Tobacco Documents Online

---

DEADLINE DATE : REPORT NO.: REPORT TITLE: PROJECT No.: f~ A 'HOR: GROUP LEADER: ~ ~T ISSUER: ~i ~ " DISTRIBUTI ON : OtS..r,' [" ~- FILE NO., I ~t ~B/) . Check List Bindin~ s : Fi Vorop.s Photographs No. : Reductions : \ Total Pa~es: SJG (q-) : IWH (1) : RAS (2) : RSW (3) : Mngr., R.&D.D. Aust. (2) : .s , (,) : Library (2) : File Tape Typing Checking ref. Date rec'd Da~e ret'd Author I Da~e Issuer Date I l /. <,,.72. god.. .,/., Draft for =ypin~ (I) (~) (3) i Prepare master: , Corrections: (i) ' (2) '~>repare l Bound Copy: ,, £ .:- i .], ;: I Issue: j S o'~'72. i" -;'/~.? 77 ~q r,,s~ t.q.li [ This pro forma is intended to facilitate and detail the progress of R. & D. Reports ~rom ~he typescript sta~e until the report is finally approved and issued. The pro forma ~hould be commenced at the time of the initial typin~ and accompany ~he typescript (and Imaster a~te~ preparation) at all subsequent stages. It must be sent to central filing with the master when the report is issued. Note: I. Wherever possible, drafts for typin~ should be written in ink. 2. After typing, the typescrip~ should normally be returned to the Author, who will subsequently obtain ~he Group Leader's and Issuer's approval. Circulation should be determined by the ~ssuer. A. The typescript mus~ not be presented for printing until the Issuer has initialled ~he pro forma agains~ "Prepare Mas~er". 5. After printing, one bound copy must be returned to the Issuer for final approval and the ~ssuer w-ill ensure that one cop)- is seen by the Manager. C~

mm.m. r~

CIGARETTES OF LOW BIOLOGICAL ACTIVITY REPORT NO. RD.934-R 13.9.1972. AUTHORS : R.E. Tho~ton R. Comber W.G.D. Lugton ISSUED BY: C.I. Ayres PROJECT: 5400 DISTRIBUTION: Dr. S.J. Green Dr. I.W. Hughes Dr. R.A. Sanford R.S. Wade, Esq. Manager, R. & D.D., Australia Herr H. So~torf Dr. D.G. Felton Library File Nos. 1 & 46D Copy No. I, 2, 3, 4 |l !! 5 " " 6, 7 " " 8, 9, I0 " " 11, 12 " " 13, 14 " " 15 " " 16, 17 " " IB, 19 COPY NO: CD fXO

RET/RC/W~ DL/CAL / I / 46D Group Research & Development Centre, Bricish-A~erican Tobacco Co. Ltd., SOUTHAMPTON. 13th September, 1972 CIGARETTES OF LOW BIOLOGICAL ACTIVITY (Report No. RD.934-R) SUMMARY AND RECOMMENDATIONS Specifications for a K.S.F.T. blended cigarette of low biological activity were agreed aC a meeting held in R. & D. in 1971. Features of this specification included a maximum TPM delivery of IO mg/cigarette and a reduction in biological activity per cigarette of 40% in all tests. This report describes cigarette designs which mee~ this specification, although for certain biological tests, e.g. skin painting, only predictions of results can be given. In detail, =he cigarettes have a 60 x 24.5 mm tobacco rod and a 25 m~ triple (carbon granule) filter with a filtration efficiency (for TPM) of about 60%. Cigarette paper of high porosity is used and the tobacco blend is of high nicotine content diluted with PCL and either NCF or Cytrel (1-324). The impact of these cigarettes is comparable to a typical U.S.K.S.F.T. blended cigarette but the flavour amplitude is low. Consequently it is recommended =hat development of cased and flavoured versions of the basic designs should be put in hand. ~0

U'I m,n.~

-2- INTRODUCTION The objective of this project was to design blended cigarettes of low biological activity (I). In detail, the agreed specifications were: (a) Length 84-85 mm. (b) Draw resistance not more than 14 cm water gauge (unlit, as measured under standard conditions). (c) TPM delivery not more than i0 mg per cigarette. (d) Nicotine delivery not greater than 1.2 mg per cigarette. (e) Impact comparable with a typical U.S. King-size filter tip cigarette. (f) A reduction in biological activity per cigarette of 40% in all biological tests, compared with a typical U.S.K.S.F.T. cigarette. (g) The delivery of vapour phase components, e.g. carbon monoxide, acrolein and hydrogen cyanide, =o be lower than for a typical U.S.K.S.F,T. cigarette. In addition, it was agreed chat an attempt should be made to produce satisfactory smoking characteristics without incurring unduly high blend costs. Subsequently it was also decided to make limited comparisons with the U.S.K.S.F.T. low delivery cigarette LUCKY TEN. DISCUSSION A. Design Considerations I. Dimensions Given an overall length of 84-85 ~, the only dimensions which can be varied are the circumference and the relative lengths of the

.m~w. C2~ t'~ CJ'7

-3- tobacco section and the filter. Over the circumference range 17-27 mm the TPM yield of a cigarette is related to the cross-sectional area of the cigarette. The requirement for a low TPM delivery therefore suggests a cigarette of small circumference, a view reinforced by the fact that cigarettes of small diameter have reduced biological activity as defined by a number of tests (2, 3). On the other hand the acceptability of cigarettes of small circumference is unknown in many markets and the delivery of nicotine is reduced. Balancing these factors led to the decision to make only a modest reduction, to 24.5 mm, in circumference. Two considerations apply to the relative lengths of tobacco and filter: (i) The longer the filter, the shorter the tobacco rod and hence the lower will be the delivery of TPM and nico¢ine. In practice, the longest filter yet used on a commercial U.$. K.S.F.T. cigarette is 25 nnn (4). (ii) At present it is easier =o design a high efficiency filter (say 60% for TPM) of acceptable pressure drop if the length is 25 mm rather than 20 mm, especially if 5 or 6 mm of the filter are to be used for a carbon section. On the basis of these factors a filter length of 25 .=n was chosen. 2. Filter Design It was thought that a TPM filtration efficiency of about 60% was probably the best that could be achieved at present, consistent with a draw resistance of not more than 7 cm water gauge. This would permit the draw resistance of the tobacco section to be up to 7 cm water gauge without the total cigarette draw resistance exceeding the specified 14 cm. CI>

C

-4- It was also decided that the filter should incorporate IOO mg of carbon which would be expected to reduce substantially (i) biological activity in those tests in which vapour phase constituents are known to be responsible for the major part of the activity and (ii) vapour phase components such as acrolein, formaldehyde, acetaldehyde and hydrogen cyanide. It was found impractical to make a "Dalmatian" type dual filter with a sufficiently high loading of carbon on the paper, so it was decided to employ a triple filter with the following specification: Houth Section I Carbon Section Tobacco Section I 7= I 6= 12~n i Cellulose Acetate ICarbon, i00 mg Jw Paper I Because equipment for manufacturing 25 ann triple filters was not available at the start of this work, quadruple filters (Filter No. l) were made as a substitute. These had a 7 mm cellulose acetate mouth section, a 5 n~ carbon bed and 6 * 7 mm paper sections. Due to the extra cutting involved the use of these substitute filters resulted in a small increase in draw resistance without a concomitant increase in filtration efficiency. Sutcliffe Speakman activated carbons (grades 203B and 205C) were considered because both appear to have a minimum adverse effect on taste (5). A Paramecium test gave a fractionally (but not significantly) better result for 205C and this carbon was used for the first filters. Because of difficulties in manufacture, possibly associated with the softness of this carbon, and because of a higher aldehyde efficiency, rxo rk9

