Relationship Between Mainstream Cigarette Smoke “Tar” and Nicotine Yields

The purpose of this research was to examine the relationship between mainstream “tar” and nicotine yields and refine the commonly used linear model with a positive intercept to incorporate non-linearity and product-toproduct differences in filler nicotine content. “Tar” and nicotine yields are examined for a wide range of cigarettes (U.S. and international) using the Cambridge filter (CF), ISO, and the more intense MDPH, and HC smoking methods. Particularly at very low machine yields, a nonlinearity is observed that can be more accurately modeled by a power law relationship, and can be further improved by incorporating the concentration of nicotine in the cigarette filler into the equation. The resultant power law relationship is the better statistical fit to the available data, avoids the physical implausibility of positive nicotine yield at zero “tar” yield and lack of dependence on filler nicotine that are inherent in the simple linear model relating nicotine yield to “tar” yield alone, and explains the nonconstancy of the “tar”-to-nicotine ratio. The relationship between “tar” and nicotine can be affected by the use of very long or very short puff intervals, and, with the same tobacco blend and the same “tar” yield, longer cigarettes tend to have a slightly higher nicotine yield than shorter cigarettes.