Skip to main content
NCBI home page
RSC Advances logoLink to RSC Advances
. 2026 Oct 7. Online ahead of print. doi: 10.1039/d6ra04858g

Ir-based electrocatalysts for facile ethanol oxidation reaction: effect of support, loading, and alloying with Ru

Alaa Soliman a,†, Ahmed S S Elsayed a,†, Shaimaa E Abd El Hamid a,b, Hany A El-Sayed a, Ehab N El Sawy a,✉
PMCID: PMC13644396  PMID: 42846925

Abstract

Direct ethanol fuel cells (DEFCs) offer a compelling pathway for sustainable energy conversion, yet sluggish kinetics and severe surface poisoning of conventional platinum-based anode catalysts hinder their practical adoption. Here, we present a systematic investigation of iridium-based electrocatalysts for the ethanol oxidation reaction (EOR) in acidic media, deconvoluting the independent and cooperative contributions of support identity, metal loading, and Ru alloying within a unified experimental framework. Structural and surface characterization by XRD, TEM, FT-IR, and XPS confirmed successful catalyst synthesis. It revealed systematic changes in the electronic environment of Ir upon support substitution and Ru alloying. ATO-supported Ir exhibited lower EOR activity than Ir/C, while increasing the Ir loading to 17–33 wt% partially recovered the catalytic performance. Ru alloying on carbon produced the strongest enhancement in the low-potential region, shifting the EOR onset potential to ∼0.08 V vs. RHE and delivering ∼25–30 A gIr−1 at 0.1 V. Benchmarked against commercial Pt, all Ir-based catalysts exhibit onset potentials 300–400 mV earlier and peak potentials 150–200 mV lower than Pt. These results establish electronic, geometric, and bifunctional descriptors as an integrated design strategy for next-generation DEFC anode catalysts, positioning IrRu/C as the superior candidate among the series investigated.

Direct ethanol fuel cells (DEFCs) offer a compelling pathway for sustainable energy conversion, yet sluggish kinetics and severe surface poisoning of conventional platinum-based anode catalysts hinder their practical adoption.graphic file with name d6ra04858g-ga.webp

1. Introduction

Direct ethanol fuel cells (DEFCs) have garnered significant attention as electrochemical energy-conversion devices that can utilize ethanol directly as a liquid fuel, especially when the ethanol is produced from renewable resources.1–4 This interest stems from ethanol's advantages, including its storage safety, ease of transportation in liquid form, relatively low toxicity, and potential renewability when produced from biomass.5 However, this attention has recently been hindered by the lack of active, cost-effective anode catalysts. The ethanol oxidation reaction (EOR) is a complex process involving multiple intermediates, including acetaldehyde, acetic acid, carbon monoxide, and other dehydrogenation products.1,6 There are two main pathways (C1 and C2) through which ethanol oxidation can occur.7 In the C1 pathway, ethanol completely oxidizes to CO2 through one carbon-based intermediate (COad and CHx species), formed through C–C bond cleavage and transferring twelve electrons. However, selectivity toward this pathway requires high temperatures because of sluggish kinetics and the high activation energy needed for C–C bond cleavage.8–11 In the C2 pathway, ethanol undergoes partial oxidation to two carbon-based intermediates, such as acetaldehyde/acetic acid (or acetate), transferring only two/four electrons without cleaving the C–C bond. Platinum has been widely used as an effective electrocatalyst for EOR because of its high catalytic activity in dissociating small organic molecules (SOMs). However, Pt is susceptible to poisoning by strongly adsorbed CO, acetaldehyde, and other carbonaceous intermediates, especially as its surface oxides form at high potentials. Consequently, researchers have explored alternative electrocatalysts with high catalytic activity and poisoning resistance to achieve high EOR rates in DEFCs.12–14 Accordingly, other noble metals that is such as Ir, Ru, and Rh have been considered as model and functional electrocatalyst materials to significantly increase activity, improve poisoning resistance, and possibly improve metal utilization.15–19

Extensive theoretical and experimental efforts have focused on improving catalytic performance, selectivity, and electrocatalyst design for ethanol oxidation.1,20–22 Density functional theory (DFT) studies have demonstrated that, among various pure metals, iridium exhibits the highest stability against dissolution in acidic environments.23 Ir is particularly attractive for EOR not only because of its high stability in acidic media but also because it strongly resists poisoning by carbonaceous intermediates; it supplies active hydroxyl species more readily than Pt at lower potentials, despite stronger CO adsorption.15,24,25 Ir performance improves significantly when alloyed with oxophilic or electronically modifying components such as Sn, Ru and Pt.12,26 In Ir–Ru systems, Ru can facilitate the formation of oxygenated surface species at relatively low potentials, while alloying can also modify the electronic environment of Ir. These combined effects make Ir–Ru catalysts promising candidates for improving EOR activity and poisoning resistance.1,27–30

To achieve cost-effective, efficient electrocatalysts, suitable catalytic supports are essential for enhancing nanoparticle dispersion. Key characteristics of effective electrocatalyst supports include strong interactions with the nano-catalyst, high electrical conductivity, affordability, and corrosion resistance.31,32 Furthermore, the support material must remain stable under operating conditions (acidic or alkaline electrolytes), have a high surface area, and prevent agglomeration of catalyst nanoparticles. Carbon black is among the most widely used electrocatalyst supports due to its high surface area, low cost, and excellent electrical conductivity. Despite these advantages, carbon supports can undergo electrochemical oxidation during high-potential excursions or long-term operation, leading to loss of catalyst-support contact, nanoparticle agglomeration, and changes in catalyst-layer wettability, which can ultimately cause water flooding, gas-transport limitations, and performance losses in fuel cells.31 As a result, alternative support materials with higher corrosion resistance and larger surface areas are being explored to replace carbon black in electrocatalysts. One of these support materials is antimony-doped tin oxide (ATO), a metal-oxide support that offers improved corrosion resistance compared with carbon and higher electronic conductivity than undoped SnO2, although its conductivity and surface area remain strongly dependent on synthesis conditions.33

Although Ir-based catalysts and the individual effects of support, metal loading, and alloying have been widely investigated for EOR, comparatively few studies have examined these catalyst-design parameters together within a common experimental framework. In this investigation, a series of Ir-based catalysts was prepared and evaluated for EOR. Carbon black and antimony-doped tin oxide (ATO) supports were compared to assess their influence on catalyst dispersion and electrochemical behavior. The effect of Ir loading was then examined to evaluate the balance between catalytic activity and metal utilization. Finally, Ir–Ru catalysts were investigated to assess whether Ir alloying can enhance activity and poisoning tolerance. Together, these comparisons provide insight into how support material, metal loading, and alloy composition govern the EOR performance of Ir-based electrocatalysts.

2. Materials and methods

2.1. Materials

Iridium(iii) chloride hydrate (IrCl3·xH2O, Alfa Aesar, 99.8%), ruthenium(iii) chloride (RuCl3, Alfa Aesar, 99.8%), Vulcan Carbon (VC, XC-72), antimony-doped tin oxide (ATO) prepared at the Technical University of Munich (TUM)34 ethanol (EtOH, LanXESS, ≥99.8%), sodium hydroxide (NaOH, Loba Chemie, 97%), and sulfuric acid (H2SO4, Sigma-Aldrich, 98%).

2.2. Synthesis of antimony tin oxide (ATO)

ATO was prepared according to the synthesis procedure previously described by Hartig-Weiss et al.34 and used as received. In detail, 2 g of granulated tin and 130 mg of antimony(iii) oxide were added under vigorous stirring to 180 mL of ≈ 2.5 M HNO3 solution, forming a yellowish colloid and releasing NO2 gas. The mixture is then transferred into a PTFE-lined autoclave, sealed, and heated to 140 °C for 10 hours at a ramp rate of 2 °C min−1. After the reaction is complete, the autoclave is allowed to cool to room temperature, and the bluish powder is collected by centrifugation. The collected powder is then washed and filtered with deionized water until the pH reaches 6, then washed with ethanol and dried at 70 °C overnight. The dried powder was calcined in a tube furnace under a continuous flow of an 80/20 argon/oxygen gas mixture at 400 mL min−1, heated to 600 °C at 5 °C min−1, and then held at 600 °C for 3 hours. The agglomerated ATO powder was ground by ball milling at 200 rpm by suspending 2 g of the agglomerated powder in 6 mL of isopropanol and pouring it into a ZrO2 milling jar (45 mL) containing 20 ZrO2 balls (Φ 10 mm). Six milling cycles of 10 minutes each were carried out, with a 1 minute break between cycles. The ground powder was then centrifuged to isolate the product and dried overnight at 70 °C under static air.

