Studies on the size distribution, number and mass emission factors of candle particles characterized by modes of burning

This study presents a model-based method to determinate the size-specific number and mass emission factors of candle smoke particles characterized by three modes of burning. A series of candle burning experiments are carried out in a test chamber. Two different emission models are employed to translate the concentration data into particle number and mass emission factors. The number concentration is processed by an aerosol number emission model, in which the effects of coagulation, deposition and ventilation on particle change are considered. A single-compartment mass balance model is used to compute the mass emission factors. Candle emission is associated with three different modes of burning, which are defined as steady burn, unsteady burn and smoldering. The number emission values of candle smoke particles of the steady burn, unsteady burn and smoldering modes are , and , respectively, and the corresponding PM10 emission factors are , and , respectively. The size-specific emission rates reveal that the size distributions of the aged particles have undergone a big change comparing to the freshly emitted ones during the emission period, especially for particles of the steady burn and unsteady burn modes. The advantage of the current method is that it can provide detailed size-specific information, since particle growth due to coagulation is taken into account in the number emission model. With this method, the size distribution, CMD and geometric standard deviation (GSD) of the freshly emitted particles are acquired, and the optimization technique ensures that the computed concentrations with the emission rates are close to the measured results. The current work not only presents detailed candle emission data for numerical simulation of indoor contaminates dispersion and air quality evaluation, but also suggests an approach of how to interpret experimental data into emission based on aerosol dynamic models.