I

C r'~ r',,j t.j-i

-5- 203B (20-50 mesh) was used in a second batch of filters. These were triple filters, conforming to the specification given on page 4, and are referred to in this report as Filter No. 2. Because the combined use of (i) a high efficiency filter and (ii) a tobacco diluent was expected to reduce i~act, it was decided to load the paper section of the Filter No. I with 10% sodium carbonate. This is known to increase impact (6). However, using these filters, it was found that impact was higher than from a typical U.S.K.S.F.T. cigarette, and sodium carbonate was omitted from Filter No. 2. The No. i filter was tested for TPM and nicotine filtration efficiencies when attached to 70 ~m cigarettes of blend 102. The results (Table I) showed that the filter had a satisfactory efficiency but a rather high pressure drop. Nevertheless, this quadruple filter was accepted as the best approximation to the target values available at the time and was used (with limited numbers of Filter No. 2) in the manufacture of the cigarettes described in this report. TABLE I PRESSURE DROP AND FILTRATION EFFICIENCIES OF FIRST BATCH OF FILTERS Property Pressure drop (cm W.G.) TPM Efficiency (%) Nicotine Efficiency (%) Target Actual 7 60 I0 62 5O 3. Cigarette Paper The delivery of cigarette smoke components may be reduced by increasing the quantity of secondary air drawn into the cigarette

-6- either in =he tobacco rod section or the filter, or both, by the use of highly porous or perforated cigarette paper or perforated plug wrap and =ipping paper. In the presen= work, it was decided to use a high porosity cigarette paper, at least for the first designs. Bollore 511C paper which contains 0.9% of sodium citrate was selected. Recent work has suggested that, for a given porosity, papers treated with citrate salts as burning accelerators produce a somewhat lower delivery of TPM than do either untreated papers or those containing other salts (7). The porosity of this paper (590 ml/min) is significantly higher than that (typically 260 ml/min) of papers used in commercial U.S.K.S.F.T. cigarettes, although citrate-treated papers are often used. 4. Tobacco Blend A tobacco blend was devised (8) bearing in mind =wo constraints: (i) It should be similar to that in a blended cigarette with respect to con=enC of flue-cured, Burley and Oriental tobaccos. (ii) It should have a nicotine content of 3%, because of anticipated dilution with non-tobacco materials. The blend used is given in Appendix I. A normal stem content equivalent to 20% of the tobacco blend was incorporated. Rather than import stem specially for the purpose, locally available stem was used. The expected TPM delivery from cigarettes made with 1OO% of this tobacco blend to the above specification was calculated using the Cigarette Design Handbook (9). These calculations (given in Appendix II) suggested that 100% tobacco cigarettes would have a TPM yield greater than lO mg, indicating the need to use tobacco diluents. CD r~D CO

C2~ (._r'1 r'~ r'~

-7- 5. Tobacco Diluents The first requirement of any tobacco diluent in this application was that it should result in a final TPM delivery of not more than I0 mg per cigarette. Other considerations were (i) the effects on specific biological activity, (ii) effects on carbon monoxide delivery, and ~iii) effects on acceptability. The tobacco diluents considered were: (i) Non-combustible filler (IO, ii). (ii) Cytrel 1-324 (12). (iii) PCL (Brown & ~illiamson, X-PCL-5). It was proposed to use these diluents as follows: Cigarette Tobacco A B C D 50 SO 50 50 PCL NCF 1-324 O 50 0 25 25 0 O O 50 25 O 25 The anticipated reductions in TPM delivery on using each of these three diluents are discussed later, in the section on smoke chemistry; the calculations are given in Appendix II. It should be noted that only the original high density NCF was available; lower density material could have been advantageous. The composition of the NCF used was 76% calcium carbonate, 7% SCMC, 15% E2E lamina extract and 2% glycerol.

,baw~ cj~ r',,j

-8- B. Cigarette Manufacture After conditioning, the selected blends were cut at 30 c.p.i. The high density of NCF has already been found to require careful setting of the making machine (II) to avoid rejection as winnowings. This problem may be greater with tobacco cut at 30 c.p.i, than with tobacco cut at 60 c.p.i., and in making the first experimental cigarette (RAg2) from a 50:50 mixture of tobacco and NCF, more than half of the latter was rejected. When a second portion of this mixture was made into cigare=~es (RAgl), the making machine was re-set and rejected very little NCF. Five different tobacco rods were manufactured; for technical reasons, 69 nrn rods were made and then cut to 60 mm. Some of each of these five had the first filter attached, and some of one design (RA85) also had the second filter attached. The six different cigarettes produced were then as listed in Table 2. TABLE 2 COb~OSITION OF SIX EXPERIMENTAL K.S.F.T. BLENDED CIGAP~TTES Cigarette Design A B C D A D Code t iRA82 RA83 RA84 RA85 RA91 RAg5 Filter i I I i 1 2 Tobacco Rod, Weight % Composition Tobacco 78 50 &8 50 51 50 PCL NCF O 25 0 25 I 0 25 22* 25 0 0 49* O 1-324 O O 52* 25 o 25 Filter Additive on Paper Section Sodium Carbonate, 10% t! tt I! I# None Carbon 205C IV 11 I! I! 203B *These figures were determined by blend separatzon; see Appendix III. CID

• 2.ob-~ ~C-a~of

-9- C. Examination of Experimental Cigarettes I. Physical Characteristics The weights and pressure drops of the cigarettes are given in Table 3. TABLE 3 WEIGHTS AND PRESSURE DROPS OF E.kTERI~NTAL CIGARETTES RA82 RA83 RA84 RAg5 RA91 RAg5 Filter i i i i I 2 Diluent (by weight) Weight of Tobacco Rods (g)* Draw Resistance (cm Water Gauge) u 22% NCF 25% PCL, 25% NCF 52% 1-324 ' 25% PCL, 25% 1-324 49% NCF 25% PCL, 25% 1-324 0.93 0.97 0.88 0.90 i. O5 0.90 ' Tobacco Rod* Whole Cigarette i I 4.5 14.4 3.3 13.1 4.3 j 14.1 i 3.~ 13.4 ! I 3.5 ' 14.7 3.S i 13.4 , [ *60 mm lengths. The pressure drop of the completed cigarettes is close to the specified upper limit of 14 cm water gauge, and in the case of RA82, RA84 and RA91 is slightly higher. Had triple filters instead of the substitute quadruple filters been used, however, the draw resistance of all cigarettes would have been within the specification. 2. Smoke Chemistry The TPM and nicotine (TNA - Total nicotine alkaloids) deliveries were determined for the five experimental designs by smoking 40 (unselec=ed) cigarettes under standard conditions. The values are given in Table 4 and may be compared with the deliveries recorded for a typical U.S. K.S.F.T. cigarette. Only cigarettes fitted with the first filter design were examined. I

