A GLOSSARY OF TERFS R-rLZV.Z`T TO PRODUCT DES10 JkND DEVELOPMENT C> C> ,10 2. jwx? a.% T L k *n-1 Lm-wu Thi$ MUU r®De C001Cd Or ShO-M L,>uft3utN>Ml~ed Orr-Ons (.j4 BATCo document for Legal Services: Health Canada 20 October 1999 50~E BASIC DE'~INITIONS The mcde:-,i cigarette represents a very complex system and there are a number of --e-ms and some standard measurements which are used to describe it. As an ai~e-memoire and to avoid any confusion and misunderstanding, a glossar7 of 100 terms normally associated with cigarettes is given in the next few pages. 1. AEROSOL Cigarette smoke is an aerosol, consisting of a dispersion of minute droplets of liquid TPM (q.v.) dispersed in a vapour phase (q.v.). AFTPE TASTE --R The taste sensation which remains in the mouth for a period of time after smoking. 3. AIR-CURED A TOBACCO TY?Z (q.v.) 4. ALKALOIDS A chemical "framily" of alkaline organic substances, the members of which occur naturally in many plants. Nicotine belongs to this 'family'. 5. AMBIEhT CONDITIONS The conditions in the room and environment in which the analytical smoking is carried out. 6. AMBIENT SMOKE The smoke in a room or localised environment. Ambient smoke is a mixture of sidestream. smoke and exhaled mainstream smoke. 7. AMELIORANTS C:) Additives/chemicals - usually acids or alkalis - added to particularly CD r1l) harsh grades or blends of tobaccos to reduce the irritation. 110 U4 (-M fr) --I a - T -V V I L,M,Id The% must not be enoted or sho-n to on2webonsed oeftnns BATCo document for Legal Services : Health Canada 20 October 1999 8. kX?L1'7 UDE 7he overa2l intensi:v of the smoke flavour (often called "tatal amount of taste"). 9. ARO~Lk The fragrance or odour of a material as registered by the nose. 10. ASH Residue of incombustible solids left after combustion. BLEND (Tobacco) A mixture of individual tobacco CRADES (q.v,) of various TOBACCO TYPES (q.v.). 12. BLIND TEST Evaluation of coded samples whose identity is known only to the investigator. 13. BURN ADDITIVES Chemicals added to cigarette paper to control the burn rate of the paper. The additives also reduce paper discolouration on storage and help prevent ash flaking. 14. BUTT LENGTH The length of unburnt cigarette remaining at the moment the smoking process ceases. 15. CAMBRIDGE FILTER HOLDER An airtight, chemically inert, chamber of low dead volume for holding a filter disc of glass fibre material. 16. CAMBRIDGE FILTER PAD A circular pad of glass fibre used to collect the TP4 delivered by a cigarette. i-l*& CD c:::) r%j CD M) %.0 (--4 C31% C* IQ-7 8 -k 7 U X &ni E%c>,me L.m,ted Thes mu* R DO4 be CoPied Or sho~n tu unautre)rlscd :,Cfmns BATCo document for Legal Services : Health Canada 20 October 1999 - 3 - 17. CARBON 40NOXIDE - CO A toxic gas formed during incomplete combustion, and by thermal degradation, of orgamic substances. It is present in smoke and measured both as a volume concentzation M and total- del'-very (mg/cig). 18. CAS ING S Materials added to tobacco in the course of manufacturing for the purpose of influencing the smoking quality of the finished product. Typical examples include sugar, liquorice and cocoa. 19. CELLULOSE ACETATE In fibrous form, this is the most -commonly used filter mater~al. A collection of cellulose acetate !ibres (tow) are formed into filter rods which are eventually cut into cigarette filters (plugs). 20. CIGARETTE HOLDER The device for holding the burt end of a cigarette during the smoking process. 21. CIY-4,RING PUFF Any puff taken by a smoking machine after a cigarette has been extinguished. 22. COAL The glowing high temperature (800 - 900*C) zone at the end of a burning cigarette. 23. COAL RETENTION Normally, the coal is retained intact during smoking. Occasionally coal f all-out occurs i. e. the coal falls f rom the end of the cigarette; this reduces the coal retention index when determined C) under standard conditions. C) U4 CC IvP7 9 A T 4 L K anj Eipuns Unata This must = be ctipic! or sho-ft W unsu&-Wnsed Persons. BATCo document for Legal Services : Health Canada 20 October 1999 - 4 - COEFrF:C-7--NT OF VARIATION Melative standard deviation) A mathematical expression of relative degree of variation w-.Z--n a set of results or data. It is expressed as a percentage and given by: 100 x Standard Deviation Average. 25. CONDITIONING The process whereby tobacco/cigarecr-es are allowed to equilibrate it a specified atmosphere. 26. CRIMP A wave-like pattern imposed on, e.g. filter tow or filter wadding. It increases -filling power and improves filtration efficiency by increasing the PRESSURE DROP (q.v.) obtained from a given mass of mater-'al. 27. CRS Cut Rolled Stem. 28. CURING A method of drying freshly harvested tobacco leaves in such a way that the desired qualities are produced in the dried leaf. There are four main methods of curing, each of which produces a different TOBACCO TYPE (q.v.) 29. CYTREL Synthetic smoking materials manufactured by the Celanese Corporation of America. 30. DELTA This is an assessment technique which is used to evaluate low delivery products. A number of cigarettes are evaluated over a range of product attributes/descriptors using a rating scale. The J:!b C> 'raw' ratings per cigarette are analysed by a statistical method. CD N) C) -10 C74 co 91 -~-TBAT -L K 3njEiixsniL,m%sej ThLs rhmst w tc cnvied orshountognautr.)nsed persons BATCo, document for Legal Services: Health Canada 20 October 1999 - 5 - -1. D E N I ---.R -ila=ent denier is the weiec (in gra=es) of a single f-bre 9,~)OO metres in length and hence an indirect measure of fibre thickness. Total denier is the weight (in grammes) of a collection of fibres (tow) 9,000 metres in length. Hence total denier/fl-lamenr- denier gives the number of fibres in the tow. 3:. DENSITv Is the weight of tobacco at a given moisture content per unit volume expressed as mg/c=3 and is sometimes referred to as Nett Nett Density. When the weight includes the weight of the cigarette paper and adhesive then the density is referred to as Nett Density. DIET A high order expansion process in which lamina is impregnated with liquid C02, chilled and expanded by the rapid release of CO-) on heating. DIET stands for Dry Ice Expanded Tobacco. DISTRIBUTION A term used in statistics to show the pattern of the frequency of any given occurrence, event or measurement. 15. DRAW RESISTANCE (Pressure Drop) The value of the pressure drop (expressed in mm. Water Gauge or cmWater Gauge) between the two ends of an unlit cigarette or a filter rod when air is drawn through at a constant flow rate of 17.5m]/s. 36. DFM Stands for DRY PARTICULATE MATTER which is equivalent to,TP4. (q.v.) minus water in Scandinavia. In Canada DPM is TPM minus water and minus nicotine. CD -.0 L.1"I Z t4p,7 9 ,A T iC K ai.,l Eirnj Limiii:4 This r-must r4c be enoted or SP&O-n to iinautrx:insea persons. -.0 BATCo document for Legal Services: Health Canada 20 October 1999 - 6 - 37. DUOf"= TEST This test is a forced choice method and is used to deter=,:-,e whethe: two prod-acts differ in any way. Three samples are presem:ad, two are the same and the third may be different. One of the idenrical samples is identified as the "control" and the subje:-- is required to pick the unidentified sample which is the same as the "control". 38. ELEC---~ ~- STATIC PERFORATION (EP) A meaz-z of obtaining ventilation in cigarette and tipping papers by fc=ing large numbers of small holes, not easily visible to the --~ked eye, by, electrostatic means. 39. END S:ABILITY A measure of how well the tobacco in the end of an unlit ciga-,e:--e is rezained whe= subjected to a standard shaker test. 40. EsnV: Trade name (EAS7~LkN) for cellulose acetate filter tow. 41. F-UA-N--7-D TOBACCO A te-.- for cut lamina that has been treated by an expansion process usual-*7 DIET (qv) or C-13 (qv). 42. FILL:NG DENSITT See -wensity. 43. Fill--NG POWER The capacity oZ- a given weight of tobacco to produce a leve-- of fir--ess when used in a cigarette. The filling power of tobacco is greatly influenced by processing techniques and blend nature. 44. FIV-7-UTION EFFICIENCY C) Def!=ed as the percentage of smoke, or a inaterial in smoke (e.g. C) r*1) nic-.:--ne) which is retained on a filter plug under standardillsed C) smokiag conditions. %.0 --7 B A r I L K a nt E %&%in ~ Umncd Thts must mcx be copied or sho n jumnaultbansed pv-son, BATCo document for Legal Services : Health Canada 20 October 1999 45. FIRE-CURED ATOBACCO TYPE (q.v.) 46. FI-VOUR Sensation caused by any substance, taken in the mouch, which stimulates the taste and/or smell senses. 47. FLAVOURS (Top dressing flavours) Aromatic materials added to tobacco usually at the final stage of primary manufacture, or more recently during secondar7 manu-acture. Small quantities of flavours are applied in a solvent, usually alcohol, propylene glycol or water. Typical flavours are Spice Oils, Fruit Oils, etc. 48. FLUE-CURED A TOBACCO TYPE (q.v.) 49. FM SMOKING The condition that exists when the burt end of a cigarette is completely exposed to the atmosphere between successive puffs. 50. G-13 A high order expansion process in which lamina is impregnated by an organic agent, Freon-11 (a Fluorocarbon) and then expanded by rapid heating. 51. GRADE (Tobacco) A subdivision of a TOBACCO TYPE (q.v.) based mainly on the position of the leaf on the stalk of the tobacco plant but also based on other factors including ripeness, colour, texture, quality, etc. 52. HISTOGRAM A statistical term given to the plot of frequency versus C) characteristic. I'J C::) N) I'D r- ,,*a s ®F - i K a-J F i-n, L~m,iej This mu-9 MKX he C mO®d Of SNO~ Ob C0 uftauttbOrlsed Penons BATCo document for Legal Services : Health Canada 20 October 1999 - 8 - 53. HOT COUAPSr. The tendency of the cigarette or filter to soften considerably when being smoked. 54. RbMEG-.,MNTS Additives to the blend - c3ost usually glycerol and/or propylene glycol - which assist moisture retention, especially important in hot climates for packs without ovet-wraps. Not usually used on flue-cured cigarettes in which high natural sugars perform a similar function. 55. IMPACT (On inhaling) A shar?, sudden, but short-lived sensation which is exverienced immediately smoke =akes contact with the back of the throat. 56. IRRITATION When used as a taste. and flavour term this describes a smoke sensation which usually builds up in the mouth, throat and nose and dies away slowly. 57. LEAGUE TABLES A table produced b~, a Government Body or Independent Authority ranking, listing or grouping the brands on a given market according to the deliveries of smoke components (PM'WN`F, nicotine, carbon monoxide, and occasionally other smoke constituents). 58. MAINSTREAM SMOKE The smoke leaving the butt end of a cigarette during the smoking process. 59. MOUTHFULL OF SMOK-1 The impression of volume or amount of smoke entering the mouth and occupying the mouth during normal puff taking. A physical impression and not to be co=used with any temperature, irritation or taste sensations. '~.C' _P_ -gl ivg? 9 k T t L K mi E~,,m i Lanaed Thts must n(X be cooted of shown it) unaujh~)rjed Demons. N; BATCo document for Legal Services : Health Canada 20 October 1999 - 9 - 60. N I COT 7~ YE- "'he ma'or ALKALOID (q.v.) in tobaczo. Nicor-'ne in cigarette smoke is often expressed as milligra=es of nicocine per cigarette. Nicotine in tobacco is often expressed as a '. on a dry weight basis. 61. NICOTINE - EXTRACTABLE Lhe amount of nicotine which disscives in chloroform when TPM is extracted with a mixture of chloroform and water. 62. NITRIC OXIDE OR OXIDES OF NITROGEN (NOX) These are toxic gases known to be present in smoke. They are measured both as a volume concem:ration (parts per million) and total delivery (ug/cig). 63. OMEGA !ETROD The group standard sensory evaluation method designed to accommodate a sensory evaluation of all cigarettes, irrespective of blend type, deliver- y or construction. It represents a package of attributes and procedures which can be used during paired-comparison or monadic tests to describe both the sensory characteristics of cigarettes and assign intensity ratings to each attribute. 64. OVERTT'~V ING LENGIB The length of cork or white (or other) tipping paper (q.v.) used to attach the filter to the tobacco rod. The anchorage is the length of the tipping which overwraps the tobacco rod. Thus anchorage plus filter length equals overtipping length. 65. PASSIVE SMOKING Usually regarded as the inhalation, in the course of the breathing Cycle of ambient smoke (q.v.). CD C:) rN > C) r1j ~10 -9~2. U4 7 8 A T , L K aAJ E vc-,m I Limited This must rKX be crti-ed of sho- - ;,7 inaulhonsed mr BATCo document for Legal Services : Health Canada 20 October 1999 - 10 - Eiz. PERHEAB I L 1 -:Y is the tendencv of zhe Daper to allow air to passthro-L:gh it. 1: is measured as the flow of air (in cm3/min) which passes through a given area of paper ~:hen a 'Known pressure drop is applied across the paper. In the case of cigarette paper, plugwrap and aaturally porous tipping paper, an area of 1cm_' is used at a pressure drop of 10cm WG yielding per-meability in CORESTA Units. In the case of perforated tippings, the flow is measured through a I= length of the paper in the perforated region at 10cm WG pressure drop. 65. ?JS Peter J. Schweitzer reconstituted tobacco manufactured in the USA by the Schweitzer division of Kimberley-Clark Products. It is made on a pater making machine to various specifications for cigarette filler and for cigar binder/wrapper. 66. PLUGTYRAP A layer of paper directly enclosing the circumfe rent ia 1 face of the cigarette "Liter material. 67. ?)M-WNF Stands for Particulate !!atter, Water and Nicotine Free, or `TPM (q.v.) minus water and minus nicotine. 68. PRESSURE DROP See Draw Resistance. 