Tom 12 (1984): Zeszyt 4 (July 1984)

Mathematical models were derived that describe the variability of ventilation and pressure drop as functions of cigarette design specifications and the variabilities of cigarette components. These models and the graphical isovariation contour representation have several important potential applications for quality control and improvement. For example, the models can be used in cigarette performance studies to determine the relative contributions of each source of variability and to identify the largest source of variability. The models can be used in quality improvement to target programs for maximum effectiveness and to set realistic goals for quality improvement. The quantitative variability values that the models produce can also be used to establish acceptance limits for the variabilities of the cigarette components which control ventilation and pressure drop variability within specified limits. These applications and the insight into the causes of variability that these models can provide should make them a valuable tool for improving the uniformity of cigarette performance.

The amount of particulate matter captured by the upstream and downstream segments of ventilated filters was directly determined by a UV method. With 25 mm 3.3 dpf/35000 total denier filters the dry particulate matter efficiency of the upstream section approximately doubled and the nicotine efficiency increased by about a half in going from 0 to 70 % tip ventilation. The efficiency of the downstream segment showed only minor variations. This resulted in an increase of total filtration efficiency from 48 % to 63 % for dry particulate matter and from 40 % to 49 % for nicotine. The dry particulate matter/nicotine ratio decreased from about 15 for non-ventilated cigarettes to less than 10 at 70 % tip ventilation because the tobacco column produced smoke containing relatively more nicotine and the difference between dry particulate matter and nicotine filtration efficiencies became successively larger as ventilation increased.

Extensive details are presented for the laboratory preparation of specific glass capillary columns for the gas chromatographic analysis of various phenolic and acidic compounds of leaf and smoke. The chromatography of derivatized and underivatized mono- and dihydroxybenzenes, volatile acids, phenolic acids, and fatty acids is evaluated on both Pyrex glass and fused silica columns prepared by our methodology. The stability, efficiency, and low surface activity of columns prepared from different liquid stationary phases (SE-54, Superox-4, SP-1000, Silar 10C) are illustrated by numerous separations.

Yields in sidestream smoke (SS) and mainstream smoke (MS) and sidestream smoke to mainstream smoke distribution ratios (SS/MS) of 34 bases were examined from four types of cigarettes made from bright, Burley, Turkish and domestic (c.v. Matsukawa) tobacco. Irrespective of tobacco variety, major bases in sidestream smoke were ammonia, nicotine, 3-vinylpyridine, pyridine and 3-hydroxypyridine, and those in mainstream smoke were nicotine, 3-hydroxypyridine, ammonia, 3-picoline and pyridine. SS/ MS ratios of the bases were generally larger than 1.0 and the largest SS/MS ratio was that of ammonia, followed by those of 3-vinylpyridine, 3-cyanopyridine, nicotyrine and pyridine in that order. The preponderance of bases in sidestream smoke over mainstream smoke might be one of the major causes for the differences in aroma and irritation between sidestream smoke and mainstream smoke.

The HLC deproteinized product [depro] with 34 % of its original dry weight removed as heat precipitable protein fractions, shows a reduction in the levels of most of the cured product constituents and the major pyrolyzate constituents while the levels of polycyclic aromatic hydrocarbons and volatile nitrosamines in the pyrolyzates from the HLC deproteinized product were increased. This indicates that the precursors of the polycyclic aromatic hydrocarbons in the pyrolyzate from the HLC deproteinized product were not removed with the protein precipitates. Another possibility is that some constituents in the protein precipitates that were removed from the HLC deproteinized product may have had an inhibitory effect toward the formation of the polycyclic aromatic hydrocarbons in the pyrolyzates of the HLC control and flue-cured reference during their pyrolysis, since the protein precipitates were not removed from these products. Generally, the levels of polycyclic aromatic hydrocarbons from the pyrolyzates of the HLC products were higher than those from the pyrolyzates of the flue-cured reference. This could be a result of the tobacco maturity or curing methods. The level of the volatile nitrosamines in the pyrolyzates of the HLC control was lower than those levels in the pyrolyzates of the HLC deproteinized and flue-cured reference products. This could be due to pyrolytic interaction between some constituents in the sample and the protein precipitate which was not removed from the HLC control product. Additional work is needed to clarify these differences. A reduction in the level of solanesol was evident in the HLC deproteinized product probably due to the association of solanesol with the chloroplastic protein precipitate that was removed from that product. In addition, the level of solanesol was highest in the flue-cured reference, which is in agreement with previous reports that solanesol concentration increases with tobacco maturity. This report demonstrates that HLC can be used to manipulate the chemical composition of tobacco. The levels of some major constituents were decreased while the levels of polycyclic aromatic hydrocarbons were increased in the pyrolyzate from the same tobacco product.

Ammonium sulfate fra.ctionation, gel permeation and cation-exchange column chromatography were employed for panial purification of proteases from leaf laminae and callus tissues of Samsun NN tobacco (Nicotiana tabacum L). The predominant proteases in the leaf and callus are acidic sulfhydryl proteases which are activated by 2-mercaptoethanol and ethylenediamine tetraacetic acid, completely inhibited- by iodoacetic acid, and partially inhibited by phenylmethane sulfonyl fluoride and pepstatin A. With hemoglobin and tobaccl:l Fraction I protein as substrates, leaf and callus proteases showed a pH optimum of 5. However, specific activity was significandy higher in the callus than in the leaf. Tobacco proteases digested hemoglobin more effectively than Fraction I protein and showed the least activity with casein. Gel permeation resolved three -protease variants in leaf extracts but only two in callus samples. Rechromatography of the large molecular weight fraction in a cation-exchange column produced three and two variants for leaf and callus, respectively. The present results suggest that there are at least five variants of sulfhydryl protease in tobacco leaf and three in callus tissue and that tobacco Fraction I protein can be metabolized by both leaf and callus proteases.