Noble Metal‐Free Electrocatalysts for OER and ORR Based on Phytic Acid

Abstract

This study presents a versatile, eco‐friendly approach for synthesizing noble metal‐free electrocatalysts by combining phytic acid (PA) with transition metals (M: Co, Mn, Ni, Cu). The resulting materials (MPA) feature metal phosphides or phosphates intricately embedded within an amorphous carbon matrix, with composition and structure highly dependent on the selected metal precursor. MnPA displayed notable oxygen reduction reaction (ORR) activity (E _ONSET = 0.85 V vs. RHE), while NiPA showed strong oxygen evolution reaction (OER) performance (E _OER = 1.62 V vs. RHE), revealing the potential of phosphorus as a powerful dopant in carbon‐based systems. This study highlights the tunability of electrocatalytic properties through metal selection and provides a sustainable pathway for designing high‐performance ORR and OER electrodes. The results underscore the role of phosphorus chemistry in advancing next‐generation energy conversion technologies.

Phytic acid enables the sustainable synthesis of noble metal‐free electrocatalysts containing transition metal phosphides/phosphates embedded in carbon frameworks. The choice of metal governs structure and electrochemical performance, yielding Mn‐based catalysts with excellent oxygen reduction activity and Ni‐based catalysts with outstanding oxygen evolution performance.

1. Introduction

The development of efficient and sustainable energy conversion and storage systems is a crucial area of research to meet the growing demand for clean and renewable energy [1]. Among various energy conversion processes, the oxygen reduction reaction (ORR) and the oxygen evolution reaction (OER) play key roles in fuel cells, metal–air batteries, and electrolyzers [2]. However, both reactions possess significant challenges in the advancement of environmentally friendly energy systems [3, 4]. The conventional benchmark electrocatalysts utilized for these reactions, such as platinum nanoparticles supported on carbon black for ORR and IrO₂− or RuO₂‐based compounds for OER, suffer from issues of limited availability and prohibitive costs. Consequently, there has been a shift in scientific research toward the development of cost‐effective electrode materials to address these concerns. Therefore, the design and synthesis of cost‐effective, highly active, and stable catalysts for ORR and OER are essential for the advancement of electrochemical energy devices.

In recent years, significant attention has been focused toward exploring alternative catalysts derived from abundant and environmentally friendly sources. Transition metal‐based catalysts have garnered significant attention in electrocatalysis due to their unique electronic properties and catalytic activity [5, 6, 7], specifically those based on Fe and N carbon materials [8, 9]. These catalysts, including cobalt, nickel, iron, and their alloys, exhibit exceptional performance in various electrochemical reactions, including the ORR and the OER [10]. Moreover, the abundant availability of transition metals makes them attractive alternatives to noble metal catalysts, addressing the issues of scarcity and high cost [11].

Recent literature has demonstrated the strong potential of transition metal–phosphorus‐based materials as noble‐metal‐free electrocatalysts. Transition metal–phosphorus‐based materials, such as metal phosphates and phosphides, have attracted considerable attention as promising electrocatalysts [12, 13]. These compounds offer unique advantages due to their inherent stability, high conductivity, and tunable electronic properties [14]. Transition metal phosphates, such as cobalt phosphate and iron phosphate, possess catalytic activity derived from the synergistic effect between the transition metal centers and the phosphate ligands [15, 16]. Similarly, transition metal phosphides, such as nickel phosphide and molybdenum phosphide [17, 18], exhibit remarkable catalytic performance owing to their intrinsic electrochemical properties and surface reactivity.

Despite their high electrocatalytic activity, the synthesis of metal phosphates and phosphides often requires the use of undesirable reagents like phosphoric acid, phosphine, or phosphine oxide [19, 20]. A promising alternative lies in phytic acid, a natural compound found abundantly in plants [21]. Phytic acid possesses multiple phosphate groups that serve as exceptional ligands for metal ions [22, 23]. Additionally, it can act as a carbon source, eliminating the need for additional carbon‐containing reagents during synthesis [24]. Although phytic acid has been widely used as phosphorus (P) source during different synthesis routes [25, 26], it has been poorly studied in combination with metal precursors [27]. Furthermore, the conventional method of preparing these materials by simply stirring the precursors has proven to be inefficient. Mecanosynthesis, which emerge as one of the technologies which will revolutionize the future of chemistry all over the world [28], involves the use of high‐energy mechanical forces to induce chemical reactions between the reagents, surpassing the efficiency of traditional stirring methods [29, 30, 31]. This approach represents a solvent‐minimized, scalable, and sustainable strategy, enhancing the mixing and interaction of precursors and facilitating the efficient formation of uniform metal–phosphorus phases, making it particularly attractive for the preparation of phytic acid‐derived electrocatalyst. To the best of our knowledge, transition metal–phosphorus‐based materials has not ever been prepared by phytic acid and mecanosynthesis processes.

