Tobacco Documents Online

F nt, L .e _. L Gr-lc<l% ,. x W I~TTROS~ITION OF NORNICOTINE AND NICOTINE IN ~'` GASEOUS PWZ'IXTURES AND AQUEOUS SOLUTIONS ,. G.B.~NEMTH,. M~ MCER & F.G. PEIN L_Mi~croanaZyticaZ Laboratory, FlamEncrg, FederaZ' RepubZzc o,f Germany .. . -; . . , . .. .:, :... . ..., .. , . . •., .. , . .. ..-: :,: :. , ; . ,,. :'9' •': : Rxus &' Kuhn (1973) reportedithe occurrence of 40'ng,nitrosonorni'cotine (NNN) in smoke originating,from cigarettes containing tobacco particularly richlin nornicotine (1.957)'. Hoffman et al. (1975) found 137.1 ng/cigarette NNN in mainstream smoke from a US blended cigarette (85 mm). These authors also discussed the formation of NNN from nicotine in relation to the re- . sults of Lij'insky & Singer (,1975),, w.io reported that tertiary bases, in- c]lud'ing,several widely-used drugs and other bases„ react with nitrite to form 1V'-niltroso compounds. Hoffmann et al.(';1974) also reported' the pres- ence of NNN'in unburned tobacco. We therefore decided to study the kinetics of the reaction by wfi3ch NNM'i's formed from nornicotine an& nicotine in gaseous mixtures containing, the reactants in concentrations and'proportions cliosely approaching,those of tobacco smoke. The setting,up of •such~a model system is not too diffi- ¢ult with respect to nitric oxide and oxygen, but it is almost impossible to d'o so for the alkaloids, which in the original smoke occur partly as salts and free bases, partly dissolved' in other smoke constitutents andi partly distributed in aerosol particles. This distribution of the alka- loids, of course, influences their vapour pressure and,, consequently, their access to the nitrosating agent, nitrogen trioxide„ which is formed from the oxidizing nitric oxide and its reaction product, nitrogen dioxide. As a first approach, we decided to use the alkaloids as free bases and according to their vapour pressures at the respective temperatures;, these cond'i'ti;ons rather favoured the nitrosating reaction. The nicotine involved, wasspecially purified and contained less than 10img/kg niornicotine. 0 The unprotectedinornicotine - free base not retarded by intersolubil- ~ ity with other smoke constituents - was nitrosated to abomt-20%-a£ter five minutes:the available time under normal smoking conditions i'sverymuch~ ~ shorter« Of purified n~icoti'neunderth~e sameconditi'ons„ gn1X ©'.6 ha Q~d reacted to form NNN after 1'S minutes at room temperature: during in11.a- tion and exhalation of one smoke puff the rate would be less by several~ ' orders of magnitude. : -227- ' . r ,; }- ! i ; r,

Y 228 NEWRATH ET AL. ~ At 80 and 220°C no detectablle amounts of NNN were formed'. This is obviously due to the negative temperature coefficient of the oxidation oE' nitric oxide (Bodenstein, 1!9'18; Neurath et al.1)1and to the almost complete cleavage of'nitrogenitrioxide, the nitrosating,agent, at those temperatures and concentrations (G'nelin, 1936)4 In aqueous solutions the formation of NNN'was found to be highest at pH 3'.8'; at pH 1!.5' and lower,, and' a't pFT' 5.2' and' higher,, almost no reaction occurred,. • The reaction rate was found tolbe slower in 1 NS buffer than in 0.2'M'buffer solutions. Nicotine,'like'many other tertiary amines (Lijinsky 8' Singer, 1975), is nitrosated in'aqueous solutions but only at a rate measurable in'parts per thousand over a period' of several hours. With longer reaction_times there seemito be two pH-depend'ent maxima. . . , a ,. _ .~ . 's< NITROSATION OPNORN'ICOTINE' AND NICOTINE IN' GASE00& M'ZXTURhTB' BY' NITRIC OR'IIDE'~ AND~ OXYGEN t In order tolapproach the conditions of nitrosation which occur in tobacco smoke aerosols, a number of systems were d'eveloped'. ~ connected one to another with Teflon Twolfour-necked 540 m1'flasks, 1 000 pS/'1 nitric oxide in nitrogen; and 101ul nornicotine or 50 yl1 tubing, were evacuated to about 0.1 mm Hg. "The first was filled with. other and' air let in through the second, so that the whole system was ° through a septum into the second. The two flasks were opened'toleach nicotine (with a nornicotine content of less than 10 mg/kg) were injected adjusted to atmospheric pressure and to an oxygen concentration of about Nornicotine andinicotine, as free bases, were introduced in concen- luK:, corresponQing to that ot clgarette'smoKe. was an excess of nitric oxide at 22'0C and an excess of tobacco alkaloids trations according,to their vapour pressures at 22 and18'0°C, i'.e.„ there at.80pC. The reactions were followed by gas chromatographic determination of' NNN, which was isolated by extraction of the gas mixtures with 10'ml' di- chloromethane at various time intervals, a new run being made for eachh measurement. To verify whether all the NNN'was thus extracted, sequential extractions were carried out,,which revealed' that nolfurther tANN was present in the extracted systems. 22°C less than 1% hadibeen nitrosated'. dary bases. Nicotine was also found to form,NNN, but after 15'minutes at the gas phase, reacts fairly rapidly in a manner similar to other secon- phase as to the proportion of salt to free base and access of free base to components, i.e., und'er optimal' conditions for reaction'in the gaseous ti'ne, as the free base and not hindered by'intersolubility with other The results are.summarized in Tables 1' and 2, which show that nornieo- to the fact that because of the negative temperature coefficient of the At 80 and 220PC no NNN is formed from nicotine. This seems to be due 1 See p. 219. rt t . K~. »'r .~~ , r c.

