Lipson, Menzie (Highton, Victoria, AU)
Mayfield, Robert John (Newtown, Victoria, AU)
Morgan, Irene Mary (Grovedale, Victoria, AU)

05/386325

04/22/1975

08/07/1973

Download PDF 3878853       PDF help

Click for automatic bibliography generation

Commonwealth Scientific and Industrial Research Organization (Campbell, AU)

131/334

A24D3/16; A24D3/00; A24B15/02; A24F7/04

131/262,267

Michell, Robert W.

Millin V.

Bacon & Thomas

We claim

1. A method for producing a tobacco smoke filter which will selectively remove ciliatoxic smoke components, which method comprises forming an aqueous solution containing a non-volatile polyamine component having a molecular weight in excess of 500 and a cationic component which contains one or more quaternary ammonium functional groups per molecule, and having a proportion by weight of cationic component to polyamine component of from 10:1 to 1:10; and treating a filter material with said solution to deposit an amount of each of said cationic and polyamine components from 1 to 30% by weight of said filter material.

2. A method as claimed in claim 1, wherein the cationic and polyamine components are functional groups of separate compounds.

3. A method as claimed in claim 1, wherein the cationic and polyamine components are functional groups contained in a single compound.

4. The method of claim 3, wherein the polyamine is poly(ethylenimine) quaternized with cetyl bromide.

5. A method as claimed in claim 1, wherein the filter material is treated with an aqueous solvent, or a mixture of an aqueous and a water-miscible organic solvent, containing said components.

6. A method as claimed in claim 1, wherein the polyamine component is non-volatile, has a molecular weight in excess of 500, and contains a high proportion of primary and/or secondary amino groups.

7. A method as claimed in claim 6, wherein the polyamine is a poly(alkylenimine).

8. A method as claimed in claim 6, wherein the polyamine is a poly(ethylenimine) or poly(vinylamine).

9. A method as claimed in claim 4, wherein the polyamine contains a cross-linking agent.

10. A method as claimed in claim 1, wherein the polyamine is produced by reacting a vinyl polymer having pendant groups selected from the class of epoxy, carbalkoxy, acyl, and isocyanate with an amine.

11. A method as claimed in claim 1, wherein the polyamine is produced by reducing a vinyl polymer containing nitrile groups.

12. A method as claimed in claim 1, wherein the polyamine is produced by reaction of an epoxide or an alkyl polyhalide with ammonia, a diamine or a simple polyamine.

13. A method as claimed in claim 1, wherein the cation is an organic compound containing at least one functional group carrying a positive charge.

14. A method as claimed in claim 13, wherein the cationic component is selected from the group consisting of polyelectrolytes having an ammonium functional group on the polymer backbone or on the side chains.

15. A method as claimed in claim 1, wherein the amounts of each said component deposited is from 2% to 12% by weight of the material.

16. A method as claimed in claim 1, wherein the said proportion is from 5:1 to 1:2.

17. A method as claimed in claim 1, wherein the filter material is treated by polymerising a polymerisable amine in the presence of the filter material.

18. A tobacco smoke filter for the selective removal of ciliatoxic smoke components comprising a filter material having associated therewith a cationic component which contains one or more quaternary ammonium functional groups per molecule and a non-volatile polyamine component having a molecular weight in excess of 500 wherein the amount of each cationic and polyamine component associated with said filter material is from 1 to 30% by weight of said filter material; and the proportion by weight of said cationic component to said polyamine component is from 10:1 to 1:10.

19. A filter as claimed in claim 18, wherein the polyamine component is a poly(ethylenimine).

20. A filter as claimed in claim 18, wherein the said components are provided by a poly(ethylenimine) which has been modified to contain cationic groups.

21. A filter as claimed in claim 18, wherein the filter material is wool or cellulose acetate.

22. A cigarette or other tobacco-containing article comprising the filter of claim 18.

23. The method of claim 1, wherein the polyamine component is poly(ethylenimine) and the cationic component is selected from the group consisting of mixed alkyldimethylbenzyl ammonium chlorides, alkyltrimethyl ammonium bromide, a mixture of methyl dodecylbenzyl trimethyl ammonium chloride and methyl dodecyl xylene bis(trimethyl ammonium chloride) and quaternized poly(ethylenimine).

