Tobacco Documents Online

Z~l,c) m Sul)ject; To: WC ~1 Yr Prediction of Tar Yields From Measurements of Cigarette Pressure Drop and/or Filter Ventilation DISCUSSION: i ri rv~Er,-1i r'1~;~: :nu rn Date: June 26, 1980 w From: D. A. Calleson Cigarette tar yield basically is a function of the amount of tobacco burned during the cumulative series of puffs taken on the cigarette. With unfiltered, un-venti- lated cigarettes, tar yield for a given brand will correlate well with tobacco density in the tobacco rod. Addition of filters introduces the variable of smoke filtration, but the prediction of changes in smoke delivery is straight forward as there is a high correlation between „ filter pressure drop and filtration efficiency. ~ The further introduction of highly porous cigarette papers and burn-rate additives has made the prediction of smoke delivery less accurate. The introduction off ventilated filter systems, however, has created a significant variation in tar yield simply because the ventilation level is highly variable in itself. This is complicated by the fact that filter ventilation is not an independent variable -- it is in fact dependent upon the previous factors of tobacco - -- ~ density (draTt resistance)-,-filter pressure arop, an ---- cigarette paper porosity, As well as the actual ventilation path variables. Smoke delivery, therefore, is dependent upon variability of the smoke produced by the tobacco column and delivered to the filter system. The prediction can be no better than this. We must know the expected tar level. presented to the filter system for each brand, as well as the variability of this number. This can be obtained by routine smoking of each brand with filter s sy t.ems removed. This will be called tar producc:d-. - RJRI FORM 74?2-Rev. 7/70

2 Data must be obtained for tar prc>cjuced as a function of percent of puff volume as seen by t.he tobacco column. Each brand must be smoked without the filter and at. various puff volumes on either side of the nominAl ventilation level of that brand. For example, a cigarettc with a target ventilation level of 30% should be smoked without the filter at a puff volume of 70% of the standard 35 cc puff, since the other 30% is pulled in through the filter vent..ilation path. This measured tar value may correlate with the tar produced with a standard 35 cc puff, multiplied by a factor of (l-o ventilation), and if so, will confirm if tar delivery is proportional to puff volume. Tar delivered to the smoker will be the tar produced, less that portion filtered out in the filter. Since filtration efficiency is affected to a small degree by air flow rate through the filter, and this in turn is related to degrce of ventilation, some means of measuring the degree of ventilation of individual cigarettes must be obtained. The preferred method is to measure percent ventilation directly, as this approach requires a single measurement, and the measurementt accuracy is limited only by the precision of the instrument. It is possible to predict percent ventilation by measurement of draft resistances of various portions of the cigarette, such as the cigarette draft with the ventilation holes covered and uncovered, but the !' variability of this prediction should be significantly greater, than the direct measurement. Basically, the amount of tar delivered to the smoker is closely related to (1) tar produced, (2) percent ventilation, and (3) percent filtration in the filt:er. If tar produced is proportional to puff volume, then the following relation should exist: Tar delivered = tar produced (1 - % vent.) (l. - % F. E.) The percent air dilution can be measured. Percent filtration efficiency is both a function of air dilution and filter pressure drop, and can be predicted very closely for any given filter system. The value for tar produced, i.e. delivered to the filter, can be measured for each brand, and this value must then be adjusted for any known changes in the tobacco rod, such as tobacco density or cigarette paper porosity. Prediction of tar delivery using only cigarette draft measurements, without knowledge of tar produced, would be highly inaccurate and dangerous.

