Calcium-, magnesium, and yttrium-doped lithium nickel phosphate nanomaterials as high-performance catalysts for electrochemical water oxidation reaction

: Electrochemical water oxidation reaction (WOR) lies among the most forthcoming approaches toward eco-conscious manufacturing of green hydrogen owing to its environmental favors and high energy density values. Its vast commoditization is restricted by high-e ﬃ ciency and inexpensive catalysts that are extensively under constant research. Herein, calcium, magnesium, and yttrium doped lithium nickel phosphate olivines (LiNi 1 − x M x PO, LNMP; x = 0.1 – 0.9; M = Ca 2 + , Mg 2 + , and Y 3 + ) were synthesized via non-aqueous sol-gel method and explored for catalytic WOR. Lithium nickel phosphates (LNP) and compositions were characterized via Fourier transform infrared, scanning electron microscopy, X-ray di ﬀ raction, and energy dispersive X-ray di ﬀ raction techniques for the structural and morphological analyses. Glassy carbon electrode altered with the LNMPs when studied in a standard redox system of 5 mM KMnO 4 , displayed that yttrium doped LNP, i.e. LNYP-3 exhibits the highest active surface area (0.0050 cm 2 ) displaying the lowest average crystallite size ( D avg ) i.e. ∼ 7 nm. Electrocatalytic behavior monitored in KOH showed that LNMP-2


CV
Undue consumption of fossil fuels is a major contributing factor that has increased pollution and greenhouse gases that led to global warming and ozone depletion by emitting hazardous carbon compounds as major concerned outcomes [1,2].To overcome these drawbacks, efficient and green energy production technology is necessitated to meet the decarbonized energy goals, thereby minimizing global climate changes [2,3].Green energy includes clean power sources, e.g.solar power, biomass, wind, hydropower, and geothermal.Green energy technology is a collaborative strategy that combines conventional fossil fuel systems and updated renewable technologies in terms of accessibility, efficiency, energy security, equity, and economic sustainability [4,5].Keeping in view the significance of clean energy and low carbon environment, findings of alternative energy resources are key research areas nowadays [6].Pollution-free, clean, and green energy is produced by renewable energy sources but for sustainable energy supply, electrochemical energy is the economical storage system [7].Electrochemical methods like hydrogen evolution reaction (HER), oxygen evolution reaction (OER), and oxygen reduction reaction (ORR) are used for the production of viable energy [8].HER and OER in alkaline media are usually slow reactions that need the use of an electrocatalyst to increase their efficiency and potentiality.
Electrocatalysis takes place at the electrode-solution interface where the electrode is modified by a catalyst at its surface [9,10].The development of nonprecious metal-based electrocatalysts and new electrolytic systems toward water oxidation is an active research area that still needs further improvement and perfection [11,12].Among multiple energy conversion means, direct methanol fuel cell (DMFC) possesses various advantages including the adoption of nonprecious electrocatalysts such as CeO 2 , NiCu-CO 2 O 4 , green synthesized catalysts, zeolite base catalysts, algal based catalysts, and SeO 2 [13,14].Risks of corrosion of carbon and catalysts are predominantly diminished.The performance of alkaline DMFC mainly depends on the electrocatalytic responses of electrocatalysts at the surface of the electrode, electrode/electrolyte interface, and the ion and electron transfer rate in the electrode [15,16].So, the development of novel electrocatalysts with enhanced stability and electroactivity is a prerequisite for such high-efficiency conversion technologies [17].Various electrocatalysts for water oxidation include perovskites, halides, selenides, layered double hydroxides and nickel, cobalt-based oxide, algal membranes, each having their recompenses and drawbacks.Perovskite oxides show the deviation from the ideal stoichiometry due to the substitution of A/B-sites leading to electrochemical stability [18].Doping and modifications in material alter the stoichiometry and create defects in the crystal structure of the material.A larger number of defect insertion points offer more oxygen vacancies and hence the active sites [19].Doping in A-site with alkaline metal ions and B-site with transition metal ions creates defects in the arrangement of crystal [20].In water oxidation, oxygen is produce OER, and reduction reaction produces a hydrogen evolution reaction (HER).
Water oxidation is a single-step reaction due to which large overpotentials are produced, and to overcome this, large overpotential efficient electrocatalysts are needed.Overpotential is the difference in equilibrium/thermodynamic potential of a reaction and the real potential where the catalyst actually initiates the reaction at a specific current under control conditions.Low-overpotential WOR is required for progressed energy conversion technology [21,22].The best catalysts so far used for OER are Pt and Ir oxide but these are expensive and less abundant [23].Nickel-incorporated nanomaterials, being used in supercapacitors, electrocatalysts, battery materials and sensors, have earned particular emphasis from researchers owing to their abundance, low cost, and environment friendliness [24,25].Owing to its large redox potentials i.e. >4.5 V vs Li + / Li, high energy and voltage, nickel-derived olivine (LNP) has acquired greater emphasis among the olivine phosphates (800 W h g 1 ) [26].Olivine edifices are constructed with MO 6 sharing octahedral sides and tetrahedral anions of phosphates at corners where octahedral holes are exchanged by lithium ions [27].LiNiPO 4 as an electrocatalyst possesses some general properties such as high thermal stability, durable potentials, unique electrochemical properties, inexpensive, environmentally friendly, and excellent cycling performance [28,29].The drawback of WOR is the reversibility of reaction, by recombination of electrons and holes which may slow down the reaction.A sacrificial agent is used for preventing this drawback which acts as an electron donor molecule by reacting with the holes.Through this method, the production of O 2 and H 2 increased by using methanol, glucose, ethanol, cyanide, and formaldehyde as a sacrificial agent [30].Methanol is reported as a sacrificing agent/facilitating agent/co-catalyst for WOR by our group that boosted the charge transfer through the electrode and decreased the onset potential [31].Sacrificial agents are basically the electron donors or hole scavengers that boost up the H 2 production by anchoring on the electrocatalyst's surface [32].A co-catalyst involved mechanism is also an uphill WOR; however, the onset potential is comparatively less [33].
In this work, a cost-effective non-aqueous sol-gel synthesis approach is adapted for the preparation of LNP and PNMP (LiNi 1−x M x PO 4 ) where M can be Mg, Ca, and Y in the perovskite structure.Physiochemical characterization is accomplished via XRD, FTIR, SEM, EDS, and CV for investigating the phase purity, particle size, surface area, and electronic properties.The voltammetric profile is established in KOH as the electrolyte because protons produced in WOR can be facilely removed in an alkaline medium, thereby restoring the catalyst's original pristine state [34].WOR is studied on different metal-doped LiNiPO 4 in KOH using glucose as the facilitating agent for the first time.

