Electrochemical Behavior of Carbon Electrodes for In Situ Redox Studies in a Transmission Electron Microscope

Electrochemical liquid cell transmission electron microscopy (TEM) is a unique technique for probing nanocatalyst behavior during operation for a range of different electrocatalytic processes, including hydrogen evolution reaction (HER), oxygen evolution reaction (OER), oxygen reduction reaction (ORR), or electrochemical CO2 reduction (eCO(2)R). A major challenge to the technique's applicability to these systems has to do with the choice of substrate, which requires a wide inert potential range for quantitative electrochemistry, and is also responsible for minimizing background gas generation in the confined microscale environment. Here, we report on the feasibility of electrochemical experiments using the standard redox couple Fe(CN)(6)(3-/4)(-) and microchips featuring carbon-coated electrodes. We electrochemically assess the in situ performance with respect to flow rate, liquid volume, and scan rate. Equally important with the choice of working substrate is the choice of the reference electrode. We demonstrate that the use of a modified electrode setup allows for potential measurements relatable to bulk studies. Furthermore, we use this setup to demonstrate the inert potential range for carbon-coated electrodes in aqueous electrolytes for HER, OER, ORR, and eCO(2)R. This work provides a basis for understanding electrochemical measurements in similar microscale systems and for studying gas-generating reactions with liquid electrochemical TEM.

Electron probing of oxygen-evolving oxide catalysts

Tzu-Hsien Shen

Water splitting is one of the cleanest ways to store energy. The production of hydrogen and oxygen gases can be utilized in fuel cells to generate electricity, power, and heat. In the water splitting process, the oxygen evolution reaction (OER), taking place at the anode, is the sluggish reaction limiting the efficiency. Electrocatalysts are used to reduce the kinetic barrier of OER. Development of high-performance and stable OER catalysts requires fundamental understanding of their electrocatalytic properties and, hence, their holistic characterization is indispensable. The objective of this thesis is to probe Co-based oxide catalysts for OER in an alkaline medium using advanced (scanning) transmission electron microscopy ((S)TEM) techniques. Starting with the surface characterization of a highly active OER oxide catalyst, Ba0.5Sr0.5Co0.8Fe0.2O3-d (BSCF), it is shown that the surfaces of the as-synthesized BSCF particles are Co/Fe rich and adopt a spinel-like structure with a reduced valence of Co ions. The Co/Fe spinel structure is further linked to the formation of the highly active Co(Fe)OOH at the BSCF surface. Further real-time probing of BSCF oxide catalysts using electrochemical liquid-cell TEM shows a switchable wetting behavior of Co-based oxide catalysts by analyzing the liquid movement around oxide particles. The interfacial interactions at the surface-electrolyte under potential cycling are correlated to electrowetting and hydrophilic surface transformation. Molecular oxygen evolution is detected qualitatively under OER conditions by identifying the O2 feature in electron energy-loss spectra. Quantification of the O2 using electron energy-loss spectroscopy is also demonstrated, pending further improvements.The work herein offers new insights into the characterization framework of oxide catalysts for OER using (S)TEM. The bulk to surface structure of the catalysts can be analyzed, and the interfacial interactions of single-particle catalysts can be probed in real-time. The findings aid in understanding the performance and stability of oxygen-evolving oxide catalysts with further prospects envisioned towards a complete quantitative probing of electrocatalytic reactions in real-time. Ultimately, quantitative characterization in electrochemical liquid-phase (S)TEM can provide understanding of surface-active sites and reaction kinetics of oxygen evolution catalysts.

EPFL2022

Préparation et caractérisation de nanoparticules bimétalliques (Pt-Au) sur un substrat en diamant dopé au bore

