Tobacco Documents Online

SUPPLEMENTAL MEMORANDUM TO THE FEDERAL TRADE COMMISSION FROM PHILIP MORRIS INCORPORATED CONCERNING MEASUREMENT OF THE RELATIVE "TAR" DELIVERIES OF BARCLAY AND OTHER CIGARETTE BRANDS THROUGH ANALYSIS OF RETAINED NICOTINE IN CIGARETTE BUTTS ? D ) 3 9 Philip Morris Research Center Richmond, Virginia July 30, 1981

I V ' , , ) 31 0 3 ) 2 ~ 3 ~ SUPPLEMENTAL MEMORANDUM TO THE FEDERAL TRADE CONMISSION FROM PHILIP MORRIS INCORPORATED CONCERNING MEASUREMENT OF THE RELATIVE "TAR" DELIVERIES OF BARCLAY AND OTHER CIGARETTE BRANDS THROUGH ANALYSIS OF RETAINED NICOTINE IN CIGARETTE BUTTS introduction •The accepted procedure for measuring the "tar" delivery of a cigarette is to smoke it on a standard- ized smoking machine and calculate the dry particulate matter, less nicotine, delivered to a collecting pad. The utility of this.method in deriving results meaning- ful to smokers necessarily depends on the assumption that cigarettes with similar "tar" deliveries on the smoking machine will have similar "tar" deliveries to any given smoker, and specifically that cigarettes with comparatively low "tar" deliveries on the FTC smoking machine will have comparatively low "tar" deliveries when smoked by a person. As the Commission is aware, the validity of this assumption has now been seriously called into question by the introduction of a new, peculiarly constructed filter on the Barclay cigarette. The problem will be greatly exacerbated by the announced intention of the brand's manufacturer to incorporate the

) 2 same filter on additional brands, and the inevitable proliferation of the new filter throughout the industry, 1/ absent prompt action by the Comaaission. To compare the "tar" deliveries of Barclay with 3 1) ) 3 I i D the "tar" deliveries of other cigarettes when smoked by r human smokers, an analysis has been made of the butts of cigarettes smoked by a panel of smokers. This "butt study"confirms our previously reported findings that Barclay -- which is unique among currently marketed cigarettes in that its high machine-smoked dilution , z.::.. percentage drops radically when it is smoked in the lips -- delivers significantly more "tar" to actual smokers than a conventionally filtered cigarette which also measures 1 mg. "tar" on the smoking machine. As discussed more fully below, in order to assess the relative "tar" delivery of a cigarette by a nicotine retention study, it is necessary to measure: 1. The amount of nicotine retained in the filter of the cigarette following smoking; 2. The nicotine filtration efficiency of the cigarette filter; and ~ 1/ It has recently been reported that Brown & william- son will incorporate the Barclay filter on the Kool Ultra brand, in both 85 and 100 millimeter varieties.

3 3. The characteristic delivered "tar" to delivered nicotine ratio of the cigarette. ) ) ) ) 1 a a a ~ Once the retained nicotine and the filtration efficiency of the filter are known, it is possible to estimate the amount of nicotine that was not retained in the filter, and thus was delivered to the smoker. That delivered nicotine is then multiplied by the."tar"-to-nicotine ratio to estimate delivered "tar."- As set forth below, it is essential to realize that both filter efficiency and "tar"-to-nicotine ratio are variables, that will depend both on the brand of cigarette and the manner in-which the cigarette is smoked. In particular, filter efficiency and "tar"-to-nicotine ratio depend heavily on the rod flow rate, that is, the velocity at which smoke, as opposed to air, is drawn through the filter. As rod flow rate increases, filtra- tion efficiency decreases and "tar"-to-nicotine ratio 2/ increases. Rod flow rate, in turn, is directly depen- dent on the cigarette's dilutior.. As dilution decreases, more smoke is drawn through the cigarette, and rod flow rate goes up. 2/ As more fully described in footnote 5, infra, filter efficiencies for "tar" differ from those for nicotine,. depending upon the flow rate.

4 ) ! I 3 : R Some earlier nicotine retention studies -- per- formed on undiluted cigarettes -- had assumed that filter efficiency and "tar"-to-nicotine ratio were constant for any-cigarette brand. This assumption did not significantly affect the results of "butt studies" performed on undiluted cigarettes, but it is patently erroneous for diluted cigarettes. Especially if the dilution of a cigarette varies depending on whether the cigarette is smoked in the lips or on a smoking machine -- which has been shown to be the case with Barclay the assumption that filter efficiency and "tar"-to-nicotine ratio are constants is guaranteed to yield distorted results. Because the amount of "tar" delivered by ciga- rettes to human smokers will vary widely from smoker to smoker depending on the physiology and puffing characteristics of the smoker and other factors, it is not scientifically meaningful to attempt to mea- sure absolute cigarette "tar" delivery through an analysis of retained nicotine in butts. It is pos- sible, however, to compare the results for several

