronoamperometric curves measured at 0.59 (V vs. RHE), (c) The durability tests performed by CVs continuous cyclization in N 2 -saturated 0.1 M NaOH + 0.5 M Glycerol solution for 500 cycles, and (d) Anodic peak current density (I p ) vs. square root of scan rates (u

Table 1 .
Structural characteristics of all catalysts obtained from X-ray diffraction analysis.

Table 2 .
Raman data showing peak positions, intensities, and intensity ratios for the Pd/V 2 O 5 -C series.

Table 6 .
Summary of electrochemical parameters extracted from different electrochemical techniques.