, `at;ti~ aion o in at a :ed ial :©- to It NITROSATION OF NORNI'COTINE AND'NZCOTINE 229 ~ " . . .. .. . .. . __. . ., ~ . - '. ,i. Table 1. Nitrosonornicotine '(iNN~N)~ formation frow nornicotine in gaseous mixture~s~~ of~ oxi'diiz~iimg~~ n~itriic oxid'e~ at 2'5°C Time~ NNN formed (ug)i (miim) ~_ run 1 trwn 2 5; . 375 390 ~ 10 490! 502 15 580 586 30! 686 60 "765 960 997: ox'3d'ationlof nitric oxide almost no ni'trogen dioxide canibe formedi. Fur- thermore, under these conditions the nitrogen dioxide which, is formed does notcombin~ewith:theabundant nitric oxid'eto forminit'rogen trioxide,the, nitrosating agent, because the equilibrium of the reaction Nfl + N02 ~" N20'3 ' at those temperatures is shifted'far to the left. 'At 100°C, for instance,, and 0.0'1 atm only 0.03% of nitrogen dioxide appears inithe form of nitrogen trioxide;, in totiacco, smoke, however, the partial pressure of nitric oxidee is only 0.0005 atm. These results wouid' suggest that the formation of NNN from nicotine in cigarette smoke makes no real contribution to the nitrosamine content of smoke during,the short time involved im the inhalation and exhalation of a smoke puff' during smoking. Table 2. Nitrosonorni'cotine (NNN) formation from nicotine in gaseous mixtures of oxidrizing ni'tric oxide at 25°C Time ~: NNNi.+' ° . (min~), ~ fiormedl (ug) % of nicotine ni't'rosat'ed,, caliculated' from vapour pressure e: 1.5 4'. 7 0.86'. 30 9.5 1'.73' 60 15.8 2'.88' Y !. .1 n ; t

------~--•-- .^---. .•E < t 230 NEURATH ET AL. VELOCITY' OF' NITROSATION OF NORNICOTINE IN AQUEOUS SOLUTIONS,AT' DIFFERENT pH VALUES' erimentaZ s =ra 0 0 1 In these nitrosation experiments 9'.5 ml of 0.2'or 1.ZVf buffer solution were stirred with~ 81.1 l:imol (T2' mg) norni'cotine and 500 }tmol (U0.5 ml) 1' N' sodium nitrite at ambient temperature. pH values were determined' us3ng,an electrometric pH meter equiped `~ with a glass electrode, before addition of the nitrite solution which causes a positive shift of about 0'.2'units. , V At different time intervals varying from 5-60 minutes, 0.5 ml samples were taken and extracted' with 0'.1 ml dichloromethane, and 2'ul were in- jectedlinto the port of a gas chromatograph for NNN determination. Gas chromatographic conditions were as follows: Varian Aerograph 1445; 3 mi glass column, 2' mm ID; 52 Silar lOC on Gaschrom P'DMCS, 100-120,mesh; column temperature, 250°C; injection port temperature, 250°C; flame .ionizati'onidetector temperature, 250PC; gas flow, 30 ml nitrogen;/min, retention tiine for NNN„ 61mini. For the kinetic evaluation, the inverse of the concentration of'un- ' reacted'nornicotine is plotted against time; the slope of'the resulting line is a measure of the reactionlrate. Resutta be 3.8'for the nitrosation of nornicotinie in 0.2'and'1 M'buffer'systems -(FY'gs 1-3). At pH 5.2' and higher, and at pH'1.5, almost no nitrosation tions at controlled pH val:ues, from 1.5 to 7.3, showed the optimal pH to Gns:chromatographic measurement of tti~eNNN' formed i~niaquieous solu- oacurred. At pH 3.8 and lower,, decomposition of the unstable free nitrous acid competes with the nitrosation reaction. A considerable inhibition, of nitrosation1was observed with L M buffers as compared withr0'.2 M sys- tems (Figs 1 &2). ~ The remarkable influence of pH on the rate of nitrosa!tion of niornico- tine„ which may be of importance in the trapping,of tobacco smoke in ~ ll eous s stems a tan a of'the -t be o io h b u ve is qu y , c mes o en v us w ~ ~ c r s . ~ [NNN ] plotted'at different pH values (Fig. 3). $1 ~ , , a. x ^f~'+6~ . I 5F3 ~ : r~~ -`r.e a i m*u ® IIi V i