) 5 I I r b 3 brands and to draw conclusions about their relative "tar" deliveries. Indeed~, if cigarette brands dis- play similar relative "tar" deliveries for a number of different smokers, there is a high probability that those relative deliveries reflect the proper rank order of the products. In the study discussed below, Barclay KSSP, Cambirdge RSSP, and Merit KSSP cigarettes were smoked by a panel of regular smokers. Cambridge was selected because, with a measured FTC "tar" delivery of 1 mg. and unlit dilution of 78%, it is similar to Barclay as smoked on a smoking machine. Merit was selected because, with a measured FTC "tar" delivery of 7 mg. and unlit dilution of 28%, it is (although somewhat lower in "tar") closer to Barclay as smoked by the human smokers in earlier studies. The nicotine re- tained in the various cigarette butts was measured, and the nicotine filtration efficiency and "tar"-to- nicotine ratio were calculated for each smoker for each brand.

T ) D ) i ) I 6 The results set forth below indicate that, for every smoker on the panel, Barclay-delivered sexeral times more "tar" than a 1 mg. cigarette, and approxi- mately as much "tar" as a 7 to 8 mg. cigarette. This confirms earlier studies demonstrating that Barclay displays anomalous behavior in a smoking'machine, and is not comparable to 1 mg. cigarettes in "tar" delivery to human smokers. Relative "Tar" Estimation from Retained Nicotine a P Z While it is not possible rigorously to calcu- late the "tar" delivery of a cigarette from compounds retained in the filter after smoking, relative com- parisons can be drawn between different kinds of cigarettes smoked by the same smoker. The analysis depends upon three distinct variables: 1. The total quantity of nicotine retained in the filter after smoking. 3/ 3/ At first glance, it might appear that retained "tar" on the filter should be related to delivered "tar." However, "tar" is unstable because of its volatile [Footnote continued on next page.]

7 2. The nicotine removal efficiency of the filter at the smoke flow rate at which the cigarette was actually smoked. 3. The ratio of delivered "tar" to deliv- ered nicotine at the smoke flow rate at which the cigarette was actually smoked. ) 4 The first variable -- retained nicotine -- can be measured directly by chemically extracting the nico- tine from the filter and analyzing it in a gas chromato- graph. The second and third variables -- nicotine fil- tration efficiency and "tar"-to-nicotine ratio -- can be derived from standard curves once the individual 4/ smoker's flow rate for the cigarette type is measured. The following method has been developed to com- •pare the relative "tar" deliveries of cigarettes to a human subject based on retained nicotine on the filter: [Footnote continued from previous page.] components, and is not chemically definable; "tar" is simply the total weight of material deposited in the filter minus the weight of water and nicotine. There- fore, butt studies have consistently relied instead on measurement of the most stable and most readily ex- tracted component on the filter, nicotine. a 4/ These curves may be obtained from controlled machine smoking of cigarettes under varying conditions.

8 ) $ First, the subject is permitted to smoke ciga- rettes in an unrestricted fashion, the butts are col- lected, the material on the filter is then extracted, and the absolute amount of nicotine is measured by conventional techniques. Second, the subject then smokes similar ciga- ~ rettes, both lit and unlit, on an instrument which : I measures the duration of a puff and the volume drawn through the cigarette rod during a puff, from which average rod flow rate can be derived: Average Rod Flow Rate ='Rod Smoke Volume (1) Puff Duration This is done enough-tisaes to establish average para- meters for each smoker for each cigarette brand under study. A Third, using the smoker's average rod flow rate, the filter efficiency at that flow rate is obtained from a curve of nicotine removal efficiency N versus rod flow rate. Knowing the filter efficiency OA O N C11 at the smoker's flow rate and the amount of retained ~ ~ ~ rP Z.

a , 9 I ) 3 t 3 2 ~ nicotine allows the total nicotine that entered the filter from the tobacco rod to be calculated: Nicotine Total Nicotine Entering Filter = Retained on Filter (2) Filter Efficiency at Smoker's Average Flow Rate The nicotine delivered to the smoker is then given by: Smoker Nicotine = Total Nicotine - Nicotine Retained (3) Entering Filter on Filter Finally, the smoker's delivered "tar" is calcu- lated by a formula using the nicotine delivered to the smoker and the ratio of delivered "tar"-to-nicotine at the smoker's flow rate, obtained from a curve: Smoker Tar = Smoker Nicotine x'Tar"-to-Nicotine Ratio (4) at Smoker's Average Flow Rate This series of tests is repeated for a panel of smokers, and the relative "tar" deliveries to smokers of different cigarettes can be estimated.