2.3. Synthesis of ATO-supported Ir and IrRu

Ir/ATO catalysts (as received from TUM)34 were prepared by depositing pre-synthesized Ir nanoparticles onto the ATO support. Iridium nanoparticles were synthesized via a polyol method. 183 mg of H2IrCl6·xH2O (0.450 mmol) was dissolved in 90 mL of ethylene glycol (EG) and purged with Ar for 60 min, then heated to 140 °C (2 °C min−1) under moderate stirring (500 rpm) to facilitate reduction. After cooling, the ATO support was dispersed in 35 mL of EG by sonication (30 min) and mixed with the required volume of the Ir-nanoparticle suspension. The mixture was stirred in air at room temperature for three days to ensure deposition. The resulting electrocatalyst was recovered by centrifugation at 11 500 rpm (5 °C), washed twice with isopropanol, and dried at 70 °C overnight. IrRu/ATO catalysts were prepared following the same modified alcohol reduction method described for IrRu/C (Section 2.4), with ATO substituted as the support material in place of Vulcan Carbon. Briefly, ATO was dispersed in an ethanol–water mixture (9 : 1 v/v) via sonication for 30 min, and the suspension was heated to ∼80 °C under an N2 atmosphere for 1 h. A metal precursor solution containing IrCl3·xH2O and RuCl3, corresponding to a nominal Ir : Ru molar ratio of 70 : 30 (0.3 mmol total metal precursor; 0.21 mmol Ir and 0.09 mmol Ru), in 10 mL of ethanol–water, was added dropwise to the boiling ATO suspension. The mixture was maintained at 80 °C for 1 h, then 0.9 mmol of NaOH was added to promote reduction. After an additional 30 min, the mixture was cooled, filtered, washed with hot deionized water, and dried at 50 °C overnight.

2.4. Synthesis of Ir and IrRu/C catalysts

Carbon-supported catalysts were synthesized using a modified alcohol reduction method. Vulcan Carbon (VC) was dispersed in an ethanol–water mixture (9 : 1 v/v) via sonication for 30 min. The suspension was heated to boiling (∼80 °C) under an N2 atmosphere for 1 h to promote the formation of reducing aldehyde species. For IrRu/C, a metal precursor solution of IrCl3·xH2O and RuCl3, corresponding to a nominal Ir : Ru molar ratio of 70 : 30 (0.3 mmol total metal precursor; 0.21 mmol Ir and 0.09 mmol Ru), in 10 mL of ethanol/water, was added dropwise to the boiling VC suspension. For Ir/C, only the Ir precursor was employed. The mixture was maintained at 80 °C for 1 h, then 0.9 mmol NaOH was added to accelerate reduction via acetaldehyde formation. After an additional 30 min of reaction, the mixture was cooled, filtered, and washed with hot deionized water. The final product was dried at 50 °C overnight.

2.5. Physicochemical characterization

The resulting powders were characterized using various techniques. Energy-dispersive X-ray spectroscopy (EDS, JEOL JCM-6000 Plus) was employed to confirm the catalyst's atomic composition. Functional group analysis was performed using a Nicolet 380 Fourier-transform infrared (FT-IR) spectrometer with KBr-grade pellets. Thermal stability and metal loading were assessed via thermogravimetric analysis (TGA) using an LTGA-A10 Full Auto (LABTRON) system. Before TGA analysis, carbon-supported samples were heated at 50 °C overnight to ensure complete dryness. Crystallographic information was obtained using X-ray diffraction (XRD) on a Bruker AXS D8 Advance Bragg–Brentano diffractometer. The morphology of the synthesized materials was examined by transmission electron microscopy (TEM) using a JEOL JEM-2100 instrument. Additionally, X-ray photoelectron spectroscopy (XPS) was performed using a K-ALPHA system (Thermo Fisher Scientific, USA) with monochromatic Al Kα radiation, and the data were analyzed with CasaXPS software.

2.6. Electrochemical characterization

Electrochemical measurements were conducted using a Biologic SP-300 potentiostat controlled by EC-Lab software. A three-electrode setup was employed, employing a Hg/HgSO4/0.5 M H2SO4 reference electrode and a titanium mesh counter electrode. The working electrode consisted of 20 µL of catalyst ink deposited onto a 5 mm-diameter glassy carbon electrode, which was then dried at room temperature. The catalyst ink was prepared by suspending 10 mg of the desired catalyst in a solution containing 200 µL of 1% Nafion (in an ethanol/water mixture), 200 µL of absolute ethanol, and 600 µL of 2-propanol, and sonicating for 30 minutes. For each electrochemical experiment, cyclic voltammetry was performed at a scan rate of 20 mV s−1 in 0.5 M H2SO4 to characterize the background surface redox and hydrogen-region behavior, and then in 1 M ethanol + 0.5 M H2SO4 solution to evaluate the electrochemical response of the synthesized materials towards EOR. Both were purged with N2.

Commercial Pt/C (40 wt% Pt, Fuel Cell Store) was used as a benchmark. Currents for the Ir-based catalysts were normalized to the Ir mass, whereas the Pt/C current was normalized to the Pt mass. To compare the susceptibility of the Ir-based catalysts to poisoning under sustained EOR conditions, chronoamperometric measurements were performed at 0.40 V vs. RHE for 600 s in N2-saturated 1 M ethanol + 0.5 M H2SO4. All potentials were converted to the reversible hydrogen electrode (RHE) scale by adding 640 mV to the potential measured versus Hg/HgSO4/0.5 M H2SO4.

3. Results and discussion

3.1. Physicochemical characterization of Ir-based catalysts

Energy-dispersive X-ray spectroscopy (EDS) was employed to determine the composition of the Ir/ATO, IrRu/ATO, and IrRu/C catalysts. Fig. S1 shows the SEM images and EDS spectra of the Ir/ATO with different Ir loadings as an example. For Ir/ATO series, Ir wt% of 10.1, 19.1, and 33.2wt% for the Ir9ATO, Ir17ATO, and Ir33ATO, respectively. For the IrRu-based catalyst, a ratio of 71 : 29 Ir : Ru is obtained, close to the desired ratio of 70 : 30. To confirm the loading of Ir and IrRu on C, TGA was performed on the carbon-supported catalysts, as shown in Fig. S2 for the IrRu/C catalyst as an example. The measured metal loadings were 18% and 19% for the Ir/C and IrRu/C catalysts, respectively, consistent with the expected value of 20%.

To assess the morphological characteristics, spatial distribution, and particle size range of the samples, TEM imaging was used. Fig. 1 shows TEM micrographs for (a) Ir/C, (b) Ir/ATO, (c) IrRu/C, and (d) IrRu/ATO. The images of Ir/C and IrRu/C show loosely aggregated, nearly spherical nanoparticles with a uniform distribution, indicating that the addition of Ru does not significantly alter the morphology but may improve dispersion. On the other hand, Ir/ATO and IrRu/ATO exhibit more densely packed and interconnected nanoparticle networks. The particle size distribution analysis (Fig. S3) revealed that Ir/C exhibits particles in the range of approximately 2–2.5 nm, while IrRu/C shows slightly smaller particles around 1.8–2.2 nm. On the other hand, Ir/ATO and IrRu/ATO samples exhibited particle sizes of approximately 2 and 1.5 nm, respectively.

Fig. 1. High-resolution TEM images for (a) Ir/C, (b) IrRu/C, (c) Ir/ATO, and (d) IrRu/ATO.

Fig. 1

Open in a new tab

To evaluate the crystal structure of the synthesized catalyst, X-ray diffraction was used, with the diffraction patterns for Ir and IrRu-based C and ATO supports shown in Fig. 2a and b, respectively. For Ir/C in Fig. 2a, the X-ray diffraction pattern shows a wide peak at approximately 25°, assigned to the (002) plane characteristic of hexagonal Vulcan carbon XC-72.35 Two additional peaks at 41° and 82° are observed and are attributed to the Ir (111) and (311) planes, respectively, indicating the fcc lattice structure of Ir. The XRD pattern matches well with the reference ICDD card No. 29-0346, corresponding to the face-centered cubic (fcc) structure of iridium. The appearance of only the (111) and (311) reflections suggests preferential growth along specific crystallographic planes, likely induced by the synthesis conditions.36 The diffraction pattern of IrRu/C closely aligns with that of Ir/C, with a noticeable shift of approximately 0.4° in the diffraction peak observed upon introducing Ru, indicating that Ru atoms were successfully incorporated into the Ir lattice, thereby maintaining the fcc crystal structure and possibly forming a solid solution without introducing additional crystalline phases.37,38 The crystallite sizes of both Ir and IrRu were determined from the (111) plane and were found to be 1.2 nm and 1.5 nm, respectively, using the Scherrer equation.39 For the ATO-supported catalysts, Fig. 2b shows both diffractograms of Ir/ATO and IrRu/ATO. Both samples exhibit the same peaks of tetragonal rutile SnO2 in ATO at 2θ = 33.9°, 38°, and 51.8°, which correspond to the (101), (200), and (211) principal planes, respectively. The (111) plane of Ir is observed at 41°, but the strong ATO background masks its signal and makes it difficult to calculate its crystal size.