m 0 aon N._ 1es 4 x- ® INIITR4SATION' OF NORNICOTINE AN© NI'COTINE' 231 FIG. 1.~ NI~TROSONO~RN~IC0T11NE~ (NNN)! FORMATIONI IN~ ®..2 MI BUFFER AT VARIOUS pHi V'ALUES 1 50 r p~H~~ 30 . .. : , . ~. . . ., pE pH 4.1 pH 4.4 3 [7 4> Y}; i ~ :o- 0.1 0:0'8' 0:06 10 20 30 40, t i m e jm iin)' pH 5:2.

I 232 NEURATH ET AL. FIG. 2'. NIITROSQNORNICOTiINE (NNN) FO'RMATI~.ONI IN : 1i.0 MBUFFER AT VARIOU5' pH VALUES 0.22 0.2 r IJ'~. . . . .. ..2 10 201 30 40 50 60 L' 50 60 IPH3.8 pH 3.6' PH 3.4 PH 4.01 pH 4:a PH 4.4 © I WHI 4.8' ~ GD, ;., E9 Q 0.18. 0.16 0.1141 i 0.12 04 1 0 08 M06 20 30 40' tilme(miml

tand([NNN]'-t) Lf IL 0 0.9 04 . 03 0.6 -0.5 03 02 NITROSATIICMJ OF NORNI'COTINE' AND NICOTINE FIG. 3. EFFECT OF"pH ONIRATE OF' NITROSATION OF NORNICOTINE decomposition of' nitrous acild' t M buFf'er p12,M1 buffer r, 5 0 4 , .4 2' 3 3.21 3.4 3!6' 3.8 4 4;2 4.4 4.6 4.8 ~ PM \

234 ResuZts VELOCITY OF NITROSATION OF NICOT7[NE' IN' AQI7EOUS SOLUTIONS AT DIFFERENT pH VALUES' For comparison with the above study, a similar one was carried' the nitrosation of nicotine in aqueous soluti'ons., eri:merttaZ For nitrosation, about 9'm1 of 0.2 M buffer solution were stirred' with 100 ul of specially purified nicotine and 0.5 ml 1' N sodium nitrilte soluition at 22'or 80°C. The pH values of the solutions were measured with an electrometric pH meter equipped with a gl'ass electrode. At the time intervals shown in Tables 3 and 4'„ 0.5 ml samples of'the solution were taken, made alkaline with aqueous potassium carbonate solution and' extracted with 100 pl dich,l'oromethane; of the extract 2'ul were taken for gas chromatographic d'etermination. The extraction procedure was carried out at the various pH values used in the experiment: noldifference due to pHl of the extracted solution was found. Gas chromatographic conditions were as followss Hewlett-Packard'H°' 1'5751 g,; 1.5m, glass bead column, washed with~1% Silar 5CR'in chlioroform;f temperature programme,, 70--120°C, at 4°C/m3.ni; injectioniport temperature, '.2'70PC; flame ionization d'etector temperature, 330°C; gas flows, 30 ml nitrog,en/min„ 3'0' ml hydiogen/min„ 500'ml' air/'mi'n. Purification of nicotine was carried out by dissolving 15'g nicotine in 1 Nc hyd'rochloric acid brought to pH 3.65 with buffer. About 2 g, sodium nitrite were added and the solution stirred'for 36 hours. After acidify- ing w3th 10', hydrochloric acid', the mixture was extracted three times with d'ichloromethane; the remainder was made alkaline with,50R potassium hy- droxide solution and extracted wi'thieLher. After removal' of the ether'the yniicotine was purified by distil'lationiin vacuo. The boiling-point was determined' as 119°C'. No traces of'nornicotine appeared in the gas chromar- togram. 'IN Hoffmann et al. ('1975) reported yields of 11-7Z NNN' after one hour, depending,onithe pHivalue; however, no data on the purity of the nicotine used was giuen,. In order to avoid'interference from nornicotine almost always present inisamples of natural nicotine, it was purified by nitrosation and separa- tion of nornicotine by d'istillation. The purity of the resulting nicotine,, determined by gas chromatography, was 9'9'.9%;' the final nornicotine con- tent was less than 10 mg/kg. Gas chromatographic measurement of the NNNiformed at 22°C in aqueous sol:utions at measured' pH values varying, from 1.7 to 6.6' showed! that after only one hour at pH 1.7' and pH' 2'.5,, 0.32' and 0'.2'0,g/'kg„ respectively, of the nicotiine,, had been nitrosated. The highest amounts measured!in the series were obtained after eight days at pH 4.0-5'.0 „ when 3'.3 g/kg NNN were found (Table 3').