24. The tobacco smoke filter of claim 18, wherein the polyamine is poly(ethylenimine) and the cationic component is selected from the group consisting of mixed alkyldimethylbenzyl ammonium chlorides, alkyltrimethyl ammonium bromide, a mixture of methyl dodecylbenzyl trimethyl ammonium chloride and methyl dodecyl xylene bis(trimethyl ammonium chloride) and quaternized poly(ethylenimine).

25. The tobacco smoke filter of claim 19, wherein the amount of each of said components associated with said filter material is from 2 to 10% by weight of said filter material.

26. The tobacco smoke filter of claim 19, wherein the proportion of cationic to polyamine components is from 5:1 to 1:2.

This invention relates to filters for cigarettes and other smoking articles. In particular, the invention concerns a filter which selectively removes ciliatoxic compounds from tobacco smoke.

Tobacco smoke, as is well known, is a two-phase system consisting of a particulate phase (spherical droplets of the less volatile components) suspended in a vapour phase containing the more volatile components. Cigarette filters fabricated from fibrous or laminated materials have varying capacities to remove the particulate phase of tobacco smoke by mechanical means. However, vapour phase components are not readily eliminated by mechanical filtering and pass essentially unaffected through these filter materials.

In recent years cigarette smoke, particularly the vapour phase, has been shown to strongly inhibit ciliary activity in systems devised to test this phenomenon. Ciliary activity has long been recognized as an important defense mechanism for the removal of foreign matter from the lungs and respiratory tract. Inhibition of these functions by the ciliatoxic compounds in tobacco smoke could predispose smokers to respiratory infections or aggravate already existing ones and may even play a role in the pathogenesis of lung cancer. Hence the development of a filter which removes ciliatoxic compounds from tobacco smoke is highly desirable.

Many of the ciliatoxic components of tobacco smoke are volatile compounds which are mainly present in the vapour phase of the smoke. A variety of adsorbents have been used in cigarette filters to remove volatile components of the vapour phase. These include activated forms of charcoal, alumina, silica gel, and diatomaceous earths. Such adsorbents suffer from the disadvantage that they are not very selective and remove a variety of smoke components thus having an undesirable effect on the taste of the smoke.

Recently, a number of filters have been disclosed which chemically react with undesirable components of tobacco smoke Such filters are usually prepared by addition of reactive polymers or inorganic compounds to fibrous tobacco materials. These filters have an advantage over adsorbent filters in their higher selectivity bbut usually are not as effective in overall reduction of smoke ciliatoxicity.

This invention provides a filter for cigarettes and other smoking articles which will remove a substantial portion of the particulate phase of tobacco smoke by mechanical means, and which, in addition, will chemically and selectively remove a substantial portion of ciliatoxic smoke components without adversely affecting the taste of the smoke. The ciliatoxic compounds referred to occur in both particulate and vapour phases of the smoke and have been well documented in the literature. Examples of such compounds are formaldehyde, acetaldehyde, acrolein, hydrogen cyanide, formic acid, acetic acid and some volatile phenols.

According to this present invention a tobacco smoke filter material is treated with a mixture comprising a cationic component and a high molecular weight polyamine component. The cationic and amino functional groups may be combined as a mixture of two separate compounds or in the form of a single compound, for example, as in a partially quaternized or protonated polyamine. The components may be applied to the filter material either from a single aqueous solution or from a single aqueous solution containing, in part, a water-miscible organic solvent. Any of the conventional means known to the art may be employed for application of the solution to the bulk filter material, for example, as by padding, spraying, dipping soaking or the like.

