Spectroscopic Characterization of Mixed Fe-Ni Oxide Electrocatalysts for the Oxygen Evolution Reaction in Alkaline Electrolytes
James Landon, Ethan Demeter, Nilay İnoğlu, Chris Keturakis, Israel E. Wachs, Relja Vasić, Anatoly I. Frenkel and John R. Kitchin, “Spectroscopic Characterization of Mixed Fe-Ni Oxide Electrocatalysts for the Oxygen Evolution Reaction in Alkaline Electrolytes”, ACS Catalysis 2(8), 1793-1801 (2012). https://doi.org/10.1021/cs3002644
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Question
The oxygen evolution reaction (OER) limits the efficiency of water electrolysis, and the best catalysts in acid are oxides of the costly metals Ru and Ir. In alkaline electrolyte, cheap base metal oxides such as NiO become reasonably stable, and earlier work had reported that mixing Fe into Ni oxide gives higher OER activity than either parent oxide, with an optimum around 10 to 20 mol % Fe. What was missing was a detailed spectroscopic picture of these catalysts that could explain why the mixed surface is more active. The paper sets out to supply that characterization, and to check whether the activity peak depends on how the catalyst is made (Landon et al. 2012).
Approach
Mixed Ni-Fe oxides spanning pure NiO to pure Fe oxide were made three ways: evaporation-induced self-assembly (EISA) with a P123 surfactant and calcination at 550 °C, nanocasting in an SBA-15 silica hard template, and dip-coating an etched Ni mesh followed by decomposition at 300 °C. Composition is reported as mol % Fe of total metal. Powders were cast with an ionomer ink onto carbon paper (about 0.5 mg/cm2) and tested in Ar-purged 1 M KOH against Hg/HgO, using slow polarization scans for Tafel analysis and 20 min potentiostatic holds for steady-state currents. The EISA materials were then characterized by BET surface area, X-ray diffraction (XRD), Raman spectroscopy with 442 nm excitation, SEM with EDS mapping, XPS, ex situ Fe K-edge XANES, in situ Fe and Ni K-edge EXAFS in electrolyte at open circuit and under OER, and temperature programmed reaction spectroscopy of adsorbed methanol (CH3OH-TPRS) to probe surface acid-base and redox sites.
Key findings
For all three synthesis routes, mass-normalized activity at a fixed overpotential of about 360 mV peaked near 10 mol % Fe and fell off above that, with the dip-coated meshes roughly an order of magnitude more active than the powders (attributed tentatively to direct electrical contact with the metal mesh). Tafel slopes dropped from above 50 mV/dec for NiO to about 40 mV/dec for the mixed oxides. BET areas of the EISA powders ranged only from 13 to 31 m2/g and did not track activity, so the enhancement was judged intrinsic.
XRD gave a NiO lattice constant of 4.18 Å regardless of Fe content, indicating Fe was not substituted into NiO. Instead a NiFe2O4 spinel peak grew with Fe content up to 25 mol %, and Fe2O3 appeared above that. Raman confirmed NiFe2O4 bands already at 5 mol % Fe and detected no iron oxide phases up to 20 mol % Fe. EDS showed Ni and Fe distributed fairly uniformly, with occasional Fe-rich spots near 30 mol % Fe. XPS showed NiO-like Ni with no measurable binding energy shift, weak Fe signals consistent with Fe2O3 references, and no surface versus bulk difference after Ar sputtering. Fe K-edge XANES of the 10 mol % sample placed the edge near that of α-Fe2O3 (Fe3+) but with a 7115 eV pre-edge feature indicating some tetrahedral Fe, as in a spinel. In situ EXAFS showed essentially no change at the Ni edge under OER, which the authors note may reflect bulk sensitivity, while the Fe edge showed a stronger Fe-O peak and changed Fe-cation shell intensities, read as increased Fe coordination, possibly Fe moving from tetrahedral to octahedral spinel sites. Currents stayed steady, arguing against Fe2O3 formation.
In CH3OH-TPRS, NiO and the mixed oxides gave mainly CO and CO2, characteristic of basic sites, and no dimethyl ether (the signature of acidic α-Fe2O3 surfaces). The number of basic sites ordered NiO ≤ 20 % Fe < 5 % Fe < 10 % Fe, the same ordering as OER activity.
Significance
The study ties the well-known activity maximum of Ni-Fe oxides near 10 mol % Fe to a NiFe2O4 spinel phase coexisting with NiO rather than to Fe dissolved in the NiO lattice, shows by in situ EXAFS that Fe responds to OER conditions and so may be part of the active site, and proposes surface basicity as a correlating property for OER activity. The authors suggest that at higher Fe contents thin, possibly amorphous Fe oxide films invisible to XRD could poison the surface (Landon et al. 2012).