US3298380A - Process for purification of tobacco smoke - Google Patents

Description

United States Patent O 3,298,380 PROCESS FOR PURIFICATION OF TOBACCO SMOKE Eldon E. Stahly, Pompano Beach, Fla., assignor to Oliver W. Burke, Jr.,' Fort Lauderdale, Fla. N Drawing. Filed Dec. 14, 1962, Ser. No. 244,572

1 Claim. (Cl. 131-462) of nickel carbonyl per pack of filtered cigarettes. On the basis of their studies, cumulative exposure to these quantities of nickel was suggested by the Drs. Sunderman to be in a broader sense the term metal carbonyl also includes v the analogous metal nitrosyl carbonyls and the metal hydrocarbonyls, which are similar compounds in which one of the carbonyl groups has been replaced by a nitrosyl or a hydrocarbonyl group, and except where the context indicates to the contrary, the term is employed in said broader sense herein.

Objects of the invention, severally and interdependently, are to provide an improved cigarette having a sub stantially reduced content of metal carbonyls in its smoke output as compared to conventional plain or filter-tipped cigarettes; to provide a process for the purification of tobacco smoke by removing metal carbonyls therefrom by converting the same to non-volatile materials; to provide an improved cigarette smoke from which metal carbonyls are substantially eliminated; to provide an improved method in which volatile metal carbonyls present in tobacco smoke in small amounts are converted to nonvolatile materials in a manner which etfectively prevents their leaving a cigarette in the smoke; and to provide an improved method in which a ligand material is reacted withvolatile metal compounds in the tobacco smoke and converts them into non-volatile compounds.

Other objects and advantages of the invention, and of preferred embodiments thereof, will be apparent from the following-description and from the illustrative examples appended thereto. 1

The invention resides in the new and useful methods and products herein disclosed, and is more particularly defined in the appended claim.

GENERAL DESCRIPTION Conventional tobacco smoke, more particularly conventional cigarette smoke, contains substantial traces of metal carbonyls, and especially of nickel, cobalt and iron carbonyls, and such materials are reported to be toxic and carcinogenic to animals.

Thus, it has been reported in a paper by the Drs. F..W. Sunderman (Sr. and Jr.), based on tests of six different brands of cigarettes, that nickel carbonyl containing from 0.4 to 0.6 microgram nickel per cigarette (corresponding to about 20% of the total nickel of the tobacco) passes through the butt and filter of conventional plain or filter cigarettes, reaching the smoker (Medical Science, page 617, May 25, 1961; American Journal of Clinical Pathology, 35, 203 (1961)). In studies with rats, small amounts of nickel carbonyl were found by the Drs. Sunderman to be carcinogenic. The nickel delivered in the smoke drawn from the butt ends of the cigarettes amounted to up to 8 micrograms nickel per pack of 20, or 23.5 micrograms nickel carbonyl per pack of 20 of unfiltered cigarettes; and up to '12 micrograms nickel or 35 micrograms a possible cause of the so-called smokers pulmonary cancer. Based on the reported figures, a person who smokes a pack of cigarettes per day over a period of a year subjects himself to about one and one-half times the amount of nickel required to induce pulmonary cancer in rats, which are considered to be notably resistant to pulmonary cancer.

Aside from the nickel carbonyl reported, I have found that traces of volatile cobalt and iron compounds are also present, apparently as carbonyls, in cigarette smoke. The cobalt is present in somewhat smaller amounts than the nickel, about 1.3 micrograms of cobalt passing the filters per pack of filter cigarettes; and thte iron is present in about twice the amount of nickel in the smoke of some brands of cigarettes.

Aside from the carcinogenic aspects of metal carbonyls reported by the Drs. Sunderman, it has long been known that such compounds are highly toxic and dangerous materials even in trace amounts. Thus Sax, Handbook of Dangerous Materials, published in 1951 by Reinhold Publishing Company, New York, prescribed a maximum allowable concentration of cobalt in the air as 0.4 part per million, and the Twenty-Second American Conference of Government Hygienists in April 1960 placed the maximal atmospheric concentration of nickel carbonyl for a working day at 1 part per billion (AMA. Arch. Environmental Health 1, 140-144, year 1960). Iron carbonyl is also considered to be toxic, although less toxic than nickel and cobalt carbonyls. Accordingly, the cumulative toxicity effects of these three metal carbonyls in tobacco smoke can be expected to be greater than that reported for nickel carbonyl alone.

