Electroreduction of Nitrogen on Pd, Rh, and PdRh Catalysts: An Online Mass Spectrometry Study

Abstract

The nitrogen electroreduction reaction (NRR) has emerged as a promising and sustainable alternative to the Haber–Bosch process for NH₃ production. This study investigated the NRR in alkaline medium using Pd/C, Rh/C, and PdRh/C electrocatalysts, employing online electrochemical mass spectrometry (OLEMS) for gaseous-product detection and ultraviolet–visible spectroscopy to confirm NH₃ formation. To our knowledge, no previous reports have simultaneously detected H₂, N₂H, and N₂H₂ intermediates and monitored N₂ consumption as a function of applied potential for Pd and Rh catalysts. The bimetallic PdRh/C catalyst showed superior NRR performance compared with the monometallic catalysts, exhibiting higher faradaic charges, more pronounced generation of nitrogen intermediates, and selectivity for NH₃. This work provides key insights into the NRR mechanisms and underlines the strategic importance of the bimetallic catalyst design for more efficient, sustainable electrochemical NH₃ synthesis.

1. Introduction

Ammonia (NH₃) is a widely used compound, primarily as a source of nitrogen in fertilizers. Additionally, it stands out as a promising energy vector due to its high hydrogen content (17.8 wt.%) and high energy density (4.32 kW h L⁻¹), with annual production reaching approximately 200 million metric tons [1,2]. An alternative to the Haber–Bosch process for NH₃ production is the electroreduction of nitrogen gas (N₂) to NH₃ (the nitrogen electroreduction reaction [NRR]), which can be carried out at room temperature and atmospheric pressure. When powered by renewable energy sources, this process can avoid carbon dioxide (CO₂) emissions, offering a more sustainable pathway for NH₃ synthesis. However, the NRR faces several challenges, such as the high stability of N₂, the low solubility of N₂ in aqueous systems (approximately 0.66 mmol L⁻¹), the slow reaction kinetics, and the hydrogen evolution reaction (HER), which competes with the NRR for the catalytic surface [3]. In aqueous systems, the HER often competes directly with the NRR, as both processes involve proton-coupled electron transfer steps. The HER tends to dominate due to its more favorable thermodynamics and faster kinetics. Compared with non-aqueous systems, which offer higher N₂ solubility and reduced HER competition [4], aqueous electrolytes have been widely used in recent NRR studies due to their practicality and alignment with environmentally friendly approaches.

Noble metal-based catalysts, such as gold (Au), platinum (Pt), ruthenium (Ru), palladium (Pd), and rhodium (Rh) [4,5,6,7,8,9], have been widely used in the NRR, primarily due to their highly active surfaces, high conductivity, and ability to form bonds with various chemical species [5]. Among these metals, Rh stands out for presenting, according to density functional theory (DFT) calculations, a predicted performance close to the top of the volcano curve, indicating high catalytic activity in the initial step of N₂ activation [6]. Furthermore, Rh-based catalysts with different morphologies and supports have been widely explored for NRR [7,8,9,10,11,12].

Zhang et al. [11] investigated the NRR by using Rh-based nanoparticles (NPs) and evaluated the effect of the support on catalytic activity. NPs supported on carbon nanotubes (CNTs) showed better performance, which was associated with the interaction between Rh NPs and CNT, with an NH₃ yield of 26.91 µg h⁻¹ mg_cat⁻¹, compared with unsupported NPs (12 µg h⁻¹ mg_cat⁻¹) and bulk Rh (4 µg h⁻¹ mg_cat⁻¹) at −0.1 V vs. reversible hydrogen electrode (RHE) in 0.1 mol L⁻¹ phosphate-buffer solution (PBS).