0 rxo

-10- TABLE 4 TPH AND NICOTINE DELIVERIES J Cigarette J. I Filter I 1 1 1 I Predicted* TPH (mg/cig) m, RAg2 I RA83 i RAg4 RAg5 RA91 i U.S.K.S.F.T. lO. 7 9.8 6.4 8.3 9.2 IPM (mg/cig) Mean Puff Numbe~ 12.4 11.5 7.3 10.2 12.4 TNA (mg/cig) 1.12 0.83 O.55 0.72 1.12 ! TPM/TNA ii.i 13.9 13.3 14.2 ii.i TPM/puff (rag) 1.38 1.32 1.00 1.29 i 1.29 I C.A.** - 21.O 1.22 17.2 2.44 * Based on data obtained from The Cigarette Design Handbook (9) and calculations given in Appendix II. ** Cellulose acetate filter. From Table 4 it is seen that, while only the cigarettes RA84 and RA85 fall within the specification for TPM delivery, "the deliveries from all The cigarettes are close to, but consistently greater than, those predicted on the basis of the Cigarette Design Handbook (9); these calculations are given in Appendix If. The greatest discrepancy noted is for the cigarettes containing 49% NCF (RA91). The TPM figure (29.4 mg per cigarette) reported for another batch of K.S.F.T. cigarettes containing 50% NCF (ii) can also be used to predict the delivery for RAgl. Using Cigarette Design Handbook data (9), the reduction in circumference reduces the delivery to 26.0 mg, while the effect of the higher porosity paper further reduces the TPM to 19.O mg. Finally, allowing for a 60% efficient filter (in place of a filter of 30% efficiency) a predicted TPM of 10.8 mg is obtained which is again lower than that observed. 9.0 8.7 7.3 7.9 9.6 8.6 L..r~ r'-,~ P,o Cr-,

0 r~ 0~

-11- The nicotine delivery in all cases is below the upper limit specified, and although cigarettes ,RA84 and RAB5 in particular have rather low nicotine deliveries, this must be considered in relation to their impact rating, which is discussed later. The alkali-treated filters used reduce the nicotine filtration efficiency, and also increase the proportion of "extractable nicotine" present (6). In all cases, the TPM/nicotine ratio and the delivery of TPM per puff are significantly lower than those of a typical U.S.K.5.F.T. cigarette. The changes in cigarette puff number re=orded are likely to reflect differences in the burning characteristics of ~he tobacco diluents used but, with one exception, =he values are ~ithin one puff of that for a typical U.S.K.S.F.T. cigarette. Data from smoke analysis was also used to evaluate the efficiency of the filter with respect to the various rods used (Table 5). Although the filter efficiency had been estimated at about 60% for TPM, using flue-cured tobacco cigarettes (blend 102), separate determinations for each of the experimental cigarettes gave rather higher values with a mean of 68%. The mean filtration efficiency for nicotine was similarly determined at 56%. O ~m CY', CO

---

-12- TABLE 5 FILTRATION EFFICIENCIES Ill Cigarette P~82 RA83 RA84 IRA85 P~91 U.S.K.S.F.T. Filter i 1 TPM (tipped) (mg/ciB) 13 ii 7 9 ii TPM (plain) (mg/cig) i l iC.A.* 2O Fi1=er Efficiency (%) (TPM) 65 7O 73 70 61 68 56 37 36 26 30 28 Mean 45 TNA ! TNA (tipped) (plain) (mg/cig) (mglcig) • j, 1.O9 2.41 o. 75 1.95 0.52 i i. 35 0.67 t 1.56 1.c,3 ! 1.82 i 1 1.22 I 1.89 Me an Filter Efficiency (%) (TNA) 55 62 61 57 44 56 35 *Cellulose acetate filter The deliveries of vapour phase components, including carbon monoxide, are given in Table 6. The delivery of acrolein, referred to in the original specification, could not be determined owing to ~echnical difficulties involving the analytical method. O r-,o

-13- L%BLE 6 VAPGUR PIb~.SE C0,%?0"~Ef:T$ Total Vola:[|e Carbon ~onoxide Formaldehyde Aldehydes Hydrogen Cyanide Cigarette F£1cer i = Reduc:ion (Z v/v) (InG/cig)t (per c~) (Ug/ci~) Z Reduction 225 ~82 RA83 ~85 ~9L [ I. ! * , U.S.K.S.F.T.C.A.* I 3.5 z-".8 1 r 3.6 ~.~' " : i 3.5 . Io.~ 3.~ i ~:.51 z.7 10.5 i 3.9 1 13.5 .5.9 23.2 7.4 22.! (0) (ug/ctg) Z Reduction 17 81 23 75 14 85 22 76 24 74 92 (o) (mg/ciz) = Ecduccion 0.52 i 53 0.58 ~7 0.31 ?2 0.~6 ! 58 0.46 5B l. ZO (o) 47 5 ;5 95 293 , 57 84 85 7~ 68 (0) *celluLose acetate filter N ¢_r3 m

-13- TABLE 6 VAPOUR I'IIASE COIqPONENTS Cigarette ~82 ~83 ~84 ~85 RAgl U.S.K.S.F.T. Filter 1 1 1 1 1 C.A.* Carbon Honoxide (Z v/v) 3.5 3.6 3.5 3.9 2.7 3.9 Formaldehyde (mglcig) 12.8 12.7 % Reduction (per cig) 5.4 5.9 (vg/cig) 17 23 % Reduction 81 75 10.4 23.2 12.5 7.4 10.5 22.1 13.5 (O) 14 85 22 76 24 74 92 (0) Total Volatile Aldehydes i (mg/cig) ,i ii L 0.52 0.58 0.31 O.46 O.46 1.10 % Reduction 53 47 72 58 58 (0) llydrogen Cyanide j (pg/clg) % Reduction 125 47 45 75 95 293 57 84 85 74 68 (o) *cellulose acetate filter 7,/212b_ 01

um~b 0 c~r~ r~J

-14- Substantial reductions of 50% or better have been obtained with all the experimental cigarettes, for all the vapour phase components measured with the important exception of carbon monoxide. On a per cigarette basis, the carbon monoxide delivery (by weight) has been reduced by between 5 and 23%, depending on the tobacco section composition. The magnitude of the reductions obtained is disappointing since, although no actual target delivery had been set, it was nevertheless hoped that the combined effects of the high paper porosity (13) and the tobacco diluents would reduce the carbon monoxide delivery (by up to 30-40% for RAg4 and RAgl). In particular the use of 1-324 was expected =o produce a substantial reduction in delivery (14), and there was a possibility of NCF producing a reduction (!4), although other results had suggested this might not occur (Ii). 3. Biological Tests (i) Mouse-skin Tumorigenicity The experimental cigarettes have not been tested in the mouse-skin tumorigenicity test, but most of the features incorporated have been tested individually, so that predictions are possible. In addition, the weight of the nitromethane fraction in smoke condensate can be used as an indication of expected mouse-skin tumorigenicity (15, 16). To predict the activity on a per cigarette basis, the following information and assumptions are required: I. TPM per cigarette. 2. Probable contribution to total TPM from each component. C

