Tobacco Documents Online

.4 division of technoliogy for society EFFECT' OF' S;rLMPL.I'*IG LOCATION ON TEiE, "'EASURE!SEA? OF S*i:.'P.E CflNCEN- TRATION IN A GLASS MJDEL • OF' TP.E tiL'"SA.V! *11DtM by M. Hissinx 2q w C,_ 6 Htol hether7anCS o!g;irnrratiora lor apptret7 i sCtentalrC rest•arch 9= p.o. box 342: 7300 ANapeldoorn addresss laan van westenenk 501 telex 36395 tnpap phone 055 . 77 33 W' In co1i1'aboratiom with R. Bos.:.an, ' Fillie no. : 8712-13154_ W.A', de Jomg-K:!ehezg, M.M'. :e1lek~oop and' R. Verleun. Reference no.: January 1985 85-0313 Principal Bzos.zr b Williasason Tob2cco Corp. P . 0. Box 35090 Louisville, Ky, 4©232'. U. S'.A. ..No part OtithrS repp,• SB'all be rfqyl• lipliec andjor pufliiS'+e0 by way of prwnt 2, icU75t~a1 IIDpy.. mbCrplUm.on Otrierw:S'e: without TNQ'S prrOr cDn. Sent .. " ca's re"n n&s :>eern rna0e accord- ing 10~nSPrwCtuonS.o the rrQntlSs an0 o3i,r.at.on5 of the principa( sn0 TNC arE as :'a.C down in the ..StJn- CallC Co'.n©aionS ror •esearrn. a^Ij Oe.eioDr'ient insl-uCtions to TNO,. ~979._ or -n rne re.e.an, agreernenc rCaCneCC t)[".weett;r.e.7ari+es.. f ' OR! _Nrl-n..r.-,.

85-n313'lhtl _ 2 _ CONSENTS Page t . 2'. I\'TRODR)CT'ION DESCRIPTION OF EQ'h1?JEN? USED 4 34 EX,PER'I_MENQAS. PROCEDURES 5 3, 1 S®ok ing and sampl i'ng 5 342 Sample preparat ion. 3 . 3'.3 Nicotine analysis 6. 3'.4 !leasurement of tar 6 4,. RESULTS AND DISCUSSION

1. ISTR0DICTION I!n various experiments samples obcained :rom a smoker's mou~th have been taken as a`possibie measure of' the actual! human in- take of tar and nicorine. To evaluate the experimental approach of these studies an investigation was made to, esta- blilsh the effect of' sampliing locations. For our experiments we used'a glass model of the human mouth.

" 85-0313/htl' - 4 - 2. IDESCR:PI ION OF EQ'UIP^E7v? USED :n figure 1 a diagram of the outline of' the experimental' set-up is given. A 'dental dam' rubber sheet holder is attached and centered across the cigarette insertiow hole. The cigarettes tested were inserted 5 mm deep into the holder. A metal support guarantees the horizontal position during smoking. Smoking puffs are taken with the aid of a FiTtrona 302 smoking cnachine (using one port) connected to the bottom arm of the glass mouth. Smoke samples of 35 m1 each are withdrawn simultaneouslv from the front of the mouth and the throat by using,two ports of a second Filtrona machine. The pa ticulate matter is collected on glass fibre filiters mounted in Gelman filter holders. A glass three-way valve at the bottom arm, end allows adaission of air chrougha magnetic valve and purgitag of the mouth by applcing, vacuum. The operation of the smoking machines and, the magnetic valves is controlled by electronic timers. A Hec.ett Packard' gas chromatograph model 5880 is used for the - analrses of the nicotine in, the filter extracts. For the measurement of tar in the extracts a Perkin. Elmer Model 3000 spectr.fluoriaeter was used.