: on I for d to me ium Y- fth + the ea- NITROSATION' OF NORNIiCOTI'NE ANDNICOTIiNE 235 Table 3. N'iltrosonornicotine formed (g/lkg, of nicotine reacted)at 22'°C at. different pH val;ues ~ Time pH 1 1. 7, pHi 2.5 pH• 4.0' pH 4.2 pH 5.0 ' pH 6.61 30 min 0.21 0.1'0' ND, ND ND ND' , . . ,. . : . 60 mi'n 0'.32 0.20 ND ND, ND ND 2 h, 0.81 0.28' trace . trace trace N,D 4' hi 0.87 0.46 trace trace trace `' ND 21 h 1.13 01.49 1.4.. 1.4 1.0 trace 28'h - - 2.1 1 - 1.7 1.1 2'd'ays 1.21 . 0.49' 2.2' 1.6 1.8 trace 3 days - - 2.3 2.3 2'.3 - 4 days 1.38 0'.4& - - - 0.21 8 days 1.41 01.51' 3.3 3.3 3.3 0.32' 13 days _ _ 3.0 3.0 3.0' - NDl- not detected Table 4. Nitrosonornicotine (NNN)' formed and pH 3.82' (I% of nicoti~ne reacted), at 80'0C Time N,NN (min)I 5 15 30, [me 45 :nt 60 .a_ 1'00! :ne, 120 150 es 1801 s 210 300' 48 hi 1.23' 3.15 4.88 ' - 6.0'6 6.42 7.81r 8.49~ 8.491 8.49' 8!.49' 8.49 6.00

At longer reaction times, 21 hours and more,, there appear to be two pHH maxima corresponding,to two different paths of reaction; however, this finding was not pursued further. - The reaction proceeded faster at 80°C: after five minutes at pH 3.8 served being, reached af ter two hours when 8.49% NNN were f ormed' (Tab1'e 4). more than 17% of the nicotine had already reacted,,, the highest values ob- This study was supported by US Council for.Tobacco Research Grant No. 891. nornicotine in tobacco. Science,, 18fi,, 265-267 Hoffmann, D., Hecht, S'.S., Ornaf, R.M. & Wynder,, E.L. (1974) N'-Nitroso- p. 738 Gmelin (1936) Handbuch de2" anorganischen Chemie, 8.Auf1. 4/3, Weinheim„ und Sauerstoff. Z. EZektrochem., 24, 183' Bodenstein, M. (1918) Die Geschwindigkeit der'Reaktion zwischeniS'tickoxyd Research on Cancer (rARC'Ecientific Publications No.9')„ pp. 111-114. Ntitroso Compounds ire the Envirorvnent, Lyon, International Agency for amines and nitrous acid. In: Bogovski, P. & Walker, E.A., ed.,,IV- Lijinksy,, W. &' Singer, G.M. (1975) Formation of nitroscanines from tertiary Tabakregie, 14, 251-257 Rauchkondensat nornikotinreicher Zigaretten. Fachl. M2:tt. t3sterr. Klus~, H.~&Kuhn, H., (1973')~ D~ie~Bestimmung desNorn2ko~tinnitro~samines~~im tions No.9), pp. 159-165 International Agency for~Research~on~ Cancer (IARC~Scien~tif'i'c~ Publica~-~ P. & Walker,, E.A.,, ed., N-N'itroso Compounds in the Environment,,, Lyon, volatiZe N-nitrosamines and hydrazines in cigarette smoke. In: Bogovski,, smoke. XXVI. On the isolation and'i!dentifieation of'volatile and non- Hoffmann„ D., Rathkamp, G. & Liu,, Y.Y. (1975) Chemical' studies on tobacco