Tobacco Documents Online

)' i ) ) ) MEMORANDUM TO THE FEDERAL TRADE COMMISSION FROM PHILIP MORRIS INCORPORATED CONCERNING BA.RCLAY CIGARETTES AND A PROPOSED CHANGE IN THE APPARATUS USED IN THE COMMISSION'S LABORATORY FOR TESTING "TAR" DELIVERY Philip Morris Research Center Richmond, Virginia July 10, 1981

t MEMORANDUM TO THE FEDERAL TRADE COMMISSION FROM PHILIP MORRIS INCORPORATED CONCERNING BARCLAY CIGARETTES AND A PROPOSED CHANGE IN THE APPARATUS USED IN THE COMMISSION'S LABORATORY FOR TESTING "TAR" DELIVERY Introduction ) ) This Memorandum is submitted to the Commission by Philip Morris Incorporated in response to the letter from Matthew L. Myers, dated June 11, 1981, and the accompanying request by R. J. Reynolds Tobacco Company ("Reynolds") that the Commission's equipment for testing the "tar" delivery of cigarettes should be modified. In a submission made to the Commission on or about June 1, 1981, transmitted to Philip Morris by Mr. Myers, Reynolds maintained that a different cigarette holding device should be used by the Commission on its smoking machine in order to obtain reliable "tar" and nicotine delivery data for the Barclay cigarettes manufactured and sold by Brown & Williamson Tobacco Corporation. Reynolds stated to the Commission that it "strongly feels that the use of the Mk II 'Filtrona' holder will provide more accurate 'tar' and nicotine measurements for all cigarette brands and styles . " Philip Morris has conducted a series of scientific measurements and analyses of Barclay

2 1 ) -1 ) ) cigarettes and has carefully studied the performance of Barclay cigarettes under various smoking conditions. Most of the scientific tests were conducted on the king size soft pack variety of Barclay, but the same filter construction is employed on all Barclay varieties and the results for all varieties would be similar. On . the basis of these analyses, we have reached the following conclusions: First, the principal determinant of the "tar" level of a given "low tar" cigarette is dilution -- the greater the volume of dilution, the lower the "tar" yield. Of all cigarettes now on the market, only Barclay experiences a significant, indeed a phenomenal, difference between its dilution when smoked on the Comuaission' s smoking machine as contrasted with its actual dilution when smoked in human lips. The dilution percentage is very high (75% to 82%) in the machine, but low (averaging 32$) when smoked in human lips. In the case of all other cigarettes, the dilution on the smoking machine is substantially the same as the dilution experienced in a human smoker's lips. Second, this unique differential in dilution between smoking on the machine and smoking in the lips is the direct result of Barclay's abnormal filter N O N CA ~ ~ ~

1. t ) 3 0 ) 3 construction. In all other diluted cigarettes, the filter is designed so that air can mix with the smoke with the result that regardless of the manner in which the cigarette is held in the smoker's mouth, the percentage of dilution is substantially the same. Only Barclay is constructed so that the dilution air enters and exits exclusively through isolated peripheral channels that are impermeable to mainstream smoke. When those channels are open as they are in the current machine's holding device or when a Barclay cigarette is smoked by a person through a plastic mouthpiece (to eliminate direct contact with the lips) the fullest level of dilution (and, therefore, the lowest level of "tar") is experienced. However, when Barclay is smoked in the mouth, the inner surfaces of the lips grasp the perimeter of the filter and drape over some or all of the exits from the dilution channels. This conclusion by the lips results in materially less dilution and, therefore, in substantially higher actual "tar" delivery. Third, in sum, Barclay is designed differently from all other cigarettes so that, alone among all cigarettes, Barclay produces substantially greater dilution (and correspondingly lower "tar") when smoked