Fig. 2. XRD diffractograms of (a) Ir/C and IrRu/C and (b) Ir/ATO and IrRu/ATO.

Fig. 2

Open in a new tab

The chemical composition of the synthesized catalysts was analyzed by FTIR spectroscopy, and the resulting spectra of Ir/C, IrRu/C, Ir/ATO, and IrRu/ATO are shown in Fig. 3. All spectra exhibit a broad absorption band around 3400–3500 cm−1, attributed to O–H stretching vibrations from adsorbed water.40 Weak bands at 2920 and 2850 cm−1 correspond to C–H stretching that may arise from the C substrate or adsorbed organic species from air. The absorption band at 1633 cm−1 is assigned to H–O–H bending vibrations of molecular water.40 For Ir/ATO and IrRu/ATO, a strong band is observed near 600–700 cm−1, characteristic of Sn–O–Sn stretching antisymmetric vibrations in the ATO matrix.41 In the lower wavenumber region (400–500 cm−1), bands are assigned to Ir–O and Ru–O vibrations, indicating the presence of metal–oxygen bonds.42 FT-IR results confirm the formation of targeted compositions.

Fig. 3. FT-IR spectra of Ir/C, IrRu/C, Ir/ATO, and IrRu/ATO.

Fig. 3

Open in a new tab

For further analysis, X-ray photoelectron spectroscopy (XPS) was used to elucidate chemical composition, electronic-configuration effects, and oxidation states. Fig. S4a shows the survey scan for (Ir/C, IrRu/C, Ir/ATO, and IrRu/ATO) samples. All final samples showed characteristic Ir peaks, but Ru was difficult to detect because of its low content. The presence of both Sb and Sn in the Ir/ATO and IrRu/ATO samples confirms the successful incorporation of ATO as the catalyst support. The slow-scan spectra of Ir 4f, O 1s, C 1s, and Ru 3p for each sample are shown in Fig. 4, and those of Sb 3d and Sn 3d are shown in Fig. S4b and c.

Fig. 4. XPS slow scan spectra of (a) Ir 4f, (b) O 1s, (c) C 1s, and (d) Ru 3p core levels for the Ir/C, IrRu/C, Ir/ATO, and IrRu/ATO.

Fig. 4

Open in a new tab

The Ir 4f high-resolution XPS spectrum (Fig. 4a) displays two primary peaks assigned to the Ir 4f7/2 and Ir 4f5/2 spin–orbit doublets. Deconvolution of these peaks reveals the presence of two distinct oxidation states, Ir3+ and Ir4+, for each doublet component. The absence of a distinct Ir0 peak suggests a high surface-to-bulk ratio characteristic of small nanoparticles, where surface atoms are predominantly oxidized. Furthermore, the terminal addition of NaOH promotes the formation of a hydrous iridium oxide surface layer that is highly susceptible to reduction, yielding metallic Ir.43,44 Since carbon has minimal support effect and there is no Ru in the Ir/C, its spectrum could serve as a baseline for any induced changes due to alloying or the support effect. The Ir 4f spectrum of the Ir/C sample shows peaks at 61.9 eV (Ir 4f7/2) and 64.8 eV (Ir 4f5/2), corresponding to Ir3+, and at 63.2 eV (Ir 4f7/2) and 66.3 eV (Ir 4f5/2), corresponding to Ir4+.45

Even though these four peaks are observed for every sample, a gradual shift toward lower binding energies was detected after adding Ru, and this shift increased upon switching to an ATO support material, with the most significant shift of 0.5 eV occurring with IrRu/ATO. In the IrRu/C sample, a slight shift of the binding energy to a lower value by 0.12 eV is observed after adding Ru to Ir, indicating an increase in electron density around Ir due to Ir–Ru interaction. This shift might also result from the formation of new Ir–O–Ru bonds, thereby changing the local electronic environment.46–48 Changing the support from carbon to a conductive metal oxide, such as ATO, in the Ir/ATO sample shifted the Ir XPS peaks to lower binding energies relative to the Ir/C sample by 0.28 eV due to the strong metal–support interactions, which alter the Ir electronic structure.49–52 Because core electron binding energies directly reflect modifications in the valence electron environment, core-level XPS shifts provide strong evidence of relative electronic perturbations and inferred d-band shifts among the catalysts, even though the d-band center itself is not measured directly.53,54

Fig. 4b shows the O 1s spectrum for the aforementioned samples. The O 1s spectrum for Ir/C, and IrRu/C shows one broad peak at 531.9 eV. The Ir/C broad peak is deconvoluted into three peaks at 530.5, 531.7, and 533 eV, which correspond to M–O bonding (M = Ir or Ru), a surface OH group or oxygen vacancies, and adsorbed H2O, respectively.55,56 IrRu/C shows the same triplet peaks with a slight shift towards higher binding energies. The Ir/ATO and IrRu/ATO O 1s peaks are deconvoluted into only two peaks at 531.0 and 532.7 eV for Ir/ATO and 531.2 and 532.9 eV for IrRu/ATO, respectively. The two peaks correspond to lattice oxygen (M–O) and adsorbed OH on the surface, respectively. Using lattice oxygen (M–O) as a reference, we observe a gradual shift toward higher binding energies after alloying with Ru and using ATO as the support. This gradual shift in the oxygen peaks correlates with the opposite shift observed in the Ir peaks. These binding-energy shifts make sense, as if Ir gains electron density, the oxygen atoms bonded to it must lose some.

The C 1s spectrum was also studied and presented in Fig. 4c. For Ir/C and IrRu/C, the C 1s spectrum deconvolutes into three peaks at 283.4, 284.6, and 285.3 eV, corresponding to metal carbide (M–C), adventitious carbon from the atmosphere, and the C–O functional group, respectively.57–60 The Ir/ATO and IrRu/ATO only show the central peak for (C–C) adventitious carbon at 284.6 eV, alongside a peak at 286.7 eV for the Ir/ATO, which corresponds to C–OH/C–O–C arising from organic residues from the synthesis process. Fig. 4d shows the Ru 3p3/2 spectrum for the IrRu/C and IrRu/ATO samples. Both samples show two deconvoluted peaks. The IrRu/C peaks at 463 and 467.7 eV correspond to Ru4+ in RuO2 and its satellite shake-up peak, respectively. The Ru spectrum for IrRu/ATO showed the same two peaks, but they were shifted to lower binding energies due to the electronic effect of ATO.

To further confirm the synthesis of ATO, the Sb and Sn slow-scan spectra for Ir/ATO and IrRu/ATO are presented in Fig. S4b and c. Due to the overlap between the Sb 3d and O 1s core-level peaks, the Sb 3d spectrum was fitted by applying constraints on the area ratio and the energy separation (9 eV) between the 3d5/2 and 3d3/2 components. The Sb 3d spectra for both samples show splitting of the Sb 3d5/2 and 3d3/2 peaks, with the two peaks appearing at 531 and 540 eV, respectively. This corresponds to Sb2O3.61 The Sn 3d peaks were deconvoluted into three peaks for each spin–orbit doublet, Sn 3d5/2 and Sn 3d3/2. The three peaks appeared at 485.5/494 eV, 486.4/495 eV, and 487.8/496.1 eV, corresponding to Sn0, Sn2+, Sn4+, respectively.62

Collectively, the physicochemical characterization confirms the successful synthesis of all catalyst series and reveals systematic electronic modifications at the Ir active sites as a function of support identity and Ru alloying. The following section interrogates how these structural and electronic differences manifest in EOR electrocatalytic activity.