Herein, we synthetized different noble metal‐free electrocatalysts based on the combination of phytic acid and the metal precursors (Ni, Mn, Co, and Cu) through a mecanosynthesis process followed by a heat‐treatment step. The study goes beyond predictable “metal selection effects” by providing experimental insights into the role of metal‐dependent carbon gasification during PA carbonization, the temperature‐driven phase evolution between phosphates and phosphides, and the cooperative effect between phosphate and phosphide phases in Ni‐based OER catalysis, where mixed‐phase materials outperform single‐phase counterparts. These observations provide experimentally grounded structure–property correlations that are directly relevant for rational catalyst design.

2. Material and Methods

2.1. Synthesis of MPA

Phytic acid solution (50% w/w H₂O), Pluronic F‐127, and Nafion (5 wt.%) were purchased from Sigma–Aldrich. CoCl₂·6H₂O (96%–97%) and MnCl₂·4H₂O (+99%) were purchased from PROLABO and ACROS ORGANIC, respectively. NiCl₂·6H₂O (98%) and CuCl₂·2H₂O (99%) salts were obtained from Alfa Aesar.

In the experimental procedure, a mixture was prepared consisting of 2 g of the phosphorus source (phytic acid, 50% w/w in H₂O), 1 g of the metal precursor (CoCl₂, CuCl₂, MnCl₂, and NiCl₂), and 2 g of a binding agent (Pluronic F‐127). The reagents were ground using a planetary ball mill (PM100, Retsch) equipped with an agate jar and agate balls, operating at 500 rpm for 1 h. The resulting mixture, which displayed a distinctive color, was then subjected to thermal treatment in a high‐temperature tubular furnace. The heat treatment was carried out at 900 °C for 1 h, with a heating rate of 5 °C min⁻¹, under a flow of nitrogen gas (N₂) at a rate of 150 mL min⁻¹. Following the heat treatment, the material underwent sonication in 1 M HCl solution using an ultrasonic bath for 1 h at 50 °C. Subsequently, it was washed by filtering through 0.5 L of 1 M HCl solution followed by 0.5 L of distilled water. This washing process aimed to remove any remaining excess metal in the sample, ensuring that only strongly bound metal species remained and excluding the contribution of loosely bound particles [10]. Finally, the material was dried overnight at 100 °C in an oven. Materials has been named MPA, being M the metal used, Ni, Cu, Co, or Mn, and PA the phytic acid.

2.2. Physicochemical Characterization

Transition metal–phosphorus‐based materials (MnPA, CuPA, CoPA, and NiPA) were characterized by elemental analysis (EA), X‐ray diffraction (XRD), transmission electron microscopy (TEM), X‐ray photoelectron spectroscopy (XPS), thermogravimetric (TG) analysis, N₂ adsorption–desorption isotherms, and inductively coupled plasma‐optical emission spectroscopy (ICP‐OES).

EA of carbon, hydrogen, nitrogen, sulfur, and oxygen in the prepared samples was performed using a Vario EL cube analyzer (Elementar). For each measurement, approximately 2 mg of sample was combusted at 1700 °C under a helium stream containing oxygen. The combustion products were separated through a chromatographic column and subsequently quantified with a thermal conductivity detector, except for sulfur species, which were analyzed by an infrared detector. The concentrations of metal and phosphorus elements were quantified by ICP‐OES using an Xpectroblue‐EOP‐TI FMT26 spectrometer (Spectro).

Nitrogen (N₂) and hydrogen (H₂) adsorption–desorption isotherms were measured at −196 °C using a fully automated manometric adsorption analyzer (ASAP 2020, Micromeritics). Prior to the adsorption tests, the samples were degassed under a high vacuum (2–4 × 10⁻⁶ Pa) for at least 24 h at 90 °C in the degassing port. An additional degassing step was carried out in the analysis port for a minimum of 6 h before the introduction of the analyte gas, starting from a high‐vacuum condition. To minimize the possibility of helium entrapment in ultramicropores, both warm and cold free‐space volumes were determined following the adsorption–desorption cycles. The BET area (A _BET) of the materials was determined using MicroActive software (Micromeritics) based on the guidelines recommended by IUPAC [32]. Pore size distributions (PSDs) were calculated using SAIEUS software (Micromeritics) by applying the two‐dimensional nonlocal density functional theory for heterogeneous surfaces (2D‐NLDFT‐HS) simultaneously to the N₂ and H₂ isotherms.

The structural ordering of all samples was examined using a Bruker D8 Advance A25 powder X‐ray diffractometer. This instrument is equipped with a copper anode X‐ray source operating at 40 kV and 40 mA, a scintillation detector, and a graphite monochromator. Silver behenate was employed as the calibration standard.