69. PRODUCT BRIEF This is the working document which states in objective terms the target characteristics (eg smoke deliveries, length, blend style, taste) of the product which is to be developed. The brieff Is drawn C) up from discussions with key departments within the company CD r%J (marketing, product development, leaf, production and purchasing) C) 111-J and is used to guide product development activities. IF, 1-40 B.At T oL X This miiumbecnpiedorsho."toonamthorscc;prwns BATCo document for Legal Services : Health Canada 20 October 1999 % . ?R: ?aper Reconstituted Tobacco. A material derived from tobaczo waste which has been reduced to a slurry and reconstituted bv a ?a;~er making machine. A process whereby the smoke from each successive puff of a cigarette is collected separately for analysis. NoTtl: This process ran also include the collection of smoke from a group of cigarettes. In this case the smoke -from each successive puff on one cigarette is bulked with the smoke from the corresponding puff of the other cigarettes. 72. PU7_7 DMATION The interval. of time during which suction is applied at the butt end of any one cigarette. 73. Pt7.-7 FREQUENCY 7he number of puffs in a given time. 74. KFF-r INTIERVAL 7he interval of time during which no suction is applied at the butt end'of any one cigarette. 75. ?UFF NUMBER -he number of puffs necessary to smoke a cigarette to a specified butt length. 76. PUFF PROFILE A graphical representation of flow rate during the puff duration as a function of time when measured behind the butt end of a cigarette. It describes the shape of the puff taken either by a smoking machine or the human smoker. N) CD INJ B A T tL K a,..j J%p.nt Lanoej This must not t* Copied or shom to U112YINinsed Pemons BATCO docurn t f L ,-,, %,%j u0curnent for Legal Services : Health Canada 20 October 1999 - I! - 77. PUFF VOLUME 7he volume leaving c~e but: end of a c,*gare:-.e during the pu-" duration. 78. REDUCING SUGARS A group of sugars contained in tobacco (especially in oriental and flue-cured). The name st'a=s from the laboratory method used for their analysis (reduction). Glucose and fructose are the only reducing sugars in cured tobacco. Normally ex:~ressed as a Z on dry weight basis. 79. RELATIVE HUMIDITY The amount of moisture vapour in air relative to the amount required to produce saturation expressed as a percentage . The RR can greatly influence the moisture content of tobacco and cigarettes. 80. RESTRICTED SMOKING The condition chat exists when the butt end of a cigarette remains closed to the atmosphere between successive puffs. 81. ROTARY SMOKING Y-ACHINE A =u1ti-port smoking machine which has the ports arranged concentrically around a rotatable circular head. Each port is brought into line with the single puffing mechanism by the movement of the rotating head. 82. RTS Reconstituted Tobacco Sheet:- A general term covering all reconstituted sheet products such as SRT, PJS and PRT (q.v.) 83. SHREDDED STEM Produced by passing conditioned whole stems through a rotary disc mill. It can be used in place of CRS (q.v.) or WTS (q.v.) in blends CD PI) C:> N) %10 @ L K ani En-rti Urn-ted Tho. MuSt 1KX be Cnpacd or sro~n to unauthansed pem*nj BATCo document for Legal Services : Health Canada 20 October 1999 84. SIDESTREA4 SMG"-",Z :he smoke -.;h'_ch leaves a cigarette during the smoking process other than from the butt end. 85. SUIRMTAYEOUS SMOKING ~LACHINE A multi-channel, multi-port smoking machine which draws a puff from all ports at the same point in :!me. 86. SMOKE DF T 1he term pH is a measure of the acidity or alkalinity of any material. Values lie on a numerical scale from 0 to 14, with 7 as the point of neutrality, acids below 7, alkalis above. The scale is logar-; thmic and thus smoke of pR 5 is ten times more acidic than smoke of pE 6. 87. SYOKING CONDITIONS (Smoking Regime) The conventional method adopted for smoking cigarettes by machine. Typical conditions are puff volume 35ml, puff duration 2 seconds, puff frequency I per minute. 88. SMOKING ENG:NE (or MACHINE) A mechanical device on which cigarettes are smoked under standardised smoking conditions to a prescribed butt length. 89. SRT Schweitzer Reconstituted Tobacco. This refers specifically to reconstituted tobacco manufacturd by a paper process at Le Mans by Le Tabac Reconstitue (a subsidiary of P.J. Schweitzer). 90. STANDARD DEVIATION A widely-used measure of scatter, spread or dispersion In a set of numbers, values, results, etc. 9 1. SUN-CURED A TOBACCO TYPE (q.v.). C) r_j CD rQ -10 (C7'. 1*-7 B AT tL K xr~i kxco~j Umb.ej Thks must not be CF9*4 Of %P10untOwn3uthonsed persons BATCo document for Legal Services -, Health Canada 20 October 1999 - 14 - Has been variously used to mean P,~',NF, DPM and --?M. It is :'nere-fore Imoortant to establish w~at is meant by TAR before it -is used. 9- TIPPING PAPER A circumferential paper wrap which joins the filter plug to the tobacco rod. 91- TOBACCO TYPE A subdivision of tobacco based mainly on the method of CURING (q.v.). There are four main tobacco types, each with distinctive smoking charac:erisrics: Air-cured, Fire-cured, Flue-c-ured and S=-cured. 95. TOTAL SUGARS The total amount of suga----.-.pe constituents in tobacco, normally expressed as a Z on dry weight basis. 96. TPM T lotal ?articulate Matter. 'That portion of the whole smoke which is retained on the CAMBRIDGE FILTER (q.v.) pad (and holder) when a cigarette is smoked under standard conditions. This is usually expressed in milligrammmes of TPM per cigarette. 9'. TRIANGLE TEST A method of difference testing in which three coded samples are used. Two samples are identical and one is different. In cigarette smoking the task of the smoker is to iden'tify the odd sample. 98. VAPOUR PILASE That amount of smoke which is not retained by the CAMBRIDGE FILTER (q. v. ) CD r~'J CD r\J %.0 00 iD i~- 7 8 -% T i L K ani E %P.,m e Limara Thil muu no& bc cnpied or %bu-ft tu unauthonsec e-¯ons BATCo document for Legal Services : Health Canada 20 October 1999 - 115 - The amount. of air which enters the cigarette bF zeans other than through -.-e lighting/burning end of the cigaret:e. It is r--,-mally e.-;:,esse,* as a percentage of the total volume draun from the unlit cigarette. Filter/tip ventilation refers to the amount of air entering through the tipping 1plu gwrap combination and tobac::o rod ventilacion refers to the air entering through the cigarette paper. 100. WTS Water Treated Stem - it is a process carried out on CRS. It involves wetting to high moisture levels and steaming, before par-drv----g. 0 CD ~10 j~:- CC' ~4%7 BA ' -L K anj Lanard ThM Must not bc Cnpied Of $hO~n lu un3uthorised Persor.% -10 BATCo document for Legal Services : Health Canada 20 October 1999 0 WRAT IS SX0KE 40 CD Pli C) rIl.) %~0 (œ-I IYM7 B A, T Et K anj E,lxnl L.rniied This Myst rioq be cnpied or thoun au un-%uthansed penorbi. Un CD BATCo document for Legal Services : Health Canada 20 October 1999 '-RAT 15 SMOKE I';-RODUC:10X C-garetze szoke *-as been described as one of the worl(~'s most complicated chemical svste=s. This is because the smoke is der.ved from burning a natural substance - the tobacco - containing hundreds of chemical constituents. On burning, the tobacco is exposed to temperatures ranging from ambient to 950*C in the presence of varying concentrations of oxygen; resulting in thousands of chemical products being formed via several distinct routes. The quantities of the chemical products for-med are dependent on the nature of the tobaccos used as well as on a large variety of cigarette paraizeters like paper permeability which affect the combustion process. T this paper covers, very briefly, the four basic aspects of a buraing cigarette: 1. The combustion process. 2. 71he generation o! smoke. 3. Methods of measuring and reporting the most common components of smoke. 4. The way somii of these components are delivered. 1. TIU COMBUSTION PROCESS (The whole of the combustion process has been illustrated in a film made by GR&DC and part of this will be shown during the Seminar). When a cigarette is smoked, two distinct types of burning take place - puffing and smouldering between the puffs. In very general terms, only about 30% of the tobacco rod is consumed during puffing; roughly a further 551- is consumed during smouldering with the remainder being discarded in the butt. rNj U1 .~j' 1"7 BAT 1LK This inwst not be copied or shown it, unauthonseapersons BATCo document for Legal Services : Health Canada 20 October,1999 DAM. TOMACCO ROD BUTT PWWNF 35.0 OUT? - SMOULDER L During the puff, air is drawn into the cigarette and the central region of the coal reaches a temperature of between 800 and 850%. The coal has relatively high draw resistance to the air flow and consequently, during a puff, the air tends to enter the cigarette at the base of the coal, just In front of the paper burn line where a temperature of over 900% can occur. This is the reason why, during a puff, It is mainly the periphery of the coal that advances. There is a rapid fall in temperature as we move away from the coal; within about 5mm of the burn line, the temperature is down to around 300*C and about 20-25mm away from the burn line, the temperature has fallen to 25%. -C:b ID C) N) C) r~j \0 Ln Z I.V7 P AT it. K am-i Ex&,m3 Lan.oci This must 1%04 tw copied or $hO%n 1%3 unaudw)nsed perso-is N.) BATCo document for Legal Services : Health Canada 20 October 1999 - 3 - TEMPERATURES IN A BURNING CIGARETTE 25mm SS M 00 KK EE at 20-25 Oc N- -(ambient temperatur . Oc fioKE OC -2 5~ at 0 mp ribientMtemperature 8500C 3000C 250C Since the pe--meability of paper to air increases sharply at about 300%, there is a large influx of air just behind the paper burn line and therefore much of the Incoming air during the puff by-passes the central region of the coal. This results in a smaller increase in the temperature in the central region compared to the increase in the temperature at the periphery. When the puff ends, the periphery of the coal cools rapidly, since It radiates heat to the surroundings; its main source of heat is now the inner core of the coal - a temperature of almost 800%. The central region of the coal advances to re-establish a relatively flat region at the back of the coal, with a relatively constant temperature across the diameter of the cigarette. Thus, immediately after a puff, Atbi, there is often a time delay of up to 15 seconds before the paper burn C:) C:) line visibly moves. ro 1,0 (71 CC, 14117 8 A r , L K aj E,o,,n) Limarl i his must no4 be Cnpied or sboun to unautbansed persons. BATCo document for Legal Services : Health Canada 20 October 1999 '.'i:h this general picture of the combustion process, we can now ident--fy the main reactLon zones in a cigarette. REACTION ZONES IN A CIGARETTE PYROLYSIS ZONE -01'77- A'~ COMBUSTION Z FILTRATION AND IN DUE COURSE REPYROLYSIS ZONE Firstly, there is a combustion zone where the burning takes place, resultinst in the release of gases and vapours. Behind the combustion zone, material is being released from the tobacco either by pyrolysis (thermal decomposition) or by direct transfer (e.g. distillation - nicotine, amino acids). As the vapour is drawn throuen the tobacco rod, It cools rapidly resulting in condensation into aerosol particles which make up the smoke. Some of these particles are filtered by the tobacco rod and hence in due course become available for re-pyrolysis and combustion. -c~h CD CD r-%j C) r%) CC ~,sM7 8 .% T gL X anj E%t,,jeti Limieti Th~i Myst not be cnpted or thoun tu itnauIN)nscd p‡rons. (il .J>b BATCo document for Legal Services : Health Canada 20 October 1999 - GEN-ERAL in '--road ter=s, when any material - whecher it is solid, liqu-4--l or gas is blur-It, a sr-oke Is produced. The temperature at combus:~on :s to alw s be -aleasec: ,cr=a_17 enough for :he co=bustion products ay 'n vapour or gaseous -foz--, i.e. invisible. However, as soon as the envitonmental temperature decreases, some of the combustion nroduc:s condense into small droplets and some of these droplets stick together giv-4--2 a mixture of gas~'vapour and droplets /particles - The drople--s will nor--ally become semi-opaque (cloudy) so that the overall resul: is a srey*sh colour (or black when the original material contains a large number of carbon atoms, e.g. smoke from burning oil). Thus, smoke consists of a large number ci~ ver7 small particles of liquid in a mixture of gas and vapour. 71he distinction betwee= gas a=d rapour is a fine one and for most practical purposes it is best to assu=e that a smoke aerosol has only two phases - particulate and gaslIvapour. Broadly speaking, therefore, we have three states in which chemicals exist 1r. smoke: 1. n..ose wholly in the particles. 2. -Those wholly in the gas. 3. 2hose in both ph2ses. As the env*".onmencal temperature decreases, more and more substances condense out, until a degree of relative stability is reached at ambient or 'room' temverature. At this stage, we see the familiar smoke where condensation from the gas phase Is effectively complete and temperature stability exists. However, certain chemical reactions continue to occur, smaller particles continue to coagulate to form larger particles and there is an Interchange of molecules between the particles and the gas phase in'which they are dispersed. C:) CD ra t.