Likewise, aforesaid Drs. Sunderman in the above cited paper, demonstrated that cigars and pipe tobacco contain nickel in amounts similar to cigarette tobaccos, and similar amounts of nickel carbonyl per gram of tobacco tobacco, 6 micrograms cobalt per gram of tobacco and over 400 micrograms iron per gram of tobacco, and that over a microgram of each metal per gram of tobacco smoked can reach the smoker in conventional methods of smoking.

Hence it is evident that the quantities of metal carbonyls present in the smoke from conventional plain and 5 from such smoking articles is based on the removal of metal carbonyls from the tobacco smoke by the formation of non-volatile complexes by combination of the metal carbonyls with a ligand, which for the purpose of this invention is a nitrogenous organic compound free of phosphorus, that acts as a complexing agent which forms com- '9 (D plexes of low volatility with transition metal carbonyl compounds in the presence of other constituents of to bacco smoke, more particularly in the presence of moist carbon dioxide. The practice of my invention does not depend on the formation of exact empirical complexes since mixtures of such complexes may be formed with equal benefit for my process.

Thus volatile metal compounds in the cigarette smoke by the present invention are converted to non-volatile complexes by contacting the smoke with one or more ligand organo-nitrogen compounds as above defined. These compounds maybe hydrocarbon or hydrocarbonoxy substituted compounds of nitrogen. Especially suitable as ligands in the present invention are (A) the amide compound ligands which comprise (1) acid amides, namely primary, secondary and tertiary acid amides, the mono- N-substituted primary and secondary acid amides and the di-N-substituted primary acid amides, (2) the diamides namely urea and the N-substituted ureas and (3) diacidamides, and (B) the nitrile compound ligands which comprise (1) nitriles; (2) isonitriles and (3) cyanoamides.

' The acid amide ligands (A above) are represented by the formulae:

RCONR'R RRNCONR R RCONRCONR RCOR and the nitrile type ligands (B above) by the formulae;

RCN; RNC; RR'NCN; wherein R, R, and R and R may be the same or different radicals selected from the group consisting of hydrogen, and unsubstituted, hydroxysubstituted, and amino substituted alkyl, alkenyl, alkynyl, cycloalkyl, aryl, alkaryl, aralkyl, alkoxyaryl, terpenyl, pyridyl, alkylpyridyl, pyridylalkyl, pyrrolidinyl quinolyl, etc. radicals, and wherein the aforesaid radicals of the compound cannot all be hydrogen simultaneously and wherein R is selected from the group comprising piperidyl, alkylpiperidyl, pyrrolidinyl, piperazinyl, morpholinyl and like radicals, the linkage to the CO group being via the ring nitrogen atom of R The restriction placed on R, R, R and R is required because of the low volatility requirements for use in filters for tobacco smoke. As indicated it has been found that R, R, R R and R of the above formulae can have amino substituents. Although the amino group may be inactivated by carbon dioxide of the tobacco smoke, the amide moiety can still afford ligand activity for complexing transition metal carbonyls. The amino-C H CONH is an effective ligand of my invention. Ligands containing from 8 to 24 or more carbon atoms are preferred.

The complexes are formed by passing the smoke containing the metal carbonyls into contact with one or more of said ligands or over or through a filter material acting as a carrier body for the ligand material, and preferably comprising fibrous material, for example, cellulose acetate tow, prepared with the ligand therein or wholly or partly coated with the ligand or with material such as activated carbon carrying the same.