) S 4 1 2 s ) 0 on the Connnission's machine using the current holding device than when smoked in a person's lips. While a Barclay (RSSP) cigarette may deliver one milligram of "tar" when tested on the FTC smoking machine, it delivers many times more "tar" to actual smokers than any other brand that is ranked by the Commission's current testing procedure at one milligram. Indeed, Barclay delivers more "tar" to smokers than brands listed on the Commission's reports at seven or eight milligrams. Therefore, the present testing apparatus will not provide a valid, reliable ranking for Barclay. In short, we agree with the conclusion of R. J. Reynolds that the "tar" yield of Barclay is not accurately measured relative to other brands by existing apparatus, and we also agree that a different holding device should be prescribed by the Commission for its smoking machine. For the reasons developed below, Philip Morris does not recommend the substitution of a pressurized holding device as proposed by R. J. Reynolds, but proposes instead a simple modification of the holding. device. We believe it merits emphasis that the integrity of the FTC "tar" measurement program has been seriously compromised by the introduction of Barclay. The reported

0 5 ) ) ) FTC "tar" results of all cigarette brands, along with corresponding data for nicotine and carbon monoxide, will no longer provide a basis for relative comparison among all cigarette brands. While it is well known that the FTC smoking machine is not intended to simulate human smoking, it had always been the case that cigarette brands with similar FTC "tar" deliveries delivered similar levels of particulates to any given smoker, and in particular that brands with comparatively low FTC "tar" deliveries delivered comparatively low levels of particulates. Barclay's claim of one milligram "tar" by FTC method threatens to destroy those relationships. If the credibility of the FTC testing program is to be preserved, it is imperative that the holding device presently in use be modified forthwith. The Barclay Problem Barclay purports to be an "ultra low tar" cigarette. In truth and in substance, it is not. The filter on Barclay is constructed differently from the filter on any other cigarette, as Reynolds has correctly stated. Unlike any other cigarette, the Barclay filter functions one way when smoked on the smoking machine, and in a radically different way when smoked by a human smoker.

6 V ) V ) i ) If a Barclay cigarette is smoked by a human smoker in a normal fashion, it delivers a strong taste, and the filter is stained a deep yellow. However, when a Barclay cigarette is smoked by a human smoker through a plastic mouthpiece, the character of the product completely changes. It becomes mild in flavor, and there is much less staining of the filter. No other brand displays this kind of variation depending solely on whether a cigarette is smoked in the lips or through a plastic mouthpiece. To explain this phenomenon, it is necessary to examine the unique construction of the Barclay filter and to appreciate the relationship of dilution to "tar" yield. Barclay's Unique Dilution Channels , 14 In most cigarettes, there is some "dilution" or "ventilation" of the smoke with air taken in through the filter wrapping. The amount of dilution varies from brand to brand, and, as explained below, is a significant determinant -- indeed, it.is the most significant determinant -- of a "low tar" brand's measured "tar" delivery. As shown in Exhibit 1, a typical cigarette filter is surrounded by a porous plug

7 0 1 0 I , wrap, which in turn is surrounded by tipping paper with one or more rows of ventilating perforations. When the cigarette is puffed, air is drawn through the perforations in the tipping paper, through the porous plug wrap into the filter where it mixes with the mainstream smoke. It is this mixing of air and smoke within the filter that causes the smoke's dilution. As shown in Exhibit 2, the construction of a Barclay filter is fundamentally different. The Barclay filter is surrounded with a nonporous plug wrap that prevents mixing of air with mainstream smoke in the filter. Four peripheral dilution channels are impressed during manufacture into the nonporous plug wrapped filter. Those channels extend frqm the mouth end of the cigarette and terminate slightly beyond the perforations in the tipping paper, short of the tobacco rod. With Barclay, only the perforations over the channels contribute to dilution; all the other perforations are completely functionless. The Barclay filter is thus designed so that smoke will flow through the filter, a-iid air will flow through the peripheral channels, with no mixing possible in the filter. While a few other cigarettes on the market, such as Philip Morris' Multifilter brand, also

