Use of electronic cigarettes (e-cigarettes) is rapidly growing around the world. E-cigarettes are commonly used as an alternative nicotine delivery system, and have been advocated as generating lower levels of harmful chemicals compared to conventional cigarettes. Cigarette smoke-like aerosols are generated when e-cigarettes heat e-liquids. The main components of e-cigarette liquids are propylene glycol (PG) and glycerol (GL) in a varying ratio, plus nicotine and flavor chemicals. Both PG and GL are considered safe to ingest in foods and beverages, but the toxicity of these chemicals in aerosols is unknown. Current studies of e-cigarettes have mainly focused on dehydration and oxidation products of PG and GL. In this study, the other degradation products that can be generated during the vaping process are discussed. In addition, the gas/particle partitioning of chemicals in vaping aerosols is determined.
This work finds that the formation of benzene in electronic cigarettes depends on the wattage, types of coils, and devices. To simulate commerical e-cigarette liquids, mixtures containing equal parts of PG and GL by volume were made with the following added components: benzoic acid (BA), benzoic acid with nicotine (Nic), benzaldehyde (BZ), band enzaldehyde with nicotine. PG only, GL only, and PG and GL mixtures were also made for comparison. The data presented here demonstrate that more benzene is generated as the wattage of a device increases. The results also seem to support the importance of ventilation in the generation of benzene. More benzene is generated from the mixtures containing benzoic acid when using the EVOD device with a smaller vent. However, benzaldehyde yields more benzene when using the Subtank Nano device with a larger vent. Findings also indicate that more benzene is produced from GL rather than PG.
This thesis also addresses the chemical formation pathways of degradation compounds found in the aerosols formed from isotopically labeled e-cigarette liquids. Mixtures of both 13C-labeled and unlabeled PG as well as GL were made. The mixtures were vaped and gas-phase samples were collected to determine which chemicals were in the gas-phase portion of the aerosols. With the use of GC/MS methods, these isotopic labeling experiments provided evidence that the majority of the benzene, acetaldehyde, 2,3-butanedione, toluene, xylene, acrolein, and furan found in e-cigarette aerosols originates from GL in the PG plus GL mixtures. It was also shown that the majority of propanal is derived from PG: while hydroxyacetone can be formed from both PG and GL. Possible mechanisms for the formation of acetaldehyde, benzene, 2,3-butanedione, toluene, and xylene formation are proposed.
Last, this study investigated the gas/particle partitioning of nicotine and flavor-related chemicals in e-cigarette fluids. The gas/particle partitioning behavior of chemicals in e-cigarettes fluids is highly dependent on the chemical volatility. A total of 37 compounds were examined. The target compounds were divided into 3 groups based on their vapor pressures: high, medium, and low. Headspace gas samples were collected and analyzed to determine the concentration of a compound in equilibrium with the liquid phase. The gas and liquid concentrations were used to calculate the gas/particle partitioning constant (Kp) for each compound. In an e-cigarette aerosol, volatile compounds have smaller Kp values and tend to be found in greater proportion in the gas-phase, whereas the less volatile compounds are likely to stay in the particle phase. General agreement with theory was found for compounds with known activity coefficients in PG and GL, indicating that theory can be used to predict Kp values for other compounds.
Identifer | oai:union.ndltd.org:pdx.edu/oai:pdxscholar.library.pdx.edu:open_access_etds-4953 |
Date | 11 September 2017 |
Creators | Kim, Kilsun |
Publisher | PDXScholar |
Source Sets | Portland State University |
Detected Language | English |
Type | text |
Format | application/pdf |
Source | Dissertations and Theses |
Page generated in 0.0015 seconds