Tobacco Documents Online

MEl1ORA\D(:M Internal Memo # 9 613 September 18, 1996 TO: M. A. Sudholt FROM: J. M. Johnson SUBJEC_ : Summary of the Effects o° Arnmoniuia Carbonate, Ammoniula Bicarbonate, Urea and Diammonium Phosphate on Smoke pH, Smoke Data and Leaf Chemistry PROJ"ECT NO. : Q 449 Analytical Support :IND--NGS: Tobacco nicotine was not affected by the additives. The four additives studied, urea, ammonium carbonate, ammonium bicarbonate and diammonium phosphate, were found to modify the smoke delivery of nicotine and corrected particulate matter. Urea, ammonium carbonate and.ammonium bicarbonate additives were found to increase the -comparative pH or the collected smoke particulates. INTRODUCTION: The smoke p:i, mairistream smoke tielive_y and leaf chemistry data for cicarettes with different concehtration levels of the additives urea, amraonium carbonate (AC) , ammonium bicarbonate (ABC), and diammonium phosphate (DAP) were determined (1,2). Statistical evaluations of the data to determine significant differences between the control and additive cigarettes were performed (3). Group and pairwise analyses were performed on mainstrearn smoke delivery data to determine if mean differences among the cigarettes at all concentrations for a single . additive were greater than would be expected by chance and, if so, which concentration level cigarettes were different from the control. Trend analyses,.were also performed to determine if the results increased, decreased or did not correlate as the percent additive was increased. Trend analyses were performed on smoke pH, mainstream smoke delivery, leaf chemistry and nicotine transfer data. This memorasidum provides a summary of the results obtained from the statistical evaluations. An add=t=ve was determined to have an impact if a trend,or threshold ef 'ect was evident. In some cases, lower additive,level czgarettes were significantly different from the control while.-higher:additive level Ciga:ettes were not significantly differer.t from.the control, This indicated that the additive was not the cause for the-di_'ferences which were found.

In evaluating smoke pH data, it should be noted that the reported smoke pH is a measurement of the pH of an aqueous extraction of collected smoke particulates. The Henderson-Haselbach equation (4) relates the pH of..aqueous extractions of collected smoke particulatej to percent levels of protonated and unprotonatad nicotine in the ;~ aqueous solutions, Increases in smoke pH can result in.inereases ih, the calculated percent unprotonated nicotine in an aqueous sblution. It is not known whether a similar effect would occur in the smoke aerosol. E,IGERIMENTAL : The control and additive cigarettes were made by the Product Development Department. Cigarettes containing urea, ammonium carbonate (AC) or azcmonium bica=bonate. (A3C) were made with an Old Gold 85 construction. The additives were applied to the tobacco at 0% (the control),, 0. 5i, 1. 0t, 1, 5% and 2. 0-1 levels. The cigarettes containing diammonium phosphate (DAP) were made with a Newport 85 construction and the additive was applied at 0.4% (the control), 1.03, 2.03 and 3, 02s levels. Leaf Laboratory and Smoking Laboratory personnel performed the leaf chemistry and mainstream smoke delivery analyses. Comparative smoke piH analyses were performed by members of the Research Analytical Development Section. The percent unprotonated nicotine in aqueous solutions and the percent nicotine transfer from leaf to smoke was calculated from the experimental data. The group statistical analysis method used was One Way Analysis of Variance (ANOVA) and the pairwise stat3stica: analysis method used was either Dunn's o= Dunnett's Pairwise Multiple Comparison. The statistical analyses were run at a 95* confidence level. Dunn's and Dunnett's tests are advanced forms of the student t-test. Dunnett's test was used when the.sample sizes were equal and Dunn's test was used when the sample sizes were not equal. These statistical analyses were performed on mainstream smoke delivery data only. The results of these,statistical tests are displayed in Table 1. In the cases where a statistical difference was found using the pairwise test, the average percent d;fference relative to the control is given. The Pearson Product Moments Correlation run at a 95% confidence level was used to establish trends, Trend analyses were performed on smoke pH, mainstream smoke delivery and leaf chemistry and nicotine transfer data. Data used in these analyses are shown in Tables 2 th..ough 5. The following Smoking Lab data were evaluated: 1) mg nicotine per cigarette, 2) mg corrected particulate matter (CPM) per cigarette, 3) puff count per cigarette, 4) mg nicotine per puff, 5) mg CPM per puff, 6) nicotin e per CPM and 7) nicotine per CPM per puff. The following leaf chemistry data were evaluated: 1) percent nitrogen, 2) percent total volatile.bases (Tv'8), 3) percent nicotine, 4) percent total reducing substances (TRS) and 5) tobacco pH. - - ~~ , ..2_