8 i ) ) ) ) 3 > have peripheral dilution channels, the channels in those cigarettes, unlike those of Barclay, are open to the mainstream smoke at the end of the filter furthest from the mouth end of the cigarette. The result, as shown in Exhibit 3, is that dilution air is free to mix with the smoke within the filter. Barclay is the only cigarette brand that completely isolates dilution air from mainstream smoke in a set of peripheral channels. The Significance of Cigarette Dilution The principal purpose and effect of diluting cigarette smoke is to reduce deliveries of "y•~r," nicotine, and gas phase components of cigarette smoke. If a cigarette is highly diluted, each puff will contain a high percentage of ambient air and a correspondingly lower percentage of particulate-containing smoke. There is a clear inverse correlation between the measured dilution of cigarette brands and their measured "tar" delivery. On Exhibit 4, percentage dilution is plotted against "tar" delivery for 138 brands of cigarettes on the market. All of the brands with very low "tar" delivery have very high percentages of dilution. For example, all of the brands with measured "yal•" delivery of six milligrams or less have measured

) 9 1 I It ) ) dilution of 50% and above, and all of the brands with measured "tar" delivery of two milligrams or less have measured dilution of 74% and above. As the exhibit shows, cigarettes with a low percentage of dilution are in the cluster of cigarettes with a higher "tar" yield. The inverse relationship between dilution and "tar" delivery can also be demonstrated for individual cigarette brands. The tipping paper of a cigarette filter can be modified to vary the degree of ventilation. Exhibits 5 and 6 show graphically the effect of such varying dilution on the "tar" delivery of a number of cigarette brands. As the percentage of dilution is decreased, "tar" delivery is increased; conversely, as dilution is increased, "tar" yield is reduced. Barclay is no exception to this general pattern. When measured on standard testing apparatus, Barclay r has a dilution of 75% to 82%, an extremely high level. It is this high level of ventilation that causes Barclay (KSSP) to achieve a "tar" delivery by current FTC method of only one to two milligrams. The importance of N dilution to Barclay's measured "tar" delivery is no N secret. The Brown & Williamson Barclay patent recognizes N that "[t]he dilution of the smoke stream reduces the ~ ~ ~ ~

) • - 10 - ) 3 1 ) ) 1 quantity of smoke particulates as well as gas phase components which are delivered to the mouth of the smoker," and expressly represents that the Barclay "invention further provides a cigarette filter for lowering tar predominantly by ventilation instead of filtration." U.S. Pat. No. 4,256,122 at 1 (March 17, 1981). When Barclay is tested on the FTC machine, its full dilution potential is realized since the current holding device leaves all four peripheral dilution channel exits fully open. Therefore, it tests with very high dilution and concomitantly low "tar." However, as described below in detail, when actually smoked in human lips, the channel exits are partially or.completely occluded. This results in less dilution (since the only avenues available for dilution air in Barclay are those dilution channels which are open) and therefore more "tar." The same result does not occur in any other cigarette since; in all other brands, dilution air is not physically segregated from the smoke and is drawn in through the filter rod regardless of the drape of the lips over the end of the filter.

) I I ) ) I Dilution Experienced by Barclay Smokers It is possible to measure the dilution of cigarettes smoked by human smokers. A schematic diagram of the equipment employed is shown in Exhibit 7. When a cigarette is puffed by a human smoker,-the measuring equipment calculates and compares the volume drawn through the cigarette rod and the volume of ventilating air drawn through the perforations around the filter. The ratio of the volume of ventilating air to the combined volume, expressed as a percentage, is defined as the percentage dilution. While measurements of cigarette dilution have traditionally been made on unlit cigarettes, the dilution measuring equipment can be used with either unlit or lit cigarettes. Using a group of 45 test subjects, Philip Morris has measured the dilution experienced by human smokers for a large number of cigarette brands, including Barclay. The test subjects were regular smokers, and included both men and women of varying ages. Measurements using the identical procedure were conducted on a number of different days, at various times, and under varying atmospheric conditions. In every case, care was taken that the test-subject not observe the results calculated by the dilution measuring equipment.

ti ) T ) 1 ) ) , The results of these dilution measurements are clear, dramatic and reproducible. The data are displayed in Exhibit 8. In the case of every cigarette brand exce Barclay, the measured dilution for human smokers (1) was very close to the measured dilution in a standard smoking machine utilizing the current holding device, and (2) remained essentially the same whether the cigarette was held directly in the smoker's lips or in a plastic mouthpiece. The results for Barclay, however, are strikingly different. When a Barclay cigarette was held in the lips as in normal smoking, no person experienced a dilution even close to the 75$ dilution encountered by the smoking machine using the current holding device. The variance from smoker to smoker was very large, ranging from a low of 12% dilution to a high of 55% dilution. The average, 32$ dilution, was less than half the dilution encountered by the smoking machine. When a Barclay cigarette was puffed by smokers through a plastic mouthpiece to prevent lip contact with the filter, the results were different. With a plastic mouthpiece, the measured dilution of Barclay was slightly lower than, but very close to, the dilution encountered by the FTC smoking machine with the current