#N7 B % T , L. K anj Eipoxi This must not be cnpltd of Sho-M 1%) unauthansed ;wtsons, -10 V1 V1 BATCio document for Legal Services : Health Canada 20 October 1999 Z- CIGARME SM01M Smoke fror- cigarettes may be split up into two broad categories: mainscrea= smoke and sidescream smoke. Xains:rea= smoke -s generated during puffing when the air -;s drawn into the cigarette through the coal. In the Interval between puf!s, a natural convection flow of air around the coal In an upward direction sustains the smoulder and forms the sidestream smoke. Thus, the mainstream and the sidestream smoke have been generated under quite different condi-ions and have characteristics as shown below: j LARGE INTER-PARTICLE DISTANCE SIDESTREAM SMOKE RELATIVE SIZE STAaILITY SMALLER PARTICLES 2750 20* FILTRATION ONTORACCO MAINSTREA SMOKE SMALL INTIER-PARTICLE 01 E )-2 FILTRiJ ON a CN Toe LOW SIZE STASILITY LARGER PARTICLES 95%r.h. EXHALED SMOKE LARGE INTER-PARTICLE DISTANCE SYSTEM RELATIVE SIZE STABIU'rY LARGE PARTICLE Because of these differing characteristics, the reflection of light from these is different and, hence, the mainstream smoke appears grey/white whereas the sidestream smoke appears grey/blue. The mainstream smoke is Inhaled by the smoker and goes Into the respiratory system. Rere the CD N) C) I-C, INX? 8 A T iL K ani Eirwrti Lemi,c%i This must not be cnPled Of shO%n to 6anautN3nsed permns, E3ATCo document for Legal Services : Health Canada 20 October 1999 - 7 - humidity is very '--;gh and some of the s=oke Is recained and the remainder e-haled. Since -t is the mainstream deliveries that are most commonlF reported and ut-1-sed in product development work, we shall concentrate on the =ains:rream s=n,e In paper. qowever, sidestream smoke is becoming an issue amongst anti-smoking lobbies in terms of room pollution, smell etc. To recap, mainstream smoke, containing thousands of compounds, may be divided --ito a particulate phase and a gas/vapour phase. One cubic centimetre of smoke can contain anywhere between ten million (107) and one billion (109) particles ranging in size from 0-1-1micron. (I ricro, ~ 1/106 =etre or 1 millionth of a metre) and each particle may have up to 10,000 substances in it. If we want to separate the particulate phase from the gas/varpour phase, we require a filter- which is 100% efficient in trapping the particles contained in the smoke. The CAMBRIDGE filter pad, consisting of glas's fibres consolidated by a resin, is Just such a filter. Therefore, If the mainstrea= smoke (or any other smoke for that matter) meets a CAMBRIDGE filter pad, the T07AL PARTICULATE MATTER or M. will be collected on the filter pad, and or-ly the gas/vapour phase of the smoke will pass through it. 0 J>l c:::) CD r,~-J ~10 Cg 14,17 0 A T iL Kar,4 Eiixirti L;o-.~te-jTh'$ Muil M be cnpicd or shouft io tin3uthonsed persons- Ul ~j BATCo document for Legal Services : Health Canada 20 October 1999 C A M B R I D G E S M 0 K E FILTER . 0 0 0 .0 00 . a AEROSOL VAPOURor GAS One thousand million particles 00') per cc. Particle size: 0.1 - 1.0 microns (io-metre) E3Ch particle may have-10,000 substances 7.74. sim-.-'-r represents the total weight of all the particles collected from a c-~parette when I-- is smoked on a machine under a given condItion. Physical-1--, TFM appears as a brown stain on the CAMBRIDGE filter Dad and It incl-,:des both water and nicotine. 7here are various other ways in which the deliveries of tllie particu-la:e matter are quoted: DPM --his is an abbreviation of DRY PARTICULATE MATTER and Is calculated by subtracting the weight of water in the condensate from T?M. MN F This stands for P-4-RTICULATE MATTER, WATER and NICOTINE FIREE and Is again calculated by subtracting from TPM the weight of nicotine and water In the condensate. TAR Tar has had widespread use In the context of cigarette smoke although It Is really a misnomer. For example, In certain CD rs') C) :--7 8 A T 1 L X ..,j E,P-,n i Urnars This must rKX be cnpicd or sho-n to nautitonseed pe-sons. un co BATCo document for Legal Services : Health Canada 20 October 1999 - 9 - c~rcumstznces, :he word has been used b-r independent bodies and Government ;rublications to mean: FM T icotine ,PM minus a- FX minus n~cocine minus water. Most cigarette smoke 'figures' published today are of the 117.1VN-z type. There are two advantages in expressing delivery In this way: 1. Water concent in smoke can be very variable, thus affecting 7_r*1 Ngures, resulting in a wide range of results over a series of tests. By subtracting the water, a --uch lower range is achieved. 2. ?ublished figures in league tables are seen to be low or lower. ;.%,e portion of the smoke that has gone through the CAMBRIDC-- filter pad is the gas/vapour phase and it contains carbon monoxide, nitric oxide, etc. Carbon monoxide is a natural by-product present in the human body whether one s=okes or not and it can be absorbed from the environment, e.g. from car exhaust fumes. It is toxic in large amounts because It affects the oxygen-carrying capacity of the blood. Nitric oxide has been implicated by, some people in the development of chronic obstructive pulmonary disease and emphysema in smokers. Because of this, removal or reduction of CO and NO is a serious problem facing cigarette manufacturers. Methods of reducing these components will be discussed briefly later. However, there are lots of compounds which occur in both the particulate phase and the vapour phase, e.g. acetaldehyde, acrolein, hydrogen cyanide, etc. The first two have been associated with Irritation -Ch. C) and HCN Is a highly poisonous gas. C) One point you have to be aware of is that different units are C) used in reporting deliveries of the various components: Url %.0 @ 114147 8 AT i UK &nJ Ei pun P Um-wj This m,st r*' tot~ cr~Pscd Or shO%%n to unjuthonsed penon, BATCo document for Legal Services : Health Canada 20 October 1999 * 0 QUOTED UNITS OF MEASUREMENT TAR OPM PMWNF mg 000 TPM 00 gram NICOTINE ~ 2 co 1 0 m NO etc. ~Ljg 1,000,000 gram * VAPOUR PHASE -o put the cocpositic= of smoke Into context: WHOLE SMOKE COMPOSITION (% shown are on a weight/weight basis) 0 S NITROGEN Ilycfro2en, Argaft. metftwto TPM OXYGEN VAP 0 16 rMst 10 Is -Ster co, C:) Cc. I.-' , 3,.k 7' 1 L K inj Ett%in3 Umort Th3s must " t-e CrV~ed or sho-" I,) una0ghonSec: persons BATCo document for Legal Services : Health Canada 20 October 1999 Beffc:e =o%,4, ng oz zo discussing the methods of measuring smoke 'eliveries, le: ~.,s su=a:-se what has beer: covered so far. '-nthe -igarette, the s=oKe -s generated by burn;ng tobacco which, 1,ke any other natural su--s tance , contains hundreds 0 f compounds. *;rh e n any single substance ~s burnt, it gives rise to several other substances: thus cigarette s=oke contains thousands of chemicals. It has been estimated that c-garette s=oke contains in the region of 10,000 compounds, some of which are present In such minute amounts as to be Just detectable but at present Immeasurable. At least 3,600 of these compounds have been ideut~fied- Some compounds present in the tobacco give rise to acids, (e.g. sugar~, others to bases, (e.g. protein). 3owever, nicot-,Me is transferred virtually unchanged into the smoke. As a rough guide, a maximum of 13-14Z of the nicot-,ne in the blend is transferred Into -zhe smoke. This =eans the higher the nicotine in the blend, the higher- will be the nicotine in the smoke. Tobaccos f rom higher plant pos~ Jons have higher nicotine, and high plant position tobaccos tend to have more flavour as well. Thus, we have high nicotine blends being associated with high flavour character. The smoke, containing thousands of substances, is made up of millions of particles dispersed in gas/vapour. The particulate phase of the smoke can be separated from the gas/vapOur phase by passing it through the CAMBRIDGE filter pad. The particulate phase is collected on the filter pad and it includes nicotine and water. By subtracting the weight of nicotine and water in the condensate from the veight of the Total Particulate Katter collected, we arrive at PYWNF, which Is normally the figure used in most publications. 3. MEASUREIMN'T OF SMOKE DELIVERIES - MAINSTREAM (Part of a film Illustrating the determination of deliveries using a CD 11 FILTRONA 300 smoking machine will be shown during the Seminar.) -10 @ 14%7 B AT tL K anj Eiaoni Ltmiie-.- Th:s must not be enp3ed or shown io unauthansed persons C)1% BATCo document for Legal Services: Health Canada 20 October 1999 -he ac:-ual amount of s=oke generated from a cigarette detiends on a large number of factors; a) the cocditions under which the cigarettes are smoked - altitude, temper-=:ure, relat--ve humidity, draught: b) the condition of the cigarettes when they, are smoked - moist-are content; c) the wav the cigarettes are smoked - puff volume, puff durat-Jon, puff frequency, puff number, butt length. Therefore, in order to compare deliveries across brands, it is Important that these deliveries are obtained under some sort of agreed standard conditions. Suc, standard conditions have been defined for: i) conditioning the cigarettes prior to scoking (hurzidity, te=perature, time), ii) conditions in the laboratory during smoking (humidity, draught), and most i=portantly, iii) smoking regime to be followed. 1he most common smoking regime is: %.ff volume 35cc Puff duration 2 sec. _Puff frequency 1/min (inter-puff period - 58sec). The final parameter to be defined is the butt length for which there are 9 a multiplicity of standards. CD r'l'i CD r~ ~10 cyl% r\j @ j y -7 B.-\ T - L K &.J F %cm,rs a L.m-ceu This fnust not be cripied or sho¯n tu unaut.-x)msed perlons BATCo document for Legal Services: Health Canada 20 October 1999 13 - BUTT LENGTH (mm) PLAIN FILTER STANDARD CIGARETTES CIGARETTES I.S.O. 23 Max (F+8,OT+3) Minimum 23mm CORESTA 23 F+8 or OT+3 Minimum 23mm UK (DHSS/LGC) Cig Length$ 7Smm 20 Max (20,OT+3) Cig Length > 75mm 20 Max f20,0T+5) USA (FTC) 23 Max (23,OT+3) CANADA/JAPAN/ AUSTRALIA/ 30 Mak (30,OT+3) NEW ZEALAND F - FILTER LENGTH (mm) OT - OVERTIPPING LENGTH (mm) Thus, for axa=:1e, for a king size cig2rette with rod length of 64mm, filter length o-- 20mm and tipping length of 24mm, the CORESTA standard would result In a butt length of 28mm (filter + 8mm), i.e. 56mm of the tobacco rod being smoked. The UK Government nemist's standard, on the other hand, would result in a butt length of 29mm (Cipping + 5=), I.e. 55mm of the tobacco rod being smoked, and hence this will give a slightl7 lower de-livery. Deliveries can be reduced simply by increasing the tipping length which results In a decrease In the length of tobacco rod being burnt during smoking. This has of ten been done coamercially and a notable e.~rample is MERIT which uses white-edged cork tipping to maintain the apparent balance between the tobacco rod and the filter. C:) It should be remembered that machine smoking Is largely artificial but it is a reasonable way for comparison of deliveries across %0 01% L.14 Cc: 1%#!17 9 A T tL K and Etponj L.inoted This must not be cnpied or shown to unaut.*K>ns4ed persons. BATCo document for Legal Services : Health Canada 20 October 1999 brands. lowever, we can be ser-fously misled by assuming that the machine deliver-:es are those obtained brr the smoker. 7'he human smoker Is a very varia.le, whimsical machine who most probably smokes different c4zareEtes in differpnt ways, ever. -aryi=z his puffing process down the clearettes In order to satisfy his/1-er req-_-'re=ents- 4. Pr-FF-3y-PUFF DELIVER'-E-Es From the produc: point of view, it is not only the total amount that is delivered which --s important but also how it is delivered on a puff-by-puff basis. D'ELIVERY PROFILES FOR TPM TPPA IM91 10. 9. Air-cured 7. Flue-cured 4 3- 2. I z PUFF NUMBER we notice that subsequent puffs deliver greater amounts of TF4. There are three reasons for this: a) Filtration by the tobacco rod during the earlier puffs means that there Is more =aterial available for re-pyrolysis in later puffs. C) C) NJ C) N) -10 ON @ I vF7 8.,.k T tL X and LAinatina, This must i%cit be copied or sho.n tu wriauthansed persons. J~_ BATCo document for Legal Services : Health Canada 20 October 1999 - 1~ - b -s the cigarec:e is consumed, there is a shorter tobacco c--lumn available for filcracion. C) As the cigarette Is consu=ed, there is less diluting a-*r coming through the cigarette pa?er and therefore --ore air is coming through the coal thus generaring more s=oke for a f-xed puff volu-me. P_14WNF follows the same trend as TF11 and so does nicotine Nicotine is considered to be the 'active' or 'physiologically active' ingredient of tobacco. When a cigarette Is sroked, nicotine is released as a combustion product and also as a result of a form of distillation process Caused by the hot, moist smoke passing through the cut tobacco. So if, during smoking, the smoke/air movement through the tobacco rod is restricted, as in a conventional smoking machine (viz- the but-, end of the cigarette remains closed to the at=osphere ir. between puffs), this will reduce the tendency for the nicotine to distil back. However, if the butt of the cigarette Is exposed to air between puffs, ie. a free-smoking condition, as would occur in the case of human smoking, then the nicotine will tend to distil back. This results In a greater amount of nicotine being available for re-pyrolysis; this means that a higher delivery of nicotine will be obtained under free-smoking conditions than under restricted smoking conditions. -Pa. C) C) f%~j CD r~1j Cc.' j-4m7 B.,~ T ~L K ami Etpomb L..-,ted Thts mu st not be CAPied or sho- IN to unauthorised penons. BATCo document for Legal Services : Health Canada 20 October 1999 - 16 - NICOTINE DELIVERY PROFILES FOR FLUE-CURED CIGARETTES Nicotine (mgi 0.6. Free concittions] Nicotine 0.5- Restricted condifions~ 0.4 0.3 0.2 0- i ; 1~ 1 4 6 -1 I'l PO number la the presence of cip ventilation or higher permeability paper, we are approaching a free-smoking condition and therefore we get a lower tar-nicotine ratio. Nicotine, which is an alkaline substance, exists in two forms - the bound or salt form and the free-form. However, only, in the free- form is It relatively volatile. In tobacco, it is present mainly, in the salt form. When a cigarette Is smokedv nicotine is released momentarily in the free-form. In this form, nicotine is more readily absorbed through the body tissue. Hence it Is the free nicotine which is associated with 11VACT, i.e. the higher the free nicotine, the higher the IMPACT. If there are acids present, then these will react with the free n1cocine to form salts andv in this form, nicotine Is not very easily absorbed. This means that the greater the amount of acids present, the Zb. CD lower will be the amount of free nicotine which Is finally delivered and 4= r*j zhe lower will be the LOAM 11~0 0\ ()N 14X7 B A r iL K anj E%pon) Ltmact This must not be cnpied or sho.n to wisawit"nsed persons. BATCo document for Legal Services : Health Canada 20 October 1999 - 1 .7 The ac-,d--.