In the practice of my invention for example in making tobacco smoke filters one or more of the ligands or complex-forming compounds or components with or without a solvent or plasticizer are preferably dispersed on solid adsorbents; for example, the liquid or complex-forming components may be vaporized onto an adsorbent material, or a solution of a ligand may be applied to adsorbent material such as carbon, e.g., activated carbon, silica, pumice, vermiculite, clay, asbestos, polyesters, polystyrene, and cellulosic materials, e.g. cotton, cellulose, cellulose acetate, cellulose acetate-butyrate, cellulose propionate, tobacco, and other absorbing materials having a high surface area per unit weight or per unit volume.

The ligand material of the present invention may be incorporated in or with the smoke-permeable bases, carriers or filters which themselves may embody various adhesives, adsorbents and surface area augmenting materials, and such incorporation of ligand material also may be made during the manufacture of the filter ma terial or'filters and'thus such ligand materials can be in"- corporated with the materials and during the processes as disclosed, for example, in US. patents as follows:

Types of ligands useful in my present invention and typical complexes believed to be formed thereby from transition metal carbonyls are exemplified by the following in which x=1 or 2.

Ligand (Y) Typical complex CNR Ni(CO) (Y) RCN Fe(CO) (Y) RRNCN Ni(NO)(CO) (Y) RCONR'R Co(CO) (Y) RR'NCONR R Co (CO) (Y) Likewise, the ligands employed herein form non-volatile complexes with other metal carbonyls, such as:

While carbonyls such as Fe (CO) and Fe (CO) are not volatile at room temperature they decompose giving volatile Fe(CO) at temperatures of C. and higher. Heating above 100 C. aids in the formation of the nonvolatile complexes, particularly with iron carbonyls. The heat of the tobacco smoke which contributes to the formation of the volatile iron carbonyl thus also aides in the rate of formation of the complexes in the cigarette filters of the present invention.

Vapor pressure observed for representative metal carbonyls and indicatlve of the volatility thereof are:

' Ni(CO) 134 mm. Hg at 0 C.;

238 mm. Hg at 15 C.

Co(CO) COH 760 mm. ca. 10 C. C0(NO)(CO) 91mm. at 20 C. Co (CO) 0.72 mm. at 16 C. Fe(COH) (CO) 11 mm. at -10 C. Fe(CO) 26 mm. at 16 C. Fe(CO) (NO) 760 mm. at C.

The alkyl groups preferred as R substituents (which term severally includes R, R, R and R in the organonitrogen ligand are the alkyls containing not more than 24 carbon atoms (i.e., methyl to stearyl). Both primary and secondary alkyls, straight chain and branched chain alkyl groups may serve as R in the ligand.

Alkenyl groups preferred as R in the ligands are vinyl, .allyl, butenyl, etc., up to alkenyl groups containing not more than 24 carbon atoms.

Alkynyl R groups .of the alkynyl ligands are exemplified by propargyl, butynyl, pentynyl, hexynyl, heptynyl, octynyl, etc., up to those containing preferably not more than 24 carbon atoms.

Cycloalkyl R groups of the cycloalkyl ligands are exemplified by cyclo-C -C -C c-yclododeoyl and the like cycloalkyls, ac-tetrahydro-naphthyl, decahydronaphthyl,

mono and dioctylbenzyl.

contains .not more than 24 carbon atoms and may be phenyl, naphthyl, ar-tetra-hydronaphthyl, alkoxy or alkyl phenyl, alkoxy or alkyl naphthyl, or alkoxy or algyl substituted ar-tetra-hydronaphthyl. Finally, R may be an arylene compound.

The aralkyl substituents R in the ligand which are useful in the present invention may be unsubstituted or hydrocarbon substituted and range from phenylmethyl to phenyldecyl, and from mono and dimethyl-benzyl to Aralkyl and arylene groups containing not more than 24 carbon atoms are preferred.

Advantages such as non-toxicity accrue to the use of amide-s and of cyano .compounds of carbohydrates such as the cyanoethylated derivative of cellulose, sugars, starch, dextrin, .and the like as complexing agents, particularly when softened by a plasticizer or solvate.

If the ligand material is water-soluble then it is preferable to use a-recessed filter or a filter free of ligand material in the portion contacting the mouth. For filters that contact the mouth non-water soluble ligand material is preferred. In either case it is desirable to use low-volatile ligand material.