3.2. Electrochemical characterization

Fig. 5a shows the cyclic voltammograms of Ir/C, Ir/ATO, IrRu/C, and IrRu/ATO catalysts in N2-saturated 0.5 M H2SO4 at a 20 mV s−1 scan rate. Prior to electrochemical analysis, all electrodes were subjected to 50 consecutive potential cycles between 0.05 and 1.0 V vs. RHE to ensure electrochemical stabilization and bring all catalysts to a consistent, well-defined oxidation/reduction state. As shown, all Ir-based catalysts exhibited a defined hydrogen adsorption/desorption region. Notably, the comparable Hupd charges across all four catalysts indicate similar electrochemically active surface areas (ECSA), justifying normalization of all measured currents to the mass of the Ir catalyst, since they are assumed to share a similar area and allow a direct, meaningful comparison of intrinsic catalytic activities. The oxide-reduction features provide comparative information about the formation and reduction of oxygenated surface species on the different catalysts. Changes in the reduction potential indicate that Ru incorporation and ATO support modification affect Ir surface oxidation behavior. The reduction potentials follow the trend Ir/C > Ir/ATO > IrRu/C > IrRu/ATO. This trend is consistent with differences in how the catalyst surfaces interact with oxygenated species. However, voltammetric peak positions alone do not provide a direct quantitative measure of OH adsorption energies or a unique measure of surface oxophilicity.63–65 The IrRu/ATO catalyst exhibits the most negative reduction potential, indicating that the combined effects of Ru incorporation and ATO-induced electronic modification synergistically stabilize surface oxygenated species.

Fig. 5. (a) CVs of Ir-based electrocatalysts in N2-saturated 0.5 M H2SO4 and (b) corresponding oxide-reduction peak potentials.

Fig. 5

Open in a new tab

The electrocatalytic activity of Ir-based catalysts toward the ethanol oxidation reaction (EOR) is assessed in N2-saturated 0.5 M H2SO4 + 1 M EtOH solution at a scan rate of 20 mV s−1. The following section discusses the effects of Ir support, loading, and alloying on the EOR. The EOR voltammetric profiles for ethanol oxidation represent a complex convolution of three primary mechanistic steps: (i) the dissociative adsorption and dehydrogenation of ethanol into carbonaceous intermediates, (ii) the electrochemical activation of water to generate reactive surface hydroxyls (OH*), and (iii) the eventual formation of a passivating oxide layer (IrOx) at higher potentials.

Fig. 6 illustrates the impact of support on Ir by showing the Ir/C (18 wt% Ir) and Ir/ATO (17 wt% Ir) activities towards the EOR, with differences observed between the two voltammograms. Ir/C achieves a sharp, well-defined peak of 62 A gIr−1 at 0.62 V, whereas Ir/ATO exhibits a significantly suppressed and broadened response of ∼34 A gIr−1, shifting to a more anodic potential of 0.75–0.80 V.

Fig. 6. CVs (forward scan only) of Ir/C and Ir/ATO in 1 M EtOH + 0.5 M H2SO4 solution at a scan rate of 20 mV s−1 and RT, illustrating the support effect.

Fig. 6

Open in a new tab

The lower EOR activity of Ir/ATO relative to Ir/C is accompanied by a negative shift in the Ir 4f binding energy, indicating modification of the local electronic environment of Ir by the ATO support. The corresponding changes in the oxide-reduction features also indicate altered interactions with oxygenated surface species. These observations suggest that metal–support interactions may influence the balance between ethanol-derived adsorbates, oxygenated species, and surface oxide formation during EOR. Such an interpretation is consistent with established electronic and bifunctional concepts reported for alcohol oxidation catalysts.66,67 However, the present XPS and CV measurements do not directly determine OH adsorption energies or identify the adsorbed poisoning intermediates.

Having observed that the ATO support modifies the electronic and electrochemical behavior of Ir and lowers its EOR activity relative to Ir/C, we next investigated whether changing the Ir loading on ATO could recover activity by altering catalyst surface density and interparticle spacing. To study the effect of Ir loading on the EOR, different loadings of Ir were prepared (9 wt%, 17 wt%, and 33 wt%), incorporating ATO support, to enhance the activity of the Ir/ATO catalyst. The electrocatalytic activity of the different loadings of Irx/ATO is illustrated in Fig. 7.

Fig. 7. EOR CVs of Ir/ATO catalysts with different Ir loadings in 1 M EtOH + 0.5 M H2SO4 solution at a scan rate of 20 mV s−1 and RT, illustrating the loading effect.

Fig. 7

Open in a new tab

The EOR mass activity of Ir/ATO catalysts showed a strong dependence on Ir loading, indicating a geometric effect governing the fate of reaction intermediates at the catalyst surface. As shown in Fig. 7, Ir33/ATO achieves the highest peak mass activity (∼38 A gIr−1 at ∼0.60 V vs. RHE), indicating the highest active-site density. This superior activity at a lower onset potential is particularly relevant for direct ethanol fuel cells (DEFC), where maximizing current output at minimal anodic overpotential is critical for overall cell efficiency. In contrast, Ir9/ATO displays monotonically increasing current across the entire studied potential window with no observable passivation peak, while Ir17/ATO exhibits an intermediate behavior with a broader, more gradual peak centered at ∼0.70 V. These trends are interpreted in the context of inter-particle spacing and its influence on the re-adsorption probability of primary oxidation intermediates, as illustrated schematically in Scheme 1. At low Ir loadings (Ir9/ATO), the nanoparticles are well dispersed and spatially isolated on the ATO support, so the mean distance between neighboring Ir sites exceeds the surface diffusion length of desorbing intermediates, namely acetaldehyde. Under these conditions, primary oxidation products efficiently escape into the bulk electrolyte before encountering an adjacent active site, effectively suppressing re-adsorption and the subsequent further dehydrogenation to acetic acid. Although surface poisoning may be minimized, ethanol oxidation efficiency drops dramatically. At intermediate (Ir17/ATO) and high Ir loadings (Ir33/ATO), the increased proximity of neighboring Ir sites increases the probability of re-adsorption of desorbing intermediates before they can escape into the bulk electrolyte. This interparticle intermediate transfer promotes further ethanol dehydrogenation on adjacent sites, leading to further oxidation of the aldehyde to acetic acid.

Scheme 1. Proposed mechanism of EOR during (a) low Ir loading and (b) high Ir loading incorporated with ATO support.

Scheme 1

Open in a new tab

Above 0.6 V, the current decline for Ir33/ATO is attributed to early Ir passivation and reduced susceptibility to ethanol dehydrogenation. At low (Ir9/ATO) and intermediate (Ir17/ATO) loadings, active-site passivation occurs at higher potentials, allowing further ethanol dehydrogenation and acetaldehyde oxidation to acetic acid, resulting in a nearly steady current up to 1 V. The high current density delivered by Ir17/ATO at potentials below 0.60 V vs. RHE (similar to Ir33/ATO), and its steady high current up to 1 V, make it interesting for further studies on poisoning resistance. Ir17/ATO inter-particle spacing balances improved acetaldehyde re-adsorption and oxidation efficiency with enough metallic Ir sites for ethanol dehydrogenation, even at high potentials. Taken together, these results show that Ir surface density on ATO serves as a geometric descriptor of EOR performance, analogous to the Sabatier principle: an optimal loading exists where intermediate desorption and site proximity balance to maximize sustained catalytic turnover. This finding further underscores that catalyst design for EOR cannot be reduced to intrinsic site activity alone; the mesoscale spatial arrangement of active sites is an independent and equally critical parameter governing both activity and durability.

Alloying is a widely adopted strategy in electrocatalysis to enhance the performance of metal-based catalysts by tuning their electronic structure, surface properties, and stability. Introducing a second or third metal can modify the binding strength of key reaction intermediates and promote synergistic interactions that improve catalytic activity and selectivity. In addition, alloying often enhances resistance to poisoning and corrosion, extending the catalyst's lifetime.1,3,16,28 The influence of Ru alloying on the EOR activity of carbon-supported Ir catalysts is presented in Fig. 8a, comparing Ir/C (18 wt% Ir) and IrRu/C (19 wt% metal) with similar metallic loading across the anodic potential sweep. The incorporation of Ru into the Ir lattice produces a marked cathodic shift in EOR onset potential from ∼0.15 V to ∼0.08 V vs. RHE, accompanied by significantly elevated current densities across the low-to-moderate potential region. This enhancement in low-potential activity is consistent with the commonly proposed bifunctional role of Ru. Ru sites, being substantially more oxophilic than Ir, facilitate water activation and OH* generation at potentials as low as ∼0.2 V vs. RHE,3,68 well below the water-activation threshold for monometallic Ir (∼0.6–0.7 V). The formation of oxygenated species on Ru at relatively low potentials may facilitate the oxidative removal of adsorbed ethanol-derived intermediates from neighboring Ir sites via a Langmuir–Hinshelwood pathway, thereby accelerating EOR kinetics at potentials where Ir/C remains largely inactive.69 This bifunctional enhancement is further modulated by an electronic effect arising from Ir–Ru orbital interactions within the alloy. X-ray photoelectron spectroscopy reveals a negative shift of 0.118 eV in the Ir 4f binding energy of IrRu/C relative to Ir/C, indicative of increased electron density on Ir upon Ru incorporation through ligand and strain effects. The improved low-potential response is therefore consistent with combined electronic and bifunctional contributions. However, the present measurements do not independently quantify OH adsorption energies or identify the intermediates removed during EOR.