-. or akalinity of a material is defined by Its 'PH' value - The pF of pure water if 7.0. If the -,F of a solu-.-o- Is less than 7, the solution Is said to be acidic, i.e. a PH of 1 would Indicate a strong ac4d. If the pH of the solution I'S !,an the greater : solution Is said to alkallue. It should be noted that the scale for pH Is logarithmic: a solution with a PH of 5 is ten times more acidic than another with a PH of 6. When sugars burn, they give rise to acids. Therefore, the higher the sugar 'content of the tobacco, the more acidic the sroke will be. %SUGAR IN LEAF SMOKE pH 20+ 4.5 15 5.0 10 5.0-5.5 5 5.5+ 0 6.0-7.0 40 Air-cured tobaccos and therefore blended cigarettes have lower sugars than flue-cured tobaccos. Thus the smoke from blended cigarettes is less acidic. This, in turn, means that there will be a greater proportion of free nicotine present in the smoke from a blended cigarette, which explains why these types of cigarettes tend to have higher impact than a flue- cuTed cigarette with the same nicotine delivery. P-- C) Finally, It must be pointed out that, along with nicotine, N.) tobacco and tobacco smoke contain small proportions (20;: or less) of C) il-i -j Cc., I y?. 7 a ..k T, t L K sai E ~~,n i L~m.sej This -ust not be emPled or vrio- n io unauthonsed ptr%ons BATCo document for Legal Services : Health Canada 20 October 1999 - is - other comoounds of the nicot-ne a~kalcid fa=i1y. The :out-,ne laboratory method of analysis Is not sensitive enough to distinguish berweezi these individual components. Therefore, %-th a t is measured by the laboratories within and ou-,s-:--=- the tobacco -rdustry is -077.kL NICOTINE ALKA'~LOIDS (TNA ) and for s-Mnlicity this is quoted as 'nicotine'. Briefly curning to the ocher conponents In smoke, typical dellveries of the more common1v measured vapour phase components are: is TYPICAL VAPOUR PHASE DELIVERIES FOR FULL FLAVOURCIGARETTES ( TAR : 15 - 20 mg/cig. ) FLUE-CURED U.S. BLEN CO (Mg) . 10-20 NO (,Ug) 40-90 150-'300 HCN ()Jg) 180-300 ACROLEIN ('Uq) 60-80 ACETALDEHYDE ('Pg) 600-800 AMMONIA (Mg) 20-30 41 Notice firstly that the CO de2-very is in mg per cigarette, the rest In pg per cigarette. Secondly, cote that for most components, we have a similar range of deliveries from flue-cured and US blended products. The one exception is the n1tr-c oxide (NO) delivery. This Is because the nitric oxide delivery Is directly related to the nitrate content of A:b. C:) the tobacco and typically air-cured tobaccos have a high nitrate content C) cocoared with flue cured tobaccos. N) OD ,Z, 19!-7 3 \ T oU X .,,j E,-,,nt Th-s must riot be cnii~eo or stto~n iv on3ulhOnse'd ;-crsons BATCo document for Legal Services: Health Canada 20 October 1999 - 19 - Nitric Oxide Delivery -v- Blend N itrate 300- No delivery (Ag/cig) X 200- X 100- X 0. ~ I . I I I I I 0 2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Blend Nitrate (%) E=p~asis has been placed in this talk on mainstream smoke deliveries, for co=-narison the deliveries of sidestream smoke COMDonents are as follows: SMOKE DELIVERIES: MAINSTREAM v SIDESTREAM MAINSTREAM SIDESTREAM PARTICULATE PHASE TAR (mg) 20 35 NICOTINE (mg) 1.5 4.0 VAPOUR PHASE CO (Mg) 15 55 AMMONIA (Aig) 25 2800 DMN (jug) 0.005 0.25 I v-7 8 AT iL K ani Eip.,m Urnaea This must AM be CmP)cd or sho-m to unauMmsed pe-".s BATCo document for Legal Services : Health Canada 20 October 1999 0 EXPANDED TOBACCO 0, CD r-11) CD C) (C-'. I qu? 9 AT #L K ans E%oa~ri I Urnite~j This must "be c"pwd or shoun io %inawthansed persorts. BATCo document for Legal Services : Health Canada 20 October 1999 -X2.4'"NDED TOBACCO - AN INTRODUCTION :n chis short discussicz paper, we provide an incroduct-4,:a to expanded :~bacco. ';e shall consiz4er four basic questions: 1. Why do we expand cobacco? 2. How do we expand tobacco? 3. What is the effect of expansion on tobacco? 4. What is the effect of incorporating expanded tobacco in cl-garet:es? I Whv do we exDand tobacco? During curing, tobacco loses the vast majority of its voliz=e: !resh leaf at harvest cured leaf Tobacto leaves lose: 10-15Z of their length 25-35Z of their width and 70-80% of their thickness. Compared with fresh or green tobacco on the plant in the field, :he cured tobacco we buy, has only 1/7th of its original volume, due to the collapse of the cell structure of the leaf. The major reason for expanding tobacco is to attempt to restore some of the orig-nal cell structure, thereby regaining some of the "lost" volume and so reducing usage of tobacco in cigarettes. 2. How do we exuand tobacco? 71%ere are two common processes - G.13 and DIET - which produce what is called high order expansion; other processes such as WTS and the use of the Haunl Tunnel which give approximately 10 - 20Z' expansion are termed "low-order". Both high-order processes use C) CD r*J C) rNJ -_10 i9id? a A T it. K zAJ Ei-ni Limitiri This must ACH be Mpled of Sho%n to unauttksmsed pemons BATCo document for Legal Services: Health Canada 20 October 1999 - 2 - the sa=e basic principle: G.113 uses the highly volatile liquid refrigerant Freon and DIET uses liquid carbon dioxide, (CO7). T ,obacco is impregnated with either liquid F-7eon or C02 and subsequently, heated to induce expansion. B.A.T. principally uses DIET and we will discuss this process in more detail here. (Exceptions in the Group include the G.131 plant in Australia and the smaller G.13C units installed in Mex-co and more recently In Singapore. 7he process for DIET (DRY ICE-EXPANDED TOBACCO) can be szu=arised as follows: 1. Cut lamina at 20 - 23% moisture content is put into an impregnation chamber, which is essertially a pressure vessel. 2. Licuid CO-) is Introduced into the cha=ber at a set pressure Cc. 30 atmospheres) and temperature (-78*C). 3. When the cut lamina has absorbed the C02 (mainly during the filling period of about I minute), the excess liquid C02 is allowed to drain off. 4. The pressure is released from the cha=ber, converting the licuid C02 to solid C02 or 'Dry Ice. 5. The tobacco is then introduced to a hot gas stream (BOX steam and 20Z air and C02) at about 320*C which evaporates the dry ice, inside the cell structure, thereby expanding the tobacco. 6. The moisture content of the tobacco at this stage can be as low as 3Z atid the tobacco has, to be re-ordered to about 13.5Z moisture content using very fine droplets of water. -c~b CD C:) CD N) -~-J tu Denon$ rl%) BATCo document for Legal Services: Health Canada 20 October 1999 Haw to expand tobacccO.' - - -- - DIET-Dry lee Expanded.Toba=0 By compar.son, the process for C13 expanded tobacco is as f c-'-ows: 1 Cut lamina, at about 25%, is conveyed to a pressurised impregnator. 2. Freon, vaporised via a hot water heated vaporiser enters the impregnator (60*C). . 0 As the tobacco is cooler than the Freon vapour the vapour condenses and is forced by the pressure (1.35 atmos.) into the tobacco cells. The retention time for impregnation is 15 - 20 minutes. The impregnated tobacco is then introduced into a hot vapour stream (120 - 130*C). As the Freon evaporates rapidly it expands the tobacco. The expanded tobacco is conveyed to a separator where the tobacco is removed from the vapour stream. 6. The tobacco then enters a stripping drum where residual Freon is removed. rQ C> r"i !---sej Thr% must not he cooled or sho- Ito wn2ulwnsed oenons BATCo document for Legal Services : Health Canada 20 October 1999 - 4 - 7. 7he :obacco then encers a cooling and condi:-'oning cyl4nde: w~-.zre it is re-ordered to about 13.5%. N.S. The G!-: process used in Mexico and Singapore follows the same prindip-es but is a small scale batch process, in which izpregnation and ex;E::sion are both carried out in a single pressure vessel. G-13 PROCESS CUT TOBACCOI FEEDER FREON HEATE LIQUID SCREw SUPPLY IMPREGNATOR FREON HEA RECOVERY RAPID STEAM ERI N D I' F, R~E 0 N L P ~ou N E ATED V"Pou D S REW G T ~Y NA OFI~j E 0 V 'IT .AT F 5TR ER ~v = 13'P T [IRECC CARBO BED TRAP 3. What is the effect of the expansion process on tobacco? A-z =entioned earlier, the expansion process attempts to first restore and then retain the collapsed cell structure of the tobacco. C:) r%j 1-.0 --j This must not be cnafedarsha%n X1- BATCo document for Legal Services : Health Canada 20 October 1999 7he illustrations below, show the ef--ec:- V-V - Z>o - "-7qb Transverse section of cured lattina before expansiot. Noce the collapsed ce-11 structure. r -M~k 1% L-4 OL A lic -d! I-ER C> Transverse section of cured lamina after expansion. C) rQ "or e the 'puffed' cell structure. C) I,,-, %'D --Ij rC-; 1408,%T iLk anjE%P~,r-,L-m.iej This Mot not De -,,P,Cdor sho.niounaueftmiec PC-%oni BATCo document for Legal Services: Health Canada 20 October 1999 As the carbon dioxide or Freon vapourises, che tobacco cells are puffed up dramatically and during the si=ultaneous drying stage the 'puffed' cells are set into this new shape. Depending on the conditions of the expansion process, the volumetric expansion of the tobacco can be from 50Z to over 100%. The use of expanded tobacco can most obviously allow reductions in the weight of tobacco used in the cigarette. However, doubling the tobacco's volume does not necessarily, result in 50% weight savings. In manufacture, cigarettes should always be made to equal firmness to ascertain the actual weight savings, but experience has shown that weight savings can be expected to follow the relationship shown below. (Note, G13C tends to give slightly lower weight savings.) Expected average weight saving when including expanded lamina at high levels of incorporation Net Net weigrit saving at eoua, firmness Nk so so 40 30 20 10 0 10 20 30 4 . 0 jo 60 70 80 90160 Content of Ficoandea Lamina (%) C) @ I v- 7 B k T L k n-j E msi,n i Um,ttt This muss *%0( be C"Pled Of so Y113-thonsed persons BATCo document for Legal Services : Health Canada 20 October 1999 - 7 - T 4: f Ihe Dl'.:'- process also a..ecws r.~!e tobacco chemistr-! and in particular, as is s,o-wn below, 3LEEND N"ACOTIME is reduced by 20-30Z and REDUCING SI)Gkv.S are reduced by approximately 20%. Effect on blend nicotine 3.z. Atier 3.0. 2.01 1.0. Attar 0.73 x Before 0.-- 0 0.! 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Before Exciansion Effect on reducing sugars 16, After Exosinsion 14- 12- 10 & :. 6- a .0 4 After = 0.8 x Before Zz. 2 C) '0~ a 2 4 6 a Ib 1'2 1'4 1'6 1'8 20 Before Exr)ansion r*I) '.0 @ I v!j? a AT iL X ani L%p,.n; Lmaiea T%ts must M be cnooed or showw to unauthonsed persons BATCo document for Legal Services: Health Canada 20 October 1999 - a - comparison One C-2 process has a much less signt-licant effect or the tabacco. che=istr7, (of the order of 5 - !GZ nlcotine reduction). The G13C process has virtually no effect on ---:!ot.ne and sugar. W-hat is the effect of incorporatinR DIET exDanded tobacco in cigarettes? If the total lamina blend is expanded using the DIET process and incorporated into the cigarette at different levels, then one finds that: i) VIM DELIVERY reduces proportionally with weight and ii) NICOTINE DELIVERY reduces more than the weight, because of the combir.ation of blend nicotine change and gross weight reduction. T 7his is illustrated below. 0 a x TNA -. P M, M imgt tmg!ctg.) 1,40 22 8801 1.27 20.=-, 800. IJI 1.8.22 700- 0195 1 e.53 600 Walght 'l"PM 0,80 13.07 500. 0.64 10,45 A00. 0.48 Nicotine 0.32 e.22 200 0.16 2.61 100 a 0 06 Exoancedr=accp C) r~J CD r%j --0 -J Co 9, - .. - - a s T - 1 L. 4n -- L i -% - t ~m.i ej This 19 us, not be eno-ed or show n lo unaut -.)n scid pe-sor. s BATCo document for Legal Services : Health Canada 20 October 1999 - 9 - ii4-.' Compounds assoc-,a:ed with !FLRITAT:0'; (eg: acyclic kerores) reduce less than :he weight reductj-on. i,) Some FLAVOUR compounds (e-: pyrazf.=es, pyridines) reduce more I C, than the weight reduction. The :esult of these two factors is that smoke from expanded tobaccos may be perceived as more irritating and blander than smoke from une-panded tobacco. V) Cigarettes containing expanded tobacco burn faster and there is a reduction in PUFF NUMBER, proportional to the weight reduction. However, at a constant total smoke delivery, lower puff numbers result in higher deliveries per puff and hence more taste per puff. Conversely less puffs could be seen as poorer value for money, unless steps to counter this are taken by cigarette design. vi) ROD PRESSURE DROP increases markedly with increasing inclusion of expanded tobacco. This is due to the smaller void fraction of cigarettes containing expanded tobacco, resulting in greater resistance to drawing smoke through the rod. It has been convincingly demonstrated that the inclusion of viil/ expanded tobacco increases IMPACT, that a greater MOUTHFUL of smoke is obtained and the cigarettes have betteroverall 0 SMOKING MECHANICS. The advantages in (vii) and that of puff number/raste (v), are particularly useful in low and ultra low delivery products, as has been demonstrated in commercial brands (mainly in the USA). vii-r)There are several manufacturing considerations to take into account when using expanded tobacco. These are well known C) now and it would be fair to suggest that satisfactory products C) rQ C9 341(7 3 AT I L K ani Etpuni Urniied This must not be cnpicd or shown j%) naug. ~'^)nsed persons BATCo document for Legal Services : Health Canada 20 October 1999 - 10 - can be manufactured without too =ch difficul:-- Spec-fically 1: is wort~: mentioning that DIETI expanded tobacc-- has a lower equiilibriu= moisture concent (emc) than un--:eared tobacco, which means it tends to dry more rapidly chan ordinary tobaccos when 'being handled. There may then be effects or, the packing moisture content, firmness and the perceived irrf-tation in the smoke which require control. In Su-arv, the effects of expansion of tobacco and the inclusion of chat tobacco in cigarettes are: 1. Expansion by the DIE--_ process reduces the nicotine and sugars in t*-e tobacco by 20 - 30%; the G13 process produces a smaller '0Z) decrease in rLicotine and sugars. The C13C method blas a negligible effect. 