Recently, in operando optical investigations of high temperature solid oxide electrochemical cells (SOECs) have gained significant traction with several research groups motivated by a need to directly correlate electrochemical performances of SOECs with their associated and complex electrode processes. These techniques (including Raman spectroscopy, ambient x-ray photoelectron spectroscopy, near-infrared thermal imaging, and Fourier transform infrared emission spectroscopy) have demonstrated a significant advantage over traditional, indirect experimental methods by providing direct, non-invasive information in real time along with high temporal, spatial, and/or molecular resolution.
Fourier transform infrared emission spectroscopy, recently implemented for in operando testing of solid oxide fuel cell (SOFC) anodes by Pomfret and coworkers, has become a novel means to directly measure the radiative emission of hot gases within the volume over an anode surface, including CH4, CO2, CO, and H2O (see Figure). In these measurements, a FTIR spectrometer was adapted by removing an infrared glow source from the optical path and aligning in its place the anode surface, located within a furnace cavity. Ideally, the detector will only see the infrared emission of the hot anode surface and any gases in the anode head space, though in reality broad-band, nonmolecular (and frequency dependent) emission sources (e.g. furnace walls) contribute to the majority of the signal reaching the detector. These must be subtracted from each spectrum collected during an experiment to leave behind the difference that shows only molecular contributions.
However, recent emission experiments have indicated that for the FTIR spectra a simple background subtraction may not account for all background contributions. These measurements have been made on a variety of fuels including 100% CH4 and mixtures of CH4/Ar and simulated biogas (50% CH4 and 50% CO2, balance Ar) at various operational temperatures which represent different SOFC operating conditions.
In operando IR emission spectra collected during operation with both fuels described above show that CO2 emission is significantly attenuated relative to other gas constituents. On the other hand, ex situ IR absorption measurements from the same experiments have demonstrated that CO2 absorption is optically thick and very intense relative to other gas constituents in the spectrum.
Additionally, attempts to calibrate the emission intensity of CO2 with concentration in a high-temperature environment indicated a nonlinear response, while FT-IR absorption measurements of the same gases directed into a flow cell showed a linear response as expected.
Interestingly, in the emission experiments, as the concentration of CO2 increased, the normally weak CO2 combination bands at 3600 cm-1 and 3700 cm-1 grew in intensity while the asymmetric stretch at 2350 cm-1 remained relatively unchanged. We expect that the gas composition over the anode headspace is comparable to the exhaust composition, despite the rather large apparent concentration differences inferred from the emission and absorption spectra. To better understand these differences, a model has been implemented to describe how the frequency-dependent substrate emission could affect the emission of a hot gas. This report describes a model where the background radiation is coupled to molecular emission, and cannot be subtracted linearly. Furthermore, the geometry of the test apparatus greatly influences any background radiation through the radiative form factor and reflections from the anode surface.
This work was done by Harold D. Ladouceur, John D. Kirtley, Syed N. Qadri, Jeffrey C. Owrutsky, and Daniel A. Steinhurst for the Naval Research Laboratory. NRL-0067
This Brief includes a Technical Support Package (TSP).
The Effect of Substrate Emissivity on the Spectral Emission of a Hot-Gas Overlayer
(reference NRL-0067) is currently available for download from the TSP library.
Don't have an account? Sign up here.