. EXAMPLES The invention as above disclosed will be more fully understood by reference to the following examples which are to be taken as illustrative and not restrictive of the invention, and which are conducted in three series, namely:

Series A, in which a synthetic gas mixture of volatile metal carbonyl and carbon monoxide was prepared containing small quantities upward of 100 parts per million of the metal carbonyl (at least equal to the maximum proportion thereof in tobacco smoke), and in which the ability of typical organo-nitrogen ligand materials to extract such traces from the diluent gas was established.

Series B, in which the same synthetic metal carbonyl gas mixtures were employed to test the eflicacy of typical organo-nitrogen ligands of this invention in single pass cigarette filters.

1 Series C, in which the synthetic metal carbonyl gas mixtures were replaced by tobacco smoke and the advantage of the invention in reducing the metal carbonyl contact .of such smoke was demonstrated.

For the series A and series B example, metal carbonyls were prepared, or obtained commercially, and put into a gas stream from which they were subsequently removed by the ligand materials employed in the present invention. These tests were followed by the Series C actual cigarette smoking tests wherein ligands were employed as filter components.

Series A Example 1 (Cntr0l)..Nickel tetracarbonyl was generated by passing carbon monoxide of 98% purity from a cylinder over 3.75 grams Raney nickel (2.625 g. Ni on bing solution. Two simple tests were used to demonstrate the presence of the nickel tetracarbonyl in the gas: (1)

the flame of the burner displayed a grey luminosity when the nickel carbonyl was present in an amount as low as 1 p.p.m. and (2) pinpoint heating of the glass outlet from the generator deposited a spot nickel deposit or mirror on the wall of the glass tube. As litle as 10 mole of nickel tetracarbonyl gave an observable metallic nickel deposit in this test, and thus was used as a sensitive test. Also, the weight of a mirror formed in a small glass tube served to establish the metal carbonyl content of a measured quantity of gas.

When the gas at 200 cc./minute was shown to contain at least p.p.m. of nickel tetracarbonyl (by mirror test) Example 2 was carried out. The flame of the burner showed a strong grey color in this range of concentration of nickel tetracarbonyl. This test demonstrated that the scrubbing apparatus itself did not decompose the nickel carbonyl, and did not remove the nickel carbonyl from the gas.

Example 2 (Control) .-100 ml. of benzene was placed in the scrubber of Example 1 and the gas rate from the nickel tetracarbonyl generator was held at 200 cc./ minute. Appreciable removal of nickel tetracarbonyl was not observed either by mirror test or change in intensity of the grey color after about 10 minutes of passing the gas through the scrubber. This example demonstrated that a solvent alone in the absence of a ligand does not remove the nickel carbonyls from the gas stream to an effective degree because of the appreciable partial pressure of the nickel carbonyl in the solution.

Example 3.A solution of 10% stearic amide in benzene was substituted for the benzene of Example 2 with complete removal (by mirror deposit test) of the nickel carbonyl from the gas stream.

Example 4 .--A mixture of cobalt tetracarbonyl hydride and cobalt tetracarbonyl was generated according to the precedure of Gilmont and Blanchard (Inorganic Syntheses, vol. II, pp. 238-243, McGraw-Hill Book Co., Inc., 1946). The hydride in the glass trap was volatilized at about 30 C. by passing a stream of carbon monoxide into the trap, the outlet of which was attached to the fritted glass scrubber of Example 1 containing a 20 wt. percent solution of hydrogenated t-allo-w acid amide in benzene. The exit gas from the scrubber contained nil cobalt compounds by the mirror test and the flame test.

Example 5.Example 4 was repeated employing cis, trans, trans-1,5,9 cyclododecatriene for the benzene in preparing the 20 wt. percent tallow acid amide solution. The exit gases showed complete removal of volatile cobalt carbonyl compounds by mirror and flame tests.

Example 6.Example 4 was repeated except that the scrubber was replaced with an absorber tube containing about 5% stearamide on finely divided decolorizing carbon (20 g.) (No. 1551 from General Chemical Division of Allied Chemical Corp.) At a gas rate of '50 ml./min ute, substantially all of the cobalt compounds were removed from the gas.