2. Generally, expanded tobaccos tend to produce a more irritating and -mande: smoke. 3. L*-;;er deliveries can be achieved by the use of expanded tobacco. 4. Comparison of cigarettes containing expanded tobacco with cigarettes w-*:houc expanded tobacco show that: a ) Impact is increased b) A greater mout~-.ful of smoke is obtained c) Cigarettes have better overall smoking mechanics. 5. Use of expanded tobacco leads to a net tobacco saving. The temptation to classify the effects of expanded tobacco as positive and negative should be avoided. For example, at first sight, the reduction of puff number would appear to be a disadvantage. However, if the tar delivery is constant, the product would have much becter C) C:) taste per puff, which is very much a product plus. N) M) -'0 co C) 3 -AT i L X I Limiud Th3S MUSE f4t be Cripled of Sh~j en r%c BATCo document for Legal Services: Health Canada 20 October 1999 - il - ~-erefore, the most objective approach is first to lea.--. a--I understand :he properties o~ expanded tobacco, and then to use these exploit its effects to our advantage. As a final point, in view of the unique physical and che--:4-1 properties of expanded tobacco, the Blender treats it as a rather special grade or 'blending block' and usually has to balance the rest of ::*-e blend to ensure good smoking quality. This is obviously particularly true at hish leveis of inclusion. 0 0 J> CD CD r%-) CD r%j ~10 Co This Muss not be cnvied or shown io wisautlumset! pe- ."s BATCo document for Legal Services : Health Canada 20 October 1999 0 CIGARETTE DESIGN 0 CD r,13 ~.D rS- IYP7 El A T I L X ani Liportl Urnat,- This must M be C"Pwd orshown to unauthonsed persons. BATCo document for Legal Services : Health Canada 20 October 1999 AN OVERVIEW !\--RODZ:C--ION 'bed as a four component A c-~garzette can convenf-enti! be descri syszel=: i. Cigarette paper 2. Filter 3. Tip ventilation 4. Blend. in order to meet the defined market brief, these are the four components wh.ich can be modified. in this paper we will consider what changes can be mads in the first three co-mponents and the effects these changes have on the deliveries. It is extremely important to appreciate how these four components may be brought together to end up with a product or products which are likely to satisfy both the subjec:.Ve and the objective requirements given in the brief. 1. CI=ETT:-: PAPER Cigarette paper is not just simply the white wrapping material used to hold the tobacco in the cigarette. By virtue of its physical and chemical characteris tics, it has a fundamental influence upon the quality, the appearance amd the combustion of the cigarettes. Most paper"s are based on flax, hemp and woodpulp fibres and contain 20-30% chalk. The chalk serves to increase the opacity of the papers and to cause formation of an attractive white paper ash. Along with the chalk, salts are added to the paper for three main reasons, viz: a) To minimise the tendency of cigarette paper to discolour during storage: r".) r~j -.10 (c) I 1#0 8 AT iL K soi E%pon I Liensirts This must nOt be CoPied of 00-n to unauthonscd 2.crsons. BATCo document for Legal Services : Health Canada 20 October 1999 - 2 - b) To c:-.trol ash for-matIon. c) 77o modif-, the burn rate of the cigarette which in turn can Influence the _:eli-veries to a limited extent. This will be ccnsidered in 11 late. in I..-le DaDer. Me property of the cigarette paper that has the most sign-Ificant effect on deliveries is its FER~CABILITY which is usually specified in. COREST.A. Units. The CGRESTA Unit is defined as :he flow of air in =3/min through a sample of paper 1=2 in area when a pressure diz-ference of 10cm water is appl.ed across the paper. The more porous the paper is, the higher will be the air permeability as =easured in CORESTA Units. The desired level of air permeability for the paper may be ieved in two raYs., -~ 1. By =aki=g the paper inherent-, y porous during its manufacture. This is normally referred to as the Natural Permeability (N?) and the normal commercially available papers cover a permeab.lity range from 8 to 200 CORESTA Units. 2. By passing an inherently porous paper through a secocdary process where devices puncture, buru or cut holes in the paper. These holes tend to be considerably larger than the pores of the NP papers. The most common method of introducing holes is to use an Electrostatic Spark Perforator and hence these papers are referred to as EP papers. one general effect of increasing the permeability of the paper (N-P and/or E?' is to reduce the weight of tobacco burnt during the puff. The reason for this being that as a fixed volume of air is drawn through the cigarette, higher permeability paper will let in more air through ~10 CD 4-1 Œ~ i~-*, 8 A T j U X &Ai E%p~iftt Lzmurd T~ss musi nct tK' CnPed or sho-n tu unaahonsec BATCo document for Legal Services : Health Canada 20 October 1999 -he paper envelope ard thus a lower volu=e of a-r will pass through the burning cca-*. 7he effec--s of increasing perzeability, and these may be su=ar-4sed as fo-lows: 1. Sta--ic burn rate increases as we increase N? from verv low to about 70 C.U. a.-- thereafter :~e changes are small. A possible reason being that as NF is increased, more oxygen becomes available for efficient combustion but at around 70 C.U. enough oxygen is available and hence there is very limited effect thereafter. 2. Introducing EP to a base paper does not alter the stat~c burn rate. STATIC BURN RATE (SBR) VERSUS PERMEABILMY S.. PERF~,;ATED A-24- 1 38 0 20 40 60 80 *00 120 140 160 IaO 10 PERMEASIUTYI=AESTA UNrM 3. Increasing NP up to about 70 C.C. results in reduced puff number because although a lower weight of tobacco is consumed during the puffs, the weight of tobacco consumed between puffs increases more rapidly as illustrated in the Slide. Beyond this level, there is no C) C) change in SBR but the weight of tobacco consumed during the puff N) CD reduces and hence a slight increase in puff number of observed. M) 00 U1. 0 1,4x7 8 A T i L X ana E Wni Thn mus, 1109 bc ;Dpoccj ot $h%)~n C'… >t ,Ons, BATCo document for Legal Services : Health Canada 20 October 1999 - I - :n::oduc-,n;z -_? does not alter the SBR but reduces the weight o4- 4" and hence increases the puf.- number. t:.-acco consumed per puf~ 91L:~-- NUMBER VERSUS PERMEABILITY ~:;R CITRATED CIGARETTE PAPER PUFF NUMBER IZO' 10.5* 10.0. 9.5- 9-0- io 0 40 6~ 80 160 lio 1~0 160 150 PERMEABILITY (CORESTA UNrrS) 5. -,=creasing NP reduces deliver4es with a greater reduction being cbserved for CD, followed by tar and nicotine. Thus, tar to nicot-4ne :a:io will reduce. EFFEECT OF CHANGING THE NATURAL PERMEABILITY OF CIGARE-jj E PAPER ON SMOKE DELIVERIES REDUCTICK 0 Ml i FSON 1 104 30, Jtb a 20 .0 so 80 120 140 160'no PERMEABILITY (CORESTA UNITS) 00 01% @ 1-7 8 A T i L k ar4 Eitvwts L-on,Led This must not be C004ed or ShO-i'l q0 unaudkMsed penons BATCo'docurnent for Legal Services : Health Canada 20 October 1999 6. S-=-Iar effec:--- occur from increasiig r-? but the e-f-leczs are .1ower t~an for %"P because whereas NP affects both SBR and level of di:uz-~on E? only af-ects the level of dilu:-on. EFFECT OF ELECTROST ATIC PERFORATION OF CIGARETTE PAPER ON SMOKE DELIVERIES REDUCTION 0.1 .0. 301 d5lf 20- BASE PAPER NICOTINE 0 id 46 Go so 100 lio )io 16040 PERMEABILITY (CORESTA UNITS) Turning now to the additives, the following chemicals in the range 0.5-3t are in common use: .Monoammonium Phosphate is added as an ash conditioner and only slightly increases the burn rate of the cigarette compared to paper with no additives which are the slowest burning papers. Disodium Hydrogen Phosphate increases the burn rate of the cigarette relative to that found from using the monoammonium phosphate additive. Sodium and Potassium Citrates are most commonly used as burn accelerators. Lhis reduces tar delivery per cigarette by reducing the puff number as a result of more tobacco being consumed in between puffs. However, tar CD r".) delivery per puff should not be significantly affected. -10 CC) Cc., 1 yx? 9 jk T t L X &ni E ir-rT 1 LimiteJ This mu st not bc cnpsed or %hO*ft ®,) pe.%ons BATC6 document for Legal Services : Health Canada 20 October 1999 - 6 - In terms of the effect of these additives on burn ra:e, the rank Drder is: chemical free (no additive) - slowest burning < zczo-ammonium Dhosohate < diso4i-um hydrogen phosphate citrate - faszas: burning. Both Phcs--hate and Ciltlrace treated papers produce better ash .ormation than papers coatain-ing no additives. 2. FILIYER The most important role of the filter is to remove a desired proportion of certain S=oke constituents. Like the smoke aerosol wh~ch may be classified into parziculate phase and vapour phase, it -s convenient to divide filters intcy two groups - those which mainly remove smoke particles and those which are effective in removing constitue=zs from the vapour phase. A r4gid sub-div~sioa is not possible since many smoke constituents (semi-volat-4les) are present in both phases and it is possible to design filters which are effective for both phases of smoke. In this paper we will concentrate on filters for the partic-alare phase. Most cigarette filters which remove particulate phase consist of fibrous or fila--entary materials which capture the smoke particles as they pass through. over 90'. of the filters used commerc-i'ally are cellulose acetate, with the remainder being paper and viscose. Further, much of the paper used goes into dual filters with the paper section next to the tobacco and the acetate section at the mouth end. The effectiveness of a filter is measured by its filtration efficiency (FE). FILTRATION EFFICIENCY (FE) Definition : The amount of smoke or a material in smoke (e.g. TPM, Nicotine etc) which is retained on a filter plug under standard smoking condition expressed as a percentage C) rQ of the total amount of smoke or the material in smoke presented to the filter. CC) @ 1-4-117 B AT fL K &~j Ev-,r., Lmttj Thm rnw~t nol be cop-ed or shu.n w nawerinsed BATCo document for Legal Services : Health Canada 20 October 1999 S., S2 FE M , Armc~;nt retalned by lht filter X 100 Arnoumt presented to the filter SI - S2 X 100 S1 z Various filtration studies have showt that, in general, .iltral,--on efficiencies are intimately related to the pressure drops in the sense that most factors which increase pressure drop also increase filtration efficiencies. Thus, for example, increasing the length will increase both pressure drop and F---. However, the absolute increase in pressure drop and FE will not be identical. This is illustrated below: EFFECT OF LENGTH ON PRESSURE DROP AND FE F-G- L*ngth : 15fflfft . PO - 4 cm WG. FE :30% SI--- -s' Io" 3.9 c~:1 7eng la 10 7 a ioo : 30 % a- Langth = 30MM FE i 15 Is~ . 0% ; 30% L 0. _~Fœ FE Si - 3M9 2.1ms - S2 PO a C- W.G. FEM 0- 4.9 00 a 51% 10 _A:b C;I N-) N10 cc CC' W-7 8 A T f L K s-j Emxrti L,m.ttj This mus: nm be Cnoted or sho.n IJ uAAuIhOnsed persons ~10 BATCo document for Legal Services : Health Canada 20 October 1999 - 8 7'1 e r elaz: ionsh- p be tween filter pressure drop and filtraz~on e"4c~ f or CA f 4enc,.- ilters is shown below. Note that at the same pressure drop, increasing the length increases the F-E. Figher FVs for the same .:'essL:re drop can 5e achieved by using paper filters as ~.s also shown below. A-1though paper filters were very common some years ago, nowadays consu=ers appear to find cellulose acetate filters more acceptable. EFFECT OF FILTER PRESSURE DROP AND LENGTH ON FILTER EFFICIENCY FILTRATION EFFICIENCY so- PAPER FILTERS 2Smm 70- 5MM 60- 50 25MM 40. 15mrn 30. ERS 20- 10- 0 FILTER PRESSURE DROP (Cin W.G.) In both cases, CA and paper filters, the FE for nicotine is lower than chat f or TFM - see below. The difference is largely due to the fact that a substantial proportion of nicotine deposited on the filter during the smoking process is eluted from it. For US cigarettes figures of 19Z for nicotine versus 4% for TFM have been quoted. C:) M) CD .10 N-D C) i:c; 1 --7 8.~k T t L K ani E%.,-m i Lamm! This must ROI be eno-ed of sho-n to unavdkx-.sed:)e-so,.s BATCo document for Legal Services: Health Canada 20 October 1999 - 9 - FILTRATION EFFICIENCY - TPM & NICOTINE 25mm CA FiLTER FE TPM 40. NICOTINE 30. 20 10- 2 3 4 5 6 7 FILTER PRESSUkE DROP CM W.C. Some brands also use a dual -filter which consiszs of a length of filter next to the tobacco (referred to as the 'T' se::-on), f ol-lc-.--ed by a length of cellulose acetate filter at the mouch., end ':r the 'IM' se::ion). This filter combines the advantages of highly efffeczive paper ff-Iters with the acceptable white end appearance of ce-Ulose aceta:e filters. The filtration efficiency of each sec:ion :-in be adjus-zed to particular requirements and following the logic illust---=:ed in Slide i it can be shown that the combined efficiency is: dual . Fr paper + (100 - FEpaper ) FECA 100 where ---- is the percentage filtration efficiency for the par:icular secti= used in the dual filter. Note that none of the filters discussed above constf:-aents from the vapour phase of smoke. A different type of ---lrer ro --.0 ~140 ig., t--' 3 AT -L K jew Etp-,Mj L,M,ird This must not be ce)poed or sho.n io vnjvtN)nsce wsons. BATCo document for Legal Services : Health Canada 20 October 1999' is recuirec~ zo remove consfvuten:s from the vapour phase of smoke. - com-mon ap=roach is to -'=_ude an adsorbent in the filter. The mos-, effective ad~sorbent is ac:--;ated carbon which has been used in some ciga-et:e "---ers for =a-.