C~ F~

-15- 3. Specific activity of TPM from each co=p. onent in terms of tumorigenic ratios. 4. No interactions occur between the various factors. 5. All dose response curves are parallel. Each of the four materials is considered in turn: (a) Tobacco~ includin~ stem The relative TPM delivery from tobacco is taken as 1.0, and the tumorigenic ratio resulting from tobacco smoke condensate treatment is also taken as I.O. This is taken as applying both to a blended cigarette and a flue-cured tobacco cigarette. (b) NCF By extrapolation of results previously reported (iO) it has been assumed that NCF produces 0.3 of the TPM produced by tobacco, on an equal volume basis (iO). The specific activity of the condensate derived from NCF is not known, but is taken as i.O (see discussion on mouse-skin thickening testj and nitromethane fraction results). (c) PCL The expected relative TPM delivery from PCL is 0.8 (9) and the tumorigenic ratio has been taken as 0.6 (17), although on the basis of the provisional B3:BO ratio, which is 0.45 (18), the prediction now being made is probably on the conservative side. (d) 1-324 As stated earlier, no figures are available for TP}! yields from cigarettes made from i00% 1-324. Consequently the TPM yields from cigarettes made from 100% 1-308 have been used. O

inlmb 0 r~ ~umb

-16- The yield of smoke condensate from cigarettes made from 1OO% 1-308 is approximately 3.8 mg/cig, compared with 30.5 mg/cig produced by cigarettes made from i00% tobacco (19), giving a relative TPM delivery of 0.12. The tumorigenic ratio of condensate from 1-324 is estimated at (approximately) 0.5. This is based on mouse-skin painting results ta 52 weeks (12) given in Figure I, in comparison with results from a similar T.R.C. experiment which showed a similar pattern at 52 weeks (also in Figure i) and which finished with a tumorigenic ratio of approximately 0.5 (17). The amount of TPM likely to be produced by each component of each experimental cigarette was derived using the figures given in Appendix II, and scaling these up to the total TPM found for each cigarette by experiment. The calculated amounts of TPM for each component multiplied by the tumorigenic ratios for these components are given in Table 7. The sum of these gives a predicted mouse-skin tumorigenicity for each cigarette. As already stated, this ignores the possibility of interactions. However, there is an indication of simple additivity of activity for 1-308 and tobacco smoke condensates in mouse- skin tumorigenicity (12, 20). The likely effect of paper porosity can be taken into account, approximately, by extrapolation (Figure 2) of known tumorigenic ratios (21). By this means a factor of i. I0 is obtained and this has been used to make the predictions in Table 7.

-J

-rTJ q_,- p,,j r~J

-17- The effects of the filter on tumorigenic activity are difficult tO assess. Several experiments involving filters have given results in which there is a tendency for the condensates to have increased activity compared to the control (plain) cigarette, but in no single experiment was it possible to demonstrate that the increase was statistically significant. However, Gallaher (22) have recently examined bonded carbon filters. They concluded tha~, at each dose level except the lowest, the condensate from the filter tipped cigarette was significantly more tumorigenic than that from the control. This is qualified by Gallaher who state =hat "on all statistical analyses attempted the response forms of the products were different and hence no valid quantitative relative tumorigenicity can be quoted" I= is concluded that the most likely effect of the filter chosen will be to cause a small increase in specific activity. However in view of the high efficiency of this filter the overall effect is likely to be a considerable reduction in activity (calculated on a per cigarette basis). The calculations (Table 7) suggest that the cigarettes RAg3, RA84 and RA85 are those most likely to give the specified 40% reduction in activity. Using the nitromethane fraction figures, the relative weights of nitromethane fraction per cigarette can be calculated (Table 8). These determinations were made using cigarettes without filters and allowance has been made for the effect of the filter. The results in Table 8 show reasonable agreement with those in Table 7. Again, RA84 and RA85 show the specified 40% reduction in activity, but RA83 just fails to reach this specification. ~rn OO

-18- ". AP-LE 7 PPrDT, CTED :~OUSE-SI:.IN TU:'.ORIGrNIC;.T¥, FER CIGAF, ETTE BASIS C~s&tette 2~82 RA83 RA84 ,'qA85 RAgl Lq,AB5 U.S. K.S.F.T. Fitter 1 1 1 1 1 2 C.A.tt jTP~! (AcCuat) L2.4 LI.5 7.3 LO.2 ;2.4 IG.2 21 .O TPH x Tumorigenicity Ratio from: Tobacco ll.gxt. O 7.9x1.O 6.5xi.O 7.Oxl. O IO.gxl. O 7.OxI.O 19.3xi.0 PCL 0 3.1x0.6 o 2.8x0.6 0 2.8xO.6 L.TxO.~ 1-324 O O O.8xO.S O.4xO.5 O o O NCF* O.Sxl.O O.Sxl.O O 0 l.Sxl.O O O iPredLcted Tumorigcnlcity per ~igacetr.c Relative to ~.S.K.S.T,T. I~1c ludEng Paper 1, Poro~tity Factor x [.10 0.67 0.56 0.37 0.48 0.67 0.48 (l. OO) *F£|lin~ power Ca~en as 2/5 o( thac of tobacco. *eCeL|u[ose acetate ~lter. TABLE 8 NITRO~:ETI~NE FFJ, CTIO::. PER C|CARETTE CigaxetCe N~.t romethane Fraction, 7. in N.V.k'.S.C.* 8.20 7.55 8.90 7.60 8.00 5.8S i TPH [ram N.V.~.S.C. Nitro~echane CLgare~tes with ' Filter No. t m~ per Fractlon [Y~ [ram Cigare[ce pet CigJrctte C~arccces Without (no FiLter) (no Filter) F£1ters (from TabLe 5) 31.6 30.8 19.8 -~5. S 23.1 2.59 2.33 I. 76 L96 1.85 O. 34 0.32 0.28 O. 34 O. 44 Calculated Z:icroc~thane Fraccicn, mg per cigarcccc with ~£tcer Relate.re ;~eights of Nirror:mrhane Fraction per Ct ~nrecte Compared WLch U,S. K.S.F.T. O.88 0.75 0.49 0.67 0.81 1.23"* 0.72 0.61 O. 40 0.54 0.66 (I.OO) *~oo-vol~cite whole smoke condensate, **B£cromethane fcactlon d¢ce~c~i.ned on cigarettes with ~i~ters attached. _...a. w

Cigarette FiLter TABLE l PREDICTED }tOUSE-St~It~ 'fUt~)HI(;,I"-RIC|TY, PI'R C$1;AILE'ITE ~ASIS TPH (~glcig) (Actual) RAg2 | 12.4 11.9 x 1.O RAm3 I 11.5 7.9 x i.O RAg4 I 7.3 6.5 x 1.0 RAS5 1 IO.2 I.(} x 1.O RA91 1 12.4 10.9 x 1.0 RA8$ 2 IO.2 7.0 x !.0 U.S.K.S.F.T. C.A.** 21.0 19.3 x 1.0 Tobacco PCI. 0 3. I x 0.6 O 2.8 x 0.0 O 2.8 x 0.6 1,7 x 0.6 oba¢'co. TI'H x ~umorige, icity Ratio [rum: 1-32/m O [I O.8 x [).5 0.4 x O.S 0 O O NCF* (1.5 x I.O 0.5 x 1,O [) O 1.5 x I.O 0 o *Filling power taken as 2/5 of that of **Cellulose acetate filter. TABLE 8 N[TROtlETI~NE FRACrKONp PER CIGARETTE Predicted Tumorigenicity par Cigarette Relative co U.S. K,S.F.T. Including Paper Porosity Factor x 1.10 0.67 O.56 0.37 o.4B 0.67 0.48 (I.OO) Cigarette RAm2 RA83 R~84 RA85 RA9 i tJ.S. K.S.F.T. H 1 c rome thane Fraction, 1 in N.V.W.S.C,* 8.20 31.6 7.55 30.8 8.90 19.8 7.60 25.8 8.00 23. ! 5,85 N.V.W.S.C. mg per Cigarette (no Filter) Nit roost )lane Fraction per Cigarette (no Filter~ *Non-vo|atJle whole smoke condensate. 2.59 2.33 1.76 1.96 1.85 11'H f rum Cigal eLIas viLli Filter 1;o. I TPH from Cigarettes NiLhout FiLters (from Table 5) O. )4 o.32 0.28 O. 34 O.,'J4 Calcalated ~itroa~rhaae Fractiol| mg per cigarette with Filtc~ O.88 O. 75 0.49 0.67 ().HI 1.23"s Relative b/eights of Ni t ruth') hane Fraction per Ci l,,aret te ConJp.lred bli tim O.S. K.S.F.T. 0.72 O. 61 O.40 0.54 0.66 ( 1. oo) **Nitromethanc fractiou determined on cigarettes will, filtels attached. ( m I i. I

