Solid oxide fuel cells (SOFCs) are electrochemical instruments that directly transform fuels into electricity. Defined by low emission, high efficiency, and long term stability, the fuel cell device includes permeable anodic and cathodic layers with a thick, solid oxide electrolyte between them, as shown in Figure 1.
Figure 1. Schematic of operation of solid oxide fuel cell
The aim of the cathodic layer is to ionize atmospheric oxygen gradually and transfer it to the electrolyte, where it meets the fuel. The device’s overall efficiency is associated with the ability of cathode to cause this supposed oxygen reduction reaction. Strontium-substituted lanthanum cobaltites serve as fuel cell cathode layers. Using these perovskite materials, the catalytic activity is influenced by chemistry and the elemental composition of the surface. The chemical specificity of XPS makes it a suitable method for studying SOFCs.
Experimental Framework
XPS technique was applied for analyzing a sample cathode material before and after annealing in air at increased temperatures, replicating the thermal cycling of an actual SOFC device. A lanthanum strontium cobaltite (LSC) layer was then examined before and after annealing at elevated temperature. This LSC layer was subsequently deposited onto an yttrium-stabilized zirconia (YSZ) substrate with a ceria layer doped with gadolinium between them. The YSZ substrate is a dense electrolyte, whilst the LSC layer serves as a cathode, as shown in Figure 2.
Figure 2. Scanning Electron Microscope image of cross section of solid oxide fuel cell
In order to acquire both elemental and chemical data of the surface of cathode layer, the apex surface of the LSC layer was studied in a non-destructive manner using the XPS method. The speed at which oxygen is absorbed from the atmosphere (air) and changed into ions relies on the composition and chemistry of the outermost surface of the LSC.
Results and Discussion
Through a simple, non-destructive XPS examination of the surface of the LSC layer, considerable changes were observed in the film composition. Such changes were caused by annealing. The comparison of the elemental compositions of as received sample and annealed sample is shown Figure 3. In both these films, the concentration of cobalt was considerably less than the predicted value. Additionally, the ratio of lanthanum to strontium was not favorable in both the cases. Moreover, given that the ratio alters between the annealed and as received films, it suggests that the film is unstable under thermal cycling settings which could be experienced while being used.
Figure 3. Elemental quantification of surface composition of annealed and as received samples
Figure 4. High resolution spectra of a) carbon and b) strontium
High energy resolution XPS analysis of the carbon spectrum from the top surface of the LSC layer helps in detecting the presence of carbon bonding states (Figure 4a). The C-C and C-O components thus observed are from adventitiously-deposited carbon and are often found on samples exposed to air for a significant amount of time. At greater binding energies, the components are the result of inorganic carbonates. Less amount of carbonate is found on the annealed surface, and the variation in the width of carbonate peak could suggest that the carbonate present in the annealed sample forms on a surface that is more physically ordered.
Using the strontium high resolution spectrum, the assignment of a component of carbonate in the carbon spectrum is validated. The strontium spectrum acquired from the unannealed (as received) sample displays dual bonding states which can be assigned to strontium and strontium carbonate in the LSC lattice. Two XPS peak components for individual bonding state are observed following interaction between the spin and orbital angular momentum of ionized strontium.
Annealing in atmosphere at elevated temperature has largely reduced the proportion of strontium carbonate in relation to the concentration of lattice-bound strontium atoms. The carbonates negatively impact the oxygen reduction reaction that occurs at the apex surface, thereby obstructing the transport of oxygen ions via the LSC layer. Consequently, the overall performance of the SOFC device would be severely affected.
Lanthanum chemistry at the top surface of the LSC layer can be tracked using a simple and easy metric, as shown in Figure 5. Powerful interactions between orbital and spin angular momentum result in a division of lanthanum XPS peaks. The extent of this splitting together with the ratio of the split components is diagnostic of the existing chemical states. For instance, carbonate and lanthanum have splittings that are variable by 1eV.
On the unannealed surface, the splitting is as anticipated for carbonate, in agreement with the strontium and carbon information which has been already demonstrated. Following annealing, the splitting raises towards the value for lanthanum oxide, which suggests that the carbonate level has reduced but is not pure lanthanum oxide bonding. The data are rather consistent with an oxide-carbonate mixture.
Figure 5. High resolution lanthanum spectra of the surface
Thanks to the non-destructive method of angle resolved XPS analysis, the depth distribution of strontium carbonate can be detected in the top few nanometers of the LSC layer. When electrons are collected from varied photoemission angles, the XPS’ information depth changes considerably. With strontium utilizing photoemission angle normal to the surface of the sample, the information depth ranges between 0 and 6nm within the surface.
When a shallow angle is utilized, a thinner layer of the surface from 0 to 3nm is sampled. In Figure 6, it can be observed that the relative amounts of strontium in the carbonate and LSC lattice states alters considerably, with the carbonate being comparatively stronger when sampling just the top 3nm of the surface, confirming that the carbonate is a surface species.
Figure 6. High resolution strontium spectra of the as received sample a) normal photoemission angleand b) shallow photoemission angle.
The chemistry of cobalt at the apex surface of the LSC layer was studied through two techniques, observing the valence and core levels of cobalt atoms. The wide band between 0 and 6eV is the result of hybridization of O2p and Co3d valence orbitals (Figure 7a).
Figure 7. a) Valence band spectra and b) High resolution cobalt spectra from as received and annealedLSC surfaces (0-6nm).
The narrow band at 2.2eV was earlier assigned to transition metals in a +3 oxidation state in a polyhedral arrangement with oxygen. Following annealing, this narrow band increases in intensity, which suggests that the concentration of Co(III) at the top surface of the LSC layer increases with annealing. Analysis of valence bands has demonstrated that there is Co(III) on the surface of the LSC layer. However, high energy resolution XPS analysis of the core levels in cobalt (Figure 7b) denotes that Co(II) is also present. In LaCoO3 which has no strontium atoms, only Co(III) would be expected, but replacing strontium into the lattice at the cost of lanthanum raises the number of oxygen vacancies, thereby leading to the formation of cobalt +2 oxidation state. Using the ratio of the XPS satellite peaks that are diagnostic of Co(II) to the primary peak, the cobalt +2 and +3 oxidation states can be measured. In accord with the valence band analysis, the XPS core-level information reveals that the proportion of Co(III) in the top surface of the LSC layer increases on annealing in the atmosphere.
Conclusion
Thermo Scientific’s K-Alpha XPS system was used for studying the potential SOFC material. A LSC layer was studied before and after annealing at increased temperature to replicate the thermal cycling of an actual SOFC device. It was observed that the proportion of carbonate on the surface reduces during the course of annealing. These carbonates have a negative impact on the oxygen reduction reaction occurring at the surface, preventing the transport of oxygen ion via the LSC layer. Also, it was observed that the carbonates are situated within the first 3nm of the surface. Such data can be acquired only through the XPS technique, thanks to its unique chemical specificity and surface sensitivity.
This information has been sourced, reviewed and adapted from materials provided by Thermo Fisher Scientific – X-Ray Photoelectron Spectroscopy (XPS).
For more information on this source, please visit Thermo Fisher Scientific – X-Ray Photoelectron Spectroscopy (XPS).