Example 7 .Iron tetracarbonyl dihydride was prepared from iron pentacarbonyl (Antara Chemical Company) by the method of Blanchard and Coleman (Inorganic Syntheses, vol. II, pp 243-4, McGraw Hill Book Company, In-c., 1946). The trap containing the iron tetracarbonyldihydride (1 gram) was allowed to warm (from 70 C.) while passing a stream of carbon monoxide therethrough. The carbon monoxide gas containing the small amounts of iron carbonyl and iron carbonyl hydride was passed through the tube containing 0.5 g. oleic acid amide absorbed on 20 g. Philblack 0, an iron-free carbon black (produced by Phillips Petroleum Co.). At a rate of 100 ml. gas/minute, the iron compounds were substantially completely removed from the gas, the carbon black absorbate after 40 minutes showing an iron content of about 0.45% (0.092 g. iron). Substitution of activated carbon for Philblack O in this experiment gave substantially the same results.

The conclusions from the teachings of this example are that the ligands of this invention can be employed with cigarette, cigar or pipe filters containing carbon, more particularly activated carbon.

Example 8.Ten grams diallylcyanamide in 100 grams benzene were placed in the fritted glass scrubber of EX- ample 2 and Example 3 was performed using the conditions of Example 2. Substantially complete removal of the nickel carbonyl from the gas stream was observed by the flame test.

Examples 9 and ]0.Example 8 was repeated using a solution of isophthalodinitrile in xylene (2% solution) in the scrubber. After 10 minutes the scrubber had absorbed about 0.5 milligram of nickel. Repeating the example with o-phthalonitrile gave substantially the same results.

SeriesB Example 11.- Cigarette filter grade cellulose acetate tow of about 40% acetyl content, 5 denier/single filament, 100,000 total denier, was used as filter material. One and one-fourth grams were soaked in 100 g. of benzene solution containing 5 g. la-uric amide. The material was dried (1.5 g.) and placed compactly in a glass tube into which it fit snugly over a length of 6 inches. This filter tube was attached to the nickel carbonyl generator of Example 2. At a gas rate of 30 ml./minute for 30 minutes the nickel tetracarbonyl could not be detected in the exit gas from the tube. By spectrochemical analysis the filters were found to contain 0.015% nickel (vs. 0.000045% before use), or a total of 224 micrograms nickel absorbed.

Example 12.Example 11 was repeated using a 6-inch length of the uncoated cellulose acetate tow as used in Example 11. The analysis showed that only 10 micrograms of nickel were retained by the filter, as compared with 224 micrograms nickel retained by the ligand or complexing agent of Example 11.

Example 13.Two grams niacinamide were dissolved in 48 g. isopropanol and 5.07 g. of cellulose acetate tow of about 40% acetyl content (5 denier fibers) were immersed therein for minutes. The tow was drained and dried in a 70 C. oven. The dried tow showed a content of 10% niacinimide. About 0.4 g. of the tow was placed in a glass tube as in Example 11 to form a 2" long filter in the exit line from the nickel carbonyl generator of Example 1. Passage of 400 ml. of the gas during 10 minutes showed an adsorption of about 90 micrograms of nickel.

Example 14.--Example 13 was repeated using about 40% urea on the cellulose acetate tow in the filter. About micrograms of nickel was adsorbed in comparison with 4 micrograms in Example 12 for a 2" length of uncoated tow.

Series C Example 15.-1,580 g. of cellulose acetate tow described in Example 11 was coated by immersing in an alcoholic solution of octadecyl amide of acetic acid, draining and drying at about 80 C. to give a filter containing 3% of the amide. The filter material was divided into 10 parts and each portion was fitted snugly into a glass-tube holder fi in diameter and was used as a back-up filter to smoke 2 filters cigarettes (Brand B) for each coated filter section. The total nickel content of the tobacco of these cigarettes was about 190 micrograms by spectrochemical analysis. The 10 coated filter sections after use in smoke filter were combined and analyzed for nickel content. The filter sections supplied withthe cigarettes were removed from the butts and analyzed for nickel content. The uncoated filters showed a total nickel content of about 35 micrograms, and the back-up coated filters showed a total of about 65 micrograms additional nickel content which had passed the mamiiacturers filters Example 16.The test of Example 15 was repeated using uncoated cellulose acetate tow as back-up filters for Brand B cigarettes. The total nickel adsorbed was 9 micrograms on the back-up filter, showing the advantage of my ligand in Example 15.

