Metal Oxide Semiconductor sensors (MOS) for measuring Volatile Organic Compounds (VOC) - performance evaluation in residential settings

Metal Oxide Semiconductor (MOS) sensors measuring Volatile Organic Compounds (VOC) seem to be an obvious step towards broadly available Demand Controlled Ventilation (DCV). The previous research shows that MOS VOC sensors can detect high pollution events such as cleaning, painting, or high occupation density. These abilities seem to make MOS VOC sensors suitable to complement ventilation control systems, especially concerning residential ventilation. However, several questions come from the practice: “Are the MOS VOC sensors reliable and stable enough to be applied in practice?” “Are there any benefits concerning energy efficiency and indoor environmental quality?” They remain unanswered. Studies on the long-term performance of MOS VOC sensors exposed to real-life environments are lacking. Some producers test their sensors in a laboratory environment, but such data are often not publicly available. Data about the influence of ventilation control based on MOS VOC sensors on energy efficiency are also missing. The present paper reports first results from a project aiming to answer aforementioned questions having following objectives: investigate performance of MOS VOC sensors exposed to a typical residential environment. Determine sensor properties – sensitivity, linearity, hysteresis by comparing their signal with a reference measurement conducted by PID (Photo Ionization Detector). Discuss the suitability of the sensors for control of residential ventilation. We measured in a typical Danish row house occupied by a family of four. We used two sets of three commercially available sensors installed two locations-bedroom and kitchen. PID gas analyzer served as a reference measurement. The results show that all tested sensors were able to indicate the pollution events like human presence or cleaning. There was a fair agreement among the signals of the two tested sensors. These sensors produced also signals, which were in a clear relationship to the reference measurements. In the opposite, the signal from the third sensor could be clearly related neither to the reference signal nor to the other two tested sensors. This is potentially problematic for sensor’s application for ventilation control.