) ) ) 3 holding device. The dilution measuring equipment thus confirms and quantifies the subjective observation that a Barclay becomes a very different cigarette when puffed through a plastic mouthpiece. These measurements establish beyond any doubt that the Barclay filter acts very differently in a smoker's mouth than in the current FTC holding device. We understand that Reynolds has hypothesized that this difference is caused by the collapse of the four peripheral channels under pressure exerted by a human smoker. Our observations show that the lips of a human smoker tend to occlude fully or partially the Barclay peripheral channels, blocking the dilution air. This effect has been confirmed by Philip Morris through observation of a Barclay filter from inside a smoker's mouth using a fiber optic instrument (a borescope), and has been recorded on a videotape which we are prepared to make available to the Commission. "Tar" Delivery Measurements at Smoker Dilution As noted above, a series of measurements made on dilution measuring equipsnent confirms that the dilution of cigarettes of any brand -- except Barclay -- is essentially the same whether the cigarette is smoked N O N r CA ~ ~ ~ ~ CA

) - 14 - i .! ) I by a human smoker or by an FTC smoking machine. Accordingly, for all cigarette brands other than Barclay, the FTC tests, as currently conducted, measure "tar" at approximately the same dilution experienced by smokers. The situation is very different with regard to Barclay, however. The average dilution experienced by test subjects holding a Barclay cigarette in their lips was less than half the dilution encountered by an FTC smoking machine utilizing the current holding device. As noted above, it is possible to measure Barclay "tar" delivery at different dilutions. When Barclay is measured at the 32% average dilution of the test subjects, the "tar" delivery is about nine milligrams. In other words, the level of particulate matter delivered by a Barclay cigarette to a human smoker is similar to that of a brand measuring about nine milligrams "tar" by FTC method. These data are set forth in Exhibits 9 and 10. In an attempt to corroborate these data, we hava also measured and analyzed the nicotine retained in Barclay filters after smoking. While it is sometimes possible to estimate "tar" delivery on the basis of retained nicotine, such analyses are not a rigorous

} t l 3 i method for measuring "tar," because of the large number of variables involved: puff volume, flow rate, nicotine to "tar" ratio at measured dilution and flow rate, and filter efficiency at measured dilution and flow rate. A retained nicotine analysis that fails to take account of these variables (particularly the impact of changing flow rate on filter efficiency) is of no scientific value. When the entire analyais is properly performed, nicotine retention studies confirm that the particulate delivery of a Barclay cigarette smoked in an FTC smoking machine with the current holding device is comparable to a cigarette measuring one to two milligrams of "tar," whereas the particulate delivery of a Barclay cigarette smoked in the mouth by a human smoker is comparable to a cigarette measuring about nine milligrams of "tar." A Proposed Modified Cigarette Holding Device As shown above, by measuring the "tar" delivery of a Barclay cigarette in a smoking machine at a dilution level comparable to that experienced by a human smoker, it is possible to obtain "tar" delivery data for Barclay that may fairly be compared with those for other cigarette brands. This comparability can be achieved by modifying the cigarette holding device used on the

) - 16 - I ? ). 1. FTC smoking machine to reflect the occlusion caused by the lips of a human smoker. Philip Morris has designed such a modified holding device, shown in cross-section in Exhibit 11. The interposition of an annular ring in the holding device so as to abut the mouth end of the cigarette affects the dilution achieved by the Barclay filter in a manner similar to that of a smoker's lips. Tests conducted on the modified holding device indicate that there is no substantial change in the dilution of any cigarette other than Barclay, and therefore no substantial change in the measured FTC "tar" delivery of any cigarette other than Barclay. The modified holding device does, however, reduce the dilution of a Barclay cigarette to approximate the level experienced by smokers. The modified holding device thus yields "tar" delivery data for Barclay that may fairly be compared with those for other brands. These data are set forth in Exhibit 12. Use of the modified holding device proposed by Philip Morris would require only a slight alteration of the Commission's testing equipment, and no change at all in test procedures. 'The modified holding device