Fig. 8. EOR CVs comparing Ir and IrRu catalysts supported on (a) carbon and (b) ATO in 1 M EtOH + 0.5 M H2SO4 solution at a scan rate of 20 mV s−1 and RT.

Fig. 8

Open in a new tab

Despite its superior low-potential performance, IrRu/C exhibits a lower peak mass activity (∼49 A gIr−1 at ∼0.55 V) compared to Ir/C (∼62 A gIr−1 at ∼0.60 V). This apparent reversal at moderate-to-high potentials stems from two concurrent effects. First, Ru undergoes progressive surface oxidation to RuOx above ∼0.4 V vs. RHE, passivating the very sites responsible for bifunctional OH* delivery and eliminating their contribution to catalytic turnover in the peak region.70 Second, substituting a fraction of surface Ir sites with Ru introduces a site-dilution effect, reducing the density of primary ethanol-oxidation sites per gram of Ir. Critically, however, IrRu/C maintains substantially higher activity at elevated potentials, retaining ∼32 A gIr−1 at 1.0 V vs. RHE compared to ∼23 A gIr−1 for Ir/C, a ∼39% enhancement.

The combined influence of Ru alloying and ATO support on EOR activity is presented in Fig. 8b, comparing Ir/ATO and IrRu/ATO across the anodic potential sweep. Relative to Ir/ATO, the incorporation of Ru into the ATO-supported catalyst produces a modest cathodic shift in onset potential and an elevated current response in the low potential region, consistent with the bifunctional enhancement established for the carbon-supported IrRu/C system. However, this low-potential improvement is substantially attenuated relative to the Ir/C-to-IrRu/C transition, a distinction with important mechanistic implications for the interplay between support-induced electronic modification and Ru-mediated OH* delivery. As discussed above, the ATO support produces a measurable shift in the Ir 4f binding energy, indicating a substantial modification of the local electronic environment of Ir. IrRu/ATO exhibits the largest negative Ir 4f binding-energy shift among the investigated catalysts, suggesting that both the support and Ru incorporation contribute to this electronic perturbation. The modest low-potential enhancement of IrRu/ATO relative to Ir/ATO is consistent with a possible contribution from Ru-assisted formation of oxygenated species. However, the weaker enhancement than in IrRu/C indicates that the ATO support negatively influences the catalytic response. At moderate to high potentials, IrRu/ATO shows substantially lower activity than Ir/ATO, reflecting several effects on Ir activity, including the Ru partial oxidation state, the density of Ir sites, and support-induced electronic modification of Ir. The present measurements do not allow quantitative separation of these contributions or their individual effects on OH, but they indicate low efficiency of ethanol dehydrogenation to either acetaldehyde or acetic acid. Taken together, the results in Fig. 8a and b suggest that the effect of Ru alloying depends strongly on support identity. The low-potential benefit observed on carbon is less pronounced on ATO, consistent with an interplay between support-induced electronic modification and Ru-related bifunctional effects.

Fig. 9 consolidates the most representative EOR performance curves for each catalyst series, Ir/C, IrRu/C, and Ir33/ATO, alongside commercial Pt/C as the benchmark, enabling a direct comparison of key performance descriptors across the full potential window. The Pt background CV shows surface-oxide formation at more positive potentials than the corresponding Ir-based catalysts. This difference is consistent with the later availability of oxygenated surface species on Pt and may contribute to its later EOR onset.68

Fig. 9. CVs of Ir-based vs. Pt electrocatalysts towards EOR in N2-saturated 1 M EtOH/0.5 M H2SO4 solution. Ir-based catalysts are normalized to Ir wt%, while the Pt catalyst is normalized to its wt%.

Fig. 9

Open in a new tab

In contrast to Pt/C, all three Ir-based catalysts initiate EOR activity at substantially lower potentials, with onsets of ∼0.15 V, ∼0.08 V, and ∼0.10 V vs. RHE for Ir/C, IrRu/C, and Ir33/ATO, respectively, representing an improvement of 300–400 mV relative to Pt/C. The earlier onset is accompanied by differences in the surface-oxide and oxygenated-species features observed in the background CVs. These electrochemical differences are consistent with differences in surface chemistry and susceptibility to deactivation. This comparison highlights the hindering effect of ATO as a support and the enhancing effect of Ru as an alloying element (in the absence of ATO support) on Ir activity, with all Ir-based catalysts showing higher activity than the Pt-based ones.

To further assess the resistance to poisoning and the interplay effects of the support, loading, and alloying with Ru, chronoamperometric (CA) profiles for all Ir-based electrocatalysts were recorded at 0.4 V vs. RHE in N2-saturated 1 M EtOH/0.5 M H2SO4 for 600 s and shown in Fig. 10. The selected potential was deliberately chosen within the practically relevant DEFC operating window identified from the CV analysis. All catalysts display a rapid initial current decrease followed by a slower decline over the 600 s measurement. Such current decay during ethanol oxidation may arise from several contributions, including accumulation of adsorbed intermediates, changes in surface oxidation state, and mass-transport effects. Therefore, the chronoamperometric data are used here primarily as a comparative probe of current retention and susceptibility to poisoning-induced deactivation, rather than to identify specific adsorbed intermediates or establish a unique reaction mechanism.

Fig. 10. Chronoamperometry curves of Ir-based electrocatalysts in N2-saturated 1 M EtOH + 0.5 M H2SO4 solution.

Fig. 10

Open in a new tab

The poisoning-induced current decay for the six catalysts in Fig. 10 follows clear catalyst-dependent trends consistent with the structural and electronic descriptors inferred from the CV data (Fig. 6–8). The chronoamperometric measurements further differentiate the long-term stability of the investigated catalysts and support the loading-dependent behavior proposed from the CV measurements. Within the ATO-supported series, Ir17/ATO shows the highest retained current, reaching approximately 7 A gIr−1 at 600 s, followed by Ir33/ATO at approximately 5 A gIr−1. This behavior is consistent with the loading-dependent geometric effects discussed above: the intermediate Ir loading in Ir17/ATO balances sufficient active-site density with reduced interparticle separation, limiting intermediate production through ethanol dehydrogenation and thus reducing poisoning-species accumulation while maintaining a substantial population of accessible Ir sites. In contrast, the higher site density of Ir33/ATO increases the probability of ethanol dehydrogenation, more acetaldehyde production, and thus further poisoning. Interestingly, Ir9/ATO exhibits the lowest retention within the ATO series, maintaining only approximately ∼2.0 A gIr−1 at 600 s. This behavior indicates the importance of acetaldehyde re-adsorption/oxidation in maintaining Ir surface activity and enhancing resistance to poisoning. In Ir9/ATO, the locally high acetaldehyde concentration leads to high susceptibility to poisoning; thus, optimum stability is observed at the intermediate Ir loading of Ir17/ATO, where the balance between acetaldehyde production and re-adsorption/oxidation is achieved. Thus, the chronoamperometric results provide experimental support for an optimum Ir surface density on ATO, consistent with the geometric model proposed from the CV analysis.

Compared with ATO-based catalysts, Ir/C shows an intermediate degree of current retention, maintaining ∼4.0 A gIr−1 at 600 s, placing it below Ir17/ATO with similar loading. Although Ir/C shows higher activity than Ir17/ATO, it is more susceptible to poisoning. These results, along with the low-activity results, indicate that ATO weakens the adsorption of both ethanol and poisoning species such as aldehydes on Ir, hindering ethanol dehydrogenation while enhancing poisoning-species removal.

In comparison to Ir/C, IrRu/C exhibits the highest current retention throughout the 600 s, maintaining approximately ∼10 A gIr−1, which is 2.5 times higher than that of pure Ir. This sustained activity is consistent with the enhanced poisoning resistance expected from the bifunctional Ir–Ru surface, where oxophilic Ru sites can provide a continuous supply of OH* species that facilitate oxidative removal of strongly adsorbed ethanol-derived intermediates. In contrast, IrRu/ATO shows a pronounced current decay, retaining only approximately ∼2.9 A gIr−1 at 600 s. The substantially different stability of IrRu/C and IrRu/ATO indicates that the beneficial effect of Ru is strongly influenced by the support environment. In IrRu/C, the sustained current results from Ru's bifunctional role in removing poisoning intermediates with minimal support effect. In contrast, in IrRu/ATO, the negative influence of the ATO substrate on ethanol adsorption and dehydrogenation outweighs Ru's beneficial effect.

