Executive Summary

The objective of work done by the CORESTA Task Force Determination of Total Triacetin in Charcoal Filters was to develop a reference method for the determination of triacetin in an activated carbon on tow cigarette filter preferably by using solvent extraction followed by gas chromatographic analysis. The approach to develop a method for carbon filters centered on three main areas. These were finding an internal standard suitable for use with carbon and investigating extraction solvents and methods. The first collaborative experiment established that glycerol tripropionate was a suitable internal standard for use with carbon filters (at 3 g/L with carbon levels below 200 mg per 20 ml of extraction solvent) and that for rods with very low levels of carbon the triacetin could be measured by extracting 3 rods in 200 ml of ethanol with internal standard after shaking for three hours. But that irrespective of the solvent or extraction method
used highly loaded carbon filters gave lower than expected values for the triacetin content of the filter rods. Also with higher carbon levels the measured triacetin decreased over time.

Work by the Task Force found that this was principally due to hydrolysis of the triacetin in the carbon filter rods. High temperature thermal desorption followed by GC-MS analysis also suggested that very minor amounts of triacetin may have remained strongly bound to the carbon even after solvent extraction. Measurement of the hydrolysis products of triacetin (glycerol diacetate, glycerol monoacetate, glycerol and acetic acid) did allow calculation of the initial amounts of triacetin applied. However, commercial supplies of glycerol diacetate and monoacetate were not available to produce calibration standards. Since any method relying purely on the measurement of triacetin would produce a different value depending on the carbon loading and age of the rod and since measurement of all the hydrolysis products was not practical for a laboratory reference method it was concluded that a method could not be developed by the Task Force.