Shen et al. [12] used single Rh atoms supported on manganese dioxide (MnO₂). The NH₃ production rate was evaluated both in diluted electrolyte (0.1 mol L⁻¹ K₂SO₄) and in a concentrated medium of 9 mol L⁻¹ K₂SO₄, known as a water-in-salt electrolyte (WISE). The catalyst showed an NH₃ yield of 271.8 µg h⁻¹ mg_cat⁻¹ at −0.4 V vs. RHE in the WISE, significantly higher than that obtained in the diluted electrolyte (50 µg h⁻¹ mg_cat⁻¹). The authors attributed this improved performance to the suppression of the HER provided by the solvation effects of the WISE, which favors N₂ enrichment at the catalyst active sites and consequently promotes subsequent NRR steps. In addition to NH₃ quantification, the authors investigated the interaction between N₂ and the catalytic system by using Fourier-transform infrared (FTIR) and Raman spectroscopy in the WISE, observing signals between 1100 and 1500 cm⁻¹, attributed to N–N stretching and intermediates containing –NH_x,ads groups (N–H stretching, –NH₂ wagging, and H–N–H bending, among others). Additionally, Raman spectroscopy revealed a band at approximately 480 cm⁻¹, attributed to the formation of a Rh–N bond, indicating the adsorption and possible activation of the N₂ molecule on the catalyst surface.

Yao et al. [9] investigated the NRR on Rh surfaces using surface-enhanced infrared absorption spectroscopy (SEIRAS) and differential electrochemical mass spectrometry (DEMS) in 0.1 mol L⁻¹ KOH as the electrolyte. SEIRAS measurements identified N₂H_x species (0 ≤ x ≤ 2), with bands between 1997 and 2036 cm⁻¹. They observed the N=N stretching mode at approximately 2020 cm⁻¹ in the potential range of 0.2 to −0.4 V vs. RHE. DEMS detected m/z 2 (H₂⁺) and m/z 29 (N₂H⁺) signals, with the latter detected due to the drag of the species by hydrogen gas (H₂) bubbles at more negative potentials. The m/z 30 (N₂H₂⁺) and m/z 31 (N₂H₃⁺) signals were not detected.

Another standout material for the NRR is Pd due to its ability to stabilize reaction intermediates, such as the adsorbed N₂H_ads species, often identified as the rate-determining step in NH₃ formation [13,14,15]. Despite this favorable characteristic, Pd-based catalysts generally exhibit low NH₃ production rates [16,17]. Wang et al. [16] investigated the NRR using Pd/C NPs in 0.1 mol L⁻¹ PBS and reported an NH₃ yield of 4.5 µg h⁻¹ mg_cat⁻¹ at −0.1 V vs. RHE. Based on DFT calculations, the authors identified the formation of the adsorbed N₂H_ads species as the rate-limiting step in the process.

Xu et al. [13] investigated the NRR by using Pd-based catalysts, evaluating the performance of nanoporous palladium (np-Pd) and its hydride-modified form (np-PdH) in 0.1 mol L⁻¹ PBS as the electrolyte. The np-PdH catalyst showed superior performance, with an NH₃ yield of 20.4 µg h⁻¹ mg_cat⁻¹ at −0.15 V vs. RHE, compared with np-Pd, which achieved 7.5 µg h⁻¹ mg_cat⁻¹ under the same conditions. Based on DFT calculations, the authors attributed the catalytic improvement to hydrogen doping in Pd, which favors N₂ activation and conversion to N₂H_ads, which is generally considered the rate-limiting step [16] in NH₃ formation. The authors investigated the interaction between np-PdD and N₂ using FTIR and Raman spectroscopy. The results revealed a band at 1439 cm⁻¹, attributed to H–N–H bending, which, according to the authors, indicates NH₃ formation. Additionally, they observed three low-intensity bands at 1177, 1241, and 1325 cm⁻¹, attributed to the bending modes of the NH₂D₂⁺ and NH₃D⁺ intermediate species. Based on these findings, the authors suggested that deuterium atoms present in Pd-D directly participate in the formation of NH₃. Raman spectroscopy identified a peak at 1644.4 cm⁻¹, attributed to overlapping H–N–H/H–O–H bending, whose intensity increased with reaction time, suggesting progressive NH₃ adsorption on the catalytic surface.

Yang et al. [18] recently reported the combination of Pd and Rh. They investigated the NRR by using boron-doped PdRh mesoporous nanotubes (B-PdRh MNTs) as the catalyst, evaluating their efficiency in converting N₂ to NH₃ in 0.1 mol L⁻¹ Na₂SO₄ as the electrolyte. The authors reported a low NH₃ yield of 12 µg h⁻¹ mg_cat⁻¹ at a potential of −0.2 V vs. RHE, but the material demonstrated high electrocatalytic stability.

