A reliable method for the routine analysis of trace carbonyl compounds, including formaldehyde, acetaldehyde, acetone, propionaldehyde, acrolein, methyl-ethyl ketone (MEK), butyraldehyde, and crotonaldehyde, in tobacco products is presented. One gram of tobacco sample was spiked with a mixture of three isotope-labelled carbonyls used as internal standards, and extracted with water. An aliquot of the aqueous extract was derivatized with o-(2,3,4,5,6-pentafluorobenzyl)-hydroxylamine hydrochloride (PFBHA). The PFBHA derivatives of the carbonyls were extracted with hexane and analysed by gas chromatography-mass spectrometry (GC-MS). The accuracy and precision of the method were evaluated using spiked matrix samples, including the Kentucky reference 3R4F cigarette filler and the CORESTA smokeless reference products CRP1, CRP2, CRP3, and CRP4. All investigated carbonyl compounds, with the exception of acrolein, demonstrated excellent recoveries (87-114%) and precision (0.2-7.6%) for the different spiked tobacco products. Acrolein, when spiked directly on the product, was found to be unstable. The linear range of the method was from 0.08 to 27 µg/g and detection limits ranged from 0.03 to 0.08 µg/g, dependent on the compound.

The method was applied to analysis of Kentucky reference 3R4F cigarette filler and the four CORESTA smokeless reference products. Formaldehyde (0.300-6.45 µg/g) and acetaldehyde (0.849-18.4 µg/g) were detected in all tested reference products. Acetone (0.197-3.31 µg/g) and propionaldehyde (0.156-1.07 µg/g) were found in the 3R4F, CRP1, CRP2, and CRP3. Levels of MEK, butyraldehyde, and crotonaldehyde were found to be below the method detection limit for all tested reference products. Using this method, storage conditions (storage time, container, and temperature) were found to impact the yields of carbonyls detected in reference tobacco product samples (3R4F, CRP2, CRP3) being both compound and sample matrix dependent; concentrations of formaldehyde in all tested products gradually increased as the storage time increased, while the concentrations of acetone in 3R4F samples dramatically decreased as the storage time increased.