The polyamine component of the mixture utilized in this invention is essentially non-volatile, has a molecular weight in excess of 500, and may contain, alone or in combination in any proportion, primary, secondary or tertiary amino groups. Polyamines containing a high percentage of primary and secondary amino groups are preferred. Some examples of polyamines which may be used are poly(alkylenimines), particularly poly(ethylenimines), poly(vinylamines), and mixtures of the foregoing with crosslinking agents such as diisocyanates, diepoxides, alkyl dihalides and the like. Other polyamines included are those which might be prepared from certain vinyl polymers by chemical reaction. For example, vinyl polymers containing pendant epoxy, carbalkoxy, acyl or isocyanate groups may be reacted with amines, or alternatively, vinyl polymers containing nitrile groups may be catalytically reduced. In addition, the use of poly(alkylene) polyamine resins formed by the chemical reaction of epoxides or alkyl polyhalides with ammonia, diamines or simple organic polyamines is covered by this invention. It is also understood that the polyamines used in this invention can carry additional substituent groups on their alkyl components such as hydroxy, alkoxy, carboxy and the like.

The cationic component which must be incorporated with the polyamine to give the desired effect of this invention may be any organic compound which contains one or more functional groups carrying a positive charge. Quaternary ammonium, phosphonium and sulphonium compounds fall into this category.

Also considered are ionene polymers and other polyelectrolytes in general which contain functional groups carrying positive charges along the polymer backbone or on side chains. Moreover, mixtures of any of the foregoing cationics in any proportion may be used. Cationic compounds of low or high molecular weight containing one or more quaternary ammonium groups are preferred.

As previously mentioned, a single compound containing both cationic and amino groups can be used as the complete treating material in this invention. For example, high molecular weight polyamines of the aforesaid type which have been partly quaternized by chemical reaction with an alkylating agent, or which have been partly protonated by the addition of a mineral or organic acid, would fall into this category. Partially quaternized polyamines are preferred since the reaction resulting in quaternization of the amino groups, unlike that involved in protonation of amino groups, is irreversible.

The treating additives for tobacco smoke filters, as defined herein, are most readily applied to the filter material from an aqueous solution which contains from 0.5% to 50% by weight of the treating mixture. The amount of each component deposited on the filter material can vary from 1% to 30% by weight of the filter material, but is preferably in the range 2% to 12%. The proportion by weight of cationic component to polyamine component can vary in the range 10:1 to 1:10 but for optimum results is in the much smaller range of 5:1 to 1:2.

An alternative procedure which may be used to form the polyamine component on the filter material involves polymerization of an appropriate monomer in the presence of the filter material. An example of this technique, which could especially apply with the use of wool as a filter material, is the polymerization of aqueous ethylenimine or β-diethylaminoethyl methacrylate solutions in the presence of wool fibres with the aid of redox catalysts or other well known polymerization catalysts. A further procedure is to expose filter materials, which may be chemically modified or contain polymerization catalysts, to the vapour or solutions of polymerizable amines.

The pH of the aqueous solution of treating materials used in the treating procedure described herein should be greater or equal to 7. This ensures that the treated cigarette filter is alkaline and therefore is more effective in removing the undesirable acidic smoke components.

Any fibrous, solid or resinous materials which can support a coating of the treating materials described herein are suitable for use with the present invention. For example, materials suitable for the preparation of filters which consist of parallel masses of fibres such as wool, cellulose acetate, viscose, cotton, rayon, polyester or any thermoplastic fibres can be utilized. Alternatively, bonded, fabric, powdered, or other forms of the foregoing may be used in which the fibres do not form a parallel mass. Filters prepared from any form of paper also fall into this category. Additionally, the treatment can be applied as a coating to solid or resinous materials which can be in the form of beads, granules or powders. Finally, multicomponent filters, which incorporate any of the foregoing supports, treated with the materials described herein as a single component in conjunction with other components consisting of filter materials known to the art, are contained within the scope of this invention.

Wool, particularly in the form of a non-woven fabric or carded web, is the most appropriate filter material for use in conjunction with this invention since its use results in filters of acceptable firmness and pressure drop without further modification.

