Tobacco Documents Online

FREE NICOTINE: ITS IMPLICATIONS ON SMORE IMPACT T.R. SCNORI MAREETING RESEARCH SERVICES OCTOBER 22, 1979 &q &D %0

i L ABSTRACT Smoke impact is a convenient term that has been applied to the smoker's degree of awareness of the presence of tobacco smoke in the back of th~ throat. Smoke impact is probably analogous, for most smokers, to smoke strength. More than likely, the sensation is mediated by pain receptors in the back of the throat. But what constituent in the smoke activates these receptors? Pain receptors respond to various forms of stimulation, however, it is felt that the primary constituent of tobacco smoke which activates them is nicotine. However, it is not nicotine per se that is responsible but it is the amount of free nicotine. Thns, degree of smoke impact is viewed as being determined by the amount of free nicotine in the smoke. The amount of free nicotine in the smoke depends upon: (i) total nicotine, and (2) pH of the smoke. The higher the pH, the greater the proportion of free nicotine and, accordingly, the greater the smoke impact. Also related to smoke impact is the N/C ntmlber, i.e., the ratio of ~ total nitrogenous substances~ carbohydrates. As the N/C number increases, there will probably be corresponding increases in smoke impact. The explanation for this is that the nitrogenous substances, upo~ the combustion of the tobacco, contribufe to the alkalinity of the smoke while the carbohydrates contribute to its acidity. Thus their ratio influences the pH of the smoke, and, therefore the proportion of nicotine that is liberated in the smoke. This suspected relationship between free nicotine concentration and smoke impact implies that we could create a ultra-low tar cigaretfe that produces much more impact than its delivery would suggest. There- fore, it is recommended that this relationship be evaluated. Speeificially, it is suggested that we develop ~ series of four 3-mg. tar cigarettes that only vary in their free nicotine concentration and test them among full-flavor smokers in comparison to a full-flavor control cigarette (such as RALEIGH or VICEROY). Background It is immediately apparent to anyone that has ever tried i0~ burley cigarettes and i0~ bright cigarettes that the two tobacco types result in very different sensory responses. These differences are primarily Oq in smoke (or inhalation impact)--the smoke impact resulting from the ~.~ hurley tobacco being considerably higher than that from the bright ~2 tobacco. Smoke impact is a convenient term that has been applied to~ the individual's degree of awareness of the presence of smoke in the back of the throat and is probably analogous to what some smokers may describe as smoke strength. The sensory experiences that are tormed smoke impact can be depicted as shown below:

smoke - > receptor~ smoke impact From this model, it can he seen that smoke impinges on certain receptors. The activation of these receptors results in certain sensory experiences which we call smoke impact. The degree of impact (or more accurately, the degree o~ one's awareness of the presence of smoke in the back of the throat) is determined by the number of receptors activated and/or the frequency of each receptor's activation. Smoke Constituent Responsible for Impact Specifically, though, what component in cigarette smoke is responsible for activating these receptors and, thus, for giving rise to the sensory experiences that we call smoke impact? It is nicotine. Other smoke constituents may also stimulate these receptors but nicotine is the main constituent in smoke to do so and, thus, smoke impact is primarily determined by the nicotine in the smoke. Suggesting that smoke impact is due to nicotine, however, is nothing new. The important factor, though, is not the amount of nicotine in the smoke per se but rather it is the amount of free nicotine in the smoke, which determines degree of smoke impact. This appears to have been recognized long ago by Sl~nuk (1939) who utilized the "nicotine number," which equBls ~ total nicotine~% free nicotine, as an objective indicator of the strength of tobacco smoke. Thus, when total nicotine is held constant, the strength of the smoke will increase with decreases in the nicotine number--which is exactly the same thing as saying that strength (or smoke i~pact) will increase with increases in the amount of free nicotine in the smoke. More recently, Shakhnovskii (1959) also observed that smoke strength is related to the amount of free nicotine in the smoke. This discussion may be confusing unless it is apparent what is meant by total nicotine and free nicotine. Importance of pH The way in which nicotine is typically reported can be misleading. This is due to the manner in which nicotine determinations are made. For instance, cigarettes X and Y may both be reported to deliver (based upon the standard smoking machine test) 2 mg. nicotine/ cigt. However, a given smoker may actually inhale much more free nicotine from cigarette X than from cigarette y. Likewise, cigarette W may deliver 1 mg. nicotine/cigt, while cigarette Z delivers 2 mg.~- nicotine/cigt, yet a given smoker may inhale equal amounts of free nicotine from cigarettes W and Z. This paradox results from the fact that in making the nicotine delivery determinations strong bases are employed to free or release the nicotine from its bonds with other