c~ c~

-19- (ii) Mouse-skin Thickenin~ Test The short-term mouse skin thickening (hyperplasia) test (23) provides an indication of the expected long-term test result (24) as well as being a measure of irritation in its own right. The experimental cigarettes have not been tested in this assay, but most of the individual characteristics have been tested, so a result can be predicted. The effects of the various characteristics are considered separately: (a) The Filter Neither the carbon, paper nor cellulose acetate sections of the filter should have any effect on the specific activity of the condensate (25). The addition of sodium carbonate to a filter, although tested only in the presence of water, should also have no effect (26). (b) NCF This material has no effect on specific activity (27). (c) 1-324 This material has not been tested, but cigarettes made with 100% 1-308 gave an area activity rating of 108 (28). Because of the lower tumorigenicity of condensate from 1-324 (12) a lower mouse skin thickening result is likely. However, for the purpose of predicting the result in the mouse skin thickening test the activity has been assumed to be the same as that of tobacco smoke condensate. (d) PCL R-PCL-5 smoke condensate gave an area activity rating of 78 (28) C~ P~O C~

..t t', r~ Q::: 0",,

-20- compared with 94 for a typical U.S.K.S.F.T. cigaret:e (29). Using the known relationship between the dose and response in this test (30), and assuming a similarly-shaped dose- response for the two condensates, the ratio of doses to give a similar response can be derived (Figure 3). This ratio is O.8. (e) Paper porosity Higher paper porosity is expected to lead to a higher response (31). Assuming a parallel increase to that obtained for T48 and T49 (Figure 4) and ignoring the (anomalous) result for T44, an area activity rating of 109 is arrived at in comparison with 94 for a U.S.K.S.F.T. cigarette (29). This corresponds to a dose ratio of 35:28 (Figure 3), giving a factor of 1.25 for the higher porosity paper. The predicted short-term mouse skin painting results are given in Table 9. The results suggest that cigarettes RA84 and RA85 will meet the specification. OlD CO -~j

-21- PP~nlC~'ED SIIOR'r-TEF31 ??O'JSL SKIN PAXNTIT;~J ACTtVIT'f PER CTGAI~TTE C~8arette I~82 1~83 P,,A84 R,~,85 P~9 l 1C~84 I C.S. )~.S.F.T. t Tilter 1 l l 2 ITPH (aczual) L2.4 ll..5 7.3 10.2 12.4 Calculated Contribution to TPH x Activity Itacio for : Tobacco xl.O 12.0xl.O 7.9x1,0 6.§x1.0 7.0xl,O ll.lxl.O PCL x 0.8 0 3,1xO.8 0 2.8xO.8 0 21. 19.3 x 1.0 1.7 x 0.8 *FilLing value token as 2/5 that Of tobacco. 1-324 KCF x I.O x 1.0 0.4xl.O* O.5xL.O 0 0 1.3x1.0 0 O O :O.Sxl.O iO.4xl.O O FreGicced Activity per C/garotte Rel.~tive re U.S. );.S.F.T. 1.cludEn& Paper Foroslty Factor x 1.2S 0.74 O.65 O.43 O.57 O.7~ (1.00) **Cellulose acetate filter. r~ c~ co

-21- TABLE 9 PREDICTED SIIORT-TERH HOUSE SKIN PAINTING ACTIVITY PER CIGARETTE Cigarette RA82 RA83 1~84 Filter TPM (mglcig) (Actual) 12.4 II.5 7.3 ~85 ~91 ~84 U.S.K.S.F.T. I 10.2 I 12.4 2 C.A.** 21. Calculated Contribution to TPM x Activity Ratio for : Tobacco x 1.O 12.0 x 1.0 7.9 x 1.0 6.5 x 1.0 7.0 x 1.0 11.1 x 1.0 19.3 x 1.O PCI, x O. 8 0 3.1 x 0.8 O 2.8 x 0.8 0 1.7 x 0.8 m I- 324 NCF x l.O O O 0.8xl.O O.4xl.O 0 x 1.0 0.4xl.O* 0.Sxl.O 0 0 1.3xl.O 0 0 Predicted Activity per Cigarette Relative to U.S.K.S.F.T. Including Paper Porosity Factor x 1.25 O. 74 0.65 0.43 0.57 0.74 (1.00) *Filling value taken as 2/5 that of tobacco. **Cellulose acetate filter. 68 1?, ca,

C Cj r~ r~

-22- (iii) Paramecium Hanging-Drop Test The results obtained in the hanging-drop Paramecium test (32) are given in Table I0. TABLE iO HANGING-DROP PARAMECIUM TEST RESULTS Cigarette Code Filter RA82 i RA83 I RA84 1 RA85 I RA91 i RA85 2 U.5. K.S.F.T. I C.A.** (Control) ; Mean Puffs to Kill 27.2 23.3 35.0 27.5 26.7 19.0 12.0 Activity Per Puff, Relative to Control O.44 0.52 0.34 0.44 0.45 0.63 (1.oo) Puffs per Cigarette* 9.0 8.7 7.3 7.9 9.6 7.9 B.6 Activity per Cigarette Relative to Control 0.46 O.53 0.29 0.40 0.50 0.58 (l.OO) *From Table 4. **Cellulose acetate filter. The results in Table 10 show that all the cigarettes meet the specification. (iv) Solution Tetrahymena Toxicit~ Test The results obtained in the solution Tetrah~mena toxicity test (33) are given in Table II. ~n :xo

r,~ "q"q"~b r'~ ",.o r,~

-23- TABLE 11 SOLUTION TETKAHYMENA TEST RESULTS • i Cigarette Code RA82 RA83 RA84 RA85 RA91 RA85 U.S.K.S.F.T. 165A** 165A** 165A** i Filter 1 1 1 1 i 2 C.A.* Test Number Mean LDIo0 (ml) Activity per Cigarette Relative to U.S.K.S.F.T. Cigarette L, 187 3.5 187 3.8 187 4.6 188 3.0 188 2.6 2O0 I. 5 188 1 .O 187 0.68 188 0.55 200 O.&l O. 36 0.33 0.28 0.34 0.40 O.5O (l.OO) *Cellulose acetate filter. **Standard plain blend 102 cigarette used as continuity control, in these experiments smoked to 8 mm butt length. The results in Table II show that all the experimental cigarettes are well within the specification. (v) Particulate Phase Tetrah~mena Toxicit~ Test The results obtained in this test (34) are given in Table 12. rk~ rk~ k~D

