RESUMEN
Literature reports the chemical constituent yields of electronic nicotine delivery systems (ENDS) aerosol collected using a range of aerosol collection strategies. The number of puffs to deplete an ENDS product varies widely, but collections often consist of data from the first 50-100 puffs. However, it is not clear whether these discrete puff blocks are representative of constituent yields over the life of a pod. We aimed to assess the effect of differing aerosol collection strategies on reported yields for select chemical constituents in the aerosol of closed pod-based ENDS products. Constituents analyzed were chosen to reflect important classes of compounds from the Final Premarket Tobacco Product Application Guidance. Yields were normalized to total device mass loss (DML). Collection strategies that consisted of partial pod collection were valid for determining yields of constituents whose DML normalized yields were consistent for the duration of pod life. These included primary aerosol constituents, such as propylene glycol, glycerol, and nicotine, and whole pod yields could be determined from initial puff blocks. However, changes were observed in the yields of some metals, some carbonyl compounds, and glycidol over pod life in a chemical constituent and product dependent manner. These results suggest that collection strategies consisting of initial puff block collections require validation per chemical constituent/product and are not appropriate for chemical constituents with variable yields over pod life. Whole pod collection increased sensitivity and accuracy in determining metal, carbonyl, and glycidol yields compared to puff block-based collection methodologies for all products tested.
RESUMEN
1,2-Propylene glycol and glycerin, principal components of e-liquids, can thermally degrade to form acetaldehyde, acrolein, and formaldehyde when heated in electronic nicotine delivery systems (ENDS). Recently the flavors in e-liquids were suggested to be the major source of these aldehydes. We used the same 10 ENDS devices to test 5 e-liquid formulations (four flavored & one corresponding non-flavored) and measured device mass loss and levels of acetaldehyde, acrolein, and formaldehyde (30 replicate measurements per formulation). Despite finding reasonable variability in measurements of device mass loss, two out of 10 ENDS devices tested produced outlier values for aerosol levels acetaldehyde, acrolein, and formaldehyde. After removing these devices from further analysis, acceptable variability (≤20% RSD) in aerosol levels of acetaldehyde, and formaldehyde were found. The flavored formulations tested resulted in a consistent and selective increase of 150%-200% in acetaldehyde, no increase or decrease in acrolein and depending on the flavor formulation, an increase, a decrease or no change in formaldehyde levels. Comparison of our results to the literature illustrates the need for development of a standardized ENDS testing protocol. Our results further support that device variability must be fully characterized and considered before assessing the impact of e-liquid formulations.