elements (e.g., to free the nicotine from nicotine citrate). By using these strong bases, virtually all the nicotine is liberated° Nicotine delivery, then, as reported in analytical reports actually represents the total possible (or virtually so) free nicotine. It does not, though, represent the free nicotine that is actually avail- able to the smoker--since the smoker is unable to employ strong bases in order to liberate all the nicotine, that is, at least not in his mouth--which is the important thing from a smoke impact point of view. Thus, the amount of free nicotine available to the smoker is determined by the degree of alkalinity (or pH) of the smoke as well as his own degree of alkalinity. Therefore, the free nicotine inhaled by him is always some proportion of the total possible free nicotine. For our purposes, the amount of free nicotine available to the smoker is determined by the pE of the smoke, since we have no control over the pH of the smoker's mouth. As pH increases, there is a corresponding increase in the proportion (of the total possible free nicotine) that is actually liberated. This relationship is depleted in Figure i. From this figure, it can be seen that as pH increases from 5½ to 9 there is a very sharp increase in the proportion of nicotine that is freed. Thus, it should be possible to increase smoke impact by artificially increasing pH. ~mportance of Tobacco Constituents Shmuk (1939) also discussed the usefulness, as an indicator of both taste quality and strength of tobacco smoke, of the ratio of carbohydrates to proteins (% carbohydrates/% proteins) in the tobacco. This ratio is generally referred to as Shmuk's number (Abdallah, 1970). According to shmuk, as the S~muk number increases, there will be corresponding decreases in smoke strength. In other words, when carbohydrates in the tobacco become relatively more plentiful than proteins, smoke strength will decrease; the reverse will result in increases in smoke strength. This suggests that we could increase smoke impact by the addition of proteins to the tobacco. Why should the ratio of carbohydrates to proteins hmve any bearing on taste strength or smoke impact? The answer is simple. In combustion of tobacco, carbohydrates tend to form acids while proteins tend to form bases. Thus, the tendency is for the carbohydrates to reduce pH while proteins tend to increase pN. That is, carbohydrates and¢~ proteins in tobacco act as antagonists--with their ratio partially determining the total pH of the smoke° Therefore, as Shmuk's number decreases, there will be corresponding decreases in the proportion of free nicotine in the smoke and, accordingly, less smoke stren~th~ ~ one of the important carbohydrates in tobacco is sugar (Elson, et a l., 1972)o The relationship between sugars in the tobacco and pH of the resulting smoke is shown in figure 2. It is readily apparent from inspection of this figure that low concentrations of sugars are associated with high pH of the smoke and, thus, a proportionately large amount of free nicotine.

I00 9O BO 40 30 2O I0 ! I t I I I [ I I I 0 I 2 3 4 5 6 ? B 9 IO pH Figure ]. Percent free ~icot~ne as a function of pH. II 12. 13 A 14

8 F I 2 4 4 5 UJ 6 ( ! I I I I I 1 Y I 6 8 I0 12 14 16 18 20 %SUGAR IN TOBACCO Figure 2. Smoke pH as a function Of percent sugar concentration in tobacco. Adapted from Elson, et al., 1972. &a

