Evaluation of Ultra-Violet Photocatalytic Oxidation (UVPCO) for Indoor Air Applications: Conversion of Volatile Organic

Efficient removal of indoor generated airborne particles and volatile organic compounds (VOCs) in office buildings and other large buildings may allow for a reduction in outdoor air supply rates with concomitant energy savings while still maintaining acceptable indoor air quality in these buildings. Ultra-Violet Photocatalytic Oxidation (UVPCO) air cleaners have the potential to achieve the necessary reductions in indoor VOC concentrations at relatively low cost. In this study, laboratory experiments were conducted with a scaled, prototype UVPCO device designed for use in a duct system. The experimental UVPCO contained two 30 by 30-cm honeycomb monoliths coated with titanium dioxide and 3% by weight tungsten oxide. The monoliths were irradiated with 12 UVC lamps arranged in four banks. The UVPCO was challenged with four mixtures of VOCs typical of mixtures encountered in indoor air. A synthetic office mixture contained 27 VOCs commonly measured in office buildings. A cleaning product mixture contained three cleaning products with high market shares. A building product mixture was created by combining sources including painted wallboard, composite wood products, carpet systems, and vinyl flooring. A fourth mixture contained formaldehyde and acetaldehyde. Steady-state concentrations were produced in a classroom laboratory or a 20-m3 environmental chamber. Air was drawn through the UVPCO, and single pass conversion efficiencies were measured from replicate air samples collected upstream and downstream of the reactor section. Concentrations of the mixtures were manipulated, with concentrations of individual VOCs mostly maintained below 10 ppb. Device flow rates were varied between 165 and 580 m3/h. Production of formaldehyde, acetaldehyde, acetone, formic acid, and acetic acid as reaction products was investigated. Conversion efficiency data were generated for 48 individual VOCs or groups of closely related compounds. Alcohols and glycol ethers were the most reactive chemical classes with conversion efficiencies often near or above 70% at the low flow rate and near 40% at the high flow rate. Ketones and terpene hydrocarbons were somewhat less reactive. The relative VOC conversion rates are generally favorable for treatment of indoor air since many contemporary products used in buildings employ oxygenated solvents. A commercial UVPCO device likely would be installed in the supply air stream of a building and operated to treat both outdoor and recirculated air. Assuming a recirculation rate comparable to three times the normal outdoor air supply rate, simple mass-balance modeling suggests that a device with similar characteristics to the study unit has sufficient conversion efficiencies for most VOCs to compensate for a 50% reduction in outdoor air supply without substantially impacting indoor VOC concentrations. Formaldehyde, acetaldehyde, acetone, formic acid, and acetic acid were produced in these experiments as reaction byproducts. No other significant byproducts were observed. A coupled steady-state mass balance model is presented and applied to VOC data from a study of a single office building. For the operating assumptions described above, the model estimated a three-fold increase in indoor formaldehyde and acetaldehyde concentrations. The outcome of this limited assessment suggests that evaluation of the potential effects of the operation of a UVPCO device on indoor concentrations of these contaminants is warranted. Other suggested studies include determining VOC conversion efficiencies in actual buildings and evaluating changes in VOC conversion efficiency as monoliths age with long-term operation.

This report was presented at Proceedings of Healthy Buildings 2006 Conference in Lisbon, Portugal. Volume 2. Pagination 243-248.

Abstract for conference proceedings: This paper summarizes three analyses of data on building-related environmental factors and occupant symptoms collected from 100 representative large U.S. office buildings. Using multivariate logistic regression models, we found increased occupant symptoms associated with a number of building-related factors, including lower ventilation rates even at the current guideline levels, lack of scheduled cleaning for air-conditioning drain pans and cooling coils, poor condition of cooling coils, poorly maintained humidification systems, and lower outdoor air intake height. Some expected relationships were not found, and several findings were opposite of expected. Although requiring replication, these findings suggest preventive actions to reduce occupant symptoms in office buildings.