85-0313;ht1 - 5 - 3. EXFER2."CEti?al. PROCED'C'RES _ 3.1 Smok i'ng and s awp 1! ing S'tep-by-step, smoicing and sampling procecures are listed below: L. :se fresh sample probes after three.cig,aret't'es have Seen smoked. 2'. Insert cigarette into the dental Cam holder. 3. Light the c'igarette by activating the first smokia£ rar ahine. . 4. Remove the cigarette and, turn three-way valve to connect the glass mouth to the vacuum pump and remove all srroicee from the mouth. 5. Switch the g1'ass val've to position, which connects g?ass mouthi to the magneti~c valve. Replace the cigarette. 6. The electronictimer activates the second smoking machine exactly 1.5 sec after the next puff' is' finished' to take the samples of 35 m,l' in 2 sec. At the same time the mag- netic valve is opened diu'ring, 6 sec. 7. Repeat steps 4- 6 until 5 puffs have been sampled. 8. Remove and extinguish the cigarette and vacuum out at:v residual smoke f'rom the mouth. 9. Repeat steps 2 - 8 until 3'cigarettes have been smoked. 1'0. Remove and iabel the mouth and throat Ge1aan pads. 3.2 Sample~preparation Each f'il ter Yad is remotied and p:aced'l in a 20 m'_ PP •?E syringe (01N 13098 Type AY, which•is r'insed!with water and methanol successively. A 1/41 of a fresh pad is: used to wipe ~ the ir.side of the Gelman ho:'d'er after which it is addedl to ~ N the syringe. 110' ml methanol is added:. After clos:r.g, the sy- ~ ringt by the plunger and a clamped tubing,at the tip, it is (O. treated f'os 30 Qin in an ultrasonic bath. The s::s~,e::sian o5- ~ ~ cained is f'il!cezed tnrough a yil1_ex-SR 0.5 _s fii'cer uTit chicio 0 ~

85!-0313/htl, - 6 - ils mounted at the tip of the syringe in place of the ciar..ped tubing. Of' the filtrate 1I ml is used'for nicotine analysis by CC and the remainder is used for f.uorescence measurements. 31.31 Yirotine analysis The filter extracts are anal.•.sed' for nicotine by gaschrorato- graphy with the following conditions: Column: 50 m of'fused silica, i.d. 0.32 = coated with CPSil' 5, thickness 0.4 ;,m. Coiumn, temperature 100 C; after 1 minute the temperature is raised to 1180 °C in 20 minutes. Detector: F1ame.ionizati'on; temperature 300 °C. Carrier gas: Helium, flow rate 25 ml/min. Injection: 1;l, splitless during 30 sec. Injection port cem- perature 2001`C. Calibration so1'utions are prepared by dissolving known amounts of nicotine in.methanol. 3.4 Measurement of' tar The tar content of the exrsact, solutions is determinedl by spec- trof'luorescence: wavelength of excitation 475 nmy, enission, measured at 525 nm. Calibration solutions are prepared by extraction with methanol of f il ter pads with knovn, amounts of vet TPM. From these soliur- tions appropriate standards are made by dillution with methanol. Calibratilon runs are made, at least once per day. A11, ezi'ssi~on readings are corrected for blanks, which 3re equivalent to about 0.2 _g/puff. It is found that caldbration factors for dif'ferent cigarette 5rands are rather similar.

g3-03131htU tiote: The use of absorbance measurements at 447 nm ras also tried for the.dietermination of tar content of extracts. This method' suffered' from (very) high blank readings and less sen- sitivity compared'to spectrofluorescence.