Experimental section 2.1 Materials
), phosphoric acid (H 3 PO 4 ), and ethylene glycol.All the chemicals are bought from Sigma-Aldrich.Solutions were prepared in deionized water.

Synthesis
Non-aqueous wet-chemical sol-gel method is employed to manufacture the olivine orthophosphate, LiNiPO 4 , and metal-doped (Ca, Mg, and Y) lithium nickel phosphate composites [35,36].The preparation technique comprises of heating of metalloorganic gel precursors.The non-aqueous mixture was organized by stoichiometric ratio of 1:1:1 of the lithium and nickel nitrate in ethylene glycol while the (H 3 PO 4, ) was added dropwise under the continuous stirring of the solution for 2 h at 200°C.The obtained gel precursors were heated to produce the powder of nanomaterials.Synthesized powders were calcined in the furnace at 850°C temperature for 12 h under air atmosphere and cooled at room temperature.For the synthesis of the doped composition of lithium nickel phosphate, a stoichiometric amount of the magnesium, calcium, yttrium nitrate, and nickel nitrate were added to obtain the metal-doped lithium nickel phosphate where mole % of dopants ranges 0-9%.
Furthermore, mechanical ball milling was performed for all calcined powders using zirconia balls (100 g) of different diameters.Zirconia balls are used for their large mass/volume ratio; producing course-sized particles of catalyst powder [37,38].It is a grinding phenomenon utilized to produce a homogeneous particle size of the sample [39,40].Milling media (acetone) are filled according to the specific ratio of the sample and ball, which means the mass of the ball is one-fourth times greater than the weight of the samples [41].Acetone is preferred in dry-milling as the material does not stick to the balls and can be isolated feasibly [42].The ball milling was sustained for 2 h under 300 rpm, and afterward, the mixture was placed in the oven for 5 h for drying.Sample simplified representation into codes is presented in Table 1.