-24- TAEI.F 12 r0~RT|CUIATE PIS',S~ TET~.AII'CII.!IA TP, XICIT¥ TY.ST RESULTS C~£arette Filter Tes¢ Code Humber I ~Bz i i L P.t.,8 3 I R~86 ! P~es '~ Z i F.AgL I L rw85 2 U.S. ~:,S.F.T. C.A.t Control 165A** L65A 165A flean Specific Toxicity, LD]00 Relative to St3t~dard Ci~areLte 76 ~2~ 76 505 76 568 77 73R 77 ?r',O 83 3~7 77 775 76 495 77 66B 83 32~ *Conven¢~ona[ cellulose acetate f~lcer. t .L7 0.98 O.B7 0.9L 0.95 0.93 0.86 (i .oo) (l .oo) (1 .co) Specific Toxicity, Relative to Co.¢rol Cigarette 1.36 1.16 !.01 I.o5 l.L! l.n8 (i.oo) ~e i!r ht nf .¢':~uk e Col Leered (m~/cig) 0.79 O.61 o. 31 U. r, 5 O. ~'J t, 0.52 (I .oo) 9.1 E.6 6.7 6.7 9.0 7.5 15.6 l'er Ci~.Irette loxicicy Rul;,tive co C..trol P.;a 5 cd on: ~Je i I,,h r of $~oke TPIi Collected 0.80 0.62 O. 35 O.5L 0.6C, q. 52 (1.00) e*Con~inuLcy control standard blend L02 ¢izatecte, [n this test smoked to 8 ~:n but: length. c~ r~ r~

-24- TABLE 12 PARTICULATE PHASE TETRAIIYHENA TOXICITY TEST RESULTS Cigarette Code RA82 RA83 RA84 RA85 RA91 RA85 U.S.K.S.F.T. ControI 165A~i 165& 165A Filter I I l I I 2 C.A.* -----------..-.t Test Number 76 76 76 77 77 83 77 76 77 83 Mean LDI00 424 505 568 738 700 347 775 695 668 324 *Conventional cellulose acetate filter. Specific Toxicity, Relative to Standard Cigarette 1.17 0.98 0.87 0.91 0.95 0.93 0.86 (I. oo) (l .oo) (I .oo) Specific Toxicity, Relative to Control Cigarette I. 36 1.14 I .O1 I. 05 I.II I. 08 (I.O0) Neight of Smoke Collected (mglcig) |, J 9.1 8.4 4.7 6.7 9.0 7.5 15.6 Per Cigarette Toxicity Relative to Control Based on: Neight of Smoke Collected 0.79 O.61 O.31 0.45 0.64 0.52 (I.OO) TPH 0.80 0.62 0.35 O.51 0.66 0.52 (l.oo) **Contlnuity control standard blend 102 cigarette, in this test smoked to 8 mm butt length.

r~ ',,o o,,,

-25- The results in Table 12 show that cigarettes RA84 and RA85 have met the specification with respect to this test, and RA83 has almost done so. For this test, smoke condensate was collected in a small impaction trap. This trap collected on average 29% less smoke than was collected in the standard TPM determination (Table 4). Calculations of toxicity per cigarette can be based either on the weights of smoke collected or on the standard TPM fi&ures; both sets of figures are given. (vi) Chorioallantoic Membrane Test The results obtained in this test (35) are summarised in Table 13. TABLE 13 CHORIOALLANTOIC ~MBRANE TEST RESULTS Cigarette Code RAg2 RA83 RA84 RA85 RA91 RA85 U.S.K.S.F.T. Filter 1 1 1 1 1 2 C.A.** I 0.90 0.85 0.87 0.75 0.87 0.87 0.89 Mean Response, Relative to a Standard Cigarette* *Blend 102 continuity control cigarette. **Cellulose acetate filter. Mean Response, Relative to U.S.K.S.F.T. Cigarette 1.01 0.95 0.98 0.84 0.98 0.98 (I.OO) CD rx.) rx.) ~O

r,,o r'~ ',,,D cc~

L -26- The results in Table 13 show only insignificant differences between the specific activities of the various smoke condensates, except for RA85, which gave a surprisingly low result. Thus, the TPM delivery can be taken as a sufficient guide to the expected activity reduction on a cigarette basis. In this case the results given earlier in Table 4 show a reduction of more than 40% for all the experimental cigarettes, as compared with a typical U.S. K.S.F.T. cigarette. 4. Objectiy9 Sensory Panel Tests The first request to the smoking panel was to compare the impacts of the test cigarettes with the first filter with that of a typical U.S.K.S.F.T. cigarette. They were also asked for general comments on the cigarettes and for comparison with a known low-delivery cigarette, LUCKY TEN. No attempt was made to replicate tests or ensure that equal numbers of smokers took part in all tests, since indicative rather than absolute information was wanted. The following is extracted from the panel's report (36): Test i - Comparison of Impact with that of a Typical U.S.K.S.F.T. Cisarette An identified U.S.K.S.F.T. cigarette and a colour-coded test cigarette were smoked in a paired test and the smoker asked to compare the impacts of the two cigarettes, all the Impact standards being available. The results are given in Table 14. CD rk~ ~-O

c..m t,~ m.q.,b c~

-27- TABLE 14 COMPARISON OF IMPACT WITH A TYPICAL U.S. K.S.F.T. CIGARETTE Cigarette Code HA91 RA83 RA84 RA85 Filter Comparison with Impact of a Typical U.S.K.S.F.T. Cigarette Much Lower O 0 I O Lower 3 2 2 5 Equal 2 2 1 o Much Higher Higher 3 2 6 O 5 1 i Conclusion Probably a little higher than the control Higher than the control Higher than the control Probably the same as the control In general smokers found these cigarettes to vary fairly considerably as they were smoked, suggesting a variable composition within the cigaret=es. Many smokers commented on the very clear-cut impact for the cigarettes - probably a plus point. Test 2 - Comparison of Impact with Imp Standards Colour-coded cigarettes were presented singly for comparison with the Impact standards; only l~pact standards 5, 6 and 7 were available. The results are given in Table 15. TABLE 15 COMPARISON WITH IMPACT STANDARDS i ,1 Ci gate tte Code U.S.K.S.F.T. HA91 HA83 RAB4 HA85 Filter Average Panel Impact Score 5.8 6.4 6.1 6.8 6.8 Number of Smokers 19 9 9 9 9 4 C)

C U C

-28- These results indicate that the test cigarettes are all slightly higher in Impact than the U.S.K.S.F.T. Test 3 - General Impression of Acceptability of Ci6arettes First it should be noted that the following remarks are made within the constraint that the members of the Panel are not regular smokers of U.S. blended cigarettes. Each cigarette was smoked against LUCKY TEN (IO mg tar, O.8 mg nicotine) and smokers asked to comment on the relative amplitude of the cigarettes and the general impression of the coded cigarette. RAgl, Amplitude lower than LUCKY TEN - averaged about 2~. Filter No. I Generally considered to have a low flue-cured character but a burnt character comparable with SENIOR SERVICE, i.e. high - a rather unpleasant, unacceptable smoke. RA83, A~plitude lower than LUCKY TEN - averaged about 2.2. Filter No. i Very unbalanced flavour-profile - high burnt character - all other scores very low, generally ~-I. Some indications of acceptability but other smokers found it to be unpleasant.

