Aerosol Chemistry and Physics - Aerosols: Measurement, Dosimetry, and Health Effects - page 141

As more evidence has emerged about the health effects of very-small-sized PM, many research
groups around the world have been measuring size distributions of UFPs generated by indoor
sources, and the tables include some of their work. These efforts have taken advantage of the wide-
spread availability of portable instrumentation for measuring particle number size distributions. For
example, He et al. (2004) reported emission rates for many indoor activities in units of both particle
mass and particle number, and Glytsos et al. (2010) showed how quickly nanosized particles from
many indoor sources agglomerated after they were generated under the same conditions.
6.2.1 c oMbustion s ources
Indoor combustion sources typically operate less eficiently than vehicle engines or industrial boilers,
leading to higher emission factors for unburned fuel and partially oxidized products. Semi-volatile
combustion-generated gases cool, condense onto existing particles, or nucleate and coagulate with
existing particles. The aerosol community is indebted to the late Glenn Cass for leading compre-
hensive particle characterization efforts at the California Institute of Technology, and many studies
from his laboratory are cited in the following. Starting in the early 1990s Rogge et al. (1991, 1998)
began reporting detailed chemical characterization of particulate matter from combustion sources
that had been operated under carefully controlled conditions, as described by Hildemann et al.
(1991). Schauer et al. (1999a,b, 2002a,b,c) expanded these investigations by characterizing both the
gas and particulate phases of many of the same sources. In spite of this large body of work and the
contributions of many other investigators, identiied components typically account for less than half
of the ine particulate mass emitted by each combustion source.
The comprehensive source characterization efforts of the 1990s were initially driven by the need
to identify and quantify the contribution of each particle source to urban air pollution. The goal was
accurate source apportionment so that appropriate control measures could be implemented. With
growing recognition and understanding of the health effects of exposure to PM 2.5 , many of the tools
developed for the characterization of ambient PM have now been applied to measure human expo-
sure to ine (PM 2.5 ) and smaller (ultraine) particles from indoor combustion sources.
6.2.1.1 Cooking
6.2.1.1.1 Cooking with Oils and Meat
Schauer et al. (2002b) found that frying vegetables in seed oils led primarily to the formation of ine
particulate alkenoic and alkanoic acids, whereas the semi-volatile gases contained mostly saturated
and unsaturated aldehydes. Emission factors for PM 2.5 ranged from 13 to 30 mg kg −1 of stir-fried
vegetables. Andrejs et al. (2002) found that particles larger than about 4 μm were composed primar-
ily of triglyceride components of the original oil. Hildemann et al. (1991) found that charbroiling
meat yielded more ine PM (7-40 g kg −1 meat) than Schauer et al. (2002b) found from cooking with
oils. Rogge et al. (1991) and Schauer et al. (1999a) reported more classes of organic compounds from
meat cooking than from the use of seed oils (Schauer et al. 2002b). Of those, saturated and unsatu-
rated fatty acids were the most abundant, as shown in Figure 6.3. However, only about 10% of the
particulate organic carbon from meat cooking could be traced to individual compounds.
Cooking with oil produces large numbers (billions to trillions per cooking event) of UFPs (He
et al., 2004; Olson and Burke, 2006; Wallace, 2006; Wallace et al., 2006; Evans et al., 2008;
Glytsos et al., 2010). In the introduction to their results on particle dose from frying food, Evans
et al. (2008) cited studies that point to fumes from cooking hot oil as an important underlying
cause of lung cancer in women who have never smoked (e.g., Straif et al., 2006). During the last
decade, Yang et al. (2007) found that particles from frying contain mutagenic dienals (aldehydes
with two double bonds), among other compounds such as the aromatic amines and polycyclic aro-
matic hydrocarbons that were found by Schauer et al. (2002b) and To et al. (2007). These classes
of compounds cause cytotoxicity and DNA damage. Using measured particle number and mass
emission rates for frying a variety of foods in a house Evans et al. (2008) estimated that 20 min of