Physical characterization
Different analytical techniques were utilized for investigating the physical and chemical composition of synthesized materials.For the determination of the crystal assembly and phase concentration of prepared materials, powder XRD (D5005 diffractometer, Siemens) was pragmatized.All the XRD measurements were done with a sol X detector High-performance catalysts for electrochemical water oxidation reaction  3 employing a Cu kα radiation source with the 1.54 Å wavelength in the 10°-80°range for a widespread and deep understanding of the physical picture of material at the ambient temperature.FTIR analysis was carried out to measure the intensity of infrared radiation as a function of wavelength.A small portion of the sample was sandwiched in KBr for FTIR evaluation, and spectra were then collected as an average of five examinations by means of the OPUS Software Package for data gaining.SEM has been accomplished to obtain the texture and particle size of composites.A field-emission scanning electron microscope (Supra55.Zeiss) with a 10 kV quickening voltage was set to obtain micrographs for the SEM.Electrochemical Gamry 1000 interface instrument was used for the electroanalytical analysis of materials through cyclic voltammetry.

Electrode fabrication and electrochemical characterization
For electrochemical application, a three-electrode cell assembly is used which consists of three electrodes, i.e. glassy carbon electrode (GCE) as the working electrode, silver wire as the counter electrode, and Ag/AgCl as the reference electrode.GCE was modified by a conventional drop-casting technique [31,43].In this technique, the working electrode was polished on the felt paper by using an alumina slurry and then rinsed with deionized water before the modification.Polishing and ultrasonication are completed in ethanol for 2 min for the removal of any impurity present on the electrode surface.
The electrode surface was wet with 2 µL of ethanol and 0.1 mg of the synthesized powder was placed on it, resulting in the drop-casting 2 µL 0.5% of Nafion (co-polymer of sulfonated tetrafluoroethylene) acting as a binder [43].All produced nanomaterials were investigated to examine their behavior toward water oxidation in 0.1 and 0.5 M KOH used as a supporting electrolyte, and 0.05 (3.2 μM) glucose was also utilized as a sacrificial agent.The full potential range voltammogram showed only an OER peak, and no other peak appeared.Thus, it is suggested that a short window range (0-1.5, 1.6) is better for the further working of the catalyst.Cyclic voltammetry is implicated as the analytical tool to investigate the electrochemical potentials of the prepared olivine materials toward water oxidation reaction.

