Loading...
Developing ionomer detection methodologies for use in quality control systems in the manufacture of catalyst coated membranes
Jacobsen, D.
Jacobsen, D.
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2020
Date Submitted
Keywords
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Polymer electrolyte membrane fuel cells (PEMFCs) are on the verge of being produced at unprecedented levels as fuel cell vehicles (FCVs) and other renewable energy resources are set to drastically increase their market share; thanks largely to international campaigns championing climate science and a new wave of young activists and investors. In order to drive production cost and waste down, it is critical that quality control methods are in place that are capable of accurately and rapidly measuring key characteristics of the catalyst layer (CL) and membrane in real time as the CL is forming. While work done at the National Renewable Energy Laboratory (NREL) has pioneered several new methodologies for inline autonomous detection systems which are capable of 100\% area scanning, a method for checking ionomer loading in the CL and mapping out its distribution in catalyst coated membranes (CCMs) has never been reported in literature. As the ionomer's distribution in the CL is critical to fuel cell efficiency, lacking an inline or even post production method to check that intended targets are being hit poses a significant risk to manufacturers. Due to this, a series of non-destructive evaluation (NDE) techniques were chosen to evaluate their efficacy in detecting the ionomer in the precursor catalyst inks and resulting catalyst layers, in addition to ordering samples by increasing ionomer content. An initial literature review of Raman spectroscopy was conducted, and while promising in other cases, was not pursued further as CCM samples have been found to combust under typical Raman conditions. Capacitive imaging, a popular industrial NDE for detecting delamination in composites and subsurface infrastructure/damage in concrete, was investigated as it seemed a natural progression from cyclic sweep voltammetry (CV) and electrochemical impedance spectroscopy (EiS); both of which having been previously performed on Nafion/CCMs with mixed results. From basic numerical models it is shown that the measured variation in substrate thickness produces a signal greater than that which can be expected from the Nafion loading variation in the CL. Fluorescence spectroscopy was performed on several batches of the commercial Nafion dispersion, D2020, which was found to fluoresce at a wavelength range that could not be explained by the presence of the solvent. As Nafion has known absorption peaks in the UV and IR, transmission UV and FTIR spectroscopy was performed on the catalyst inks and CCMs. It is shown that in the UV the spectra is dominated by carbon scattering; whereas in the IR, the substrate absorbs the light in the spectral range of interest completely. Subsequent ATR-FTIR experiments were ultimately successful in detecting the ionomer in catalyst films and in ordering samples by ionomer content. It was found that it is possible to observe the catalyst layer forming in real time while monitoring solvent levels and the eventual Nafion polymerization. The ordering of samples by ionomer content was possible by taking advantage of an area of anomalous dispersion caused by the strong absorption peaks (CF$_2$ symmetric and asymmetric stretching) of Nafion and the Christiansen effect. Expanding on the study presented here is necessary for tuning the correlation presented into an accurate measurement of ionomer loading, along with translating the findings into an inline detection system; one which currently does not exist, but is surely needed.
Associated Publications
Rights
Copyright of the original work is retained by the author.