RESULTS; URF.A The addition of urea to cigarette tobacco was found to increase the 1) nicotine per CPM, 2) percent nicotine transfer from leaf to smoke per cigarette and 3) smoke pH. A su=ary of the effects of adding urea to cigarette tobacco, based on results from statistical evaluations of the analytical data, is shown below. Tobaceo Chemistrv The addition of urea to the tobacco did not affect the percent nicotine in the tobacco blend or the leaf pH. As the percent added urea was increased, corresponding increases were seen in the percent nitrogen, percent TVH and percent TRS in the tobacco. Mainstream Smoke Delivery The addition of urea to cigarettes.without filters did not affect the following mainstream smoke variables: 1) nicotine per cigarette, 2) CPM per c=aarette, 3) puffs per cigarette, 4) nicotine per puff, 5) CPM per puff and 6) nicotine per CPM per puff. Results for the filtered cigarette were similar, The addition of urea to unfiltered cigarettes did affect t:ae nicotine per CPM ratio i.^_ the mainstream smoke. Nicotine per CPM A positive trend was observed between the increasing delivery of nicotine per CPM and an increasing level of urea in unfiltered cigarettes. Nicotine per CPM increased from 0.077 (control) to 0.085 (2.0~k urea) which represented percent increases ranging from 5 to 112s. The corresponding filtered cigarettes did not show this increase in nicotine per CPM with urea concentration. Variations in filtration may have obscured the relatively small effect of i;rea on the nicotine per CPM ratio. Parcent Nicotitne Transfer A positive trend was observed between increasing levels of urea and increases in the percent transfer of nicotine from the leaf to the smoke on a per cigarette basis, This effect was observed in both the unfiltered and filtered cigarettes. The percent nicotine transfer increased from 12. i~ (control) to 13. 5"s (Z. 0t urea) in the unfiltered cigarettes and from 6.5t (control) to 7.0~; (2.021 urea) in the filtered cigarettes: Comyarative Smoke nFi - Siltered Cictaret tes i As the uercent of added urea was i_ creased, the comparative smoke pH increased from 6.,70 (control) to 7.19, (2.02, smoke pH increase for.urea cigarettes would percent unprotonated nicotine in an aqueous (control) to 139s _ (Z. 0t ur'ea) : -3- urea) . The observed increase the calculated solution from about 5-W ~ I N Cd ~ W C!t

AMMON S VM C3iR80rIATS The addition of ammonium carbonate (AC) to cigarette tobacco was found to increase the 1) leaf pIi, 2) nicotine per puff, 3) nicoatine per C?M, 4) nicotine per CPM per puff, 5) percent nicotine transfer from leaf to smoke per puff and 6) smoke pH. A summary of the effects of adding ammonium carbonate to cigarette tobacco, based on results from statistical evaluations of the analytical data, is shown below. Tobacco Chemistry The addition of ammonium carbonate to the tobacco did not affect the percent nicotine or the percent TRS in the tobacco blend. As the percent added asniaon-um carbonate was increased, corresponding increases were seen in the percent nitrogen, percent TVB and leaf pH of the tobacco. Maizistream smoko Deliverv The addition of ammonium carbonate to unfiltered cigarettes did not affect the following mainstream smoke variables: 1) nicotine per cigarette, 2) CPM per cigarette, 3) puffs per cigarette and 4) CPM per puff. Results from the filtered cigarettes were similar. The addition of ammonium carbonate in unfiltered cigarettes did affect ~ the following smoke variables: 1) nicotine per puff, 2) nicotine per ~ CPM and 3) nicotine per CPM per puff. Nicotine per CPM The addition of ammonium carbonate to unfiltered cigarettes increased the nicotine per CPM ratio from 0.077 (control) to 0.098. The percent.increases ranged from 4 to 143s relative to the control. The nicotine per CPM was not found to increase with increasing levels of ,additive,_ as in a trend. Rather there appeared to be a threshold concentration, less than or equal to 0.5% amtnoniutn carbonate, above which the impact of the additive was not related to concentration. This has been noted as a threshold effect (3). The corresponding giltered,cigarettes showed similar effects. Nicotine per Pu_f and Nicotine per CPM per Puff Positive t=ends were observed between increasing delivery of nicotine per_puff and.nicotine per CPM per puff with increasing levels of ammonium carbonate:in the unfiltered cigarettes. Nicotine per,puff.inereased ±rom about 0.28 mg/puff (control) to 0.31 mg/puff'aith percent inc,set4ses ranging from 4 to 10-t. The ratio of nicotine per CFM per puff increased from about 0.0092 (control) to . 0, 0105 with percent_ increases ranging f_om, 6 to 10. The corresponding filtered cigarettes.showed similar effects.