Results and discussion
Material's physiochemical characterization and catalytic studies are presented in this section.To analyze the highefficiency LNPs, deep-rooted material investigation was carried out earlier than the electroanalytical research.To understand the simultaneous analysis of electrode engineering and material properties, synthesized materials were tested electrochemically.LNPs were synthesized under an air atmosphere at 850°C and ball milled for smooth narration.Electrochemical tests were first performed with LiNiPO 4 powder.For the authentication of doping in the prepared sample and to study the accuracy of the structural features, morphology, and thermal stability of the samples, various experimental techniques were also employed.The redox and electrocatalytic properties of all the prepared samples were examined via CV.The crystal structure of electrode materials was studied via XRD.Diffraction patterns for LNP and its Ca-, Mg-, and Y-doped analogues are shown in Figure 1(a).Peaks clearly designated the orthorhombic structure of olivine with the Pnma phase group and displayed the desired agreement with reference card JCPDS card No. 81-1528 [44].There are no peak shifts observed by the addition of external new metal as its content percentage is in traces.However, the peak intensity is altered suggesting the formation of crystal structure and extent of crystallinity.The basic structure is not significantly changed but the catalytic efficiency is enhanced by defect formations.High-intensity peaks evidences the decidedly crystalline morphology, whereas the peaks shown by ball-milled samples exposed great widths owing to smaller crystallite sizes calculated using Debye Scherrer equation shown by equation ( 1) [45].
Peak positions are not altered, hence signifying the integrity of crystalline structure.D avg of LNP and its Ca-, Mg-, and Y-doped analogues are represented in Table 1, respectively.The range of D avg is from 9.24 to 10.5, 9.04 to 9.58, and 7.0 to 11.3 nm for Ca-, Mg-, and Y-doped materials, respectively.Morphological studies of the samples have been accomplished via SEM, and the micrographs are shown in Figure 1(b-d) where particles preserved the distinguishable grain boundaries and smooth surface structure.Spherical particles increase the surface area of the interface, however, show the agglomeration of the particles due to their smooth and steamy nature.Moreover, the elemental composition of LNP and its doped analogues were expounded by EDX as given in Table 2, and respective spectra are displayed in Figure 2. The determined elemental composition is about similar to the predictable ratios of elements, hence suggesting it is an adaptable synthesis route.
FTIR vibrational spectrum of LNP is shown in Figure 3(a) which reveals the presence of various functional groups.LNP showed three fundamental vibrations consisting of stretching modes [46].Symmetric stretching appeared due to internal vibration of the P-O single bond at 971 cm −1 .Owing to the asymmetric stretching, a triplet has appeared in the region of the 1,055-1,150 cm −1 .The peak was observed, in the region between 400 and 700 cm −1 due to the transition metal ions.The peak at 646 cm −1 may also be due to asymmetric stretching vibrations of Ni-O bonds in the NiO 6 octahedron.WOR catalysis proceeds at NiO 6 edges of the metal oxides, and a high catalytic efficiency is perceived [47].These observations suggested the presence of major functional groups related to Ni and (PO 4 ) 3− .Metal can accept diverse structures exhibiting different orientations of phosphate groups present in the crystal.Tunable coordination of − PO 4 group can stabilize the transition state of metal by feasible modifications in its local positions; hence, a robust redox reaction can occur on the TM surface [48].The vibrational spectra of LNCP-1, LNMP-2, and LP-3 are shown in Figure 3(b), and respective specific peaks are addressed in Table 3.
Employing cyclic voltammetry as the analytical tool, electroanalysis was performed with the potentials window of 0-1.6 V using 0.1 and 0.5 M potassium hydroxide as electrolyte and 0.05 M glucose as a sacrificial agent [49].Conducive properties of synthesized samples are analyzed by using potassium ferricyanide and potassium chloride.Potassium ferricyanide has a known diffusion coefficient which provides information about catalytic performance.Voltammetric cycling for all modified electrodes was performed by employing 5 mM K 3 [Fe(CN) 6 ] and 3 M KCl to estimate the active surface area of the modified electrodes [50,51].The potential window for all modified electrodes to measure the surface area ranges from −0.2 to 1 as shown in Figure 4.A cyclic voltammogram for Fe 2+ /Fe 3+ redox system depicts where I pa represents the anodic peak current of each catalyst in (μA), n indicates the number of electrons transferred (1 in this case), A shows the active surface area (cm 2 ), D denotes the diffusion coefficient (0.76 × 10 −5 cm 2 s −1 , 25°C),   υ stands for scan rate (100 mV s −1 ), and C is the analyte concentration (5 mM).Estimated ASA are listed in Table 4.
ECSA stands for the electrochemical surface area active for the reaction.The larger the ECSA exhibited by a catalytic material, a more facile reaction and boosted catalysis output is anticipated [50,53].Values of active surface area range from 0.0012 to 0.0050 cm 2 .By comparing the value of the active surface area of metal ion-doped phosphate, it is deduced that a vivid difference exists between these series.From the calcium-doped series, LNCP-1 showed the highest value of ASA.From magnesium-doped series, LNMP-2 attained the maximum value of ASA.While in the yttrium-doped series, LNYP-3 depicted the highest value of ASA.The higher active surface area corresponds to the better catalytic activity of nanomaterials.
WOR has been studied by modifying the electrode with the samples at room temperature, and resultant cyclic voltammograms showed a single peak; i.e.only an anodic peak was observed.The electrocatalytic potential of lithium nickel phosphate and metal-doped LNP-fabricated GCE was determined using 0.1 and 0.5 M KOH as electrolyte and 0.05 M glucose between 0 and 1.6 V potential window, and an irreversible oxidation peak has been witnessed in the resultant voltammograms.CV profiles displayed in Figure 5 were recorded at fixed 100 mV s −1 for the various synthesized  High-performance catalysts for electrochemical water oxidation reaction  7 materials.By adding different concentrations of glucose, the peak profiles showed an increase in the peak current of OER, which is the anodic half-reaction.As glucose concentration increases, the corresponding value of OER current increases indicating glucose-favored OER process [54].Among all the series LNCP-1, LNMP-2, and LNYP-3 showed the highest peak current in their respective analogues.Reinmuth equation displayed by equation ( 3) was used to estimate the rate constants for the prepared materials [55].Values of k o can be calculated from the slope by plotting a graph between I p vs [glucose].
Values of the rate constant quantified the reversibility and irreversibility of the process going on the interface [56].
where I p is the anodic peak current in (A), F is Faraday's constant in (C/mole), A is an area of the electrode in (cm 2 ), C is the concentration of glucose (mol cm 3 ), and k o is the heterogeneous rate constant in (cm s −1 ).The k o as shown in Table 5 for the WOR describes the capability of electron transfer at the electrode-electrolyte interface and the reaction kinetics of the materials.
According to the values of the k o , the insulating and conductive behavior of the materials can be distinguished.The lowest value of k o thus indicates the insulating behavior of the LNP due to the slowest electrocatalytic property.The heterogeneous rate constant value for lithium nickel  phosphate and metal ion-doped phosphate-modified GC electrode increased in the following order.

