During routing analytical smoking on a commercial smoking machine air flows through the Cambridge filter pad. It is suspected that this air flow is responsible for desorption of volatile components of TPM which can be a source of variability in measured yields. Furthermore it is known that the HCI (Health Canada Intense) method of smoking has a higher level of desorption / deposition of these volatile components.

Using a synthetic TPM mixture that is of known compensation and can be dosed accurately onto a substrate, an empirical model has been developed for the influence of puff volume and puff period on desorption. This has been further developed to compare how this might vary with semi-constant puffing as occurs in a rotary smoking machine as opposed to intermittent puffing in a linear smoking machine.

It is shown that this has particular relevance when comparing smoking machine types under HCI and ISO smoking and how clearing puffs can introduce unwanted variability if not consistently specified.

The applicability of this model is tested by comparing with experimental data gathered from smoking of cigars and cigarettes where high levels of TPM have been generated.