Abstract. The impact of a hexamethyldisiloxane (HMDSO) treatment on the response of doped SnO2 sensors is investigated for acetone, carbon monoxide and hydrogen. The sensor was operated in temperature cycles based on the DSR concept (differential surface reduction). According to this concept, the rate constants for the reduction and oxidation of the surface after fast temperature changes can be evaluated and used for quantification of reducing gases as well as quantification and compensation of sensor poisoning by siloxanes, which is shown in this work. Increasing HMDSO exposure reduces the rate constants and therefore the sensitivity of the sensor more and more for all processes. On the other hand, while the rate constants for acetone and carbon monoxide are reduced nearly to zero already for short treatments, the hydrogen sensitivity remains fairly stable, which greatly increases the selectivity. During repeated HMDSO treatment the quasistatic sensitivity, i.e. equilibrium sensitivity at one point during the temperature cycle, rises at first for all gases but then drops rapidly for acetone and carbon monoxide, which can also be explained by reduced rate constants for oxygen chemisorption on the sensor surface when considering the generation of surface charge.

Introduction The impact of siloxane exposure on SnO2 based gas sensors is highly important in many applications due to the frequent occurrence of these substances in ambient air. This process is usually called poisoning since siloxane exposure leads to permanently impaired sensors, changing selectivity and decreasing sensitivity to gases [1] except for hydrogen [2]. The often observed first increase in sensitivity led to the complex interpretation of several surface sites [3]. Two independent investigations on the effect of siloxanes on the commercially available UST GGS 1530 (Umweltsensortechnik GmbH, Geschwenda, Germany) were conducted yielding the same output: the reactivity of the sensor’s surface is continuously reduced during siloxane exposure. Experimental The first investigation focused on the effects on the dynamic operation called differential surface reduction (DSR, described in [4]), where the sensor is alternating between a high temperature oxidation phase and a low temperature reduction phase, which enables the independent measurement of both processes. The rate constant for reducing gases reacting on the sensor surface is evaluated through the first derivative of the logarithmic conductivity ln(G) on the low temperature phase. Turning the DSR method around allows the evaluation of the differential surface oxidation (DSO) through the dynamic behavior of ln(G) during the high temperature phase. The second investigation targeted gas sensors for monitoring methane concentrations (<4.4 %) in harsh environments contaminated with siloxane. The measurements were performed in a gas mixing system similar to the system described in [4] consisting of mass flow controllers and valves. During the first investigation the humidity was always held at 50 % R.H, the second investigation included measurements at 0, 10 and 65 % R.H., supplied by a water bubbler at 20 °C. further described in [5]. (table 1) and increasing sensor temperature were performed, gas measurements with 2 ppm of hydrogen were conducted in between. The second investigation used the most abundant siloxane OMCTS (octamethylcyclotetrasiloxane) and the 2 l/min dilution MFC was omitted due to the low vapor pressure. The sensor was constantly operated at 400 °C. conducted, several methane concentrations and backgrounds were measured in between as seen in figure (Fig. 2). Results The main result from the first investigation on hydrogen is given in Fig. 1. Features from both DSR (middle, 300 °C) and DSO (right, 450 C°) are declining with increasing siloxane dosage, while the sensor response (left, 300 °C) shows no distinct trend. This indicates that siloxane exposure gradually slows down both surface reactions, which corresponds to a deactivation of active sites, i.e. a lower reactivity of the sensor surface, whereas the static equilibrium is shifted in different directions depending on the relative effect on hydrogen and oxygen reactivity at each treatment step. The deactivation of active surface sites can also be observed in steady state operation by an exposure to a high concentration of reducing gas, e.g. methane. The reaction (“combustion”) of methane creates excess heat reducing the heater power at constant temperature similar to a pellistor [6]. This effect is offset by the change in thermal conductivity of the surrounding gas which increases the heater power for higher methane concentrations. Siloxane exposure reduces the reaction rate and therefore the equilibrium between both effects is shifted towards thermal conductivity, thus increasing the power consumption. This is shown in Fig. 2, where the heater power difference in different gas compositions compared to dry air is plotted vs. the OMCTS dose. Results for 1 % methane in dry air shift from the combustion dominated (up to 10 ppmh OMCTS) to the thermal conductivity dominated (from 500 ppmh) region. At 65 % R.H., the thermal conductivity effect is always dominant independent of the OMCTS dosage, but also increases with siloxane exposure. Both described methods show independently that siloxane exposure decreases the reaction rates on the sensor surface affecting reduction and oxidation process, which can be exploited for sensor self-monitoring. In temperature cycled operation the resulting deterioration could be determined by evaluating the high temperature dynamic behavior, which is mainly reflecting the surrounding oxygen concentration. In atmospheres with varying oxygen concentration or other circumstances which interfere with the DSO evaluation, a reference gas mixture could be used to determine the balance between combustion and thermal conductivity as an indicator of the sensor state. Acknowledgement Part of this research was done within the project “GasMOS” funded by the European Regional Development Fund (ERDF). We want to thank our project partner Schaller automation GmbH & Co. KG for their ongoing support. References [1] G. Korotcenkov and B. K. Cho, Sensors Actuators, B Chem., vol. 156, no. 2, pp. 527–538, 2011. doi:10.1617/s11527-012-0006-0 [2] M. Fleischer, S. Kornely, T. Weh, J. Frank, and H. Meixner, Sensors Actuators, B Chem., vol. 69, no. 1, pp. 205–210, 2000. doi:10.1016/S0925-4005(00)00513-X [3] D. E. Williams and K. F. E. Pratt, J. Chem. Soc. - Faraday Trans., vol. 94, no. 23, pp. 3493–3500, 1998. doi:10.1039/a807644h [4] M. Leidinger, C. Schultealbert, J. Neu, A. Schütze, and T. Sauerwald, Meas. Sci. Technol., vol. 29, no. 1, 2018. doi:10.1088/1361-6501/aa91da [5] M. Schüler, T. Sauerwald, and A. Schütze, J. Sensors Sens. Syst., vol. 4, no. 2, pp. 305–311, 2015. doi:10.5194/jsss-4-305-2015 [6] X. Liu, S. Cheng, H. Liu, S. Hu, D. Zhang, and H. Ning, vol. 12, no. 7, pp. 9635–9665, 2012. doi:10.3390/s120709635 Figure 1

No abstract