TEM observations were carried out with a JEM‐ARM 200F Cold FEG TEM/STEM microscope, which includes both probe and image spherical aberration correctors. For sample preparation, the powdered materials were dispersed in ethanol using mild ultrasonication, and a drop of the suspension was deposited onto a carbon‐coated copper grid (200 mesh) and left to dry under ambient conditions. SEM coupled with energy‐dispersive X‐ray spectroscopy (EDS) mapping was utilized to analyze the spatial distribution of heteroatoms across the material surfaces. These measurements were performed on a Quanta 650 FEG (FEI) microscope, with the samples mounted on Scotch tape for imaging.

XPS data were collected using an ESCAPlus OMICRON spectrometer equipped with a nonmonochromatized Mg Kα radiation source. The acquired spectra were processed using CASA software for Shirley‐type background subtraction, peak fitting, and quantitative analysis.

TG analysis was conducted to evaluate mass loss and thermal decomposition behavior during the heat‐treatment stage. The colorful paste obtained from the mechanochemical synthesis was heated up to 900 °C at a rate of 10 °C min⁻¹ under a continuous argon flow of 50 mL min⁻¹.

2.3. Electrochemical Measurements

Electrochemical characterization of the MPA materials was performed using a conventional three‐electrode configuration equipped with a rotating ring‐disk electrode (RRDE) setup, connected to a PGSTAT302N bipotentiostat (Metrohm). The primary working electrode was a glassy carbon disk with an exposed surface area of 0.196 cm², while a platinum ring functioned as the secondary working electrode. A reversible hydrogen electrode (RHE) served as the reference electrode, and a glassy carbon rod (8 mm in diameter) was used as the counter electrode. All measurements were conducted at ambient temperature in alkaline media, specifically 0.1 M KOH for ORR tests and 1 M KOH for OER analyses.

To fabricate the working electrodes, the catalyst powders were dispersed in a mixed solvent of isopropanol and water (20:80 v/v) containing 0.2 wt.% Nafion, yielding a suspension with a concentration of 4 mg mL⁻¹. Subsequently, four aliquots of 8.42 µL of this catalyst ink were carefully drop‐cast onto the glassy carbon disk, corresponding to a catalyst loading of 0.68 mg cm⁻². The prepared electrodes were subsequently immersed in the working solution under vacuum for a few minutes in order to improve its wettability [33]. Each prepared electrode was immersed in the electrolyte solution while continuously purging with nitrogen to remove dissolved oxygen. Before commencing the electrocatalytic measurements, cyclic voltammetry (CV) was conducted between 1.00 and 0.00 V vs. RHE at a scan rate of 50 mV s⁻¹ for 20 consecutive cycles in order to stabilize the electrode surface [33]. For the ORR, linear sweep voltammetry (LSV) experiments were carried out under an O₂‐saturated atmosphere at a rotation rate of 1600 rpm, using a potential range from 1.0 to 0.0 V vs. RHE and a scan rate of 5 mV s⁻¹. In the case of the OER, LSV measurements were conducted under N₂‐saturated conditions within a potential window of 0.8–1.8 V vs. RHE, maintaining the same rotation speed and scan rate. The recorded current densities were normalized to the geometric surface area of the glassy carbon disk (0.196 cm²). The potentials corresponding to current densities of −10 mA cm⁻² (E _OER) and −0.1 mA cm⁻² (E _ORR), as well as the overpotential values, defined as the additional potential required to drive the electrochemical reaction, were employed as performance metrics to compare the catalytic activity of the samples. OER polarization curves were corrected for Ohmic losses using automatic current‐interrupt (CI) iR compensation integrated into the potentiostat. In this procedure, the current is briefly interrupted to measure the instantaneous potential drop, from which the uncompensated resistance (R _s) is obtained and used to adjust the applied potentials. The long‐term electrochemical durability of the catalysts was assessed by performing 750 and 250 consecutive LSV cycles for the MnPA and NiPA samples, respectively. For benchmarking purposes, a commercial Pt/C (Sigma–Aldrich) catalyst was tested for ORR, while commercial IrO₂ was evaluated under identical conditions for OER.

3. Results and Discussion

Metal‐free electrocatalysts based on phytic acid were synthetized by a mecanosynthesis step, where phytic acid and the metal chloride were combined in a 1:1 ratio, followed by a heat‐treatment process at 900 °C. After that, the resulting material was washed with 1 M HCl to remove the metal that has not reacted with the phytic acid. Figure 1 illustrates a graphical scheme of the synthesis process. A colorful paste (purple, dark green, light orange, and light green for CoPA, CuPA, MnPA, and NiPA samples, respectively) was obtained after the mecanosynthesis step (Figure S1); meanwhile, a dark material was obtained after the heat treatment in all cases. The as‐prepared materials were named NiPA, CuPA, CoPA, and MnPA, when Ni, Cu, Co, or Mn precursors were employed in the synthesis, respectively.

FIGURE 1.

(a) Schematic representation of the synthesis process and (b) SEM image and elemental distribution mapping of MnPA sample, as representative sample.