Bahaa El Roustom

Using petrol as a main source of energy, during the 20th century, has caused considerable amounts of damage among which are atmospheric pollution and global warming. At the same time, many geologists and economists expect that the discovery rate of new petrol sources will not follow the market's demand causing a serious shortage of petrol in the near future. These alarming perspectives have motivated scientists around the world to develop new clean and renewable energy sources. It is in that objective that electrochemists intensified their research in order to develop the fuel cell technology. This latter consists in directly converting chemical energy into electricity. The main advantage of this technology over traditional energy production is that the fuel cell energy efficiency is Carnot cycle independent. The theoretical efficiency of such an energy source is nearly 90%. Direct Alcohol Fuel Cell (DAFC), fed by ethanol on one hand and by air (oxygen) in the other hand, is a very interesting option for such kind of energy production. Ethanol is a renewable energy source of vegetal origin whose CO2 production is consumed by plants during the cycle. Even though Pt has long been known to be a perfect catalyst for DAFC, it has some principal limitations, the most important ones being poisoning by intermediate products (OH on the cathodic side and CO on the anodic side) and a relatively high cost compared to other materials. These problems have pushed the experts in the field to focalize their research on the development of more suitable and better performing materials. Gold nanoparticles have shown a surprising catalytic activity, very different from the bulk, making them a preferential candidate to replace Pt. A system based on using two or more metals as catalysts (example Au-Pt) dispersed on an appropriate substrate seems to be also an interesting candidate to enhance the cell's efficiency. The substrate should be chemically inert, non-oxydable and should not present any interference with the electrochemical and electrocatalytic behaviour of the nanoparticles. In the first part, the employed substrate, which is boron doped diamond (BDD) is characterized. The substrate's main characteristics are its chemical inertness, its weak residual current and its resistance to corrosion. These characteristics contribute to making the substrate an ideal support for nanoparticles. In this work, the substrate is a diamond film, but for fuel cell applications diamond should be used as powder or dispersed particles. In chapters III and IV, we studied the effect of boron incorporation on the morphology and the behaviour of the diamond film. For this purpose, samples with different ratios of boron/carbon (B/C) in the vapour phase during preparation were characterized (B/C = 300, 1000, 2500 and 6000 ppm). Results showed that the grain size decreases with increasing boron content in the film. This is accompanied by a transition towards a quasi-metallic character. It was also found that graphitic impurities in the diamond film increase when the B/C ratio increases. We also established some graphitic impurities increase in the diamond film when the ratio of B/C increases. The identity of these impurities was established in chapter III as being surface redox couples (semi-quinone/semi-hydroquinone). Using cyclic voltametry we were able to observe a total elimination (B/C=300 ppm) or partial elimination (B/C=6000 ppm) of these surface redox couples. We also found in chapter IV that these surface redox couples are responsible of the electrochemical activity of the BDD electrode. BDD electrodes with different doping level B/C=300, 1000 and 6000 ppm) in the vapour phase are modified with IrO2 nanoparticles using the method of thermal decomposition. Our results showed that, at low potentials, the behaviour of the composite electrode IrO2/BDD with respect to the reaction of oxygen evolution is independent of the doping level. On the other hand, at high potentials, the weakly doped electrode (B/C=300 ppm) presents a supplementary overpotential due to the interface BDD-IrO2. An approach based on the double junction is developed in this work describing two distinct behaviours depending on the boron concentration in the diamond film. A new method for Au nanoparticles deposition on highly doped BDD electrodes (B/C= 2500 ppm) is presented in chapter VI. This method involves two steps: the first step consists on sputtering the equivalent of few nanolayers of Au on the diamond surface; the second step is a heat treatment in air at high temperatures. Several parameters that could influence the dispersion, the stability as well as the diameter of the Au nanoparticles are studied. In chapter VII, the behaviour of the composite electrode Au/BDD is studied in a solution containing redox couples in acid media. This electrode was simulated as an arrangement of active Au microelectrodes dispersed on a less active BDD substrate. The overall behaviour of the composite electrode (Au/BDD) has been related to the electrochemical rate constants on each material. In chapter VIII, another new synthesis method for a bimetallic system formed by Au and Pt nanoparticles dispersed on a BDD substrate is developed. This method consists in electrodepositing a small amount of Pt on a composite electrode Au/BDD. Our results showed that Pt was preferentially deposited on Au covering its whole surface. In fact, Au nanoparticles serve as germ of nucleation for Pt. In the following step, the bimetallic PtbAu/BDD is heated to 400°C in air to enhance the interaction between Au and Pt. The experiments showed that Au reappeared on the surface. The electrochemical behaviour of the bi-metallic electrode is characterized by a negative shift in the peak potential of reduction of Pt oxides. This is not the case of heterogeneous alloys obtained by the fusion of two metals at high temperatures. The composite electrode Au/BDD as well as the bi-metallic electrode Pt-Au/BDD are used for electrocatalytic applications. The studied reaction in chapter IX is the oxygen reduction reaction (ORR) in an acid medium. We found that the number of exchanged electrons increases with increasing the amount of Pt at the surface. Oxygen is reduced into H2O2 on the Au/BDD electrode and into H2O on the Pt-Au/BDD electrode. In chapter X we studied the reaction of oxidation of oxalic acid. The behaviour of both Au and Pt electrodes seemed very similar for this reaction. The bi-metallic electrode containing both Pt and Au on its surface has shown an intermediate behaviour between the behaviour of the two metals.