i Percent Nicotine Transfer A positive trend was observed between increasing levels of ammonium carbonate and increases in the percent transfer of nicotine from the leaf to the smoke on a per puff basis for the unfiltered cigarettes. The percent nicotine transfer on a per puff baaia increased fro2n 3.5% (control) to 1.6i (2.0-% AC). An increase in transfer was not observed in the corresponding filtered cigarettes. Variations in filtration may have obscured the small effect that ammonium carbonate had on the percent nicotine transfer per puff. .; Comcarative Smoke p1i - Fi2tered Giqarettes As the percent of added aruaonium carbonate was increased, the comparative smoke pH increased from S. 70 (contro~l) to 7, 38 (2. 0~ AC) . The observed smoke pH increase for ammonium carbonate cigarettes would increase the calculated percent unprotonated nicotine in an aqueous solution from about 5t (control) to 19ir (2.0% AC) . AMONIt7M SIGAABONATE The addition of ammonium bicarbonate (ABC) to cigarette tobacco was found to increase the 1) leaf pH, 2) nicotine per cigarette, 3) nu:~ber of puffs per cigarette, 4) nicotine per puff, 5) nicotine per CPM and 6) smoke pH. A summary of the effects of adding ammonium bicarbonate to cigarette tobacco, based.on results from statistical evaluations of the analytical data, is shown below, Tobacco Chemist--'Y . The addition of ammonium bicarbonate to the tobacco did not affect the percent nicotine, the percent nitrogen or the percent TRS in the tobacco blend. As the percent added ammonium bicarbonate was increased, corresponding increases were seen in the percent TVS and leaf pH of the tobacco. Mainstream Smoke Delivery The addition of,ammonium bicarbonate to cigarettes without filters did not affect the following mainstream smoke variables: 1) CPM per cigarette, 2) CPM per puff and.3) nicotine per CPM per puff. Results from the filtered and unfiltered cigarette were similar. The addition of `arunonium bicarbonate. to ur:filtered cigarettes was found to affect the following smoke variables:, 13 nicotine per cigarette, 2) number of puffs per cigarette, 3) nicotine per puff and 4) nicotine per CPM. -5-

Nicotine per Cigarette, Nicotine per Puff and Nicotine per CPM For unfiltered cigarettes, the values of nicotine per cigarette, nicotine per puff and nicotine per CPM were higher for cigarettes contain:ng a=onium bicarbonate than for the control cigarettes. There was no trend where increases in the smoke variable corresponded to increases in the percent additive. Rather there appeared to be a threshold concentration, less than or equal to 0.5-~ ammoniusn bicarbonate, above which the impact of the additive was not related to concentration. Nicotine per cigarette increased from 2.3 mg/cig (control) to 2,7 mg/cig with percent increases rangina from 8 to 17%. Nicotine per puff increased from 0.28 mg/puff (control) to 0,32 mg/puff with percent increases ranging from 6 to 13%. Nicotine per CPM increased from 0.077 (control) to 0.090 with percent increases ranging from 9 to 17%. Similar increases were seen with filtered cigarettes, Number of Puffs per Cigarette The addition of anunonium bicarbonate in unfiltered cigarettes at the 2.0t level increased the number of puffs per cigarette by 8;s from 8.3 puff/cigarette to about 9.0 puff/cigarette. The puff count of cigarettes containing lower levels of ammonium bicarbonate were not different from that of the control, Similar results were obtained for filtered cigarettes. Percent Nicotine Tr aasfer No trend was observed between increasing levels of ammonium bicarbonate and increases in the percent transfer of nicotine from the leaf to the smoke for either the unfiltered or the filtered cigarettes. Comparative Smoke pH - Eiltered Cicarattes As the percent of added ammonium bicarbonate was increased, the comparative smoke pH increased f=om 6,70.(control) to 7.26 (2,0,9s AziC). The observed smoke, pH ,increase fo,r ammoniurn bicarbonate cigarettes would increase the percent unprotonated nicotine in an aqueous solution , from about, _53 (control) to .153 (2 . 0% AaC) . ~ ~ ~