3.1. Physicochemical Properties

The selection of the transition metal showed a decisive impact on both the composition and the structural features of the resulting MPA materials. The MPA samples were widely characterized using various techniques. Elemental mapping by EDS‐SEM (Figures 1b and S2–S4) confirm the absence of metal particle formation and suggest the presence of metal–phosphorus structures in the material. TEM images (Figure 2a) further corroborate the EDS‐SEM observations by revealing a homogeneous structure without any evidence of the formation of metal particles in any sample. HRTEM analysis of the NiPA sample reveals the presence of Ni‐containing particles with a broad particle size distribution, ranging from several tens of nanometers to highly dispersed species, including isolated atoms (Figure S5). Elemental mapping confirms that these particles contain P as well as N, with minor O contributions, which is consistent with the coexistence of phosphide and phosphate phases (Figure S6). The corresponding fast Fourier transform (FFT) analysis shows lattice spacings attributable to Ni–P phases [34], together with graphitic carbon domains (Figure S7). In contrast, the MnPA sample does not exhibit large Mn‐containing particles, and Mn is instead homogeneously distributed throughout the carbon matrix (Figures S8 and S9). The presence of isolated Mn atoms suggests a high dispersion of Mn species, in agreement with its low surface concentration observed by XPS.

FIGURE 2.

(a) TEM images of CoPA (left‐up), CuPA (right‐up), MnPA (left‐down), and NiPA (right‐down). (b) N₂ adsorption–desorption isotherms of the MPA materials. (c) Pore size distributions (PSDs) of the MPA materials obtained by applying the 2D‐NLDFT‐HS simultaneously to the N₂ and H₂ adsorption isotherms (for NiPA, only the N₂ adsorption isotherm was used). (d) TG plots of the as‐prepared samples at 10 °C min⁻¹ under inert atmosphere (N₂/Ar), as well as for Pluronic F‐127 (e) and PA (f) for comparative purposes.

EA and ICP analysis were performed to determine the composition of the materials (Table 1). The metal content varied depending on the metal used in the synthesis, with the highest content observed in NiPA, followed by CuPA, CoPA, and MnPA. The carbon content showed an inverse relationship with the metal content, meaning that the higher metal content, the lower the carbon content. Such differences in the carbon content suggest that Ni could act more efficiently as a gasification agent during the carbonization than other metals under the same conditions, resulting in the gasification of most of the carbon in the phytic acid, thus explaining the low carbon content of the NiPA material. On the other hand, the phosphorus content did not show a clear correlation with the metal content, suggesting the formation of metal–phosphorus phases with different thermal stabilities.

TABLE 1.

Carbon (C) and oxygen (O) contents determined by elemental analysis and phosphorus (P) and metal contents determined by ICP‐OES, as well as A _BET determined by N₂ adsorption–desorption isotherms.

Sample	C content, wt.%	O content, wt.%	P content, wt.%	Metal content, wt.%	Metal/P ratio	A BET, m2 g−1
NiPA	4.3	14.3	26.4	55	2.08	106
CoPA	17.9	25.2	27.9	29	1.04	238
MnPA	61.4	25.5	7.5	5.6	0.75	1246
CuPA	33.5	9.5	15	42	2.8	594

The textural properties were further studied by the analysis of the N₂ adsorption–desorption isotherms, shown in Figure 2b. An increase uptake of N₂, especially at low relative pressure, and a transition from a Type I isotherm to a combination of Type Ib and Type II isotherms was observed with increasing carbon content (see Table 1). Thus, NiPA is considered a mainly microporous material (pore width < 2 nm), whereas CoPA, CuPA, and MnPA would exhibit both micro‐ and mesopores (pore width between 2 and 50 nm). This was corroborated by the PSDs (see Figure 2c), from where it is clearly seen that the contribution to the pore volume from large micropores and from the mesopores increases with the carbon content. This development of porosity is also reflected in the values of apparent surface area, A _BET, that followed the same behavior (see Table 1).

The TG outcomes of the materials synthesized through mechanochemical processes are presented in Figure 2d. The initial mass reduction, occurring at 100 °C, can be attributed to water evaporation. Subsequently, a more substantial mass decrease manifests at 220 °C, likely attributed to the removal of Pluronic [35]. Notably, the elimination of Pluronic at lower temperatures, in comparison with the individual TG analysis of pure Pluronic (350 °C) (Figure 2e), suggests plausible interactions between Pluronic and the other reagents with lower thermal stability, thus facilitating its elimination from the material. Moreover, the temperature at which Pluronic is removed varies depending on the metal precursor, with the lowest temperature of Pluronic removal found at 210 °C for CuPA and the highest at 230 °C for MnPA. As observed in Figure 2e, at 400 °C, it is expected that Pluronic would be completely eliminated from the MPA samples. Nevertheless, the Pluronic removal is not completed in all cases. Interesting disparities in the mass loss have provided valuable insights about interactions between Pluronic and other reagents. While CuPA exhibits a mass loss of around 45 wt.% at 400 °C, which aligns with the combined mass of water and Pluronic, MnPA sample shows a mass loss of 57 wt.% at the same temperature. This substantial difference is attributed to the retention of Pluronic in the resulting MnPA sample, likely in the form of carbon content. This variance underscores diverse Pluronic interactions contingent on the specific metal phosphide counterpart. Moreover, the absence of mass losses at 268 and 330 °C, associated with pure phytic acid (PA) carbonization (Figure 2f) implies complete interaction between PA and the metal precursor. The slight decrease of mass at temperatures above 400 °C could be related to structural and chemical modifications which chemical modifications culminating in the formation of the resulting metal phosphide/phosphate.

