A Model for the Determination of Diffusion Capacity Under Non-Standard Temperature and Pressure Conditions

The diffusion capacity of cigarette paper has been reported to be an important parameter in relation to the self-extinguishment of cigarettes and also in relation to carbon monoxide yields. Although the diffusion capacity is routinely measured and instruments for this measurement have been available for several years, differences between measured values obtained on the same paper sample but on different instruments or in different laboratories may be substantial and may make it difficult to use these values, for example, as a basis for paper specifications. Among several reasons, deviations of temperature and pressure from standard conditions, especially within the measurement chamber of the instrument, may contribute to the high variation in diffusion capacity data. Deviations of temperature and pressure will have an influence on the gas flow rates, the diffusion processes inside the measurement chamber and consequently the measured CO2 concentration. Generally, the diffusion capacity is determined from a mathematical model, which describes the diffusion processes inside the measurement chamber. Such models provide the CO2 concentration in the outflow gas for a given diffusion capacity. For practical applications the inverse model is needed, that is, the diffusion capacity shall be determined from a measured CO2 concentration. Often such an inverse model is approximated by a polynomial, which, however, is only valid for standard temperature and pressure. It is shown that relative approximation errors from such polynomials, even without temperature and pressure deviations, cannot always be neglected and it is proposed to eliminate such errors by direct inversion of the model with a comparably simple iterative method. A model which includes temperature and pressure effects is described and the effects of temperature and pressure deviations on the diffusion capacity are theoretically estimated by comparing the output of a model with and without inclusion of temperature and pressure effects. The results suggest that these effects may cause relative differences in the diffusion capacity up to 50% for pressure and temperature ranges of practical relevance