< − < − < − LNP LNCP 1 LNYP 3 LNMP 2
Doping of magnesium resulted in the highest value of the rate constant which proved it to be a good scheme of the catalysts.
The electroactivity of synthesized material was tested using cyclic voltammetry in the presence of 0.5 M KOH and 0.7 ml of glucose (facilitating agent), which equates to 0.05 molarity of glucose.The scan rate was varied from 20 to 100 mV s −1 within the potential window (0-0.5, and 1.6 V) at room temperature.The CV responses of LNCP-1, LNMP-2, and LNYP-3 for OER are presented in Figure 6.Voltammograms for OER using various synthesized samples at different scan rates showed the responsive increase in the peak current gradually as the scan rate was varied from 20 to 100 mV s −1 .It was also inferred that the of the synthesized material was better in the presence of 0.05 M glucose than that in KOH solution only.Analyzing the overall voltammetric profiles for all samples, LNCP-1 LNMP-2, and LNMP-3 showed the best results from all doped LNP analogues.
Furthermore, to enhance the understanding of the electrocatalytic potential of LNMPs, the diffusion coefficient "D o " was calculated using the Randles-Sevcik equation for irreversible processes.Electrocatalytic property of the materials gives information on facile electron transfer process by the electrochemical response.By using the equation ( 4), the mass transport coefficient is calculated [57].High-performance catalysts for electrochemical water oxidation reaction  9 the irreversible nature of the electrode process and the smallest for LNP, indicating it to be less efficient for OER electrocatalysis.The trend for the optimized samples is as follows: Magnesium doping shows an increase in the value of the D o due to the fast electron transfer process.LNMP-2 shows the highest values of both D o and k o , which indicates these materials to be the better catalyst as compared to the other prepared samples.
To analyze the effect of the temperature on the electroactivity of the prepared pure lithium nickel phosphate catalyst and its doped analogues, the temperature was varied from 20 to 40°C at 100 mV s −1 scan rate in 0.5 M KOH in the presence of 3.27 μM optimal glucose concentration, and the CV responses were observed for each composition.At elevated temperatures, the kinetics of water-splitting was facilitated, and Figure 7 shows the enhancement in the current by increasing the temperature for three optimized samples.
By increasing the temperature, active sites of the electrode are free from the adsorbed CO, due to this reason high temperature is required for the water oxidation for such materials.The maximum current was observed for the LNMP-2 composition, i.e. 1.33 mA for water oxidation at 40°C.
A more simplified form of the Marcus equation is given in equation ( 6).
where Z het is the collision frequency and can be calculated at any temperature with the help of the following relation shown in equation (7).
where M is the molar mass of glucose and G is the Gibbs free energy.
Respective ΔH* and ΔS* values were obtained from the slope and the intercept of the Marques Equation plots.ΔH* and ΔS* values were retrieved by plotting the graphs between the ln (k s /Z het ) vs 1/T.The values of activation energies were calculated by using the following relation in equation (8).
where ΔH* and ΔG* showed a positive value, which specifies that water oxidation in the presence of glucose is an activation-controlled process.Also, the reaction is endothermic in nature, and the spontaneous flow of the reaction takes place in forward direction [59].Hence, on operating the electrochemical cells at higher temperatures, the functions of electrocatalysts are effectively improved due to an increase in the catalytic activities.Minimum value of activation energy as observed in Table 6 was observed for the LiNi 0.97 Mg 0.03 PO 4 (LNMP-2) composition; therefore, it is suggested as the best electrocatalyst among all other analogues.Furthermore, different groups of compounds have been employed as water oxidation electrocatalysts.A comparative analysis presented in Table 7 with electrochemical parameters indicate that the presented LNMP-based composites are superior catalysts for WOR owing to the developed electronic and transport properties of the materials.In this way, particularly effective and firm WOR electrocatalysts are anticipated for future energy bids.Increase in number of active sites by tuning the structure of the electrocatalyst and enhancement in inherent activity of active sites by modifying their atomic and electronic states.Such flexibility of LMNPs makes them available as proficient catalysts for future electronic devices [60].