- years. The amount of 'gas' phase removed depends on the weight o: carbon in the filter and its activity. During the manufacturing process the carbor is activated, which means that it can adsorb gas molecules onto its surrace. Ion exchange resins, such as Duolite, remove specific chemical sroups such as aldehydes. Although carbor filters are popular in some parts of the world, e.g. Venezuela and Switzer-and, they are also associated with a change in taste which is -not generally liked. The 4-on exchange resins and carbot both reduce irritation izz the smoke but are comparatively expens.ve to assemble and manufacture. 3. TIF VEN-L-ILATION Ventilation, in simple terms, is the process of introducing holes in the wrapper of a cigarette. The holes allow air to enter the mainstrea= s=oke when the cigarette is smoked. The air thus enter.ng the cigarette is not used to burn the tobacco. Hence, less tobacco will be burnt during each pulf and the delivery of smoke reduced. Tip ventilation, more specifically, is the amount of air entering through the tipping/plugwrap combination and is normally expressed as a percentage of the total volume drawn through the cigarette. Me relationship between percentage tip ventilation and delivery reduction for the most co=on smoke constituents measured is: C) r%j CD rl".) -10 ic:. y. 1 9 % 'r it: K anj Eii>.n … L.mwd Thn musc no( be enobed or shoun lu un auchansed PC mons BATCo document for Legal Services : Health Canada 20 October 1999 OVERALL- REDUCTION 100r 80r I 60 r 20~ 0 20 ~0 60. VENTILATION ('~z) - UNLIT CIGARETTES The line for is an approximate 1:1 relationship, --.e. percenzaze tip ventilation approx-IMately equals percentage reductic= o: ?,TWN-T. ---.-r nicotine the reduction is less at any particular leve- of ventila:~-zn 'relative to tar and therefore as tip ventilation is increased the tar :j nicotine ratio reduces. The reduction for CO is greater zzhan for nico-.4-ne and the P.MWN,-r at any particular level of ventilation. :.his occurs d=e to the increased residence time of the smoke withir, the cigarette rod envelope allowing increased diffusion to occur. At ventilation levels greater than 55% the increased diffusion tends -.~ be offset '--T the increasing puff number. X-- C::;` C.,:> rQ %.0 iw-~,3A T jL K &niE%r-,MoL,rn,icj This Must not be enpwdof Shaun tounauthansed penons %D L.14 BATCo document for Legal ServiCGS : Health Canada 20 October 1999 PUFF NUMBER VERSUS VENTILATION PUFF NUMBER 0.81 O.G. 0.2, 12.0- 0.8. O.G. 0.4. 0.2- II.Q. 0.8 00000'~ 10 10 20 30 46 sia Go 70 so 9 - 0100 VENTILAnom % Additional effects of tip ventilation you need to be aware of are: a) Venti2ation of the fIlter will produce a change in the stain due tc tar deposition. This results in a concentrated deposit tend-Inz towards the centre of the filter ('Bullseye' effect). b) The unbound (perceived) cigarette pressure drop will be reduced due to tip ventilation. it is, of course, possible to compensate for this and the following rule of thumb may be used as a rough guideline: For each 12 (absolute) increase in tip ventilation, the total unbound cigarette pressure drop will be reduced by Imm. WG. C) Tip ventilation between 10% and 60% has virtually no effect on filtration efficiency for tar and nicotine when cellulose acetate filters are used. Above 60Z tip ventilation the filtration efficiency shows a slight increase. C) rIIj CD -P-- 0C 11#03AT tLK aniEtporteLsmitea This must M be cnpied or shown t,) unarith,)nice BATCo document for Legal Services Health Canada 20 October 1999 - 13 - The des-i-edd 'Ieve- of tiz venc-larion can be ac~ieved -~n two main ways: I Using natura-ly porous or pre-perforated tf?ping in co=_4unc---'on with Por:!US pIuz%;==_-:. In --h4s 'nstance a series of diffa:,ent rypes of tipping papers (see below), some with visible and somme with non- visible holes, are available. There are also a series o.- gluing techniques available to combine the ventilated tipping and plugwraps and the technique adopted can affect the --'--'nal level azd var-ability of the filter ventilation. 2. To construc: cigarettes with non-venzilated materials and to subsequently introduce ventilation on the assemble6 product. A simple means of doing this is to use a pin block method. This is readily available on Rauni tipping artac'--ments where. the number of pins (4-8), diameter and the ~setting can be varied. This method may seem rather unprofessional and lacking in control but it has been used extensively on successful brands, e.g. MERIT - USA (now appear to be converting to pre-perforated tipping). More recently, Raun~ have developed on-=achine perforation using a laser. The filter tips on completed cigarettes are perforated by a laser beam built into the tipping attachment. This method of ventilation can cover a range of 10A'-80Z with reduced variability compared with cigarettes produced from pre-perforated materials. -:-Ib CD CD ;%I~ Cc; i~!-7 8 .% T iL K ,.… Eiponi L.m.~ru Thil Mu%t not be cnpbed or sroun tu unawtt%an%ec penons BATCo document for Legal Services : Health Canada 20 October 1999 - 14, - METHODS OF VENTILATING CIGARETTES CIGARETTE PAPER 1. Naturally Porous 2. Electrostatica!lv perforated 3. Pin hofes 4. Combinations TIPPING 4with ventilated plugwrapl 1. Naturally porous 2. Electrostatically perforated 3. Mechanically perforated Kcro 4. Mechanically perforated Macro 5. Micro perforated laser B. Macro perloraled laser ON MACHINE VEIWILATION OF COMPLETED CIGARETTES Laser >Hauni Needles fElec-ostatic perforation of tipping on machine) The use of Lip rentilation is increasing world-wide and can now be regarded as a normal cigarette design feature. It is in use on commercial cigarettes covering a range o~6 0.1-17.0mg of tar. 0 _z~b CD CD K> 11.0 CCI IYY 7 8 A T and E %>,m 3 L.m.:c.; Thss musc nx te ,ped Cr BATCo document for Legal Services: Health Canada 20 October 1999 4 SMOKE SENSORY TESTING 0 CD CD r%) CD Ccj J Y K7 BA T it K arj Ei port … Lamitti Thts musi not t~r coped or st0~,% te unajt., .,arssed Pe, .sons %M I-J, BATCo document for Legal Services : Health Canada 20 October 1999 SUBJECTIVE ?RODUC"7 ASSESS]'ENT AND SCRZENING In considering subjective product assessment we are confronted with a cake which can be cut in se-7eral ways. Methods available can be broadly qualitat~ve or quantitative, in-house or external, use consumers or "experts", and so on. Each procedure has its strengths and weaknesses and none in itself provides a totally reliable indication of a product's ultimate success or failure. The aim of those engaged in developing procedures should be to equip the user with an appropriate repertoire of techniques. It is the responsibility of the product developer or mp-rket researcher to select the best method for his particular purpose, having regard to the limitations attaching to each procedure. With changing market circumstances there is a clear imperative to anticipate those developments which render existing techniques impotent, or at least restrict their utility, and to provide alternative ways of dealing with new problems. By way of example, consider responses made to the advent of low delivery products and the rapidly emerging need to expand our repertoire to enable valid and reliable subjective assessments of cigarettes in this sector. The 1400DROSE method was the first technique which achieved more or less universal exposure throughout the GrouD. It was intended to provide a common basis for product evaluation and was developed and promulgated to satisfy the demands of the middle/high tar sector. It is a method which relies on specific training of in-house personnel, is for same attributes related to reference standards, and is an overtly formal system. Properly applied it ran provide useful information for the detailed character isation of those products appropriate to it. However, it was never intended to deal with products outside its range of influence and was found to be severely lacking when 10 confronted by the need to evaluate low-tar cigarettes. That said, it remains a useful procedure and one widely employed in the assessment of middle/high tar products. Whether it is ezz)loyed fully, sufficiently, or enthusiastically is another matter. For specific evaluation of low delivery products, two approaches will be discussed. That developed in Canada is a highly qualitative and complex X~b procedure relying on extensive training but able to provide very precise C) C) characterisat ions of own and other's products. It Is specifically oriented rQ to the needs of product development and is generally acknowledged to be a CD N) N.0 %.0 co la i- IVIC7 6 A T iU K aiti Eixi-~j Limstri This Phast - be 1:91ped or sho.n lo unsutbonscd perW$ BATCo document for Legal Services ; Health Canada 20 October 1999 very successful approach. The DELTA method, developed at GR&DC, is a quantitative procedure, simple in operation although relying on computer analvsis, which, rather thaa yielding precise charact e r ~sat tons, gives relative profiles of a range of products and estimates of overall similarities between a range of products. As such its utility is demonstrated in markets where a low delivery segment exists and the information which it provides can be additionally valuable to market researchers interested in "mapping" products in terms of sensory characteristics. It is our central thesis that no one method is, or will be, available to answer all the pertinent questions asked of a product by product developers or market researchers. The aim should be to provide a basket of techniques which are complementary and which together yield the information required to assist decision making, With this in mind, a number of further problem areas have been identified. 11ese include: the need to develop a standards- based (WOODROSE type) procedure for low-delivery product assessment applicable in circumstances where an established mild sector does not exist: more detailed characzerisaciorn of the taste and aroma features of low-tar products: and the peculiar features of ultra-low delivery cigarettes. Whatever its basis - qualitative or quantitative, "expert" or consumer, any method for product assessment demands a number of individuals able and willing to be trained and subsequently available to undertake the evaluations required. The development of a procedure is usually straightforward: its implementation is almost'alwa7s more problematic. Acknowledgement has to be made of the act that equipping an operating company with reliable methods requires an investment, by that company, of time and manpower resources. In many operating circumstances it is difficult to allow sufficient effort to be directed towards the establishment of one, let alone several, panel pro- 40 cedures without affecting the normal staff work patterns. In order to achieve reliability in its evaluations, any quantitative method requires at least ten members. Qualitative procedures may need fewer panellists but their individual training will take considerably longer. For any method it is useful, perhaps essential, to have back-up personnel in order not to inhibit panel operation during members' absences. A panel leader is also required to train panellists and to co-ordinate their subsequent 4~11 activities. Clearly, these necessary conditions can impose considerable C) strain on the small operating company, particularly if we pursue our CD N) objective of offering a repertoire of procedures which may require more CD than one panel. 1\3 %10 __0 %.D rg I Yx7 8 AT ;L K ani E%ppnt Urnited This must not ~e ctitsved Of shos" to unatothanstd penons BATCo document for Legal Services : Health Canada 20 October 1999 -3- One wav of amellorat'Ing the effect of panel operations on the dav-to- day activittles of the comrair Is to consider the introduction of panel methods employing consumers. We have relatively little experience in this regard although our colleagues in Louisville have for some time involved 11 captive" consumers in a variety of product development exercises. In Southampton we are currently engaged In training a consumer DELTA panel and so far the results are encouraging. Recognising the pressures existing on manpower resource allocation, we should, I believe, give more consider- ation to the involvement of consumers in panel activities. The advantages of employing non-employees are obvious; but so are the disadvantages. Individuals are not so immediately or predictably available, on-site accommodation may not be appropriate and therefore an alternative location may have to be found: and we may have to expect moderate attrition rates. Nevertheless, all things considered, the nett cost may be less than using in-house personnel. For many situations this may prove to be the only viable means of satisfying two apparently conflicting requirements: the need to undertake reliable subjective assessment and commitment of sufficient staff manpower to achieve a viable panel activity. In short, more attention may have to be paid to populating the repertoire of techniques with procedures appropriate to this middle ground between traditional in-house panel methods and the normal consumer product tests. Whether in-house or external, any panel will only function satisfactorily if sufficient attention is paid to maintaining a high level of individual and corporate motivation. Experience suggests that monetary inducements are not sufficient - and for in-house panels are probably inappropriate. Much more important is the individual's perception of worth in the specific activity. By reinforcing his status as a valued contributor to an important function and by maintaining his involvement through discussion of the project, the risk of losing, or perhaps worse, disaffecting, trained personnel can be significantlv reduced. This requirement places