1 ) ) ) ) is easily installed, may be readily obtained, and requires no auxiliary apparatus. Conclusion Philip Morris recoAanends to the Coaanission that the modified holding device described above be substituted forthwith for the holding device currently employed on the F'RfC's smoking machine. With the use of such a modified holding device, the dilution of cigarettes smoked on the machine will closely approximate the dilution experienced by human smokers. By testing cigarettes at that dilution level, it will be possible to preserve the validity of the FTC "tar" rankings as a basis for comparison among brands. ~

) I Index of Exhibits 1. Diagram of Dil ution Mechanism of Typical Diluted Cigarette ) 2. Diagram of Dilution Mechanism of Barclay Cigarette . Diagram of Dilution Mechanism of Multifilter Cigarette V 4. FTC "Tar" Versas Dilution for 138 Market Cigarettes 5. FTC "Tar" Versus Dilution for Barclay and Cambridge Cigarettes 6. FTC "Tar" Versus Dilution for Barclay and Carlton Cigarettes ) 7. Schematic Diagram of Instrument to Measure Dilution 8. Dilution Experienced by Smokers of Various Cigarette Brands ) 9. FTC "Tar" of Various Cigarette Brands at Average Dilution 10. FTC "Tar" Versus Dilution for Barclay, Showing "Tar" and Dilution Experienced by Smokers ) 11. Diagram of Proposed Modified Cigarette Holding Device 12. FTC "Tar" of Various Cigarette Brands at Average Dilution and When Measured with Modified Holding Device 0 I

) EXHIBIT 1 h l":1t.AM _OrYDI LUr tOti Mr!:tMA\ W._,) Tl'P I Cal. DI LUTF n C I G:11tI:TT1i i :,j 1% ~ I 1:Ut:T Ii J\ PF P ~n-„yT4~J? `~ ) 0 ! SI:C'I IllN 5i101ti IN(s. P1.Otd P,tTT'ERN n!tUTION AIR j i , DILUTION AIR

EXHIBIT 2 ...:1GILAM OF DILUTION MrC11AVt..A OF IiARCLAY CIG.1RfiTTF. I I ) 0 I ti t:: l\S'I' RUC'I' I ON t- ~ r- ~' F ~„ t- r. ~ AT c •-, ~,~ Ti!-'•'rl " ~C r il , r ~ F --Of IN 00 -7 .010 ~SIIOtti'I\G FLOh' PATTERF DILUTION AIR . .4 I DILUTION AIR L^ t`'/,'. 1 ~.:K E~ . ,. .

1 EXH!IBIT 3 1, GRAM 01: nILUTION Nn:CIIaNI;. t1r Atu!.T t 1:t L•ri:l: c t GARI:TTI: `STr.uC'r 10 N PA PF: i' I ,rPERFORATI ONS 0 Q ~ PGRGUS COMB!N!NC, '.','=' P ..•. .~. . • ... •;. ...,., ... ,,. ., . ,. ...,. PLUTEED .14ON-POROUS . P'L.'-.STIC SHEATH >l.t:T! tri SHOli INc FLOW PATTEIL'V D1'LUT!ON AIR D1LUT:ON AIR C;AR6ONICk !':'ti ER FILTER IM'ON- PCROU~S ~,'; R a~

,,. ' Jf' 11 I '1 "; ] ; ~ !"- +l'! ~ G q v I I ~. '~ 0!~7~: ~ F V v s~~t'J k+w 4Z1 v fi-i Jcif sh A ^ I.. . . . : ~~ v , ~p fFv `7 0o I~p r ri r:' -i 1, .•. ~., i / 202i5'T4694 ,