Overall, the CA results demonstrate that sustained EOR activity is governed by the combined effects of catalyst composition, Ir loading, active-site spatial distribution, and support–metal interactions. In particular, the superior retention of Ir17/ATO compared with both Ir9/ATO and Ir33/ATO experimentally reinforces the loading-dependent optimum identified from the CV results. IrRu/C shows the highest retained current, indicating the lowest susceptibility to poisoning under the conditions investigated. These results complement the CV measurements by showing that the catalyst-dependent trends observed during potential cycling are also reflected in the extent of current loss under sustained polarization.71,72

To place the performance of the present catalysts in the context of the current state of the art, Table 1 compares representative Ir-, IrRu-, Pt-based, and other advanced EOR electrocatalysts reported in the literature, including their onset potential, activity, stability, catalyst loading, electrolyte, and relevant testing conditions where available.

Table 1. Comparison of the EOR performance of the present catalysts with representative Ir-, IrRu-, Pt-based, and other advanced electrocatalysts reported in the literature.

Catalyst Electrolyte/scan rate/temperature Catalyst loading E onset (V vs. RHE) E Peak (V vs. RHE) I (0.5 V vs. RHE) Stability Ref.
Ir/C 0.5 M H2SO4 + 1 M EtOH/20 mV s−1/RT ∼1 mg cm−2 ∼0.15 ∼0.60 50 mA gIr−1 CA at 0.40 V vs. RHE for 600 s; ∼4 A gIr−1 at 600 s This work
IrRu/C 0.5 M H2SO4 + 1 M EtOH/20 mV s−1/RT ∼1 mg cm−2 ∼0.08 ∼0.55 ∼47 mA gIr−1 CA at 0.40 V vs. RHE for 600 s; ∼10 A gIr−1 at 600 s This work
Ir33/ATO 0.5 M H2SO4 + 1 M EtOH/20 mV s−1/RT ∼1 mg cm−2 ∼0.10 ∼0.60 ∼31 mA gIr−1 CA at 0.40 V vs. RHE for 600 s; 5 A gIr−1 at 600 s This work
Electrodeposited Ir 0.1 M HClO4 + 0.1 M EtOH/50 mV s−1/N.R. N.R. ∼0.63 ∼1.06 ∼10 µA cm−2 Not tested 73
Ir70Rh30/C 0.1 M HClO4 + 0.1 M EtOH/50 mV s−1/N.R. N.R. ∼0.63 ∼0.90 ∼20 µA cm−2 Not tested 73
Ir3Sn/C Acidic (H2SO4-based; DEFC single-cell, 90 °C)/50 mV s−1/RT N.R. ∼0.2 ∼0.4 ∼2 mA Rapid current decay followed by stability (1200 s CA at 0.44 V vs. RHE) 74
PtIr/C (20 : 20 at%) 0.5 M H2SO4 + 1 M EtOH/20 mV s−1/25 °C N.R. ∼0.40 ∼0.8 ∼0.001 A cm−2 N.R. 75
PtIrSn/C (20 : 5:15 at%) 0.5 M H2SO4 + 1 M EtOH/20 mV s−1/25 °C N.R. ∼0.30 ∼0.8 ∼0.005 A cm−2 N.R. 75
ND-Ir3Pb/C 0.1 M HClO4 + 0.1 M EtOH/20 mV s−1/N.R. 2 mg ∼0.54 ∼0.79 ∼0.75 mA cm−2 34.3% loss in specific activity after 6000 cycles 76
PtIr/Vulcan (1 : 1, 20 wt%) 0.5 M H2SO4 + 0.5 M EtOH/10 mV s−1/N.R. ∼160 µg cm−2 ∼0.30 ∼0.87 ∼15 mA cm−2 90% of 1st CV Ipeak retained after 100 CVs; after 1 h of CA, I decreased by 15% 77
WO3-modified PtIr/Vulcan 0.5 M H2SO4 + 0.5 M EtOH/10 mV s−1/N.R. ∼160 µg cm−2 ∼0.30 ∼0.87 ∼23 mA cm−2 90% retention after 100 cycles; 15% decrease after 1 h CA 77
PtIr alloy dendritic NPs 0.1 M HClO4 + 0.2 M EtOH/50 mV s−1/25 °C 40 µg/cm2metal ∼0.7 ∼0.9 ∼0.25 mA cm−2 Reported qualitatively as “good electrochemical stability” 78
Ir91Ru9/C 0.5 M H2SO4 + 0.5 M EtOH/30 mV s−1/RT 3.6 µgIr N.R. N.R. ∼58 A gIr−1 N.R. 1
Ir77Ru23/C 0.5 M H2SO4 + 0.5 M EtOH/30 mV s−1/RT 3.3 µgIr N.R. N.R. ∼80 A gIr−1 More than 21× higher mass current density than Pt/C after 2 h at 0.2 V vs. Ag/AgCl 1
Open in a new tab

The values summarized in Table 1 should be compared with caution, as they were obtained under varying experimental conditions across different studies. Variations in electrolyte composition and concentration (e.g., H2SO4vs. HClO4, ethanol concentration), catalyst loading on the electrode, operating temperature, electrode/catalyst preparation methods, and the basis of current normalization (e.g., geometric area, electrochemically active surface area, or catalyst mass) can each significantly influence the reported onset and peak potentials. As such, the data presented here provide a general benchmarking context rather than a strictly quantitative, one-to-one comparison of catalytic performance.

4. Conclusion

This work provides a systematic comparative framework for evaluating Ir-based catalysts for ethanol oxidation in acidic media by investigating three key parameters: support identity, Ir loading, and Ru alloying. Successful synthesis of all catalyst series was confirmed through a complementary suite of structural and surface characterization techniques, including XRD, TEM, FT-IR, and XPS. Critically, XPS analysis revealed shifts in Ir 4f binding energy upon changing the support and alloying with Ru, indicating electronic modification of Ir associated with metal–support and alloying interactions. Ir/C outperformed Ir/ATO under the tested conditions. The lower activity of Ir/ATO is consistent with support-induced electronic modification of Ir and earlier formation of Ir oxide species, which may reduce the availability of active metallic Ir sites or change their ability to actively dehydrogenate ethanol during EOR. On ATO, the EOR response depended strongly on Ir loading, suggesting that particle dispersion and interparticle spacing influence the balance between active-site accessibility, intermediate removal, and oxide formation: an optimal Ir loading balances site proximity and intermediate escape, reducing the likelihood of intermediate accumulation and IrOx-related passivation. Ru alloying shifted the EOR onset to lower potentials, a behavior consistent with cooperative bifunctional and electronic effects. Compared with the Pt benchmark under the present half-cell conditions, the Ir-based catalysts exhibited earlier EOR onset and lower peak potentials. These results demonstrate that support identity, Ir loading, and Ru alloying provide complementary parameters for tuning the EOR performance of Ir-based catalysts.

Conflicts of interest

The authors declare no competing financial interests.

Supplementary Material

RA-OLF-D6RA04858G-s001

Acknowledgments

The authors thank the American University in Cairo for supporting this work.

Data availability

All data supporting this article are included in the article.

Supplementary information is available. See DOI: https://doi.org/10.1039/d6ra04858g.