Any pretreatments known to the art for improving the surface spreading of resins can be used in conjunction with the previously described treating process. This refers in particular to any pretreating processes which modify the surface of the filter material making it more receptive to polyamine and cationic materials. Such processes would include the production of anionic surfaces on the filter material by addition of anionic compounds of any nature or by chemical modification. Pretreatments which are known to improve the spreading properties of polymers applied to wool fibres, for example, are dry and wet chlorination, corona discharge, oxidative processes such as peroxide or peracid treatments, or the like. Polyamines, in particular polyethylenimines, are very substantive to cellulosic materials and chemical pretreatments may not be necessary with these filter materials.

A number of tobacco smoke filter treating materials have been disclosed in the prior art which contain amino functional groups. Some materials disclosed have been low molecular weight amines and amides which due to their volatility, have the undesirable property of being transferred in part to the mainstream smoke of a cigarette during its passage through the filter. In addition, these compounds penetrate the surface of some filter materials and are consequently removed from contact with the smoke, thus reducing the desired effect of the filter.

Other materials disclosed in the past have included high molecular weight polyamines such as poly(alkylenimines) which are non-volatile and therefore more desirable for treatment of tobacco smoke filters. Filters treated with these materials have been claimed to remove in part phenols and volatile acids from tobacco smoke.

Tobacco smoke filters treated with a polyamine component together with a cationic component in accordance with the present invention are found to be superior to filters treated only with the cationic component of the polyamine component in the removal of ciliatoxic compounds from tobacco smoke. Such ciliatoxic compounds referred to are formaldehyde, acetaldehyde, acrolein, hydrogen cyanide, formic and acetic acids, and volatile phenols particularly, but probably not exclusively. This claim has been substantiated by the results of chemical analysis and a biological assay carried out on smoke from cigarettes attached to the above-mentioned filters.

The ability of polyamine treated tobacco smoke filters to remove undesirable acidic and aldehydic ciliatoxic smoke components depends largely on the effectiveness of chemical binding of these components to the treated filter material. Polyamines of high molecular weight such as poly(alkylene) polyamines or poly(alkylenimines) are known to possess tightly coiled molecular conformations in the dry state due to intra-molecular hydrogen-bonding. Consequently, in the dry state, many amino groups are rendered unreactive because of steric factors. We have discovered that incorporation of one or more cationic compounds of the type described herein, in the polyamine treatment of the filter material, results in a significant improvement in the degree of chemical binding of the ciliatoxic smoke components to the treated filter. Apparently, electrostatic interactions resulting from incorporation of the cationic component disrupt hydrogen-bonding within the polyamine chains allowing some relaxation of the coiled conformations and thus increasing the number of amino groups available for chemical reaction.

The method of carrying out this invention will be more fully described in the following examples which illustrate some of the preferred embodiments.

EXAMPLE I

An aqueous solution containing 6% (w/w) PEI 1000 (a product of the Dow Chemical Co. consisting of a 33% aqueous solution of poly(ethylenimine), molecular weight 50,000-100,000) and 5% w/w Zephiran (a product of Winthrop Laboratories consisting of a 10% aqueous solution of mixed alkyldimethylbenzyl ammonium chlorides) was placed in the nip of a Peter pad mangle. Dry chlorinated (1% w/w uptake of cl ₂ ) 64's combed wool top (11 g/metre end weight) was padded and dried at 80°C for 10 minutes in a suction dryer. The treated top, which contained 6% PEI 1000 and 5% Zephiran by weight, was equilibrated for 24 hours at 65% RH and 21°C.

A treatment omitting Zephiran was applied to the chlorinated top in a similar manner to give 7.9% by weight of PEI 1000. The treated wool was dried and equilibrated as described above.

Filter plugs (20 mm) were prepared from the treated filter material by drawing short lengths of the sliver through paper tubes (100 mm) and then squarely cutting into 20 mm sections. The end weight of the filter material was adjusted so that the filter pressure drops were mainly in the range 5.6-7.1 cm of water. Filters having pressure drops outside this range were discarded.

Other filters were prepared in a similar manner from untreated 64's combed wool top which has been equilibrated at 65% RH and 21°C.