DznNMxzvM pxospxaTE The addition of diammonium phosphate (DAP) to cigarette tobacco was found to affect the 1) leaf pH, 2) nicotine per cigarette, 3) CPM per cigarette, 4) number of puffs per cigarette, 4) nicotine per puff, 5) nicotine per CPM, 6) nicotine per CPM per puff and 7) smoke pE. A summary of the effects of adding diaatmonium phosphate to cigarette tobacco, based on results from statistical evaluations of the analytical data, is shown below. Tobacco Chamistry The addition of diammonium phosphate to the tobacco did not affect the percent nicotine, the percent nitrogen or the percent TRZ in the tobacco blend. As the percent added diammonium phosphate was increased, corresponding increases were seen in the percent '.'v'd and leaf pH of the tobacco. Mainstream Smoke Delivery The addition of diammonium phosphate to cigarettes without filters did not affect the following mainstream smoke variables: 1) CPM per puff and 2) nicotine per CzM. Similar results were obtained from the corresponding filtered cigarette analyses for CPM per puff but not for nicotine per CPM. The addition of diammonium phosphate to unfiltered cigarettes did affect the following smoke variables: 1) nicotine per cigarette, 2) CPM per cigarette, 3) number of puffs per cigarette,'4) nicotine per puff and 5) nicotine per CPM per puff. Number of Puffs per Cigarette, Nicot+r1e per Cigarette and CPM per Cigarette As the percent diammonium phosphate in unfiltered cigarettes increased, corresponding increases were seen in both puffs per cigarette and CPM per cigarette. Puffs per cigarette increased from 8.6 (control) to 9.3 (2.0* DAP) representing an increase of 91S. CPM per cigarette increased from 30.9 mg/cig (control) to 34.0 mg/cig representing an increase of 10ss. Similar results were observed for the filtered cigarettes. Nicotine per cigarette for unfiltered c'_garettes is different from the control only for cigarettes containing 3-t d+ammonium phosphate. Nicotine per cigarette;increased from 2.3 mg/cig (control) to, 2. 6 mg/cig (31s DAP) representing a 121s increase. For filtered cigarettes, as the percent diammonium phosphate on the tobacco increased corresponding increases were seen in nicotine per, cigarette. The values of nicotine per cigare~te, CPM per cigarette and puffs per cigarette were higher for cigarettes containing diammoniuin,phosphate relative to the control cigarettes. The -7-.