w~ TABLE 1" Constituents of Major Tobacco Types Nitrogenous Constituents %Total Nitrogen ' %Volatile Bases {BV) as Ammonia %hicotine % Volatile Bases minus Nicotine as Ammonia % Ammonia % alpha-Amino Nitrogen % G1utamine % Asparagine I Glutamic Acid % Aspartl¢ Acld % Nitrate as NOa % Protein Total Carbohydrate Constituents % Reducing Sugars (Before Inversion) as Dextrose % Pectin as Calcium Pectate % Cru~e Fiber % kignin Total Acids pH Volatile Acids (VA) as Acetic Formic Acid Volatile Acids (NA) minus Formic as Acetic ~ ~ Turkish (1.G~) (9.25) (0.30) (z.17) 0.20 0.62 0.37 0,29 1,93 2,91 1,27 1.05 0,08 0.32 0.23 0.]8 0.02 0.16 0.13 O.lO 0.05 0.17 0.06 0.10 0.28 0.30 0.35 0.17 0.19 0.86 0,12 0.45 0.04 O.T2 0.11 0.09 0.02 0.34 0.04 0.23 Trace 1.70 0.00 0.23 5.69 11.06 10,05 7.44 ~-5'r- 1TT5"%'-- T2CTE- 1DUT0-- 22.09 Trace Trace 10.30 6.19 9.91 12.14 6.77 7.88 9.29 21,70 6.63 5.37 5.66 i0.26 6.43 5.45 5.00 6.60 4.90 0.15 0.10 0.09 0.19 0.06 0.03 0.02 0.08 O.OO 0.07 0.05 0.09 2.83 6.75 2.43 3.67 0.70 8.22 2.98 I ,03 0.8] 3.04 2.70 3.16 Malic Acid Citric Acid Oxalic Acid Aromatics % Volatile 0t1 % Alcohol-Soluble Resins Mineral Constituents % Ash % Calci=m as CaO % Potassium as KzO % Magnesium as MoO % Chlorine as CI % Phosphorus as P205 % Sulfur ~s SO~ Su~ Nicotine pH 5hmuk's Number N/C Number 0.15 0.14 0.14 7.25 9.08 9.27 8.94 II.2~ / 10.~1 24.53 21.98 14.78 2.22 8.01 4.79 4.22 2.47 5.22 4.40 2.33 0.36 1.29 1.03 0.69 0.84 0.71 0.26 0.69 0.51 0.57 0,53 0,47 1.23 1.98 3.34 1.40 1.93 2.91 1.27 1.05 5.45 5.00 6.60 4.90 7.30 2.25 4.39 4.33 2.06 7.46 2.91 3.13 +Adapted from Harlan & Mosley, 1955; Wynder & Hoffmann, 1967.

Shumk's number (carbohydrates/proteins ratio) leaves something to be desired. It does not take into consideration the fact thBt numerous other constituents in tobacco in addition to proteins, e.g., ammonia, nitrates, etc., also tend to for~ bBses upon~omb~stion and, thus, tend to increase the tot~l alkalinity (pH) of the tobacco smoke. Therefore, it would be more appropriate to use the ratio of carho- hydrstes to total nitrogenous substances, as an indicator of smoke strength or of smoke impact, i.e., 10x (% nitrogenous s~bstances/ carbohydrates) which will be referred to as the N/C number. Since our primary interest is in increasing smoke strength or smoke impact, this ratio was inverted and multiplied by i0 so that numbers greater than 1 would normally be achieved. AS the N/C number increases, there are corresponding increases in pH and, therefore, in free nicotine in the smoke. Thus large values of the N/C number should be associated with high strength and smoke impact while small values should be associated with low strength and smoke impact. Discussion If these assumptions are correct, then, smoke impact is due to the amount of free nicotine in the smoke. The amount of free nicotine in the smoke is determined by two factors: (i) the total potential free nicotine in the smoke (which in our usual way of viewing it, is simply nicotine delivery in mg/cigt.); and (2) pH of the smoke, which deter- mines the proportion of the nicotine that is actually liberated in the smoke. obviously, then, we should be able to increase smoke impact by increasing the total free nicotine potential (i.e., by using high nicotine blends and/or nicotine additives) in the smoke. Also, we should be able to increase smoke impact by simply increasing the pH of the smoke and, thereby, Causing more nicotine to be available to the smoker as free nicotine. We could also reduce impact by de- creasing pH (which shmuk, e t a l, 1932, did by adding acids to the tobaccos), however, this paper is primarily concerned with techniques for increasing smoke impact. It is apparent that we could add nitro- substRnces to the tobacco in order to increase the pH. However,~ genous such techniques (e.g., the addition of ammonia) have not proved too ~w successful since they tend also to produce unpleasant side effects, such as making the side stream smoke very noxious. But let us consider readily available means of increasing pH in tobacco smoke, that is, those that can be accomplished by varying our existing blends. In Table l, a breakdown of the various constituents in the major tobacco types is shown (c.f. Harlan & Moseley, 1955; and partially c.f. Wynder & Hoffmann, 1967). Tobacco Types The major tobaccos in our blends are burley and bright. Therefore, the values shown in Table 1 for these tobaccos are of primary interest°