Conclusions
In this study, the non-aqueous sol-gel method was used to successfully synthesize the single-phase LNP and doped LNMP nanoparticles calcined at 850°C in an air atmosphere.XRD for pristine LNP and magnesium doped LNP patterns with intact phases according to the peak analysis ensured the stability of the material.Single-phase orthorhombic crystalline structure was obtained for LNP and magnesium doped analogs, suggesting the appropriateness of doping into the crystal lattice of lithium nickel phosphate.Average crystallite sizes determined using XRD were found to lie in the 7-10 nm range.FTIR spectra indicated the corresponding peaks of phosphate ion and metal oxide in lithium nickel oxide.No extra band appeared in the doped analogs thus conforming to the purity of synthesized samples.SEM micrographs showed nanoparticles to be nearly spherical in shape with a homogeneous distribution of particles.The EDX results showed that the stoichiometric amount of magnesium added was in good agreement with the theoretical values.The maximum ASA and the highest peak current were associated with LNMP-2 composition.All materials showed potential applications in the WOR catalysis and electrocatalysis the oxygen evolution reaction in 0.5 M KOH and glucose.In a basic medium, all the prepared samples responded to water oxidation, but LNMP-2 produced the best results, with the highest D o (5.2 × 10 −5 cm 2 s −1 ) and k o value (3.2 × 10 −2 cm s −1 ) for WOR.LiNi 0.95 Y 0.05 PO 4 i.e.LNYP-3 is found to be the optimum catalyst among its series.For LNMP-2, the high peak current value and low onset potential (0.90 V vs Ag/AgCl) demonstrated the catalyst's potential for the water oxidation reaction.In conclusion, all electrocatalysts have a strong electrocatalytic potential for WOR and high-performance DMFCs.The synthesized catalysts can be recommended as promising candidates for further electroanalytical research.
Funding information: The experimental work has been accomplished at Fuel Cell Lab, Department of Chemistry, QAU, Islamabad.This work was funded by the Researchers Supporting Project Number (RSP2024R441), King Saud University, Riyadh, Saudi Arabia.We are highly thankful

Table 1 :
Coding of samples into simpler abbreviations LiNi 1−x M x PO 4 [M = Mg, Ca, and Y]

Table 2 :
Calculated crystallite sizes and EDX analysis of LNP and its Ca-, Mg-, and Y-doped analogues

Table 4 :
Estimated active surface area for all synthesized samples

Table 5 :
Calculated values of heterogeneous rate constant, diffusion, and mass transport coefficients

Table 6 :
Thermodynamic parameters of LNP and its doped series in 0.5 M KOH + 3.27 μM glucose at 100 mV s −1 scan rate and 20-40°C