a considerable onus on the panel leader, who should be selected on the basis of his ability to deal with panel members sympathetically and in ways which sustain a high personal commitment to the panel operation. Trained panellists, whether staff or consumers, are a valuable asset whose worth must be acknowledged, and whose activities supported, at the highest level. CD r\) C:) (.A CD C) 0C I Yv? R AT tL K ani ktpwti Limej This MuM nix be copied or sho~n to unautWnsed pertcfts CD BATCo document for Legal Services: Health Canada 20 October 1999 I _4_ f panel proceduzes appropr'ate deve~.opment o- - to cons=ers =ay also be one way of reconciling the dIfferences in output from traditional in-house 1. expert" panels and consumer tests- ':he major impediment in this area derives from the differential use of language employed by "experts" and constr--ers. Somewhat provocatively, one might suggest that the "expert" does not know vhat the consumer likes but can describe the experience very well, whereas the consumer does know what he likes but cannot describe the experience very well! One feature of future evaluation techniques which ought to be borne in mind is the need to facilitate the product developer's understanding of what the consumer is telling him. It is quite possible to use consumer vocabulary as the basis of a panel method. Consensus amongst consumers regarding (a) the most useful parameters of discrimination, and (b) the particular word to be employed to describe that parameter can, through careful elicitation procedures, be reliably achieved. By using consumer-based terminology the training of the naive panel is facilitated because the burden of inculcating the members with descriptors and terms outside their own ir=ediate experience is removed. The result Is a highly reliable operation based upon an albeit limited number of parameters. The penalty to be paid is the lack of detailed information on the more subtle features of the product(s) which the developer may require to effect appro- priate modifications. Again we have to trade-off the benefits of panel reliability following short training against the lack of fine detail in its assessments. This in turn argues for a range of complementary procedures which again unfortunately Imposes higher demands on manpower resources. Once more we enter the vicious circle described by the need to achieve a high standard of product evaluation at zzero cost. In order to maximise the effectiveness of subjective product assess- ment methods, it Is probably necessary to re-examine the purpose of such procedures, to objectify their aims and, if possible, to quantify their effectiveness. It could be argued that, given a sufficiently precise and reliable range of internal methods, traditional consumer product tests could assume a lesser Importance In the product development /marketing process. Whether or not confidence In such methods could ever be suffi- cient to totally remove, or even drastically reduce, the need for consumer product testing is debatable. It is. however, possible to contemplate some re-distribution of resources, more towards panel methods, particularly consumer panel methods, and away from increasingly expensive field trials. C) r1J C:) U4 rc) sy-*7 8 A T iL X mo E,:.-mi L.mnej This mw%t riot be copied or $has" it) Unluthansed per%ofts, CD CD h 1999 BATCo document for Legal Services : Health Canada 20 October -5- Wh3tever r:Lx of interna:/exte=al evaluat-'on schemes is arrived at we must endeavour to ensure t~zt the most cost-effective and reliable techniques are encouraged in order to proride the best advice for competitive product development. 0 4~!¯ C> c~> 0 1'0'7 B A, T 'U K …nJ Ei>,Ms Li mur-' This m vil ffl be cmpted or ~h0~r,, 1, "UtpujIsed C) N) BATCo docurnent for Legal Services : Health Canada 20 October 1999 O~EGA -- AIN-D '11 INTERNAL PAYEL -nLS---SSMENT Be-fore going into detail about the sensory evaluation process the first question that needs answering is "wby do we do it when we have ob ective, chemical analysis available to us?". 0 OHP 1 - SI1,TLE ANALYSIS: TWO CIGARZ:TES OBJECTIVELY COYff'ARED PM14h1F mg/cig 9.7 9.7 Nicotine mg/cig 0.88 0.87 C.O. mg/cig 11.6 12.2 Ventilacion Z 18.1 17.7 Z Nicotine in Tobacco 2.17 2.13 Z Reducing Sugars in Tobacco 10.0 9.7 X Total Sugars in Tobacco 12.7 11.7 Manufacturer GALL&HERS R.J. REYNOLDS Brand B&H SILK CUT11I.K. VANTAGE/U.S. 0 Looking at the above slide the results of a simple chemical analysis on two cigarettes, A and B, can be seen. From these results it would be thought that the two cigarettes were very similar indeed. However, the products A and B are quite dissimilar with one being a U.K. flue cured product and the other a U.S. blended product. The second question that may be asked is "where does internal panel assessment fit in?" C) CD 14-7 B --k T ;L: K xnJ Etr-ml Lir--trj This r"ust ncx be Copied or Shown 1%3 unawtMmsed persons BATCo document for Legal Services : Health Canada 20 October 1999 0F?_-2 - -=- POSITION OF SU3J---C71VE ASSESSM-rN-L PAINELS EXPERT PAYEL(S) F PRODUC."L DEV----DPMENLT MARaTING CONSMCER L From this slide it can be seen that the large box contains the world of the tobacco industry. The consumer lies outside the large box in the real world and it is the job of the marketing man to interpret what the consumers needs are. Having done thiSx the product developers will then attempt to develop products which meet these needs. Their success, or otherwise, is assessed using an expert panel. The results of this assessment are then passed back through to marketing/ product developers for i=ter- pretation. 0 Sensory evaluation is a process and as such it involves several stages. It is Important to be aware that because of the intervening stages things are often lost in the translation. It is a little like translating a poem from French to English, some of the meaning is lost in the translation. If the translations are continued from one language to another then by the time the last translation is made much of the meaning is lost. In sensory evaluation one should be aware of each of the stages and their associated problems. N) U4 CD C) Cc! 1%iOSAT jLK jniE%L-niL.m*ieJ This 'bust not bc cnptcdor shoun wonsuawnsed t::- BATCo document for Legal Services : Health Canada 20 October 1999 L-- 13 OF--- 3 - CO\7ROL PROBLEMS IN SENSORY EVALUATION I CONTROL PROBLEM I I TRANS~`ORX!-- 1ONS COMPLEX PRODUCT PHYSICO~ VARLkBILITY CHEMICAL EN'rITY SYSM NOISE PANEL SMOKER LANG1jAG--E SYSTEM STATISTICAL LANGUAGE CHARACTERISTICS DATA ANALYSIS JUDGEMENTAL REPORT LANGUAGE CRI'-"-RIA JUDGEMENTAL PRODUCT DEV./ LANGUAGE SELECTIVITY MARICETTNG INTERPRETATION INADEQUATE BRAND LANGUAGE 5 UTILISATION IMAGE OF iLNOWLEDGE INADEQUATE CONSUMER LANGUAGE 6 KNOWLEDGE C) "What. apparatus is needed to perform sensory evaluation?" C) rQ CD (Jr. j. w17 8 A T tL )~ anJ F.-ml This must not be ce*Kd or ss,)un 1,) wnawtWn,rd Vem BATCo, document for Legal Services : Health Canada 20 October 1999 14 OHP L --7:--- 'HEAD Nasal ca ity Teelh The head:- All the panellist needs to evaluate a product is his head and in particular certain parts of the head i.e. a brain. This is where interpretation of the sensory experience takes place. However, in the CD experience itself, the mouth and nose are of prime importance because TIJ the nerves stimulated are located here. CD tA CD #F' I Y*7 B A T I L k 3mJ Eir-1 Lsmsiej This must n0t be ceawd or sha~n tu una,jNjn,e'j CD (N BATCoAocurnent for Legal Services : Health Canada 20 October 1999 J, z --- - CF---~UCAL SENSES 1WOLVED IN SENSORY EVALUA-TION Or: :!,--s slide the chemical senses stimulated during sensory evaluat-on of zhe product are shown, the location of these receptors, the nerves t-ese receptors are attached to, and, lastly, the type of sensation sz:~=ullation of these nerves is likely to produce. From the slide it can De seen that the prime sensorial experiences produced by smoking are assoc-*ated with flavour and irritation. NAIE TIPE/LOCATION NERVE SENSATION COMECTION ;.rue" Taste Taste buds Gustatory Sweet (Custation) tongue Salt Sour Bitter Smell Special. tells Olfactory Odours (Olfaction) in nose Aromas Common Free ner-ve Trigeminal Irritation chemical endings in (various types) sense mouth,aose etc. I I- -- I I 1.-7 Lb A T 1L K a,,j E%r-ni L-m-,ej Thit Must not be copied or sho.n lo un3utti,-., BATCo document for Legal Services : Health Canada 20 October 1999 6 - SENSATIONS PRODUCED 3T SMOKING FLAVOUR TAST E + AROMA L !RRITAN'T IMPACT 1FRITATION SENSATIONS + 0 taste and aroma. Irritant he lavour experience is a com!~_nation oZ sensations, on the other hand, are a combination of both impact and 4rritarion. Impact is described as the immediate effect, upon inhaling, in the back of the throat/top of the chest known as catch or hit. Irritation zs the lingering, tingling sensation experienced in the throat, nose and =outh after inhaling. -here are various types of evaluation techniques available to us in the group. The traditional technique is known as Woodrose which was developed about 25 years ago. This technique is an absolute technique (i.e. cigarettes can be evaluated on a one-off basis) where the cigarettes can be evaluated with reference to standards and_exemplars. The Woodrose method was originally used to produce a detailed profile of a cigarette's Irritant and flavour characteristics and commercial cigarettes were used as exemplars of cigarette attributes. Tmpact was dealt with using specially manufactured standard cigarettes designed to produce a range of 5 levels of impact with subjectively equal increments between them. ~his technique is now to be replaced by an updated evaluation method which is designed to deal more effectively with the current trend towards lower deliVery products. This has beer. done; firstly, by expanding the attribute base on which the cigarettes are evaluated; and secondly, by providing a greater range of standards which have been designed specifically to de=onstrate controlled levels of a given attribute. C) CD IS: 1-40 B A T , L K ani E %r%,n) Lamm.: This inusl WC be cnpocd of sho.n Io unauthonscd persons., L14 C) C) co BATCo document for Legal Services : Health Canada 20 October 1999 /7 omega is currently being instituted in the following territoria: networks. OHP 7-- TERR14-ORIAL NETWORKS AFRICA FAR EAST/INTDIAN LATIN AMERICA EUROPE SUB-CONTINENT Sierra Leone Malaysia Panama Belgium Ghana Bangladesh Argentina Canary Islands Eong Kong Brazil Cyprus Y.--lawi India Chile Denmark Indonesia Costa Rica Finland Nigeria New Zealand El Salvador Germany Zaire Pakistan Guatemala Holland Zimbabwe Singapore Honduras Rep of S 1--rica Sri Lanka Mexico Switzerland Venezuela Trinidad 9 By the end of 1984 Omega would have been instituted as the predo=inant evaluative technique used throughout the group. A series of taste and flavour workshops have been set up to achieve this. CD CD r,~-J CD Lbi Ci Cc. jwc? B-%T it- K ana Ei‡xffleLimalr&J Ths'¯ mysinmbecnPiedorsboumiownauch-jm%cdpc.-Cns C) %10 BATCo document for Legal Services: Health Canada 20 October 1999 /8 The vocabulary structure used in Omega is detailed in the follca;,;ing slide. OF2 8 - VOCABULARY STRUC-,UU Mechanics irritations Impact Amplitude Top Notes Modifiers Base Notes Supplet:entary Flavours Aftertastes Visual/Tactile Elements The cigarettes are evaluated using descriptors designed to characterise the product in terms off these attributes. A more detailed look at the vocabulary structure is seen in: 10 4:::. C) C) U4 CD CC) 14~7 B A T L K anas E%t-ns Limtcz Thai must not be ceplcd or sho~n ju u,,,,utK,,.,sce -Wns . P',, (Z) BATCo document for Legal Services : Health Canada 20 October 1999 19 O'~-~ 9 "OCAB!~; ARY DETAIL Draw Resistance Draw Bffrt CH Mou'n4:u UkIN I C S - - 11 of Smo1ke Ease of Lighting Bur= Rate Mouth Irritation TFLRITATITONS Nose Irritation Throat Irritation "ACT Impact A~LITUDE Amplitude Flue Cured TOP NOTES Air Cured Or4ental Fire Cured Green/Grassy 140DIFIERS Musty Earthy Tarry 5ASE NOTES Roasted/Toasted Bur-at SUPPLEMENTIARY Supplementary Flavour (15 item checklist) 7--AVOUR Mouth Coating Mouth/Throat Drying &FTIERTASTES Residual Throat Roughness Aftertaste Amount Aftertaste Quality Packing Firmness VISUAL/TACTILE Ash Stability ELEMENTS Ends Stability X~N Hot Collapse C) 2g~ ;.j-7 B k T LX &,,,j F.,t.-nt L;~riru Thn musc not bc cnp.cd or lu uniuchomsed >e. .sons BATCo document for Legal Services : Health Canada 20 October 1999 /"G- Mec;=-cs refer to rhe way 4-n which the cigarette smokes. 1r:4.:-=:4.on and _~-Pzc: have a'-ready been- described. Amplit-ade refers to :'-z- a=ount of ar=a rather than the quality, of it. Qualit-y is assessed u54-ng the attributes described under 'fop Notes, Modifiers, Base Notes, am-; Supp- la-enzary Flavour. Aftertastes, like mechanics and visual uc:-~:e eiiements, have been included within the Omega technique and are additiona: to the type of assessment performed in Woodrose. This is because cons=er research has shown that lower delivery smokers feel these to be par:'cullarly useful in the evaluation of low del'-very products. is especially true for visualltactile elements where it is believec that since many of the traditional sensations associated with smokf=i have been attentuated then the look and feel of the product is incre-zs-ngly important. PrevLously it was indicated that Omega represents an improvemen". over Woodrase because rather t1hat using cozzmercial brands to exempi.--F a part-cular prc4uct attribute a series of standards have been deve-~oped to represent successively incremental levels of a particular attr_,~-_:e. So far -"- tie following cigarette standards are available. OEP 10 CIGARETTE STANDARDS 1. Draw Resistance 2. Draw Effort 3. Mouthfull of Smoke 4. Mouth/Throat Irritation 5. Impact 6. Flue Cured Amplitude 7. U.S. Blended Amplitude 8. -lop Notes r%-) CD L.4 CD ~c;. iv-7 3 .