Comparison of PM_2.5 emission rates and source profiles for traditional Chinese cooking styles

The number of restaurants is increasing rapidly in recent years, especially in urban cities with dense populations. Particulate matter emitted from commercial and residential cooking is a significant contributor to both indoor and outdoor aerosols. The PM_2.5 emission rates and source profiles are impacted by many factors (cooking method, food type, oil type, fuel type, additives, cooking styles, cooking temperature, source surface area, pan, and ventilation) discussed in previous studies. To determine which cooking activities are most influential on PM_2.5 emissions and work towards cleaner cooking, an experiment design based on multi-factor and level orthogonal tests was conducted in a laboratory that is specifically designed to resemble a professional restaurant kitchen. In this cooking test, four main parameters (the proportion of meat in ingredients, flavor, cooking technique, oil type) were chosen and five levels for each parameter were selected to build up 25 experimental dishes. Concentrations of PM_2.5 emission rates, organic carbon/elemental carbon (OC/EC), water-soluble ions, elements, and main organic species (PAHs, n-alkanes, alkanoic acids, fatty acids, dicarboxylic acids, polysaccharides, and sterols) were investigated across 25 cooking tests. The statistical significance of the data was analyzed by analysis of variance (ANOVA) with ranges calculated to determine the influence orders of the 4 parameters. The PM_2.5 emission rates of 25 experimental dishes ranged from 0.1 to 9.2 g/kg of ingredients. OC, EC, water-soluble ions (WSI), and elements accounted for 10.49-94.85%, 0-1.74%, 10.09-40.03%, and 0.04-3.93% of the total PM_2.5, respectively. Fatty acids, dicarboxylic acids, n-alkanes, alkanoic acids, and sterols were the most abundant organic species and accounted for 2.32-93.04%, 0.84-60.36%, 0-45.05%, and 0-25.42% of total PM_2.5, respectively. There was no significant difference between the 4 parameters on PM_2.5 emission rates, while a significant difference was found in WSI, elements, n-alkanes, and dicarboxylic acids according to ANOVA. Cooking technique was found to be the most influential factor for PM_2.5 source profiles, followed by the proportion of meat in ingredients and oil type which resulted in significant difference of 183.19, 185.14, and 115.08 g/kg of total PM_2.5 for dicarboxylic acids, n-alkanes, and WSI, respectively. Strong correlations were found among PM_2.5 and OC (r = 0.854), OC and sterols (r = 0.919), PAHs and n-alkanes (r = 0.850), alkanoic acids and fatty acids (r = 0.877), and many other species of PM_2.5.