Nafion Based Two Electrode CO Sensor

The aim of this work is to determine the optimal platinum loading for the preparation of membrane-based electrochemical sensors for carbon monoxide detection. Platinum is a required constituent of the electrode. Its function is to catalyse CO as it enters the sensor. Initial results show that a Nafion based sensor hydrated with small amounts of sulphuric acid can result in a response in the 30 nA/ppm range. Nafion based sensors were also stored in 10% RH for 63 days, to investigate what effect this would have on sensor output. Sensor response decreased on average by only 11%. Our results show that Nafion is an excellent ionic membrane for CO sensing. Keys words: Platinum Catalyst, Electrochemical Sensor, Carbon Monoxide Oxidation, Solid Electrolyte. Introduction Pt is a very effective catalyst used for both oxidation and reduction in electrochemical sensors and fuel cells [1, 2]. It is used almost exclusively in the residential carbon monoxide (CO) detection industry for oxidizing CO which occurs through the steps shown in equations 1-3. The industry standard for residential CO detection is the liquid electrolyte CO sensor. This sensor type is very reliable due to its stable and ultra-low drift current output. However, the sensor requires a liquid electrolyte reservoir which takes up much of the size of the sensor. Currently, there is research in the area of sensor miniaturization, this would allow for new technologies such as wearable CO detectors and sensors inbuilt into mobile technologies such as smartphones. For this to be achieved the liquid electrolyte will have to be replaced with an alternative electrolyte such as a polymer. This has other benefits such as increased mechanical stability, and it would also reduce the risk of leaks. A well-known polymer proton electrolyte is the copolymer of poly(tetrafluoroethylene) with poly(sulphonylfluoride vinyl ether, known as Nafion. It is manufactured by Dupont [3]. It is a cation-exchanger containing hydrophilic sulpho groups firmly bound to the hydrocarbon backbone, whose charge is compensated by counterions (mostly H + ) [4]. This electrolyte was tested as a replacement for pure sulphuric acid-based electrolyte and the results are presented in this work. Results Initially, 1 mm thick dry Nafion was used in the sensor and it was tested in 100 ppm CO. This sensor type showed a poor response to CO. The Nafion was then hydrated with a small quantity of sulphuric acid. This increased the response of the sensor significantly. The output of three of these sensors is shown in Fig 1. These sensors have a Pt loading of 0.2 g/cm 3 . When the CO was pumped into the sensors, their outputs rose quickly. The average output of the sensors was 27 nA/ppm. The output remained steady until the CO was removed. Their output then quickly returned to the baseline. Three additional sensors were then built and stored in 10% RH. The sensors were removed from the storage conditions every two weeks and left in ambient conditions for four hours before being tested in 100 ppm CO. Their response and the average response of three liquid electrolyte sensors are shown in Fig 2. The initial average output was 28.5 nA/ppm. Every 14 days the sensors were removed from the storage conditions and allowed to settle in ambient conditions for 4 hours. The sensors response was then tested in 100 ppm of CO. This was done for 63 days. The graph shows sensors outputs decrease steadily with time while in the storage conditions. However, the overall output of the sensor is still relatively high and comparable with a standard liquid electrolyte sensor. The components of a Nafion sensor are shown in Fig 3. Conclusions and Future Work We have shown that at high Pt loading a reliable two electrode CO sensor can be made. The output of these sensors decreases when stored in low humidity. Further investigation will be carried out to find the optimal catalyst loading for the sensor. The sensors will also be tested in other accelerated conditions such as high humidity, high temperature, and low temperature. References [1] J. Wu, H. Yang, Acc. Chem. Res., vol.46, 1848, 2013. [2] F. M. F. Rhen, C. McKeown, J. Phys. Chem. C, vol. 121, 2556, 2017. [3] F. Opekar, K. Stulik, Analytica Chimica Acta, vol. 385, 151, 1999. [4] C. Yu et al., Sensors and Actuators B: Chemical, vol. 86, 259, 2002. Acknowledgements This work is supported by the Irish Research Council (EBP/2017/435). Figure 1