3 Use of Preysurc l)ro~p to Predict Percent Ventilation Percent ~zii• dilution is defined as the ratio of air volume pulled in throus;li the filter ventilation holes to the total puff volume (35 cc.) lt may also be defined as the air flow rate through the ve,:tilat.ion path divided by the standard flow rate of 17.5 cc/sec. J.t: is generally accepted that the relationships between flow r.:ztc and pressure drop of tobacco rods and filters are analogous to el ectri.c ci.rcui ts, and follow ohms law. Pressure drop i s aiia.logous to vol tage drop and flow rate to current. The rat~.o of pressure drop to flow rate (resistance) is a constant-. Therefore, if the pressure drop of each component is measured at the standard flow rate of 17.5 cc/sec., the measured pressure drops may be used as relative resistances of these componeiits. Knowing the relative resistances, ohm s law may be used. to caic.ul.ate the unknown values, and flow rates throu-Vii the ventilation path may be inferred. The accuracy of this calculati«n obviously is dependent upon the accuracy of the measurement of individual resistance values (such as filter draft, cioarette draft with holes open, with holes closed, tobacco column draft, etc.), and the completeness of the mathe- matical model. A s:implif.i_ed approach is made by measuring only two values of total cigarette draft, one with ventilation holes open and" one with venti lat ion holes closed. The difference in these two values would be a coarse estimate of percent filter ventilation. The degree of ventilation depends upon the relative values of three pressure drops (or resistances): tobacco column resistance, resistance of the filter upstream of the filter vents, and the resistance of the filter vent holes. The measurement of the resistance of these individual components requires destruction of the cigarette and is time consuming as well. A non-destructive test requiring a minimum of measurements is desired and by measuring c.igarett.e draft both with holes open and holes closed and taking the difference, an inferred value of the resistance of the ventt holes is obtained. It is necessary in doing this to assume a pressure drop of the portion of the filter tip upstream of the vents. 7'his may be obtained from previous data on filter tip drafts, but the accuracy will depend upon the variability of filter draft and upon the variation in position of the band of vent holes on the filter tip. The measurement of holes closed - holes open draft gives a value, M. Using ohms law analogy, the prediction equation for the resistance of the vent holes, Rv , reduces to: Rv = Ru ( P.u where 12u = pressure drop of the cigarette with holes closed -- pressure drop of the portion of the filter downstream (mouthend) of the vents

4 Since % ventilation = Qv qu +Qv where Qv and Qu represent flow rates through the vent and tobacco column respectively then, % ventilati.on may also be represented by: $ v a IZtI _ ltu + Rv or, using a measurement of holes closed - holes open draft: ~ v= M x 100 Ru where M is the difference in cigarette pressure drop measurements, holes closed - holes open and R is the pressure drop of the tobacco column and abuttinguportion of the filter tip. If the pressure drop of the filter portion could be assumed t%" be constant for a given brand, the holes-closed - holes open '~ draft, 1.1, would be proportional to percent.ventilation.. The calculation of percent ventilation in this manner provides an estimate only; however, because a difference in measured values is required, the accuracy of the result is highly dependent upon the precision of the pressure drop measurement. It is for this reason that a direct measurement of ventilation is preferred, if available. Predict ion of Smoke Del. ivery . The result desired, after all, is a prediction of smoke tar delivered to the smoker. There are many variables associated with this, but as mentioned earlier, the two factors most influential are puff volume through the tobacco column and filtration efficiency of the filter. Puff volume through the tobacco column is directly related to percent ventilation in the filter. But it must also be remembered that. porosity of the cigarette paper influences both the degree of.ventilation in the filter and the air flow through the tobacco column. It is possible, for example, to have a cigarette with both high tobacco density and high porosity paper. In this case, the pressure drop of the tobacco column may not change, and therefore, the percent ventilation will not change. The air flow through the burning cone will be lowered, however, and smoke delivery will al.s~ change significantly. Only a measure of encapsulated cigarette pressure drop will indicate this condition.

5 In the prediction of smoke delivery, the filtration efficiency of the £ilte.r must be considered. As percent ventilation is increased, the filter efficiency w:i.ll increase because of lower flow rates. An average filtration efficiency for each brand should be determined and the effect of changes in ventilation also determined. This may be used as a correction. Practically, for the purposes of quality assurance, the filter pressure drop must be known in order to assume that efficiency has not changed. In order to then be relatively assured that smoke delivery is as specified, the fol.lowing measurements are required: 1. percent ventilation, direct measurement 2. tobacco rod weight and density (size) 3. cigarette pressure drop, totally encapsulated 4. cigarette pressure drop, holes closed 5. filter tip pressure drop, encapsulated From these measurements, it should then be possible to rapidly indicate if any of these parameters are out of specification, ,., and to assess the relative effect on smo}:e delivery. Only the measurement of filter tip pressure drop requires destruction of the cigarette and this may not be required once it is d established whether or not the filters being used are within '~ specification. The measurements of cigarette pressure.drop ~ with holes closed and with total encapsulation will give a measure of tobacco column pressure drop and degree of cigarette paper porosity effect. ~ onal . ~`al- e son DAC:jcs