EPFL2006

Electrochemical promotion of Pt catalysts for gas phase reactions

Arnaud Jaccoud

The subject of this work is the study of the electrochemical promotion of catalysis (EPOC), also called non-Faradaic electrochemical modification of catalytic activity (NEMCA). In this phenomenon, application of small currents or potentials on catalysts in contact with solid electrolytes leads to very pronounced strongly non-Faradaic and reversible changes in catalytic activity and selectivity. This work concerns the system composed of platinum film electrodes, deposited on the YSZ electrolyte (ZrO2 doped with 8% Y2O3), in contact with an oxygen containing atmosphere (the O2(g),Pt/YSZ system). The aim of this work was to focus on two points in the phenomenon of EPOC: 1) The nature of the species present on the surface of the catalyst and their impact on the catalytic activity. 2) The phenomenon of permanent promotion of catalysis (P-EPOC), for which the catalytic rate enhancement is not totally reversible, leading to a long lasting activated state. The Pt electrodes were prepared via three different methods (sputtering, thermal decomposition and screen-printing) and, as expected, the resulting microstructure, observed by scanning electron microscopy (SEM), was found to depend greatly on the preparation technique. The electrochemical response of each electrode was characterized by the technique of cyclic voltammetry. Due to the occurrence of different site-related electrochemical reactions, the voltammogram shape was highly influenced by the microstructure of the Pt/YSZ contact. Based on the preliminary studies, the cermet electrode was chosen for additional electrochemical experiments, using chronoamperometry, chronopotentiometry, cyclic voltammetry, steady state polarization and impedance spectroscopy. The results of all these techniques suggest that under anodic polarization, two parallel reactions take place. The first reaction is the oxygen evolution, giving rise to a steady state current under a constant applied potential. The other reaction is the oxygen storage (oxygen injection) in the form of Pt-O species. This second process gives rise to a time dependant current under a constant applied potential, and is therefore responsible for the pseudocapacitive behavior of the electrode. Moreover, linear sweep voltammetry measurements indicated that, by application of an anodic potential, at least three types of Pt-O species were stored, following distinct kinetics. Based on the amount of stored Pt-O and on the corresponding storage kinetics, they were attributed to three different locations on the electrode: 1) at the Pt/YSZ interface, 2) diffusing from the tpb toward the Pt/gas interface, 3) diffusing from the Pt/YSZ interface toward the bulk of the platinum electrode. It has been shown that the Pt film exhibited various activity states toward the catalytic oxidation of ethylene, depending on the oxygen species present at its surface. As evidenced by XPS measurements and in agreement with the thermodynamic data for the given system, the inactive (deactivated) state has been attributed to the formation of surface platinum oxide at T > 550°C. Decreasing the reaction temperature to 525°C caused the slow spontaneous reactivation of the catalyst and several days were necessary to reach the active state, where most of the catalyst was present in its metallic state. The catalyst activity recovery could be accelerated by the application of anodic potential pulses. In agreement with the mechanism of EPOC, this was explained by the oxide destabilizing effect of the anodically produced Oδ- species. Anodic polarization also caused reversible electrochemical promotion of catalysis at unusual high temperature (600°C). In addition, very intriguing catalytic relaxation transients were observed after prolonged anodic polarization times. Indeed, it was found that enhanced catalytic activity could be maintained for up to 10 hours after current interruption. Combination of these latter results, the related literature, the state of the art model of EPOC, and the results obtained by electrochemical techniques led to the proposition of an original model based on the processes of Pt-O storage/release at various locations of the O2(g),Pt/YSZ system. According to this model, anodic polarization produces Oδ- back-spillover species, promoting the catalytic activity, in agreement with the mechanism of EPOC. In parallel, hidden oxygen species are stored at the Pt/YSZ interface and diffuse toward the Pt phase. When the polarization is switched off, these hidden oxygen species diffuse to the gas-exposed surface and cause non-Faradaic promotion, as back-spillover Oδ- do. The large amount of stored charge and its slow diffusion controlled emergence causes the rate enhancement to last for hours.

EPFL2007