Based on the studies reported to date on Rh and Pd catalysts, the NRR is predominantly associated with the associative mechanism, as supported by theoretical studies [11,13,16,19,20]. Although these theoretical models offer important insights, there is a lack of experimental evidence to confirm or refine the proposed reaction pathways on Rh and Pd surfaces. Recently, we investigated the NRR by using molybdenum disulfide (MoS₂) electrodes in an alkaline medium with both labeled and unlabeled compounds using online electrochemical mass spectrometry (OLEMS) and ultraviolet–visible (UV–Vis) spectroscopy [21]. Based on experimental data, we discarded certain reaction steps frequently proposed in DFT-based theoretical studies. Our utilization of isotopic labeling and OLEMS allowed us to identify the N₂H⁺ and N₂H₂⁺ species as the primary NRR intermediates, correlating NH₃ generation with N₂ consumption as a function of the applied potential. Despite these advances, experimental studies specifically focusing on Rh and Pd-based catalysts remain scarce, particularly those capable of directly identifying surface intermediates under operating conditions.

Herein, by combining OLEMS and UV–Vis spectroscopy, we analyzed Rh/C, Pd/C, and PdRh/C electrocatalysts for the NRR in an alkaline medium. We aimed to identify key intermediates and to elucidate the reaction pathways in these catalysts. We hypothesized that the association of Pd with Rh enhances NRR performance. Our findings contribute to the rational design of more efficient and sustainable electrocatalysts for NH₃ synthesis under mild conditions.

2. Materials and Methods

2.1. Polyol-Based Synthesis

The PdRh NPs were synthesized by a modified polyol method [22,23]. The synthesis was performed in a three-mouth flask equipped with a reflux column, with the addition of the precursors Pd(acac)₂ and Rh(acac)₃, together with 1,2-hexadecanediol and dioctyl ether as a solvent under an argon atmosphere. The mixture was heated to 290 °C with the addition of oleic acid and oleylamine, and then refluxed for 30 min. After cooling, the NPs were purified, supported on Vulcan carbon (10 wt.%) by stirring for 12 h, vacuum-filtered, and dried at 80 °C.

2.2. Physical Characterization

The physical characterization of the electrocatalysts was carried out by X-ray diffraction (XRD, D8 Advance, Bruker, Hardtstrabe, Karlsruhe, Germany), transmission electron microscopy (TEM, JEM-2100 JEOL, Peabody, MA, USA), energy-dispersive X-ray spectroscopy (EDX, LEO-440, Cambridge, Cambridgeshire, UK), and thermogravimetry (TG, METTLER Toledo, Colombus, Ohio, USA). XRD measurements were performed using Cu Kα radiation (λ = 0.15406 nm), with a 2θ range from 20° to 100° and a scanning rate of 1° min⁻¹. The diffraction patterns were analyzed using the Crystallographica Search-Match software (version 2.1.1.1) to identify the crystalline phases. Transmission electron micrographs were acquired at an acceleration voltage of 200 keV. Particle size distributions were determined by measuring particle diameters using the ImageJ software (version 1.4.3.x.). TG analysis was conducted under a synthetic air atmosphere with a gas flow rate of 50 mL min⁻¹, a heating rate of 10 °C min⁻¹, and a temperature range of 30 to 1000 °C.

2.3. Working Electrode

The NRR was evaluated by using a gas diffusion electrode (GDE) made of carbon cloth, with a geometric area of 1.9 cm², on which the catalyst was deposited. The Rh/C and Pd/C catalysts used were acquired from Fuelcell Store (Rh/C) and ETEK (Pd/C), and the PdRh/C catalyst was synthesized by using the modified polyol method. For the preparation of the working electrodes, the catalyst mass was adjusted to obtain a metal loading of 0.5 mg cm⁻². Subsequently, 37 µL of Nafion (5% w/w) and five drops of isopropyl alcohol were added to the material, which was subjected to ultrasonic bath for 15 min. Afterwards, the obtained catalytic ink was transferred to the GDE, which was dried at 80 °C for 1 h.

