Additively Manufactured Mixed Potential Electrochemical Sensors for Natural Gas Detection

Abstract

Emissions of methane from natural gas infrastructure account for over 13 Tg/yr of CO2 equivalent in the US with an expected cost of 2 billion dollars to the industry.[1] Mitigating the environmental effects and financial costs will require a robust, widely-deployable, low cost sensing technology to provide continuous monitoring of natural gas infrastructure so leaks can be identified and repaired. An additional challenge is that methane emissions also originate from agricultural or wetlands sites and sensors will need to distinguish between natural gas or one of these other sources. Mixed potential sensors operate on the principle of a difference in catalytic activity in their electrodes (Figure 1), where the mixed potential in the presence of both reducing and oxidizing gases is used as a sensing parameter. Recent advances in portable computing hardware have made deployment of machine learning technology more feasible. In our previous studies, MPE sensors coupled with artificial neural networks have successfully quantified and identified mixtures of hydrocarbons, NOx, and NH3.[2,3]Additive manufacturing is a promising route to the fabrication of mixed potential sensors, particularly at small prototyping scales where large volume production is not needed. Previously, we have demonstrated that LSCO/YSZ/Pt two electrode devices made by syringe dispense could detect NOx, C3H8, and NH3 at the 100 PPM level [4]. We will report on the printing of sensor components using syringe dispense or aerosol jet printing using materials selected for sensitivity to CH4, heavier hydrocarbons, other subcomponents which allow the fingerprinting of methane emissions. Parameters optimized include paste formulation, sintering schedule, and device geometry. We will also report on hardware interfaces developed to perform data acquisition and the application of machine learning algorithms for mixture quantification and identification.This project was supported by the Department of Energy, Office of Fossil Energy, through award DE-FE0031864.