Iso- and Anteiso-Alkanes: Specific Tracers of Environmental Tobacco Smoke in Indoor and Outdoor Particle-Size Distributed Urban Aerosols ILIAS G. KAVOURAS, NIKOLAOS STRATIGAKIS, AND EURIPIDES G. STEPHANOU* Environm ental Chem ical Processes Laboratory (ECPL), Division of Environmental and Analytical Chemistry, Departm ent of Chem istry, University of Crete, 71409 Heraclion, Greece Branched iso- and anteiso- alkanes were conjointly used, with n- alkanes and PAHs, as specific molecular markers to trace environmental tobacco smoke (ETS) in particle- sized aerosols collected in the indoor and outdoor urban atmosphere. GC/MS and GC-FID were used for the de- termination of iso- , anteiso- , and n- alkanes and PAHs. The branched alkanes (ranging from C 29 to C 33 ) were detected only in particles in the accumulation range mode (<1.5 μm) in both indoor and outdoor samples. The concentrations of iso- and anteiso- alkanes in the indoor aerosols (0.75- 8.53 ng/m 3 ) were higher than those measured in outdoor samples (0.77-1.51 ng/m 3 ). The indoor aerosol pattern of iso- , anteiso- , and their calculated diagnostic concentration ratios were characteristic for ETS. The compound distribution pattern of indoor n- alkanes (ranging from C 21 to C 33 ) was of biogenic origin, and the use of odd-to- even predominance running ratio curves indicated their cigarette smoke origin. The corresponding outdoor pattern and concentration ratios, although less characteristic than the indoor ones, also indicated ETS as the main source of these compounds. The distribution study of the branched alkanes between gas and particulate phase in indoor aerosol demonstrated their presence only in the particles. On the other hand,PAHs in the gas phase gave a compound pattern more characteristic of ETS components than the PAHs present in the particulate phase. Iso- and anteiso- alkanes, due to their specificity, their nonreactive character, their association with the accumulation range mode particles, and therefore, their long atmospheric residence time, are the most suitable tracers for particulate ETS emissions in the indoor and outdoor urban atmosphere. Introduction Environmentaltobacco smoke (ETS)is the materialreleased into the ambient atmosphere bysmokingtobacco products. Cigarettes are the primary concern because of their wide- spread use (e.g., 600 billion cigarettes are consumed in the United States per year) (1). Cigarette-generated ETS com- pounds derive primarily from side-stream smoke emitted between puffs and differ as the smoke mixture is diluted and aged. Released reports (1, 2) include sufficient evidence to conclude that concerns about respiratory disease and the impairmentoflungcancerdevelopmentduetoETSexposure cannot be ruled out. Therefore, in the above-mentioned reports, research was required to determine the physical properties and chemical composition of ETS respirable suspended particulate matter (RSP), determining the gas- particulate phase distribution of ETS constituents and identifying suitable molecular markers for ETS exposure. During the past decades, studies focusing on the gaseous and particulate smoke constituents have been conducted (3-11). The most important problem for environmental tobacco smoke indoor monitoring was to identify the correlation between ETS species and those compounds responsible for the observed human health effects (1, 2). There are many factors for the selection of molecular markers or tracers ofETS. The knowledge ofgas-to-particle distribution of the compounds is useful because of the different removal rates of each phase (12). On the other hand,indoorparticlesoriginatingfrom ETSwillbeexchanged very rapidly with the outdoors, influencing the external particulate composition (13). Recently,researchersreported that cigarette smoke accounted for about 2.7% of the urban fine organic aerosol material (14). Therefore, appropriate molecular markers should be used to trace ETS presence outdoors as well. Tracers of ETS used in the past include CO (5), nicotine (5), N- nitrosamines (5, 6),urinaryconcentrationsofnicotine and cotinine (5), and sterols and sterenes (9). Nicotine is a unique gaseous compound and a major constituent of ETS. It could be a good biological marker as it metabolizes to cotinine. Cotinine is stable in the body, at detectable levels for several days (9). Nitrogen - containing compounds such as N- nitrosamine and 3-ethenylpyridine were also deter- mined in several indoor environments in gaseous form (4- 7). But the gas-to-particle distribution ofnicotine and other nitrogen - containing compounds is a disadvantage to the suitability of these compounds as molecular tracers for ETS (4, 5). RSP has been suggested as a potential tracer of ETS. However, RSP is emitted from several sources; therefore, it is neither specific for ETSnor representative for the exposure to the ETS gaseous constituents (8). PAHs are not source- specific and are susceptible to photodegradation in the atmospheric environment. Solanesol, a trisequiterpenoid alcohol (C45H74O), was also considered as a suitable par- ticulate compound to trace ETS in the indoor environment (10). This compound is an alcoholwith nine double bonds, and it was found to be very reactive under atmospheric conditions (e.g., O3 and photooxidation). If the above- mentioned compounds are not the most appropriate ETS tracers for the indoor atmosphere, they are certainly not suitable for the same use in the outdoor environment. 2-Methyl- (iso- ) and 3-methylalkanes (anteiso- alkanes) ranging from C29 to C34 had been proposed and successfully used as specific ETS tracers outdoors (11, 14, 15). These compounds show a characteristic concentration pattern in tobacco leaf surface wax and, consequently, in tobacco smoke, which is different from the one found in the leaf surface of other plants (11). In addition, none of these compoundswasdetected in crude oilsasa result ofdiagenetic and catagenetic processes having occurred during oil forma- tion (11). Therefore, iso- and anteiso-alkanes demonstrate * To whom correspondence should be addressed; e-mail: stephanou@chemistry.uch.gr. Environ. Sci. Technol. 1998, 32, 1369-1377 S0013-936X(97)00634-2 CCC: $15.00  1998 American Chemical Society VOL. 32, NO. 10, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1369 Published on Web 04/08/1998