Measurement and evaluation of gaseous and particulate emissions from burning scented and unscented candles.

It has been known for a long time that incomplete combustion processes produce by-products that are harmful to human health. Particularly high concentrations of such by-products can arise in indoor environments when operating open flames without venting. The emission behavior of many combustion sources, including candles, has already been examined in detail. However, to date there are no studies in which the chemical composition of the candles is known exactly or where the candles were specifically manufactured for comparative measurements. In this respect, the study presented here, which was designed in collaboration with candle manufacturers and fragrance houses, demonstrates new insights into the emissions of burning candles depending on their composition. All investigations were carried out under controlled climatic conditions in an 8 m stainless steel chamber. Combinations of four different fuels (waxes) and five different fragrances in addition to one set of unscented control candles were examined. This resulted in 24 experiments, 20 with scented candles and four with unscented candles. The typical combustion gases carbon monoxide, carbon dioxide and NO, organic compounds, such as formaldehyde, benzene, and polycyclic aromatic hydrocarbons, PM and ultrafine particles were monitored in the chamber air and the emission rates were determined. The data were statistically evaluated using parametric and non-parametric methods as well as hierarchical cluster analysis. Exposure scenarios typical for indoor environments were calculated from the emission rates and the results were compared with indoor guidance and reference values. As expected, a multitude of gaseous and particulate emissions were detected. These were typical combustion products as well as evaporated constituents of the fragrance mixtures. In most cases, the calculated indoor concentrations were well below the respective guidance and reference values. The exceptions observed in some cases for nitrogen dioxide, acrolein and benzo[a]pyrene are discussed critically.