XRD analysis provided further insights into the crystallographic phases present in the materials (Figure 3a). The results showed that different phases, such as phosphides and phosphates, were generated depending on the metal used in the synthesis. In the case of MnPA and CuPA samples, the resulting materials only show the contribution of one metal phase, Mn(PO₃)₂ for MnPA and Cu₃P for CuPA.

FIGURE 3.

(a) XRD patterns and (b) XPS spectra of P 2p of all samples. (c) XPS spectra of Co 2p, Cu 2p, Mn 2p, and Ni 2p of the corresponding material.

However, both Co and Ni are composed of several metal phases: CoPA contains 24.3% of Co(PO₃)₂ and 75.7% of CoP, and NiPA is composed of 8.6%, 24.4%, and 67% of Ni(PO₃)₂, Ni₂P, and Ni₁₂P₅, respectively. To better understand the effect of the thermal treatment, additional XRD measurements were performed at 800 and 1000 °C on the Ni‐based sample (Figure S10). The results reveal a clear temperature‐dependent phase evolution: At 800 °C the material is predominantly composed of nickel phosphates, at 1000 °C, it mainly consists of nickel phosphides, while at 900 °C, a coexistence of both phases is observed. Moreover, a notable change in the relative distribution of phosphide phases is observed, with a lower Ni₁₂P₅/Ni₂P ratio at higher temperature, as evidenced by the increased intensity of Ni₂P reflections in the NiPA‐1000 °C sample. This evidence confirms that the annealing temperature plays a key role in determining the phosphate/phosphide balance of NiPA. Additionally, MnPA shows a broad contribution in the XRD spectra at 20°–30°, suggesting the presence of amorphous carbon [36].

XPS was used to further analyze the surface chemistry of all the transition metal–phosphorus‐based materials. Table S1 provides a summary of the carbon, oxygen, and phosphorus contents detected on the material surfaces. The results showed distinct variations in the surface composition compared to the bulk composition obtained from EA and ICP. Notably, the metal content is significantly reduced on the surface (even negligible in the case of MnPA), while the carbon content is noticeably elevated.

These findings imply that carbon is mainly located at the surface of the materials, while the metal phosphide–phosphates conform the bulk structure. P 2p spectra show the peak related to P—C bonds at 132.6–133.1 eV in all samples (Figure 3b) [37], suggesting that the transition metal–phosphorus‐based phase is linked to the carbon surface. As expected, the peak associated to P—O bonds at 134.3–134.8 eV was also observed [37], according to the phosphate phases determined by XRD. It is worth mentioning that a peak linked to polyphosphate groups at 136.5 eV has been identified in MnPA [38]. Regarding the oxygen groups, two peaks at 531.5 and 532.8 eV associated to C=O or P=O bonds and C—OH or C—O—P groups were observed (Figure S11a), respectively, in all carbon materials [36]. Moreover, a peak at 534.2 eV related with P—O—P species was observed in O 1s spectrum of MnPA, which corroborates the presence of polyphosphates [36]. Interestingly, a peak at 529.1 eV related with lattice O [39] was observed in O 1s spectrum of CoPA. To evaluate the oxidation states of the metal species in all samples, the Co 2p, Cu 2p, Mn 2p, and Ni 2p spectra (Figure 3c) of the corresponding materials were analyzed. Co, Cu, and Ni oxides were detected at 779.8 [40], 934.6 [41], and 854.3 eV [42], respectively, which are characteristic of oxidized metal species, typically associated with Co³⁺, Cu²⁺, and Ni²⁺ surface states. Such species are in agreement with the crystallographic phases observed by XRD, taking into account the surface‐sensitive nature of XPS, as well as with the well‐known oxidation capability of metal phosphide surfaces when they are exposed to the air [43]. Therefore, XPS mainly reflects the oxidized surface layer, while the bulk composition is dominated by phosphide and phosphate phases as determined by XRD. Moreover, a peak related to Cu—P bonds was also detected in the Cu 2p spectrum at 932.5 eV [41], confirming the coexistence of metal–phosphorus bonding beneath the oxidized surface in this sample. Remarkably, no peaks were detected in the Mn 2p spectrum, according to the low amount of Mn in MnPA sample. Furthermore, the composition of the carbon phase, determined through XPS of C 1s (Figure S11b), was similar in all MPA. In general, the peak related to a graphitic carbon at 284.1 eV was mainly observed [44]. Besides, peaks at 285.6, 287.3, and 288.3 eV were also identified, related to aliphatic carbon [44] and carbonyl and carboxyl groups, respectively [45].