2.4. OLEMS

The distribution of gaseous products was monitored by OLEMS, which involves the analysis of gases generated during electrochemical polarization. The equipment used was from OmniStar^®, model GSD 301 (Pfeiffer Vacuum, Frankfurt, Hesse, Germany), Prisma QMS 200, operating with an ionization energy of 70 eV, an emission current of 1 mA, and a pressure of 10⁻⁶ mbar. The mass spectrometer capillary was coupled to the outlet of the cathodic compartment of the electrochemical cell. The detailed scheme of the cell can be consulted in Figure S1. The monitored masses, corresponding to the gaseous products, were m/z 2, m/z 28, m/z 29, and m/z 30, attributed to the species H₂⁺, N₂⁺, N₂H⁺, and N₂H₂⁺, respectively. To ensure reproducibility, the OLEMS experiments were repeated multiple times, and the curves shown are representative of the consistent trends observed across all measurements.

2.5. Electrochemical Characterization

A bipotentiostat (PARSTAT 3000A-DX, Princeton Applied Research, Oak Ridge, TN, USA) was used for the electrochemical characterization of the electrocatalysts. For the electrochemical measurements, the catalyst was deposited on the GDE. Cyclic voltammetry was obtained in the potential range of 0.1 to 1.0 V vs. RHE, with a scan rate of 5 mV s⁻¹, in 1.0 mol L⁻¹ sodium hydroxide (NaOH). Chronoamperometry coupled with mass spectrometry (CA-MS) was conducted in an H-type electrochemical cell, separated by a Nafion 115 membrane, using a Hg|HgO (1.0 mol L⁻¹ NaOH) reference electrode and a graphite counter electrode (Figure S1). The dried Nafion 115 membranes were activated by immersion in a hydrogen peroxide (H₂O₂) solution (3% wt.%) at 80 °C for 1 h. Then, the membranes were washed to remove any H₂O₂ residue and subjected to a similar procedure in sulfuric acid (H₂SO₄, 0.5 mol L⁻¹) at the same temperature. Finally, the membranes were stored in Milli Q water for later use.

The measurements were performed in solutions saturated with nitrogen 6.0 (99.9999% N₂) and, as a blank, in a helium 6.0 atmosphere. The applied potentials for Rh/C, Pd/C, and PdRh/C during the CA-MS experiments were 0.0, −0.1, −0.2, −0.3, −0.4, −0.5, and −0.6 V vs. RHE. Between each potential step, polarization at 0.05 V vs. RHE was performed for 600 s to stabilize the catalyst surface before proceeding to the next potential.

2.6. Determination of NH₃

NH₃ was quantified with the indophenol blue method [18,24,25], in which alkaline phenol and sodium hypochlorite react with NH₃ to form the indophenol blue compound, whose concentration is proportional to the NH₃ concentration present in the solution. At the end of the polarization step, 2 mL of the solution from the cathodic compartment was collected for subsequent analysis. The absorbance measurements were performed using a UV–Vis spectrophotometer (UV-2600, Shimadzu, Malaga, Australia).

2.7. Measurement Protocol

In studies involving the NRR, Suryanto et al. [26] highlighted the difficulty of conclusively proving that the detected NH₃ results solely from the electrochemical reduction of N₂, rather than from contamination originating from the gas feed, electrolyte, catalyst materials, or the separation membrane. To address this, NH₃ was initially quantified on the electrolytes, revealing no signs of contamination (Figure S2). Prior to polarization, the system was purged with either N₂ or He gas (blank control) for 60 min to ensure a clean environment. The Nafion 115 membrane was replaced after each experimental run.

3. Results and Discussion

3.1. Characterization of the Physical Properties

Figure 1A1 shows the XRD patterns obtained for the Rh/C, Pd/C, and PdRh/C electrocatalysts. The diffraction patterns are compatible with the Joint Committee on Powder Diffraction Standards (JCPDS) and in agreement with the materials described in the literature [27,28]. The analyses revealed that all materials exhibited a face-centered cubic (FCC) structure. For all catalysts, the peak at 24.5° on the diffractogram can be attributed to the presence of carbon. The observed peak for the Pd/C catalyst at 34° on the diffractogram can be attributed to the presence of PdO·H₂O [28]. TG analysis was performed on the Rh/C, Pd/C, and PdRh/C catalysts, whose mass loss curves are shown in Figure S3. The temperature range from 25 to 260 °C corresponded to the removal of wastewater from the synthesis process. From 280 to 550 °C, there was oxidation of the carbon carrier in the three catalysts. Above 650 °C, with the stabilization of the mass, it was possible to determine the metallic content of Rh, Pd and PdRh present in the samples. TG analysis indicated metal loading of approximately 30% for Rh/C, 20% for Pd/C, and 10% for PdRh/C. EDX indicated an atomic ratio of Pd₅₅:Rh₄₅ for the PdRh/C catalyst, a value close to the expected nominal composition of 50:50 (Table S1).