increased puff count is probably due to the smolder retardant properties of diammonium phosphate changing the burn characteristics of the cigarettes (5). The i:icreases in observed nicotine per cigarette and CPM per cigarette appear to be mainly due to the increase in the number of puffs per cigarette. Nicotine per Puff, Nicotine per CPM and Nicotine per CPM per Pu°f The results seen for these three smoke variables were possibly influenced by the increased puff count and modified burn chemistry caused by the smolder retardant properties of diammonsum phosphate (5). For the unfiltered cigarettes, nicotine per puff was about 4% less than the control for the 1.0% and 2.025 diammonium phosphate cigarettes and about 323 higher than the control for the 3.01 d.iammonium phosphate cigarettes. For filtered cigarettes, nicotine per puff is slightly higher for the 2.0~; and 3,0ss diaatmonium phosphate ciearettes relative to the control. No trends or threshold effects relating the small changes in nicotine per puff to added DAP could be estaalished. Nicotine per CPM and nicotine per CPM per puff for unfiltered cigarettes were either less than or not different from the control at all diammonium phosphate additive levels. Nicotine per CPM decreased from 0.074 (control) to a minimum of 0.069 representing a-7~; change. Nicotine per C?M:per puff decreased from 0.0086 (control) to minimum of 0,0077 orith'decreases ranging From about -63.r to -11%. Results for the filtered cigarettes did not correspond to the changes observed for the unfiltered cigarettes. Percent Nicotine Transfer No trend was observed between percent nicotine transfer and increasing levels of diammoniumlphosphate in unfiltered cigarettes. For filtered cigarettes, a positive trend was observed between increasing levels of diazmnoniusn: phosphate and increases in the percent transfer of nicotine from the leaf to the smoke on a per cigarette basis. The percent nicotine transfer on a per cigarette basis increased from 8.21 (control) to 9.9-'k (3.0§ DAP). Comparativa Saioke aH - Filtered Cic7arettes As the eercent, of added diamznortsu~~n phosphate was increased, the compara~tive smoke pH of the cont_ol at 6.74 was not significantly different from those of the diammonium phosphate cigarettes which ranged from 6.68 to 6.79. The zalculated percent unprotonated nicotine in an aqueous solutioi,,would,be a.cout 53 for collected particulates from.the control end the diammonium phosphate cigarettes. Gn -8- ~ Cd N V O

CQNCLCSION: The addition of urea to cigarette tobacco was found to increase smoke pH, but not leaf pH or leaf nicotine. Nicotine per CPM delivery increased with increasing levels of urea in the tobacco of unfiltered cigarettes. The corresponding filtered cigarettes did not show this increase indicating that variations in filtration may have obscured the relatively small effect that urea had on the nicotine per CPM ratio, The addition of either antmonium carbonate or aatmonium bicarbonate to cigarette tobacco was found to increase both smoke pH and leaf pH, but not leaf nicotine. For both additives, increases were observed in nicotine per CPM for both unfiltered.and filtered cigarettes. Also, increases in nicotine per puff were observed for both unfiltered and filtered cigarettes containing added amrnoniuttt carbonate or ammonium bicarbonate. The addition of diammonium phosphate was found to increase leaf pE, but not smoke pH or leaf nicotine, Changes in the nicotine delivery of the diainmonium phosphate cigarettes appeared to be related to changes in puff count per cigarette that were observed for this additive. For exampie, a three percentn add:,tion of diammonium phosphate resulted in producing about one additional puff per cigarette. The nicotine per CPM per puff in the diammonium phosphate additive cigarettes appeared.to decrease compared to the control by 6 to 10~s .. Diammoniuin, phosphate is knosrn to be a smolder retardant (5), and this property may be a factor in the differences observed between it and the other three additives. /1m:v908 XC: T. D. Jessup 5. T. Jones V. Norman J. R. Reid _R. M. Striegel Library GD CD -9- Cp F+ C!T ~ N

REFERENCES: 1. Johnson, J. M.; Striegel, R. N. (Memorasadum to M. A. Sudholt) The Effect of Ammonium Carbonate, Aaaitoniu:n Bicarbonate and Urea on Smoke pX, Smoke Data and Leaf Chemistry, Analytical Support project Q-449, 14 June 1996. 2. Striegel, R. M. [Me:aorandum to M. A. Sudholt] Effect of Diasrononium Phosphate on Nicotine Delivery, project B-446, 4 October 1994. 3. Johnson, J. M. [Memorandum to M..A. Sudholt] Statistical Evaluations of the Effects of Amtttonium Carbonate, Amtnoniurtt Bicarbonate, Urea and Diammonium Phosphate on Smoke pH, Smoke Data and Leaf Chemistry, Analytical Support project Q-449, 24 July 1996. 4. HF and Ph. F. Reemtsatn Gatbh & Co., 2000 Hamburg, DE, German Patent 3 150 582, June 30, 1583. 5. The Merck Index, Tenth Edition, Merck & ,Co,_Znc., 1983, p. 561.