References

  1. Du W. Deskins N. A. Su D. Teng X. ACS Catal. 2012;2:1226–1231. doi: 10.1021/cs3002308. [DOI] [Google Scholar]
  2. Acres G. J. K. J. Power Sources. 2001;100:60–66. doi: 10.1016/S0378-7753(01)00883-7. [DOI] [Google Scholar]
  3. Sawy E. N. E. El-Sayed H. A. Birss V. I. Phys. Chem. Chem. Phys. 2015;17:27509–27519. doi: 10.1039/C5CP04634C. [DOI] [PubMed] [Google Scholar]
  4. Sawy E. N. E. Molero H. M. Birss V. I. Electrochim. Acta. 2014;117:202–210. doi: 10.1016/j.electacta.2013.11.109. [DOI] [Google Scholar]
  5. Ito Y. Takeuchi T. Tsujiguchi T. Abdelkareem M. A. Nakagawa N. J. Power Sources. 2013;242:280–288. doi: 10.1016/j.jpowsour.2013.05.064. [DOI] [Google Scholar]
  6. Lai S. C. S. Kleijn S. E. F. Öztürk F. T. Z. van Rees Vellinga V. C. Koning J. Rodriguez P. Koper M. T. M. Catal. Today. 2010;154:92–104. doi: 10.1016/j.cattod.2010.01.060. [DOI] [Google Scholar]
  7. Wang Y. Zou S. Cai W.-B. Catalysts. 2015;5:1507–1534. doi: 10.3390/catal5031507. [DOI] [Google Scholar]
  8. Yaqoob L. Noor T. Iqbal N. RSC Adv. 2021;11:16768–16804. doi: 10.1039/D1RA01841H. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Lv F. Zhang W. Sun M. Lin F. Wu T. Zhou P. Yang W. Gao P. Huang B. Guo S. Adv. Energy Mater. 2021;11:2100187. doi: 10.1002/aenm.202100187. [DOI] [Google Scholar]
  10. Kavanagh R. Cao X.-M. Lin W.-F. Hardacre C. Hu P. Angew. Chem., Int. Ed. 2012;51:1572–1575. doi: 10.1002/anie.201104990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ghumman A. Li G. Bennett D. V. Pickup P. G. J. Power Sources. 2009;194:286–290. doi: 10.1016/j.jpowsour.2009.05.020. [DOI] [Google Scholar]
  12. Du W. Wang Q. Saxner D. Deskins N. A. Su D. Krzanowski J. E. Frenkel A. I. Teng X. J. Am. Chem. Soc. 2011;133:15172–15183. doi: 10.1021/ja205649z. [DOI] [PubMed] [Google Scholar]
  13. Huang W. Ma X.-Y. Wang H. Feng R. Zhou J. Duchesne P. N. Zhang P. Chen F. Han N. Zhao F. Zhou J. Cai W.-B. Li Y. Adv. Mater. 2017;29:1703057. doi: 10.1002/adma.201703057. [DOI] [PubMed] [Google Scholar]
  14. Lv H. Sun L. Wang Y. Liu S. Liu B. Adv. Mater. 2022;34:2203612. doi: 10.1002/adma.202203612. [DOI] [PubMed] [Google Scholar]
  15. El Sawy E. N. Handal H. T. Thangadurai V. Birss V. I. J. Mater. Chem. A. 2016;4:15400–15410. doi: 10.1039/C6TA06161C. [DOI] [Google Scholar]
  16. Sawy E. N. E. Birss V. I. J. Electrochem. Soc. 2017;164:F1572. doi: 10.1149/2.0611714jes. [DOI] [Google Scholar]
  17. Hoang A. El Sawy E. Blackburn A. Ketabi S. Goledzinowski M. Comeau F. J. E. Birss V. ACS Appl. Energy Mater. 2020;3:8423–8436. doi: 10.1021/acsaem.0c01061. [DOI] [Google Scholar]
  18. El Sawy E. N. Brueckner T. M. Pickup P. G. J. Electrochem. Soc. 2020;167:106507. doi: 10.1149/1945-7111/ab98f1. [DOI] [Google Scholar]
  19. El Sawy E. N. Pickup P. G. Electrocatalysis. 2016;7:477–485. doi: 10.1007/s12678-016-0328-8. [DOI] [Google Scholar]
  20. Wang Y. Ruiz Diaz D. F. Chen K. S. Wang Z. Adroher X. C. Mater. Today. 2020;32:178–203. doi: 10.1016/j.mattod.2019.06.005. [DOI] [Google Scholar]
  21. Antolini E. J. Power Sources. 2007;170:1–12. doi: 10.1016/j.jpowsour.2007.04.009. [DOI] [Google Scholar]
  22. Cao L. Sun G. Li H. Xin Q. Fuel Cell. Bull. 2007;2007:12–16. doi: 10.1016/S1464-2859(08)70142-1. [DOI] [Google Scholar]
  23. Gu Z. Balbuena P. B. J. Phys. Chem. A. 2006;110:9783–9787. doi: 10.1021/jp063447o. [DOI] [PubMed] [Google Scholar]
  24. El Sawy E. N. Birss V. I. ACS Appl. Mater. Interfaces. 2018;10:3459–3469. doi: 10.1021/acsami.7b13080. [DOI] [PubMed] [Google Scholar]
  25. Zou S. Gómez R. Weaver M. J. Langmuir. 1997;13:6713–6721. doi: 10.1021/la970853d. [DOI] [Google Scholar]
  26. Sahin O. Kivrak H. Karaman M. Atbas D. Am. J. Mater. Sci. Eng. 2015;3:15–20. [Google Scholar]
  27. Lee M. J. Kang J. S. Kang Y. S. Chung D. Y. Shin H. Ahn C.-Y. Park S. Kim M.-J. Kim S. Lee K.-S. Sung Y.-E. ACS Catal. 2016;6:2398–2407. doi: 10.1021/acscatal.5b02580. [DOI] [Google Scholar]
  28. Courtois J. Du W. Wong E. Teng X. Deskins N. A. Appl. Catal., A. 2014;483:85–96. doi: 10.1016/j.apcata.2014.06.029. [DOI] [Google Scholar]
  29. Huang M. Yang H. Xia X. Peng C. Appl. Catal. B Environ. Energy. 2024;358:124422. doi: 10.1016/j.apcatb.2024.124422. [DOI] [Google Scholar]
  30. Mangoufis-Giasin I. Piqué O. Khanipour P. Mayrhofer K. J. J. Calle-Vallejo F. Katsounaros I. J. Catal. 2021;400:166–172. doi: 10.1016/j.jcat.2021.05.028. [DOI] [Google Scholar]
  31. Natarajan S. K. Hamelin J. J. Electrochem. Soc. 2008;156:B210. doi: 10.1149/1.3033517. [DOI] [Google Scholar]
  32. Trogadas P. Fuller T. F. Strasser P. Carbon. 2014;75:5–42. doi: 10.1016/j.carbon.2014.04.005. [DOI] [Google Scholar]
  33. Zhang J. Gao L. Inorg. Chem. Commun. 2004;7:91–93. doi: 10.1016/j.inoche.2003.10.012. [DOI] [Google Scholar]
  34. Hartig-Weiss A. Miller M. Beyer H. Schmitt A. Siebel A. Freiberg A. T. S. Gasteiger H. A. El-Sayed H. A. ACS Appl. Nano Mater. 2020;3:2185–2196. doi: 10.1021/acsanm.9b02230. [DOI] [Google Scholar]
  35. Silva-Junior L. C. Maia G. Passos R. R. de Souza E. A. Camara G. A. Giz M. J. Electrochim. Acta. 2013;112:612–619. doi: 10.1016/j.electacta.2013.09.030. [DOI] [Google Scholar]
  36. Liu X. Li Z. Zhou L. Wang K. Zhao X. Li Q. Deng Y. Nanomaterials. 2019;9:1145. doi: 10.3390/nano9081145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Kumara L. S. R. Seo O. Kim J. Wang F. Chen Y. Gueye I. Song C. Ohara K. Kusada K. Kitagawa H. Sakata O. ACS Appl. Nano Mater. 2023;6:16653–16661. doi: 10.1021/acsanm.3c02860. [DOI] [Google Scholar]
  38. Maulana A. L. Han S. Shan Y. Chen P.-C. Lizandara-Pueyo C. De S. Schierle-Arndt K. Yang P. J. Am. Chem. Soc. 2025;147:10268–10278. doi: 10.1021/jacs.4c16638. [DOI] [PubMed] [Google Scholar]
  39. Holzwarth U. Gibson N. Nat. Nanotechnol. 2011;6:534. doi: 10.1038/nnano.2011.145. [DOI] [PubMed] [Google Scholar]
  40. Amutha A. MSEJ. 2022;9:1–8. doi: 10.5121/msej.2022.9301. [DOI] [Google Scholar]
  41. Mazloom J. Ghodsi F. E. Gholami M. J. Alloys Compd. 2013;579:384–393. doi: 10.1016/j.jallcom.2013.06.063. [DOI] [Google Scholar]
  42. Balu P. Asharani I. V. Thirumalai D. J. Mater. Sci.: Mater. Electron. 2020;31:10669–10676. doi: 10.1007/s10854-020-03616-z. [DOI] [Google Scholar]