85-03'13 /ht1 4. A,T_S'L1TS A.'~D DI'SCCS'SI'0'h Nine brands of'cigarettes, obtained on the Swiss carket, verrd u~s~~ed ~ in the intiest~igat'io'~n- The~ tar and nicotine~ de~~l iveries~ (mr~Icis.) were: labeled as follows: Tar laicot ine B'arclap• 1 0'.2 Select Extra 1 0'.1 *fary Long, ao. 1 i 0.1 Peter' Stuyvesaint C1tra Mild 1 0.1 Flint L'ltra 1 0.1 Stel.a Super 9 0.9 Kent 8 0.8' Parisilenne Douce 1!0; 0.8. wuratti Ambassador 10, 0.8' Six replilcate experiments (3 cigs./f'il'ter)~ were carried' out for each brand. All results' for tar an& nicotine are expressed in micrograms per puffx.. Table : shows the mean result's' and the standard deviations. Total amoants and mouthitYnroat ratios for tar and nicotine as well as tar/nicotine ratios for mouth and throat are g;iven in Table 2. From these results it can be observed that the total amounts of tar and nicotine found' are generally in agreement with the labeled deliveries. This is true in spite of striking diffe- rences in the mouth/:hroat ratios. In particular it is noted that these ratios for the Barclay and!Flint cigarettes are dif- ferent from those for the other brands. Visual observation, oi the smoking and sampling process makes clear that this is dlue to a, significant difference in the flovw pattern of the smcKe in the mouth. The smoke from a Barclay' cigaret,e is d.spersed tiomogeneous,y before the start of the sazrling operation due to the turbulence induced by the filter.

85'-0 313 V hr 1i Table 1. Means and' standard deviations for tar and n4cotine iin mouth and threat' samples Mouth samples Throat samples Tar Nicotine Tar Nicotine Brand x s x s x s x s Barcla5 • 2!0 8' 1.91 0.72 II, 2.8 1.6, 0.48 I 0.1I9. z;e1'ect Extra 5.5 1.2 0.52 0.17 ' 24 4 2.81 0.62 : ary Long So. 1 12 3' V.05 0.20 24 9 1.88 0.55' Peter Stuy~vesant Ultra 12 3' 0.85 0.34 i t 6 ~ 0.92 0.24 Flint C'lirra 20 2 1.63 0.15 ' 4.5 i 1.8 0.38 0:1i7 Stella Super 29 7 2.46 0.40 23!9 36 18.4 3.34 Kent 59 8 3.96 0.22 116 21 7'.72' 1.69 Parisilenne Douce 45 3 3.95 0.38 97 22 i 8.10 1.49 .uratti Ambassador 132 33 . 7.66 1.92 141! 601. 18.45 4.00 . All results ir. ;:glpuff'. For the Flint cigarette the fairly homogeneous distribution of the smoke in the mouth appears to result from a rebound ef- fect fros the back walls of the mouth. This rebound effectt seems to be due to the low density of the smoke of a Flint ci- garette. For the other brand's the smoke column penetrates to a large extent down into the throat during puffing. Although during sampling the flow is returning to some extent, the net result is a cuch lower mouth/thrcat ratio than for Barclay and Flint. During,a number of experiments a video recording was mad'e of the processes occurring,in the mouth. A study of this video tape explains d!iff'~erences observed in the mouth/throat ratios of' some brands. Apparently the density of the smoke aerosol plays

85-0313/ha1. Table 2. Total amounts, mouth/throat ratios and tar/nicotine ratios. - 10 - Blrand Total amount Mouth/throat ratio Tar/nicotine. ratio Tar Nic. Tar Nic. mouth throat Barclay 23' 2.39 7.0 4.0 10 6 Seliect Extra 29 ' 3.33 0.23' 0.19 11 8 Mary Long No. 1 36 2.93 0.50: 0.56 11 13 ,Peter Stuyvesant Ultra 23 i 1.77' 1.0 0.92 I 14 12 F1int Ultra 25 i 2.01 4.5 4.3 13 12' Stella Super 268 20.86 0.12 0.14' 12' 13 Kent 175 11.8i6 0.50, 0.51 15' 15. Parisienne Douce 142 12'.05 0.46' 0.49 tlt 12' Huratti Ambassador 272 16.11 0.94 0.91 17 17' All results in jig/puff'. a significant role in the flow behaviour in the mouth. It is aoc'evor'thq that the taz/nicoti'ne ratio for the mouth, and throat samples for a particular brand generally agree very vell. Also between the various brands the agreement is generally gooC For the few exceptions to these general rules there is no exii nation up to now. In suam ary it can be concluded that the design of experiments -with in-mouth samrpling of ci garetre smoke must have..diue cons ide- rat'ion. The interpretation of the results obtained:must be made with care.