% T 4 L J~ ami Exs>,n i I..miictl Tho%, must " be copie d or s-to.n tu BATCo document for Legal Services : Health Canada 20 October 1999 7heme are five cigarette standards for each attribute except for top notes where there are ten. Cigarettes are to be evaluated on an ele,.en scale of sensory magnitude with reference to the standards. OH? 11 - ELEVEN POINT SCALE OF SENSORY MAGNITUDE 0 Zero Just Perceptibl 1-~ Low 2' 2t Moderate 3 3-1 High 4 4-~ Very High 5 Extremely High It can be seen that the scale ranges from I to 5 and that half scores can be used in the evaluation. Each successive cigarette standard for each attribute has been designed to subjectively represent the points 1 through to 5 on this scale. 7he Top Note Standards have been achieved by using single grade tobaccos. 0 -9b. CD c:~ r\J CD LNJ C) l~#w7 B A T et. K jn%j f.%L-,ne L~.ed Thit mw%4 M bc COP3ed or thoun iti una 1 > Se,! u N M BATCo document for Legal Services : Health Canada 20 October-,1999 / 1 12 ORP 11 TOP \O--!' STANDARDS FLUE CURED VIRGINIA (~' U.S.A. R3W FLUE CURED VIRGINIA (-) Thai BCBT FLUE CURED AMARELINHO Brazilian AFE AIR CURED BL"RLEY U.S.A. BIB AIR CURED K-42Y Malawi MBX0 DARK AIR CURED P-1-RAGUA-!AN Paraguayan Largo S. AIR CURED HLRYLA-N-D U.S.A. ML4 FIRE CURED LATAKI-k (4-) Cyprus CFL FIRE CURED Malawi ZSD SUN CURED ORIENTAL Greek CAB SUN CURED ORIEN7AL Greek AF 2 1-4 In some cases a good and a poor version of each top note have been produced to give a more accurate picture of the tobacco quality. The tobacco grades used to achieve the standards are also shown on the slide. The panellists are also given a list of Supplementary flavours again which to check off the cigarette. CD C) r\.) CD (J'i CD CC' ti)- 7 B …, T # L K amJ E %-%,m This M"51 ffl bc ct~Pled Of sho-n w uniutN>i,.sed pe-,ofis. J~:. BATCo document for Legal Services : Health Canada 20 October 1999 OH? 1 -1 Sl:?P'U~TNTARY FLAVOURS s P l CI Y 3. ACRID 4. SWEr--- 5. TANGY 6. B 171 ME R 7. COOLLNG 8. FRUITY (HEAVY) 9. MM (LIGHT) 10. ~ETALTLIC 11. WOODY 12. MALTY 13. MIhTY (MENTHOUPEPPERRINT) 14. DRY 15. ARTIFICUL This checklist is to be used as an early warning device that flavours may have been added to a cigarette rather than to provide a sophisticated flavour profile of the cigarette. A second technique has been developed which is known as Omega 11 to provide accurate and detailed flavour profiling of the product. This technique uses people who have been highly selected and trained. It is not designed that they provide a typical consumers viewpoint of a product but that, in some ways, they act as "human gas chromatographs". The aim of this panel is to give a precise profiling of the effects of additional flavours added to products and to provide us vith profiling of competitors products in terms of flavour additives. In order to develop this panel, therefore, a fairly comprehensive screening process is necessary. Initially, potential panellists are screened to determine whether or not they have any specific anosmias. An anosmia is where an individual is totally unable to detect a certain smell. The screening agents are chosen on the basis that they are known to provide difficulties for less sensitive individuals. Such a screening exercise has been performed at GR&DC Southampton. U4 ; -F; 140 aAT iLk xAjF*,-.-m3Lon,tej Ths- mul-t not be cnpsedorsho-n ao.nawtynsea BATCo document for Legal Services : Health Canada 20 October 1999 OR? .146 SPEC: C .00SMIA S =- ---NING FRI-~LARY ODOEWN-, ?RIXARY 0DOUR Pyridine Burnt, Smoky iso-Valerfc Acid Sweaty 1-?yrrolize Spermous Trimethyla=ine Fishy iso-Butyra-ldehyde Malty 5cE-Androst-116-en-3-one Urinous W-Pentadec,alactone Musky '(- -Ca--voae Minty 1,8-Cineole Camphor On this slide a list of pri=ary chemcal odorant, and the specific odours associated with them, used in the R & D test are shown. From the results of this exercise it can be seen that over fifty percent of the totall population sampled were anosmic to 2 or more of these odours. .C:- CD CD r,~'.) CD L~14 'ZE: lu* 1 B A T #U K in-1 E %pin t Ihn Ml‚sI Mc bc CnP,ed or ihoum 90 un…%JCN)Msed c-erson, BATCo document for Legal Services : Health Canada 20 October 1999 OHP 15 ANOSMIA SCREENINC SUBJEC-7S DET--C::NG THE SCREENINC CONCENTRATION Number of Total Women Women Men Men Odorants Subjects Non-Smokers Smokers Non-Smokers Smokers Detected (N=214) 'IN-39) (N=59) (N=59) (N--57) (Z) 9 16.6 20.5 12.9 18.6 14.0 8 32.4 36.0 35.5 28.9 31.5 7 23.7 25.6 2.2.6 22.0 24.6 6 13.4 12.8 16.1 20.3 5.3 5 7.0 5.1 9.7 1.7 12.3 4 4.3 0.0 0.0 6.8 7.0 3 0.5 0.0 0.0 0.0 1.8 2 1.6 0.0 3.2 0.0 3.5 1 0.5 0.0 0.0 1.7 C). 0 0 0.0 0.0 0.0 0.0 0.0 Only those who detected at least 8 out of the 9 odours were selected to go on to the next stage. It can also be seen from the slide that non- smokers were better at detecting the odours than smokers and that women tended to be better than men. The next stage of the screening process, for those who passed the first stage, was where subjects were required to identify and correctly label eighteen different flavours from their aroma and taste. C) C) Q4 Cc, 14-7 8 -% T - L K am,; hxg~,Mj L.n-iet Thm mwst not be c"P,ed or mown lo uniutr.)r-,$e,: BATCo document for Legal Services ; Health Canada 20 October 1999 /16 ORP 16 SCREENING FLAVOt7RS 'A.C GRA.PEFFRUIT COO LZMON C'-\N.;MON ORANGE ALMOND MANDARIN RUM L LM-7 CL Or. Z PEPPERMINT BERGHOT SPEAJZMIN`T NUT.'"-G AN-ISE CLTM7-N GINGER CELERY Froat the res--,Its it can be seen that no subjects could correctly detec: and label all 18 of the f lavours. Only 60Z of the already highly selected population were able to label --' or more f:.avours. These individuals were selected for the flavour pane:. Again, nor smokers, were better than smokers with a slight tendency for women to be better than men at this type a.-- task. 0 C) C) N-) C> tA C) @ I w-7 B AT I L K A" FxV.)n% Unitsed This must root be cnpied or shown to unauthansed persons co BATCo document for Legal Services : Health Canada 20 October 1999 T 1~e next stage in the develo;=ent of such a panel is to der-ermine how indi-viduals zend to cluster .11avours. The subjects were therefore give: 4-~ flavours to place inro groups on the basis of similarity to others within that group and dissimilarity from those in other groups. Fort'! four flavours were used in all' and are shown in the slide. OHP 18 - --ORTY FOUR FLAVOURS 1. CUMINI 23. GINGER 2. HZLIOTROPIN 24. BUTYRIC ACID PHENi"--ACETIC ACID 25. LAVENDER PERU BALSAM 26. MUSK 5. CADE OIL 27. ORANGE 6. ISO-VALERIC k C,? D 28. MENMOL 7. BERGAMOT 29. MACE 8. ROSE OIL 30. ff-LONONE 9. YLArG YLANG 31. BLA.CKCURRANT io. COSTUS 32. jASMIN 11. SAGE OIL 33. PEPPERMINT 12 BENZALDEHYDE - D-DECALACTIONZ 34. 13. RASPBERRY 35. CLOVE OIL 14. SPEARMINT 36. ALDEHYDE C-8 15. CEDAR OIL 37. APPLE 16. CIN1&;AMY`L ES-1'-R 38. CASTOREUM 17. COMIARIN 39. 2-ETROXY- 18. ANCELICA 6-METRYLPYRAZIM 19. FLOUVE OIL 40. AMBROXIDE 20. ANISE 41. BENZYL BENZOATE 21. TOLU BALSAM 42. VANILLIN 22. MAPLE LACTONZE 43. PEENETRYL ACETATE 44. GERANIOL Resalu-.1-on of this similarity data was achieved through multi-di=ens-onal scaling. From the analysis it can be seen that subjects have the al-~Iity to disting-u-ish between the following clusters of flavours. C) iNJ C:) C:) not be copted or sho.,, tu unawo )nsc,4 . ~w t--l 8 k T I L N anj h%:-~ P Umotj Thij, mA%t BATCo document for Legal Services : Health Canada 20 October 1999 CL /17 0 ci Olip 17 CI) NUMBER TOTAL MMEN WOMEN MEN MEN 0 OF CORRECr POPULATION NON-SMOKERS SMOKERS NON-SHOKERS SMOKERS ®2 i IDENTIFICATIONS (N-91#) (1) (N-26) (Z) (N-18) (Z) (N-26) (Z) (N--24) M MI 25.5) 42.3) t6.7) 26.9) 12.5) 0 15 13.8) 3.8) 22.2) 15.4) 16.7) (A 14 7.4)60.5 3.8)61,4 16.7)66.7 0.0)57.7 12~5)58.4 13 13.8) 11.5) 11.1) 15.4) 16.7) CD z 12 8.5 11.5 5A 7.1 8.3 M il 8.5 11.5 11.1 3.8 8.3 7. 2 10 12.8 7.7 5.6 19.2 16.7 C) 5 ci) 0 9 2.1 3.8 0.0 3.8 0.0 0) E 8 1.1 0.0 0.0 3.8 0.0 7 4.3 3.8 5.6 3.8 4.2 6 1.1 0.0 5.6 0.0 0.0 CD 5 0 a -21 4 0.0. 0.( 0.( 0.0 (.0 0.0 (.0 0.0 0. 0 0.( Cr CD 2 0A 0.0 M -a: --- sL;ce"~* ---p --a- cc, C:-.es .. .. - -. casm, Va" La-.D~ I -"CC, I se~-2-cen~ e<3 I ke - S s 0.,ame F~,Zaafe r--- Gilrmct -7. 77'r' - ?.4 "3,t .Nm M91.0110 aft alidim "heriemyl 3 E3 At-: V Cole 13 = MUSK L A~a Vanaliw sai&ame #A,.ry Minty Fru ity Vanillic Floral Balsamic Musky Smoky S icy CD N) ;--7 3 % T -L K arij ftcb~; L-m-ici This must nOt be CoPled Or s~" 14) unsuckeansed Ptrnons BATCo document for Legal Services : Health Canada 20 October 1999 1, 0 e Z 7herefore, in the developme=: of the flavour panel :ec"nnique a set of flavour standards have beet developed for each of these clusters. ORP 21 FLAVOUR STANDARDS spicy CUKIN FRU = BLACKCURRAXT MiN-,L-V. PEPPERMINT 0 FLOP-kL JASMIN SMOKY CASTOREEUM VAKILLIC VANIL' LIN BALSAMIC PERU BALSAM BENZALDEME HUSKY MUSK BASE CIGARETTE STANDARD REFERENCE 10mg U.S. K-7~= CASED BUT UNFLAVOURED For each cluster a representative* flavour has been used. The base 40 cigarette on which the standard is placed is a 10 mg US blended product. In order to assess the flavour profile the panellist would use the following flavour circle. J>- CD CD r-J CD Lili CD r'l-J @ 1,4197 9 A T tL K anJ kit-rr, L,m,tra This mu,-% ffl bc cop.ed or sho.m tu p, r*-i .M, BATCo document for Legal Services : Health Canada 20 October 1999, " C I R C'- 0H.P 22 - FILAVOUR PRO--~-E ,"=Y BALSAMIC NILLIC RUI 71Y. FLORAL 14IN7Y For each flavour standard there is a five point scale. Using this scale the panellist would mark off the extent to which they felt that any cigarette contained that particular flavour. By joining together all the points on the circle a flavour profile is obtained. Profiles between cigarettes can be compared by overlaying one profile over another. Areas C:) of overlap represent co=onality between cigarettes and areas of non overlap represents non commonalivy. @ I #X7 B A T iL K ar-j This must C44 be mplea or sho-ri to urisymmsed pei-son, BATCo document for Legal Services : Health Canada 20 October 1999 BAT GROUP RESEARCHS7~ATEGYAND "HE ROLE OFF BAT (UK&E) RES--AR.CX A.ND DEEYEILO~--ENT- C-ENTRE Prior to 1985, BAT Group research activity was organised on a centralised basis with the Group Research and Development Centre (GR&Y,) being located in Southampton, U.K. This Centre did not have a direct reporting respons- ibility to any single Operating Company, being devoted principally to fundamental researches to be regarded as s.-L:rce information for the Group as a whole. In addition to the central R&D Laboratory each of the leading Operating Companies in the Group (Australia, Brazil, Canada, tories within which investig- Germany, U.S.A.) maintained research laborat ations were conducted with somewhat more of an emphasis on locally driven requirements. This pre 1985 scenario can be summaried as follows: A CO g PROCE. CA L Se"~ j~:b CD L)q CZ (C:-' Poy? B ¯. T -L K &AJ t%>,Ml L,m~icj Thsi 'n"t n'>' le cnp-ed or stw~--b r~1j _c::¯ BATCo document for Legal Services : Health Canada 20 October 1999 - 2 - ~ere it can be seen that central R&D devzzed fuily two-thirts of activity to fundamental research on product and pr.-cess research, whicri was largely interfaced to the major research companies together with their own partic- ular research activities. The remaining one third of activity was devoted to technical support services to other BAT Associate Companies (Ithe majority of this technical support be*,ng in the form of routine/semi- routine brand analysis and product develoDment consultancy). Since 1985, this scenario has been subs,:antiallv altered. Although this change is entirely in keeping with evolving BAT requirements it should also be observed that the nature of these changes is clearly reflected in similar changes in the role of R&D in most major commodity industries. The essence of this change is one of de-centralising the R&D flunction and increasing the ratio of applied and technical research to fundamental research. In practice, the Group Research and Development Centre was re-assigned to a role in which it now becomes the R&D resource for BAT U.K. and Export and BATCo. Companies. The re-named BAT (UK&E) R&D now becomes o-.e of six major R&D laboratories which in aggregate represent the BAT Group Research effort: I 4 APPLIED RESEARCK AND TECHICAL SUPPORT OCALI RISEARCI F;RDUP. UNDAMENT& 01 APPLIE1 cr. Mr-A _r__ i Jt:b C:) 0 Ll ILA CD Jqw7 8 ~A T #L k ami ~.jr-n$ L,M,8C;1 i hl! Inusi, nkX IX ;0PCJ W ~p wn t~ Qr. BATCo document for Legal Services: Health Canada 20 October 1999 -ri this pcs-, 19E-'-: scenario, the tallance cf fun~ameentat to aDplied research has been redressed to a more even distribution in a si'tuation where each major laboratory devotes subs-.antial prooortions of time to [a) the specific avplied needs of the Operating Company concerned and W contributions to the sort of commonly shared fundamental knowledge requirements which a Group research progran-.ne implies. This re-distribution of the balance between fundamental and applied researches may be seen in the case of BAT MW R&D by examining the allocation of effort in this Centre pre and post 1985: '40 ALLOCATION OF R&D EFFORT PRE 1985 FUNDAIVIEN'rAL RESEARCH 78% TECHNICAL SERVICES 22% (PRODUCT TECHNOLOGY/TECHNICAL SUPPORT) 1986 FUNDAMENTAL RESEARCH 47% APPLIED RESEARCH & DEVELOPMENT 36% (PROJECTS RELEVANT TO BATCo) TECHNICAL SUPPORT 17% (PROJECTS SPECIFIC TO OPERATING COMPANIES) The BAT Group research programme is, therefore, now assembled f-om the aggregate efforts of the six major Research and Development laboratories and represents a proportion of the total research activity of each of these centres. As such this Group programme is organised against two broad categories of objective, one of which is largely pro-active and strongly influenced by marketing requirements (i.e. PRODUCT RESEARCH) and CD Cc] 10-7 5 A~ 1 iL Ka" Lm®J Th$$ mu*¯ M be Cnp-cd or sno-n i,, perso', c::) (-)-i CD r1l) 01_ BATCo docurnent for Legal Services : Health Canada 20 October 1999