3.2. Electrochemical Characterization

The ORR electrocatalytic performance of the transition metal–phosphorus‐based materials synthesized in this study was assessed in an O₂‐saturated 0.1 M KOH solution. Prior to the electrocatalytic testing, the CV profiles were recorded in an aqueous solution containing 0.1 M KOH, saturated with N₂, and a scan rate of 50 mV s⁻¹ (Figure S12). The sample with the lowest carbon content, NiPA, displayed a negligible CV profile. Conversely, for CuPA, CoPA, and MnPA, an increase in the measured current is observed, which could be ascribed to the enhancement of both electrical conductivity and surface area development favoring the formation of an electrical double layer, according to the observed CV profiles [46]. Additionally, MnPA and CuPA displayed discernible redox peaks, indicating the occurrence of Faradaic reactions promoted by the presence of accessible metal active sites on the materials surface [47].

LSV experiments were conducted from which the E _ORR values for the transition metal–phosphorus‐based materials were determined. The outcomes are illustrated in Figure 4a,b. NiPA exhibited an E _ORR value of 0.73 V vs. RHE, while CuPA and CoPA displayed a similar E _ORR value of 0.80 V vs. RHE. In contrast, MnPA exhibited a higher E _ORR value of 0.85 V vs. RHE. The combination of carbon and metal oxide phases has proven to be promising for different electrochemical reactions [48]. To further assess the role of Mn in the ORR activity, a comparison can be made with a metal‐free phytic acid‐derived carbon previously reported by our group [49]. Despite also containing P—C functionalities, this material exhibited a lower onset potential (E _ORR = 0.82 V vs. RHE) under similar experimental conditions. This observation suggests that the presence of Mn leads to an enhancement of the ORR performance beyond that provided by P‐doped carbon alone. However, based on the available evidence, it is not possible to unambiguously attribute the catalytic activity to a specific Mn species or to identify a well‐defined active site. Therefore, the improved performance of MnPA is more appropriately ascribed to the combined contribution of the P‐doped carbon matrix and embedded Mn‐containing species.

FIGURE 4.

(a) LSVs, (b) E _ONSET values obtained from LSVs, and (c) Tafel slopes of all MPA materials in an O₂‐saturated 0.1 M KOH solution at 1600 rpm. (d) LSVs, (e) illustrative E _OER, and (f) Tafel slopes of all MPA materials in 1 M KOH, N₂‐saturated solution at 1600 rpm.

To further support the observed trends in ORR activity, the electrochemically active surface area (ECSA) was estimated from the double‐layer capacitance measured in the non‐Faradaic region. The capacitive current (ic) was calculated following Equation (1):

$$\mathrm{ic}=\frac{J_{\mathrm{a}}-J_{\mathrm{c}}}{2}$$ (1)

where J _a and J _c are the anodic and the cathodic current density, respectively, and the capacitance was obtained as

$$\mathrm{Cdl}=\frac{\mathrm{ic}}{V}$$ (2)

where V is the scan rate. ECSA was then calculated as

$$\mathrm{ECSA}=\frac{\mathrm{Cdl}}{\mathrm{Cs}}$$ (3)

Using a typical specific capacitance Cs = 40 µF cm⁻² for carbon/metal–phosphorus hybrids, the estimated ECSA values are MnPA = 720 cm², CuPA = 235 cm², CoPA = 188 cm², and NiPA = 10 cm². The significantly higher ECSA of MnPA is consistent with its highly developed porous texture and large specific surface area. This is expected to favor electrolyte accessibility and increase the number of exposed active sites, which is particularly beneficial for ORR processes where surface‐driven reactions dominate.

Moreover, the electrocatalytic behavior of the Mn‐based phosphate is in line with the general trend reported for Mn phosphates and phosphides, which have been widely recognized as promising candidates for ORR catalysis [50, 51, 52, 53, 54].

This consistency suggests that the activity observed here reflects the intrinsic catalytic features of Mn–P systems, further supporting their relevance as nonprecious metal electrocatalysts.