The cyclic voltammograms of the Rh/C, Pd/C, and PdRh/C in 1.0 mol L⁻¹ NaOH solution, recorded in the potential range of 0.1 to 1.0 V vs. RHE and at a scan rate of 5 mV s⁻¹, are shown in Figure 1A2. The curves show the typical electrochemical behavior of Pd and Rh [29], with regions attributed to hydrogen adsorption and absorption, as well as the formation/reduction of the metal oxide layer. The peaks observed between 0.1 and 0.4 V vs. RHE correspond to hydrogen adsorption/absorption and were used to determine the electrochemical surface area (ECSA) [30] for the Rh, Pd, and PdRh catalysts: 91.0, 56.0, and 46.0 cm², respectively.

The transmission electron micrographs of the electrocatalysts (Figure 1A3) revealed a predominantly spherical morphology, with a homogeneous distribution of the particles on the carbon support and no visible agglomeration. The average particle size was 2.9, 4.5, and 4.1 nm for the Rh/C, Pd/C, and PdRh/C catalysts, respectively. In electrocatalysis, particle sizes in the range of 2 to 5 nm maximize the active area. In addition, all catalysts showed good dispersion. The current was normalized by ECSA to compare the electrocatalytic activity.

3.2. Electrocatalytic NRR Performance

Figure 2 shows the CA-MS results for the Rh/C (Figure 2A), Pd/C (Figure 2B), and PdRh/C (Figure 2C) electrocatalysts. The monitored m/z signals were 2, 28, 29, and 30, corresponding to the species H₂⁺, N₂⁺, N₂H⁺, and N₂H₂⁺, respectively. For all catalysts, measurements carried out in the absence of N₂ (in a He atmosphere) are shown in blue, while the NRR-related signals are represented in wine, orange, and green for Rh, Pd, and PdRh, respectively.

The CA-MS results for the m/z 2 signal, attributed to hydrogen, are presented in Figure 2A1–C1 for Rh, Pd, and PdRh, respectively. For all catalysts, in the presence and absence of N₂, H₂ was produced, with onset at 0.0 V. Additionally, there was an increase in H₂ production as the applied potential increased, indicating that all catalysts are active for the HER.

The m/z 28 signal, attributed to nitrogen, exhibited a consumption band starting from 0.0 V vs. RHE for the Rh and PdRh catalysts (Figure 2A2,C2) and from −0.1 V for Pd (Figure 2B2), indicating anticipation for the NRR in the Rh-containing catalysts. This anticipation may indicate that the initial step of adsorption and activation of N₂ on the catalyst surface at a low potential is favored [6,12]. All materials showed an increase in consumption as the potential increased in the cathodic direction.

The CA-MS data for the m/z 29 (N₂H⁺) and m/z 30 (N₂H₂⁺) signals are presented in Figure 2A3,4–C3,4 for Rh, Pd, and PdRh, respectively. Consistent with the onset of N₂ consumption, N₂H and N₂H₂ were detected starting from 0.0 V vs. RHE for Rh and PdRh and from −0.1 V vs. RHE for Pd. Yao et al. [9] detected the m/z 29 signal (N₂H⁺) on Rh surfaces starting from a potential of −0.7 V vs. RHE, and they did not detect the m/z 30 signal (N₂H₂⁺). Shen et al. [12] observed the adsorbed N₂ and NH_xads species in the FTIR spectra at a potential of −0.4 V vs. RHE for Rh₁/MnO₂ catalysts, while the Raman spectra showed a vibrational band at 480 cm⁻¹ that was attributed to the Rh–N interaction.

The detection of H₂, N₂H, and N₂H₂, with N₂ consumption as a function of the applied potential on the Rh, Pd, and PdRh surfaces, is one of the main highlights of this work, given the scarcity of published studies that have employed coupled techniques to monitor NRR products and intermediates as a function of potential. In the absence of N₂, the signals related to the NRR species were not detected, as expected, confirming the purity of the reagent gases, electrolytes, and system cleanliness [26].