I : a.. . !. t ; , a, ,:!r : A V1J~ -4'd8 7••---47 l4C);!fl-1]al .~ . . , 2021574695 a . r a G

a. w. . .. . ~ . ~ a: ._ i 0 9 . _a-.._.._ 20215744696 2+17 PE 02 01 la _ 1 _..... .. ~_ .... ..... _ ._.l. -_....._.... l - : .._l = f-~ • ... y_` ' I -4 ~ _. .. ~ .. .J .~ A Ab ol

f 0 11 J r EXHIBIT 7 SCl!l:'•1:1T1C DIAGRAM 0(: INSTRUNt[i\T TO tiiPASURf: 1)I LUT l(1N ~- i ----- 1 1050 , Ftow aATS c.e-fN.1K 35 DENTAL DA4'S r 1 -P 8 vtt,=toK, .f. s i

) EXHIBIT 3 ) ) I ) ) I 4 DILUTION EXPERIENCED BY SMOKERS ON VARIOUS CIGARETTE BRANDS Human Smoking Dilution Under Machine Smoking; Cigt. Type* Conditions Barclay 75 Cambridge 78 Carlton 72 ' Marlboro Lights 20 Merit 28 B 6 H Multifilter 25 Vantage Ultra.Lights 34 Winston Lights 16 Y Cigarette Held In Plastic Cigarette He1d In Mouthpiece Smoker's Lips Average Average (Ran ge ) (Ran ge ) 69 32 (60 - 75) (12 - 55) 72 72 (68 - 76) (68 - 76) 68 68 ' (63 - 74) (63 - 74) 20 . 21 (15 - 23) (15 - 23) 30 28 (25 - 31) (25 - 31) 28 27 (22 - 32) (21 - 30) 33 34 (25 - 37) (25 - 37) 21 21 (16 - 24) (16 - 24) *A11 brands are KSSP.

EXHIBIT 9 0 FTC TAR OF VARIOUS CIGARETTE BRANDS AND AVERAGE DILUTIONS Ave. Tar At Ave. ) Average Brand* Dilution Tar By FTC Method** Smoker Dilution Smoker Dilution** Barclay 75 2.1' 32 9.3 ) Cambridge 78 1.3 72 1.8 Carlton 72 0.9 68 1.8 ) Marlboro Lights 20 . 12.2 21 12.2 Merit 28 7.1 28 7.1 0 B& H Multifilter Vantage Ultra 25 11.9 27 11.9 t Lights 34 6.6 34 6.3 Winston Lights 16 12.3 21 12.8 1 *A11 brands are KSSP. **Tar measurements typically have an experimental uncertainty of 10-20$ depending upon the delivery level.

w ;",~~ ; Ut=l_ [ t/EK'( VEh SUS ie U[ LIiT [ C]~1 , -., 1 J ~ 1 A--L'BARCLnY h.ssrr R0.waa GF GaJA•ot1f- t'i C'I LwU .:?f~~ 2- ~w A:. E'_ V. A!4r.- Ci I l.t lit s t: s n 31 24 8 7 N ~ :3 c .) U®f.AYFM'TZ0(i ~. 7 Fii._itt\ L~[L!TT [UNI

EXHIBIT 11 CUT-AtVAY DTAGR.:k,'-1 OF M0I1I FI I:D CIGARL-TTL• FIUI.h I VG DIiVI C1: 1 ) 3

/ I. EXHI'$IT 12 f FTC TAR OF VARTOUS CI'GARETTE BR4NDS AND AVERAGE DILUTIONS WITH MODIFIED HOLDING DEVICE ) Tar At Tar Measured Average Ave. Tar By Smoker Ave. Smoker By FTC Method With Modified Brand* Dilution FTC Method" Dilution Dilution** Holding Device** Barclay 75 2.1 32 9.3 9.4 Cambridge 78 1.3 72 1.8 1.4. ) Carlton • 72 0.9 68 1.8 1.0 Marlboro Lights 20 12.2 21 12.2 11.8 ) Merit 28 7.1 28 7.1 7.0 B & H Multifilter 25 11.9 27 11.9 11.5 Vantage Ultra Lights 34 6.6 34 6.3 6.4 Winston Lights 16 12.3 21 12.8 12.2 *A11 brands are KSSP. **Tar measurements typically have an experimental uncertainty of 10-20$ depending upon the de:ivery level.