It will he noticed that total nicotine is higher in burley than in bright. Thus, there is a potential for greater amounts of free nicotine in the smoke from hurley tobacco than there is from bright tobacco. In addition, the pH of the hurley is higher than that of the bright. And this difference in pH is actually magnified-i~Idze smoke, since pH of smoke is considerably higher thsn that of the tobacco from which it was generated (due to the products of combustion)° Because of the higher pH, proportionately more of the nicotine is present in the free state in burley smoke than in bright smoke. These twe factors in combination, then, are responsible for the much higher impact in i0~ hurley cigarettes than in i0~ bright cigarettes, i.e., both the higher total nicotine and the high pH that are found in burley tobacco. Even if some particular burley and bright tobaccos had equal amounts of total nicotine, the burley tobacco would have more free nicotine in its smoke as a consequence of its higher pH and, therefore, its smoke impact would still be higher than that of the bright. Likewise, even if some particular burley and bright tobaccos had equal pH, the hurley tobacco would have more free nicotine in its smoke as a result of its greater amount of total nicotine. Accordingly, its impact would still be higher than that of the bright. Since it is a given fact that a 10~ hurley cigarette results in greater smoke impact than a i00~o bright cigarette, Shmuk's number (carbohydrates/ proteins) should be lower for hurley tobacco than for bright tobacco. Inspection of Table i reveals that burley does result in a lower Shmuk number than bright tobacco. Likewise, the N/C number (nitro- genous substances/carbohydrates) should be higher for hurley than for bright. And from Table 1 it can be seen that the N/C number is higher for hurley tobacco than for bright. Both Shmuk's number and the N/C number, then, should provide some predictive information concerning smoke impact. However, since the N/C number takes into consideration total nitrogenous substances rather than just proteins, its utilization may he more useful as an aid in predicting impact than Shmuk's number. Taking into consideration these known differences between hurley and bright tobaccos, we can with great ease increase smoke impact by increasing the proportion of hurley that we put in the blend. By doing~ this, we can take advantage both of the higher nicotine concentration~" in hurley as well as its higher concentration of other nitrogenous substances (besides nicotine), which are responsible for the higher pH in burley tobacco and, accordingly, for the higher proportion of free nicotine in the smoke.~ In addition, we could reduce the carbohydrate concentration in the tobacco--eliminating components that upon combustion tend to form acids~ Effectively, this would result in higher pH and therefore a higher proportion of free nicotine in the smoke. Thus, the elimination of carbohydrates would result in greater smoke impact. Easier than doing this, however, would be to eliminate some of the carbohydrates that

we currently add to our tobaccos. With bnrley, we add carbohydrates, viz., sugars, as flavoring. Actually, we know from the preceding discussion that the addition of sugar will act to reduce pH and, thus, reduce the~vailability of free nicotine in the smoke. Therefore, the addition of sugar acts, in this fashion~ to reduce smoke impact° Implications If the suspected relationships between free nicotine, pH, the relative concentrations of nitrogenous and carbohydrate constituents of tobacco and smoke impact are true, this implies that we could develop an ultra- low tar cigarette that produced considerably more smoke impact than its delivery level would suggest. Therefore, it is recommended that we consider developing a series of four 3-mg. tar cigarettes that only vary (or primarily vary) in their free nicotine concentrations, to be tested among local full-flavor smokers in comparison to a full- flavor control cigarette--such as RALEIGE or VICEROY° &1