  43. Pittkowski R. K. Punke S. Anker A. S. Bornet A. Magnard N. P. L. Schlegel N. Graversen L. G. Quinson J. Dworzak A. Oezaslan M. Kirkensgaard J. J. K. Mirolo M. Drnec J. Arenz M. Jensen K. M. Ø. J. Am. Chem. Soc. 2024;146:27517–27527. doi: 10.1021/jacs.4c08149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Quinson J. Adv. Colloid Interface Sci. 2022;303:102643. doi: 10.1016/j.cis.2022.102643. [DOI] [PubMed] [Google Scholar]
  45. Banerjee W. Maikap S. Tien T.-C. Li W.-C. Yang J.-R. J. Appl. Phys. 2011;110:074309. doi: 10.1063/1.3642961. [DOI] [Google Scholar]
  46. Taylor G. Shallenberger J. Tint S. Fones A. Hamilton H. Yu L. Amini S. Hettinger J. Surf. Coat. Technol. 2021;426:127803. doi: 10.1016/j.surfcoat.2021.127803. [DOI] [Google Scholar]
  47. Hu C. Yue K. Han J. Liu X. Liu L. Liu Q. Kong Q. Pao C.-W. Hu Z. Suenaga K. Su D. Zhang Q. Wang X. Tan Y. Huang X. Sci. Adv. 2023;9:eadf9144. doi: 10.1126/sciadv.adf9144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Zhu W. Song X. Liao F. Huang H. Shao Q. Feng K. Zhou Y. Ma M. Wu J. Yang H. Yang H. Wang M. Shi J. Zhong J. Cheng T. Shao M. Liu Y. Kang Z. Nat. Commun. 2023;14:1–12. doi: 10.1038/s41467-023-41036-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Moriau L. Smiljanić M. Lončar A. Hodnik N. ChemCatChem. 2022;14:e202200586. doi: 10.1002/cctc.202200586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Corcoran C. J. Tavassol H. Rigsby M. A. Bagus P. S. Wieckowski A. J. Power Sources. 2010;195:7856–7879. doi: 10.1016/j.jpowsour.2010.06.018. [DOI] [Google Scholar]
  51. Park Y. Jang H. Y. Lee T. K. Kim T. Kim D. Kim D. Baik H. Choi J. Kwon T. Yoo S. J. Back S. Lee K. Nat. Commun. 2025;16:579. doi: 10.1038/s41467-025-55910-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Oh H.-S. Nong H. N. Reier T. Bergmann A. Gliech M. Ferreira de Araújo J. Willinger E. Schlögl R. Teschner D. Strasser P. J. Am. Chem. Soc. 2016;138:12552–12563. doi: 10.1021/jacs.6b07199. [DOI] [PubMed] [Google Scholar]
  53. Gorzkowski M. T. Lewera A. J. Phys. Chem. C. 2015;119:18389–18395. doi: 10.1021/acs.jpcc.5b05302. [DOI] [Google Scholar]
  54. Ruban A. Hammer B. Stoltze P. Skriver H. L. Nørskov J. K. J. Mol. Catal. A: Chem. 1997;115:421–429. doi: 10.1016/S1381-1169(96)00348-2. [DOI] [Google Scholar]
  55. Marques T. M. F. Morais R. N. Nobre F. X. Rocha J. M. Ghosh A. Soares T. A. S. Viana B. C. Machado G. Costa J. C. S. Matos J. M. E. D. An. Acad. Bras. Cienc. 2019;91:e20190082. doi: 10.1590/0001-3765201920190082. [DOI] [Google Scholar]
  56. Jain S. Shah J. Negi N. S. Sharma C. Kotnala R. K. Int. J. Energy Res. 2019;43:4743–4755. doi: 10.1002/er.4613. [DOI] [Google Scholar]
  57. Dementjev A. P. de Graaf A. van de Sanden M. C. M. Maslakov K. I. Naumkin A. V. Serov A. A. Diamond Relat. Mater. 2000;9:1904–1907. doi: 10.1016/S0925-9635(00)00345-9. [DOI] [Google Scholar]
  58. Lin Y. Chen H. Shi Y. Wang G. Chen L. Wang F. Li S. Yu F. Zhang L. Glob. Chall. 2019;3:1900043. doi: 10.1002/gch2.201900043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Askari S. Haq A. U. Macias-Montero M. Levchenko I. Yu F. Zhou W. (Ken) Ostrikov K. Maguire P. Svrcek V. Mariotti D. Nanoscale. 2016;8:17141–17149. doi: 10.1039/C6NR03702J. [DOI] [PubMed] [Google Scholar]
  60. Racz A. S. Menyhard M. Sci. Rep. 2024;14:18497. doi: 10.1038/s41598-024-69495-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Bhachu D. S. Egdell R. G. Sankar G. Carmalt C. J. Parkin I. P. J. Mater. Chem. C. 2017;5:9694–9701. doi: 10.1039/C6TC04462J. [DOI] [Google Scholar]
  62. Lu F. Ji X. Yang Y. Deng W. Banks C. E. RSC Adv. 2013;3:18791–18793. doi: 10.1039/C3RA43532F. [DOI] [Google Scholar]
  63. Sun Y. Wang C. Wang Y. Xu Y. Yang X. Ouyang B. Li C. Appl. Surf. Sci. 2024;651:159222. doi: 10.1016/j.apsusc.2023.159222. [DOI] [Google Scholar]
  64. Khan I. A. Morgen P. Gyergyek S. Sharma R. Andersen S. M. Mater. Chem. Phys. 2023;308:128192. doi: 10.1016/j.matchemphys.2023.128192. [DOI] [Google Scholar]
  65. Grimaud A. Demortière A. Saubanère M. Dachraoui W. Duchamp M. Doublet M.-L. Tarascon J.-M. Nat. Energy. 2016;2:16189. doi: 10.1038/nenergy.2016.189. [DOI] [Google Scholar]
  66. Zou J. Wu M. Ning S. Huang L. Kang X. Chen S. ACS Sustainable Chem. Eng. 2019;7:9007–9016. doi: 10.1021/acssuschemeng.9b01270. [DOI] [Google Scholar]
  67. Hammer B. and Nørskov J. K., in Impact of Surface Science on Catalysis, Academic Press, 2000, vol. 45, pp. 71–129 [Google Scholar]
  68. Sawy E. N. E. El-Sayed H. A. Birss V. I. Chem. Commun. 2014;50:11558–11561. doi: 10.1039/C4CC04824E. [DOI] [PubMed] [Google Scholar]
  69. Wang H. Abruña H. D. J. Am. Chem. Soc. 2023;145:18439–18446. doi: 10.1021/jacs.3c04785. [DOI] [PubMed] [Google Scholar]
  70. Guo Y. Zheng D. Liu H. Friedrich A. Garche J. J. New Mater. Electrochem. Syst. 2025;27:101160. [Google Scholar]
  71. Ka S. Lee Y. Lim C. Kim E.-D. Bang H.-S. Kim W. Choi J.-Y. Oh H.-S. Lee W. H. Nat. Commun. 2025;16:8946. doi: 10.1038/s41467-025-63992-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Kormányos A. Ayyub M. M. Kutus B. Rashid M. Priamushko T. Samu G. F. Cherevko S. Endrődi B. Janáky C. J. Am. Chem. Soc. 2026;148:9006–9013. doi: 10.1021/jacs.5c22283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. da Silva F. R. P. Silva-Junior L. C. Camara G. A. Giz M. J. J. Braz. Chem. Soc. 2019;30:2095–2104. [Google Scholar]
  74. Cao L. Sun G. Li H. Xin Q. Electrochem. Commun. 2007;9:2541–2546. doi: 10.1016/j.elecom.2007.07.031. [DOI] [Google Scholar]
  75. Tayal J. Rawat B. Basu S. Int. J. Hydrogen Energy. 2011;36:14884–14897. doi: 10.1016/j.ijhydene.2011.03.035. [DOI] [Google Scholar]
  76. Zhang G. Zhang Z. Inorg. Chem. Front. 2020;7:2231–2240. doi: 10.1039/D0QI00233J. [DOI] [Google Scholar]
  77. Murawska M. Cox J. A. Miecznikowski K. J. Solid State Electrochem. 2014;18:3003–3010. doi: 10.1007/s10008-014-2493-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Lee Y.-W. Hwang E.-T. Kwak D.-H. Park K.-W. Catal. Sci. Technol. 2016;6:569–576. doi: 10.1039/C5CY01054C. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

RA-OLF-D6RA04858G-s001
RA-OLF-D6RA04858G-s001.pdf (1.3MB, pdf)

Data Availability Statement

All data supporting this article are included in the article.

Supplementary information is available. See DOI: https://doi.org/10.1039/d6ra04858g.

Articles from RSC Advances are provided here courtesy of Royal Society of Chemistry

RESOURCES