~aw.L r,~ c..,,j c~

-29- RA84, Amplitude lower than LUCKY TEN - averaged about 21. Filter No. 1 Low flavour profile scores, except burnt character. Some comments on "non-tobacco taste". Generally not acceptable. RA85, Amplitude lower than LUCKY TEN - averaged about 2.3. Filter No. I Low flavour profile scores, probably the least burnt character of all the cigarettes smoked. Some indications of acceptance. Smokers found all these cigarettes difficult to classify because of the very low levels of the flavour characteristics. However, fairly high levels of the burnt character were associated with all the samples. LUCKY TEN cigarettes always had a higher amplitude but this was probably associated with the added casings and flavours. From these tests it appears that the cigarettes containing PCL in the mixture may have the best chance of producing some measure of acceptance but need some added flavour to give some distinct feature to the flavour profile. No tests were carried out to measure irritation levels and few comments were made about this. Two smokers did, however, note that RA91 had a high nose irritation. The impact levels of the cigarettes are probably too high to gain much acceptance with the Group R. & D. Centre panel as the smokers seem to prefer an impact in the ran&e 4-6. In view of the slightly greater impact than for a typical U.S. K.S.F.T. cigarette, a second filter was designed and sodium carbonate was omitted from the paper section. The smoking panel was asked to compare the impact from RAg5 (second filter) with that from a typical U.S.K.S.F.T. rk> C:> L~

t~ o,,

-30- cigarette. The panel was also told chat any other comments on the cigarettes would be welcomed. The following is from the panel's report (37): Test A The cigarette was smoked against an identified typical U.S.K.S.F.T. cigarette in duplicated paired-comparison tests and the smokers were asked to compare the Impact of the test cigarette with that of the U.S.K.S.F.T. cigarette. Impact of RA85/second filter - Much lower O Lower 8 Equal 5 Higher 3 Much higher O It is concluded from this test that the Impact of the test cigarette is probably slightly lower than the typical U.S.K.S.F.T. cigarette. Test B The test cigarette was smoked against the Impact standards in duplicated paired-comparison tests. With two exceptions all assessments were at the level of Impact 5 or 6. The panel average score was 5.2. This test confirms the result of Test A. Comments on the Test Cigarette As attention had been focussed on Impact most of the comments were about this. Ocher comments made (in comparison with a typical U.S.K.S.F.T. cigarette) were: t~

(,.r t,j,j c:~

-31- (a) Less amplitude. (b) Less blended, more air-cured. (c) More burnt, or acrid. (d) Slight carbon off-taste. (e) Slight paper taste. CONCLUSIONS The work described in this report has shown that it is possible to design blended cigarettes to meet the specific design objectives listed on page 3 of this report, principally a TPM delivery of not more than i0 mg, and a reduction in biological activity per cigarette of 40%. It has been established that the impact of these cigarettes is comparable to a typical King Size U.S. blended cigarette but in the important area of "acceptability" the limitations of using a panel at R. & D. to assess blended cigarettes must be considered. Nevertheless it appears that some of the designs have a degree of acceptance although the flavour amplitude is low. It is therefore recommended that further work should be carried out to develop cased and flavoured versions of these cigarettes so as to increase the flavour amplitude. If any of these subsequent designs were to prove acceptable, attention is drawn to the fact that one of the most successful of the current designs is based on a blend containing 25% of a synthetic smoking material (Cytrel 1-324). ~0

-r.~c. x uz.~. oj ,',,,_

-32- REFERENCES i. Evelyn, S.R. Discussion on Biological Research: A Meeting held in R. & D.E., Southampton, 23.9.71, File 46D. 2. B-A.T. Report No. RD.834-R, 2.11.71. 3. T.R.C. Docunmnt J.151, 27.1.71. 4. Nall, J.F. Brown & Williamson Report on LUCKY TEN Cigarettes, 16.11.71. 5. Jenkins, C.R. B-A.T.R. & D. Report in Preparation. 6. B-A.T. Report No. RD.758-R, 30.12.70. 7. Jodl, R. Paper read to 8th CORESTA Tobacco Colloquium, Bremen, 1966. 8. Cousins, A.R. Memo to R.E. Thornton, 4.10.71, File 46D. 9. B-A.T. Cigarette Design Handbook, Second Edition, 1972. iO. B-A.T. Report No. RD.TFg-R, 6.4.71. ii. B-A.T. Report No. RD.820-R, 18.8.71. 12. Evelyn, S.R. Notes on a Meeting with Amcel Representatives in R. & D.E., 22.9.71, File 36. 13. B-A.T. Report No. RD.778-R, 16.4.71. 14. B.T.C. Document 221/7, 20.9.71. 15. B-A.T. Hamburg, Progress Report II/69. 16. B-A.T. Report No. RD.776-R, 28.4.71. 17. T.R.C. Document G.726, 15.5.69. 18. B-A.T. Report No. RD.773-R, 7.4.71. 19. Lugton, W.G.D. Memo to R. Comber, 8.12.71, File 46D. 20. B-A.T. Project JANUS Progress Report: May-August, 1971, 1.9.71. 21. T.R.C. Document J.161, 27.1.71. 22. Darby, P. Report of Biological, Chemical and Consumer Tests of Cigarettes using Carbon Filters. Gallaher Ltd., 24.7.71., circulated by T.R.C. as Document K. 126. Lr~ r~o

-33- 23. B-A.T. Report No. B-4, 15.2.67. 24. B-A.T. Report No. B-IO, 18.12.67. 25. B-A.T. Report No. B-12, 7.11.68. 26. B.T.C. Document No. 174/14, 22.12.69. 27. B.T.C. Document No. 221/15, 10.11.71 28. B.T.C. Document No. 2OO/16, 19.11.70. 29. B.T.C. Document No. 160/3, 28.5.69. 30. B-A.T. Report No. B-I, 24.6.66. 31. B.T.C. Document No. 174/8, 13.8.69. 32. B-A.T. Report No. L.157-R, 17.8o65. 33. B-A.T. Report No. L.359-R, 8.1.71. 34. B-A.T. Report No. RD.714-R, 1.5.70. 35. B-A.T. Report No. L.331-R, 24.4.70. 36. Backhurst, J.D. Smoke Panel Tests, OSP 361-378, 3.3.72, File 46E. 37. Backhurst, J.D. Smoke Panel Tests, OSP 413-416, 10.5.72, File 46E. 38. Hedge, R.W., Personal Comunication, 1972. 39. T.R.C. Harrogate, House-skin Painting Experiment No. 1.1.11.2.

c-.,l

-34- APPENDIX I TOBACCO BLEND The following U.S. type blend (8) was calculated to give an average blend nicotine content of 3%: U.S. Flue-cured Virginia strip U.S. Burley strip Oriental Grade Crop % 02E1970 12 H2E 1969 12 $2XE 1969 12 BF2XW 1969 12 M2W 1969 12 WK2 "1969 12 MK1969 12 TK1969 6 CAB - I0 T55 m CZ) x-. ~O

