Chemical reactions between ozone and pollutants commonly found indoors have been suggested to cause adverse health and comfort effects among building occupants. Of special interest are reactions with terpenes and other pollutants containing unsaturated carbon-carbon bonds that are fast enough to occur under normal conditions in various indoor settings. These reactions are known to occur both in the gas phase (homogeneous reactions) and on the surfaces of building materials (heterogeneous reactions), producing a number of compounds that can be orders of magnitude more odorous and irritating than their precursors.
The present thesis investigates the effects of ozone-initiated reactions with limonene and with various interior surfaces, including those associated with people, on short-term sensory responses. The evaluations were conducted using a perceived air quality (PAQ) method introduced by Fanger (1988). The experiments, involving hundreds of human subjects and subsequent physical and chemical measurements, were conducted under realistic indoor conditions in unfurnished office-like environments, in stainless-steel chambers and inside a full-scale model of a section of an airplane cabin.
These experiments have shown that the common occurrence of ozone and limonene at typical concentrations and ventilation rates encountered indoors can significantly reduce the perceived air quality even under conditions when these pollutants cannot be sensed if present by themselves. Many of the reaction products that are typical for ozone-limonene chemistry have been concomitantly identified with real time measurement using proton-transfer-reaction mass spectrometry (PTR-MS), at concentrations high enough to be responsible for the sensory effects reported. The stabilized reaction products of ozone-limonene chemistry including aldehydes, ketones and carboxylic acids are known to partition between the gas phase and condensed phase according to the vapour pressure specific to each compound. The concentrations of condensed phase products, which can be easily detected by ordinary particle counters, were shown to be proportional to the magnitude of the sensory responses. However, the particles themselves did not appear to be the primary causative agent, but instead are co-varying surrogates for sensory offending gas-phase species.
The experiments involving heterogeneous reactions of ozone with various indoor materials also showed that these reactions can significantly alter the nature of emitted pollutants from indoor surfaces. One set of experiments with various carpet samples showed that although ozone initiated reactions reduced to some extent the initial strong odor of a new carpet, ozone exposure of old carpets significantly enhanced the emissions of odor offending compounds that can persist for hours and days after ozone exposure has ended.
The PTR-MS measurements conducted in the simulated aircraft cabin demonstrated that the presence of ozone had significantly increased the concentrations of numerous oxidized compounds in the air of the cabin environment. The most abundant oxidation products were saturated and unsaturated aldehydes and tentatively identified low-molecular-weight carboxylic acids. Some of these compounds were detected at concentrations high enough to trigger the human olfactory sense. When the cabin contained soiled T-shirts, as well as ozone, the concentration of products derived from oxidized skin oil was significantly higher than when the cabin contained ozone alone.
Detailed measurements of ozone removal in the aircraft cabin, under systematically varied conditions have shown how different surfaces, including seats, recirculation filters and people themselves contribute to overall ozone removal. People are the largest ozone sink, removing almost 60% of ozone in the cabin and its recirculation system. The aircraft seats, that are contaminated with human bioeffluents and represent a large surface area in the cabin, were the second largest ozone sink, removing about 25% of the ozone. To a smaller extent ventilation filters (~7%) and other surfaces (~10%) also contributed to the removal of ozone.