Figure 3(1–5) present the integrated charges as a function of the applied potential for the Rh/C, Pd/C, and PdRh/C electrocatalysts, obtained from the CA and CA-MS (Figure 2A–C), to allow comparison of the electrocatalytic activity based on the OLEMS data. Among the evaluated materials, the PdRh/C catalyst, followed by Rh/C, exhibited the highest faradaic charge, suggesting enhanced electrochemical activity at the investigated potentials for both the NRR and HER.

Figure 3(2) shows an increase in the integrated charge associated with H₂ production with the application of more negative potentials, with a higher charge observed for the PdRh/C catalyst. The integrated charges associated with N₂ consumption (Figure 3(3)) also indicated greater consumption for the PdRh catalyst, followed by the Rh and Pd catalysts. These results suggest that PdRh exhibits higher activity toward both the HER and the NRR, corroborating the faradaic charge data obtained based on CA.

The integrated charges related to m/z 29 (N₂H⁺) and m/z 30 (N₂H₂⁺) in Figure 3(4,5), respectively, indicate the formation of NRR intermediates. For the Pd/C catalyst, the integrated charge related to the intermediate species indicated an increase in production. In this potential region, the HER competitive effect was less pronounced, likely due to the later onset of the NRR. Conversely, the Rh/C and PdRh/C catalysts exhibited similar behavior, starting at 0.0 V, with a plateau in intensity from a potential of −0.3 V, reaching its maximum value at −0.4 V, followed by a reduction in intensity up to −0.6 V vs. RHE. This trend suggests that at more negative potentials, the competition with the HER becomes more pronounced, intensifying the consumption of most of the available electrons, occupying active surface sites with adsorbed hydrogen atoms (H*), and destabilizing nitrogen intermediates, thereby reducing the overall catalytic efficiency for NH₃ formation [16]. The production of N₂H and N₂H₂ species was also more pronounced for the PdRh catalysts between −0.3 and −0.4 V, indicating that this is the most favorable potential range for NRR.

Finally, to evaluate and confirm NH₃ production by using UV–Vis spectroscopy, CA was performed in 1.0 mol L⁻¹ NaOH at a potential of −0.4 V vs. RHE for 1 h on the Rh/C, Pd/C, and PdRh/C electrocatalysts (Figure S2). The Rh/C catalyst showed NH₃ production of 26 µg h⁻¹ mg_cat⁻¹. These results can be compared with those obtained by Zhang et al. [11] for Rh NPs supported on CNTs (Rh/CNT) in 0.1 mol L⁻¹ KOH. The authors reported a production rate of approximately 7.2 µg h⁻¹ mg_cat⁻¹ at a potential of −0.4 V vs. RHE, indicating that the NH₃ production obtained for Rh/C was superior at the same potential.

At −0.4 V vs. RHE, the Pd/C catalyst achieved an NH₃ yield of 3 µg h⁻¹ mg_cat⁻¹. Wang et al. [16] reported similar results in their investigation of the NRR using the Pd/C catalyst. They observed the highest performance at a potential of −0.05 V vs. RHE in 0.1 mol L⁻¹ NaOH, with an NH₃ yield of 1 µg h⁻¹ mg_cat⁻¹. In another study, for np-Pd in 0.1 M PBS at −0.15 V vs. RHE, the NH₃ yield was 7.5 µg h⁻¹ mg_cat⁻¹ [13].

The PdRh/C electrocatalyst had an NH₃ yield of 64 µg h⁻¹ mg_cat⁻¹ at −0.4 V vs. RHE. This is notably higher than the yield of 12 µg h⁻¹ mg_cat⁻¹ reported for boron-doped PdRh mesoporous nanotubes at −0.2 V vs. RHE in 0.1 mol L⁻¹ Na₂SO₄ [18].

The NH₃ faradaic efficiency for all catalysts at a potential of −0.4 V vs. RHE presented the order PdRh/C (5.7%) > Rh/C (4.9%) > Pd/C (0.9%), which corroborates the results obtained for CA-MS (Figure 3(3–5)). The superior performance of the Rh and PdRh catalysts, associated with the reduction in the NRR onset potential compared with the Pd/C catalyst, observed with the OLEMS results for intermediate species at low potentials, is consistent with reports indicating high catalytic activity in the initial step of N≡N bond weakening [15]. The Pd/C electrocatalyst exhibited limited faradaic efficiency, consistent with the literature [31], despite its ability to stabilize key intermediates during the NRR (*N₂H_x) [13,16], likely due to competition with the HER.

