Tar, nicotine and carbon monoxide (TNCO) cigarette yields determined under different smoking regimes form part of a continuous linear function, with filter ventilation both blocked and open, when linked to the reduction Δt in the smoking time due to puffing (TSRC 2013). This means that the use of a single smoking regime is sufficient to characterize emissions for a particular product if the smoulder rate is known. However, this single smoking regime has to carefully chosen because the application of intense conditions can produce high water yields leading to trapping issues. In addition, the smoulder rate determination involved in the calculation of Δt can be difficult for cigarettes with low ignition propensity design which present some tendency for self-extinguishment during measurement.

Experimental issues were overcome in a novel testing scheme involving the determination of smoking times under two smoking regimes and inputting this data into a burning model. Beyond inter-puff smoulder rate determination, the model was used to provide an extensive set of information such as the weight of tobacco burnt during puffs. Good correlations were observed between mass of tobacco burnt during puffs and TNCO or benzo[a]pyrene yields derived under ISO validated methods.

It was concluded the application of two smoking regimes is required to provide smoking times and Δt but the analysis of the yields is required from only one of them; taken together, this is an optimal scheme since it provides comprehensive characterization of products at reduced cost. An appropriate choice of validated smoking regime could then overcome the experimental limitations and still fit with the regulatory purposes of product monitoring and characterization.