The performance of the electrode materials in the ORR is not solely determined by their electrocatalytic activity but also by the selectivity of the reaction, which is determined by the number of electrons transferred. A high number of electrons transferred, closer to four, results in increased energy production in fuel cells or metal–air batteries. On the other hand, a two‐electron pathway signifies the generation of hydrogen peroxide (Equations (1) and (2)), which is corrosive and can potentially harm various components of electrochemical devices. Otherwise, a two‐electron pathway can also be very useful for the electrochemical production of hydrogen peroxide, a valuable molecule in the chemical industry [55]. Therefore, to comprehend the underlying mechanisms of the ORR for each material, it is crucial to quantify the production of H₂O₂ during the reaction.

$${\mathrm{O}}_2\hspace{0.17em}+\hspace{0.17em}2{\mathrm{H}}_2\mathrm{O}\hspace{0.17em}+2{\mathrm{e}}^{‐}\hspace{0.17em}\to \hspace{0.17em}{\mathrm{H}}_2{\mathrm{O}}_2\hspace{0.17em}+\hspace{0.17em}2{\mathrm{OH}}^{‐}$$ (4)

$${\mathrm{H}}_2{\mathrm{O}}_2\hspace{0.17em}+\hspace{0.17em}2{\mathrm{e}}^{‐}\hspace{0.17em}\to \hspace{0.17em}2{\mathrm{OH}}^{‐}$$ (5)

Figure S13 presents the H₂O₂ formation data for all materials, revealing that NiPA and CuPA exhibit moderate H₂O₂ production, approximately 60%. Surprisingly, although MnPA exhibits the highest electrocatalytic activity toward ORR, it does not display the lowest H₂O₂ production, 45% compared to a 25% produced by CoPA. Tafel slopes were obtained to investigate further the kinetics and mechanisms of electron transfer at the active sites of MPA materials. Comparing the Tafel values shown in Figure 4c, it can be observed that NiPA exhibits the highest Tafel slope (418 mV dec⁻¹), indicating slow kinetics and a less favorable ORR performance. Moreover, the low current density exhibited by the NiPA sample suggests that the reduced carbon content in this sample leads to a really low electrical conductivity which inhibits the redox reactions on the active sites. On the other hand, MnPA shows a Tafel slope of 92 mV dec⁻¹, similar to the one obtained by the commercial Pt/C (89 mV dec⁻¹), suggesting a moderately fast kinetics and a considerable electrocatalytic performance in the ORR. CuPA and CoPA exhibit higher Tafel slopes of 127 and 114 mV dec⁻¹, respectively, indicating slightly slower kinetics and different rate‐determining steps compared to MnPA and NiPA. That is probably related to the poorer electrocatalytic activity of copper and cobalt phosphides. The stability of MnPA, identified as the most active catalyst for ORR, was assessed by 750 consecutive LSV cycles in O₂‐saturated 0.1 M KOH (Figure S14). Despite a slight loss of activity in the first 250 cycles, the polarization curves showed negligible changes in onset potential and limiting current density, demonstrating excellent durability under alkaline ORR conditions. The ORR activity achieved by MnPA (E _ONSET value of 0.85 V vs. RHE in alkaline media) is close to that previously reported by MnP‐based electrocatalysts [12, 56, 57, 58]. Higher activities have been reported for Mn single atoms, or Mn‐doped nanoparticles or Mn–N–C systems [59, 60, 61], benefiting from enhanced electronic structure and optimized surface chemistry. In this context, the present results place MnPA among the most efficient Mn‐based ORR catalysts reported to date, especially considering the simplicity and sustainability of the synthesis route.

To evaluate the electrocatalytic performance of MPA materials in the context of the OER, LSV experiments were conducted in a nitrogen‐saturated alkaline solution (1 M KOH). The results of these experiments, along with the calculated E _OER and the Tafel slopes, are depicted in Figure 4d–f. The highest E_OER value for the transition metal–phosphorus materials was displayed by NiPA, with an E _OER value of 1.62 V vs. RHE, overpassing, by far, the E _OER of the commercial IrO₂ (E _OER value of 1.74 V vs. RHE). In contrast to the ORR behavior, the OER performance does not appear to be directly governed by the specific surface area. Despite its much lower A _BET value, NiPA exhibits superior OER activity, suggesting that the intrinsic activity of Ni‐based species plays a more dominant role than textural properties in this reaction. To evaluate the effect of thermal treatment, additional catalytic tests were performed on NiPA samples prepared at 800 and 1000 °C (Figure S15), which predominantly contain phosphate and phosphide phases, respectively (see again Figure S10). In both cases, the E _OER values were lower than that of NiPA, highlighting that the presence of a single phase alone is less favorable for OER [62, 63]. However, given the multiphase nature of these materials, it is not possible to unambiguously assign the catalytic activity to a single active site. Nevertheless, the superior performance of NiPA strongly suggests that nickel phosphides are the main contributors to OER activity [64, 65], with Ni₁₂P₅ likely playing a particularly relevant role due to its higher proportion in this sample. At the same time, the presence of phosphate phases should not be overlooked. However, although the coexistence of phosphide and phosphate phases may be correlated with the enhanced OER performance, no direct evidence of a synergistic interaction between these phases can be firmly established. Therefore, the improved activity of NiPA is more appropriately attributed to its multiphase composition and overall structural characteristics rather than to a specific cooperative effect.