In summary, the combination of Pd and Rh enhanced NRR activity: Rh serves as the primary site for N₂ adsorption and activation, while Pd facilitates proton transfer and provides reactive hydrogen species (*H) for intermediate hydrogenation steps. Additionally, electronic interactions between Pd and Rh can modify the adsorption energies of intermediates, thereby optimizing both activation and desorption processes [6,7,10,12,15]. Thus, the Pd–Rh bimetallic system represents a promising approach for developing efficient and selective NRR electrocatalysts.

It was possible to qualitatively identify the formation of N₂H and N₂H₂, as well as the consumption of N₂, under the applied potential, and the resulting NH₃ production. The combination of Rh and Pd decreased the onset of the reaction and led to a gain in catalytic activity based on OLEMS and UV–Vis spectroscopy.

Although there is an ongoing debate regarding the active species in the NRR, N₂Hy intermediates (1 ≤ y ≤ 2) are frequently identified as crucial in NH₃ formation. In this study, (i) N₂H⁺ and N₂H₂⁺ were detected as intermediates beginning at 0.0 V vs. RHE for the Rh and PdRh electrocatalysts, and beginning at −0.1 V vs. RHE for the Pd electrocatalyst. (ii) N₂ consumption began at 0.0 V vs. RHE for the Rh and PdRh electrocatalysts, and at −0.1 V vs. RHE for the Pd electrocatalyst. (iii) UV–Vis spectroscopy confirmed the presence of NH₃. (iv) There were no signals corresponding to N₂H₃⁺ or N₂H₄⁺. Moreover, our earlier isotopic labeling experiments supported these findings, showing no trace of N₂H₃⁺ or N₂H₄⁺, and further confirming the absence of N₂H₄ based on UV–Vis spectroscopy [21]. Based on these data, we propose a plausible NRR mechanism occurring on the Pd, Rh, and PdRh electrocatalyst surfaces, as outlined in reactions 1 to 4. In this pathway, N₂ molecules adsorb onto the catalyst and undergo two successive protonation steps, forming N₂H and N₂H₂ at an electrode potential below 0.0 V for the Rh and PdRh electrocatalysts and below −0.1 V for the Pd electrocatalyst. Among these, we suggest that N₂H₂ is the principal intermediate responsible for NH₃ generation.

N₂ + * → *N₂

(1)

*N₂ + H₂O + e⁻ → *NNH + OH⁻

(2)

*NNH + H₂O + e⁻ → *NHNH + OH⁻

(3)

*NHNH + 4H₂O + 4e⁻ → 2NH₃ + * + OH⁻

(4)

*Active site.

4. Conclusions

We demonstrated the effectiveness of combining OLEMS and UV–Vis spectroscopy to investigate the Rh/C, Pd/C, and PdRh/C electrocatalysts for the NRR in an alkaline medium. The simultaneous detection, unprecedented to our knowledge, of the intermediates N₂H⁺ and N₂H₂⁺, as well as the observation of N₂ consumption as a function of the applied potential for these materials, represents a significant advance in elucidating the NRR pathways. The PdRh/C catalyst demonstrated superior electrocatalytic performance, especially at low potentials, showing higher faradaic charges and enhanced formation of nitrogen-related intermediates. Nonetheless, at more negative potentials, competition with the HER intensifies, reducing the number of active catalytic sites and compromising the stability of nitrogen intermediates. At −0.4 V vs. RHE, PdRh/C achieved the highest NH₃ faradaic efficiency among the evaluated electrocatalysts, outperforming both Rh/C and Pd/C, as supported by the CA-MS and OLEMS data. This enhanced activity is attributed to the bifunctional and electronic effects between Rh and Pd: Rh promotes N₂ adsorption and activation, while Pd facilitates hydrogenation steps. Their electronic interaction contributes to the optimization of adsorption energies and reaction kinetics. All tested catalysts followed a mechanism in which N₂ is initially adsorbed onto the surface, followed by successive protonation steps, with N₂H and N₂H₂ acting as key intermediates in the formation of NH₃. The combination of Rh and Pd yields a bimetallic catalyst with synergistic properties, facilitating improved nitrogen activation and enhanced electrochemical performance. These findings underscore the potential of bimetallic systems in advancing efficient and sustainable NH₃ synthesis via electrochemical routes.