Example 17.-Example 15 was repeated using cellulose acetate tow of Example 13 coated with niacinamide to give a filter material with 10% of the amide. The filter material (5.5 g.) in 2-0 portions was used as a backup filter for smoking two Brand B cigarettes per portion, Le, a total of 40 filter cigarettes (Brand B) showed a pick-up of approximately 1.45 microgram nickel per cigarettes (vs. 0.12 microgram in the filter supplied with the cigarette). The iron content of the back-up filter containing my ligand showed a pick-up of 78 micrograms per cigarette smoked vs. 63 micrograms per cigarette pick-up by the filter supplied with cigarette. The cobalt absorption was 0.2 microgram per cigarette smoked for the back-up filter vs. 0.1 microgram for the maunfacturers filter.

Example 18.--A filter was preparedfrom celluose acetate tow described in Example 11 by coating with diallylcyanamide. The filter material contained about 13% of the cyanamide derivative. Example 17 was repeated using this cyanamide-coated filter instead of the niacinamide-coated filter. The back-up filters picked up about 1 microgram of nickel and 10.4 micrograms iron per cigarette which had passed through the filter supplied with the cigarette.

Example 19.Cellulose acetate tow (5.000 grams) of Example 11 was immersed in a mixture of 12 g. isopropanol, 68 g. benzene and 3 g. oleic amide for 15 minutes. The tow was drained and dried to give a filter material containing 8.4% amide. The filter material was cut into 20 portions of 0.273i0.005 g. each, and each portion was used in a glass tube of *7 "'I.D. as a back-up filter for smoking two Brand B cigarettes. After smoking, the combined filters showed the following amounts of metals absorbed per cigarette smoked:

While there have been described herein what are at present considered preferred embodiments of the invention, it will be obvious to those skilled in the art that modifications and changes may be made without departing from the essence of the invention. It is therefore to be understood that the exemplary embodiments are illustrative and not restrictive to the invention, the scope of which is defined in the appended claim, and that all modifications that come within the meaning and range of equivalency of the claim are intended to be included therein.

I claim:

A process for the purification of tobacco smoke from a smoking article by the removal of volatile metal compounds therefrom, which comprises passing said smoke through a filter into the mouth of the user down stream therefrom, said filter comprising a ligand compound'containing transition metal chelating groups, said compound being a cyanoamide.

References Cited by the Examiner UNITED STATES PATENTS 134,713 1/1873 Turner 131208 2,171,986 9/1939 Poetschke 13l9 2,815,760 12/1957 Schreus et val. 131-208 2,857,249 10/ 8 Wolfe 23205 2,886,591 5/1959 Lautenschlager 252-430 (Other references on following page) 9 10 UNITED STATES PATENTS Degering: Organic Nitrogen Compounds (1945) pub. 2 933 4 0 4 /19 0 Richter 131 2 3 by University Lithoprinters, Ypsilanti Michigan, pp. 515 2,968,306 1/1961 Touey et a1 131208 and 51 6 especially cited. 3,026,226 3/1962 Touey et a1 131208 Merck Index (Seventh Edition), pub. by Merck and FOREIGN PATENTS 5 C0., 1960, p. 8 and 134 especially cited.

Mgrdichian: (Text) The Chemistry of Organic Cyano- 2 5 fi gg ggzgifi gen Compounds, 1947, published by Reinhold Pub. Co., 932,560 12/1947 France. NC/C, pp- 10 and 9,664 1907 Great Britain. 10

OTHER REFERENCES SAMUEL KOREN, Primary Examiner.

Alien Property Custodian Application of Laude, Ser. MELVIN D. REIN, Examiner. No. 261,049, published May 11, 1943.