The electrocatalytic behavior of the Ni‐based phosphide is consistent with the general trend reported for Ni phosphides in OER [66, 67, 68, 69, 70]. These materials are widely recognized as effective nonprecious metal electrocatalysts, with their activity often attributed to the formation of active Ni sites and the favorable adsorption of reaction intermediates. The results obtained for our Ni phosphide sample confirm the relevance of Ni–P phases as OER catalysts and align with the intrinsic catalytic properties observed in this class of compounds. The same activity than IrO₂ was observed in CoPA, exhibiting an E _OER value of 1.74 V vs. RHE. In contrast, both MnPA and CoPA showcased negligible activities.

Tafel slopes shown in Figure 4f provide critical insights into the electrocatalytic properties of the MPA materials. IrO₂, a commercial benchmark, exhibits a Tafel slope of 239 mV dec⁻¹, serving as a reference point for comparison. CuPA and MnPA display a considerably high Tafel slope of 940 and 1216 mV dec⁻¹, respectively, potentially due to higher energy barriers for the electrochemical reaction for those materials, as observed from the negligible electrocatalytic activity. In contrast, CoPA exhibits a relatively low Tafel slope of 263 mV dec⁻¹, while NiPA exhibits a much lower Tafel slope of 152 mV dec⁻¹, indicating fast reaction kinetics.

The stability of NiPA, the best‐performing OER catalyst, was examined by 250 continuous LSV cycles in 1 M KOH (Figure S16). The E _OER value initially decreases up to 1.58 V vs. RHE after 50 cycles, suggesting an activation process during the early stages of operation. This initial improvement in activity suggests progressive surface reconstruction of the Ni‐based active sites under anodic conditions, likely associated with the formation of highly active oxidized nickel species, such as NiOOH or NiO₂. However, a rapid loss of electrochemical activity was observed thereafter, with complete deactivation occurring after 250 cycles. Postreaction CV curves (Figure S17) remain essentially unchanged compared to the initial ones, suggesting that the overall structure of the catalyst is preserved. Therefore, the observed loss in activity is likely associated with changes at the active site level rather than with a complete degradation of the material. Additionally, the relatively low carbon content and limited surface area of NiPA may also contribute to its reduced stability. In contrast to MnPA, the less developed carbon matrix in NiPA could result in a lower degree of structural confinement of the active phases, potentially facilitating their degradation or restructuring under OER conditions. Overall, these findings highlight the dynamic nature of Ni‐based catalysts under OER conditions and emphasize the intrinsic trade‐off between high activity and long‐term durability. They also underscore the importance of stabilizing Ni active phases, for instance, through the incorporation of additional transition metals [71, 72], as a key strategy for the rational design of more robust and durable OER electrocatalysts. Regarding the activity of NiPA, it is not far from the values typically reported for NiP‐based electrocatalysts in alkaline media, which commonly reach potentials around ∼1.50 V vs. RHE [73, 74, 75]. Although the performance of NiPA does not match the best NiP systems reported in the literature, the performance of NiPA can be considered competitive within the NiP‐based OER literature, particularly taking into account the sustainable synthesis route and the absence of additional dopants or complex nanostructuring strategies.

4. Conclusion

The synthesis of noble metal‐free electrocatalysts through ball‐milling and pyrolysis of phytic acid (PA) and transition metals (M: Co, Mn, Ni, Cu) has resulted in metal phosphides or phosphates. The nature and content of these phases were found to be intricately related to the metal employed during synthesis, resulting in distinct textural characteristics and differing chemical properties. The resulting catalysts, MPA, displayed competitive electrocatalytic performance. Specifically, MnPA exhibited a remarkable E _ONSET of 0.85 V vs. RHE for the ORR, favored by the high carbon content in the material, which contributes substantially to its electrical conductivity and large surface area. On the other hand, NiPA, in which different Ni phosphide phases are identified, emerged as a top performer in the OER, showing an E _OER of 1.62 V vs. RHE. Although the OER activity of NiPA does not surpass the best‐performing Ni‐based catalysts reported in the literature, it remains within the same performance range, while being obtained through a simpler and more sustainable synthesis route. Our comprehensive characterization of the MPA materials has offered valuable insights into the influence of different transition metals on the composition and textural attributes of the resulting MPA catalysts. These intriguing findings open up exciting prospects for the development of novel ORR and OER electrodes based on phytic acid, a green phosphorus source, advancing the field of sustainable electrocatalysis.

Funding

This work was supported by the Agence Nationale de la Recherche (LUE, ANR‐15‐IDEX‐04‐LUE), the European Regional Development Fund (2019‐000214), the Ministerio de Universidades (MU‐21‐UP2021‐030 30267158), and Universidad de Oviedo (MU‐21‐UP2021‐030 30267158).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.