% t.,,j ~r7

-35- APPENDIX II CALCULATION OF TPM DELIVERIES FROM EXPERIMENTAL CIGARETTES USING CIGARETTE DESZGN HANDBOOK DATA (9) ' A. Calculation of TPM Delivery from I00% Tobacco Ci6arette BD TPM delivery from 60 ~,m cigarette smoked to B mmbutt length - 38 mg. I. Allowance for Blend, Cgmponents 1 e For 80% lamina = For 20% s=em TOTAL Allowance for Paper Porosity At 120 ml/minute = At 600 ml/minute, x 0.73 = Allowance for 60% Filtration Efficiency Without filter With 60% efficient filter - TPM, mg/ci~arette 34.96 4.18 39.14 TPM, mg/cisarette 39.14 28.57 Calculation of TPM Deliveries for Experimental Ciaarettes usin~ NCF~ ~CL and 1-324as Tobacco Diluentsr at 59%b~ Weight These calculations assume cigarettes to be made to equal firmness. I. Dilution with NCF (RA91) The NCF used in these experimental cigarettes was assumed to give about I/3 of the TPM from tobacco at equal volumes and equal firmness (IO, 38). The NCF Bade for this exercise appeared to have 1/3-I/2 of the filling value of tobacco, and the TPM calculations which follow are based on an intermediate value, viz. 0.4. c~

cZ~ q,m...,mb qmh,q.t, ",4

-36- If NCF has 0.4 of the filling value of tobacco at equal weight, a cigarette with 50% NCF by weight will contain 5:2 pacts of tobacco:NCF by volume. The tobacco will therefore contribute 5/7 of the TPM produced by a 100% tobacco cigarette, i.e. 11.4 x 5/7 = 8.14. The NCF will contribute 1/3 x 2/7 x 11.4 mg TPM since it will be 2/7 of the total volume and per unit volume will deliver 1/3 of the TPM: The total predicted EPM is therefore 9.2 mg. 2. Dilution with 1-324 (RA84) This calculation is based on the following information and assumptions : (i) The predicted EPM yield from a 100% tobacco cigarette is Ii.4 mg per cigarette. (ii) The TPM yield of 1-308 is 0.12 of that from tobacco (19). (iii) The TPM yields from 1-524 and 1-308 are similar. (iv) The filling powers of 1-324 and tobacco are similar. TPM~mg/cigarette Contribution from tobacco, 50% of 11.4 5.7 Contribution from 1-324, 50% of O.12 of 11.4 0.7 6.4 3. Dilution with PCL The expected TPM delivery from a cigarette containing 50% PCL, 50% tobacco blend will be 9OZ of that from an all tobacco cigarette, since PCL delivers approximately 80% of the TPM produced by tobacco (9). exo CO

r,~ (..,,j M3

-37- Co The calculated yield is therefore: 11.4 x 0.9 - 10.3 mg/cigarette. In practice PCL was only added at the 25% level, in conjunction with the tobacco diluents. Calculation of TPM Deliveries for Other Experimental Cigarettes i. Blend:.~g% Tobaccot 22% NCF (RAg2) The filling value of NCF is again taken as 0.4 of that of tobacco. If NCF has 0;4 the filling value of tobacco at equal weights, a cigarette with 22% NCF by weight will contain 78:22 x 0.4 parts of tobacco:NCF by volume, or 90:10 as percentages. The contributions to the total TPM will then be: 9O From the tobacco, 11.4 x IO-"O I0 From the NCF, 11.4 x ~--~ x 1/3 TOTAL 2. Blend: 50% TobaccoI 25% PCL~ 25% NCF (RAg3) !PM, mg per cigarette 10.3 0.4 10.7 The filling value of NCF is again taken as 0.4 of that of tobacco. Equal filling values for tobacco and PCL are assumed. The relative volumes of tobacco, PCL and NCF will then be: 50:25:25 x 0.4 - 50:25:10 or 59:29:11 as percentages. The contributions to the total TPM will then be: t~ rko t~v r~

c~ t~ r~

-38- . 59 From the tobacco, ~-~ x 11.4 = 29 From the PCL, ~-6~ x 11.4 x 0.8 = II From the NCF, ~ x 11.4 x 0.3 = TOTAL Blend: 50% Tobaccot 25% 1-325~ 50 From the tobacco, ~-~ x I1.4 = 25 From the PCL, -~-~ x 0.8 x 11.4 = 25 From the 1-324, ~-~ x O.12 x 11.4 - TOTAL 25% PCL (RA85) 6.73 2.64 0.38 9.8 5.70 2.28 0.3~ 8.3 <-n t~ t~o ,'x_~

---

-39- APPENDIX III BLEND SEPARATION OF CIGARETTES RA82r RA91 AN.D KA84 Cigarette Weight of NCF (g) Weight o£ Tobacco (g) % NCF RA82 m 1 2 3 He an O. 178 0.228 0.143 0.183 0.657 0.621 O.701 0.660 21.3 26.9 17.0 21.7 Cigarette Weight of 1-324 (g) Weight of Tobacco (8) % 1-324 ,,,, RA84 m i 2 3 }~an Cigarette RAgl m 1 2 3 Mean 0.377 O.405 0.394 0.392 0.343 O. 384 0.363 O. 363 52.4 51.3 52.0 51.9 Weight of NCF (g) 0.434 0.474 O. 430 0.446 Weight of Tobacco (g) 0.456 0.462 0.465 0.461 % NCF 48.8 50.6 48.0 49. I <z> t~ 4~

I=IG. I ~o 934- o.. EFFECTS 0r 300 ms/WEEK OF' SMOKE CONDENSATE5 FROM 100% LAMINA ('T2"/t AND 100% ~TEM CIGARE;TTE$ ('TSB') (REFERENCE 39) % MICE WITH TUMOUR5 10 O I 0 I0 TUMORIGF..N¢ RATIOS ('~OM R¢:~ l'/) ARE "- STEM O.E3 • • 0"4"/ LA MINA I • I i~ IOO~/e LAMINA b ao 1o- 4o so 6o wct;Kl~ b. EFIrECTS OF" 380 rag/WEEK OF' SMOKE CONDENSATrS F'ROM Z-308 AND Z-324 CrROM ~C~NCE ~) MICe WITH TUM0URS ~10a 10- I - 305 Z - 3;~,~ ~.~ • & III . I nlb- " ~ ~ 4o so ~,o r.EK$ c~ .J~ r~

RD934-R PREDICTED EFFECT OF HIGH PCW~51TY CIGARETTE PAPER ON TUMORIGENIC RATIO, BY EXTRAPOLATION TUMORIGENIC RATIO ~.P., I.I 1.0 / 0,9 o 0.8 0 T~8 J J J J J I ! 1 I I I I00 800 aO0 400 $00 600 f ! t T44 TYPICAL T49 U.SJ< .$ r.T. f CURRENT EXP£RiI',~..NTAL CIGAR£TT¢" 6 re,E/MINUTE, PAPER POROSITy..... Ct'. O~

FIG.3 RD. 93 4- R ESTIMATION OF DOSE5 OF CONDENSATE$ GIVINq EQUAL RESPONSE IN MOU~E-5KIN THICKENINq TE~T, USINq DOSE-RESPONSE FROM REFERENCE 30 AREA AC, TIVITY RATING 20o to0 ! 5O DOSE (mS). Lr~ r~

FIG. 4 RD. ~34.- R PREDIC, TION OF" AREA-ACTIVITY KATINq FOR SMOKE CONDENSATE FROM U.$. K.5.F.T CIGARETTE WITH HIGH POROSITY PAPER AREA ACTIVITY RATING 130 12o Iio I00 90 90 T49 T4.4 J 0 ! I l I00 200 ,,100 J J E',) J U.S.K.S, FT. WITH HIGH PORO,5 rry PAPER I I l l 4OO 500 60O 700 (_,,,., C~

r,,j t..Aj t~

---