Filter plugs were attached to standard cigarettes with adhesive tape and the cigarettes smoked under standard and reproducible conditions by an automatic machine at the rate of one puff per minute of 2 seconds duration and 35 cm ³ volume. Each cigarette was smoked to a constant butt length which required 8-9 puffs. The whole smoke was analysed for hydrogen cyanide and steam volatile phenols while the vapour phase obtained after passing the whole smoke through a Cambridge filter assembly was analysed for formaldehyde, acetaldehyde, and acrolein. In addition, a bioassay was performed for the estimation of overall vapour phase ciliatoxicity (and cytotoxicity) by the "hanging drop" exposure method utilizing the ciliated protozoa, Paramecium aurelia. A method for this bioassay has been published: "Cigarette Smoke Gas Phase and Paramecium Survival," W. Weiss, Arch. Environ. Health, 17, 62-64, (1968). Unless otherwise stated, smoking for the bioassay was carried out with an automatic machine set to take a 17.5 cm ³ puff volume of 2 seconds duration at one minute intervals. The vapour phase was obtained by inserting a Cambridge filter assembly in between the cigarette and bioassay chamber. The number of puffs (LD100) required to immobilize 100% of the Paramecia was taken as the bioassay result. All Paramecium survival times reported are the mean of 4 separate determinations.

Table I illustrates the results of comparative tests performed on filters described in Example I.

TABLE I ______________________________________ Filter Treatments Constituents None 7.9% 6.0%PEI 1000 removed (Untreated PEI 1000 5.0%Zephiran (%) wool) ______________________________________ HCN 5 73 73 Formaldehyde 0 63 73 Acetaldehyde 0 19 36 Acrolein 0 17 29 Phenols (S.V.) 56 58 82 Paramecia (LD 100) 19 25 37 ______________________________________

Results in Table I show that wool filters treated with the Zephiran-PEI 1000 mixture are considerably more effective in removing specific smoke components than untreated filters or filters treated only with PEI 1000. In addition, the Paramecium survival time (LD 100) demonstrates a significant decrease in overall gas phase ciliatoxicity for the filters incorporating the cationic additive with PEI 1000.

EXAMPLE II

Cellulose acetate tow (2.5 denier) was padded with an aqueous solution containing 5% (w/w) PEI 1000 and 5% (w/w) Zephiran. The treated tow contained 6.0% PEI 1000 and 6.0% Zephiran by weight. A treatment omitting Zephiran was performed in a similar manner to give treated cellulose acetate tow containing 10.7% by weight of PEI 1000.

The above treatments were dried, equilibrated, and fabricated into filter plugs (20 mm) as described in Example I. Comparative tests were made on these filters and a commercial cellulose acetate filter (20 mm) using the bioassay (Example I type). Results are recorded in Table II.

TABLE II ______________________________________ Filter Material, Treatment Pressure Paramecia and % (w/w) treating drop Survival time material (cm of water) (LD 100) ______________________________________ Cellulose acetate with PEI 1000 (6.0%) and Zephiran (6.0%) 5.6 - 7.1 31 Cellulose acetate with PEI 1000 (10.7%) 5.6 - 7.1 25 Untreated commercial cellulose acetate 6.1 19 ______________________________________

Results in Table II demonstrate that cellulose acetate filters containing PEI 1000 and Zephiran as additives significantly reduce the overall vapour phase ciliatoxicity of cigarette smoke compared to untreated filters or filters treated only with the polyamine component.

EXAMPLE III

Corona treated wool top was padded with an aqueous solution containing 5% (w/w) PEI 1000 and 5% (w/w) Vantoc N (a product of ICI, England consisting of an alkyltrimethyl ammonium bromide in powder form). The treated top, which contained 5.6% PEI 1000 and 5.6% Vantoc N by weight, was dried, equilibrated, and fabricated into filter plugs as described in Example I.

EXAMPLE IV

Corona treated wool top was padded with an aqueous solution containing 5% (w/w) PEI 1000 and 5% (w/w)Hyamine 2389 (a product of the Rohm and Haas Co. consisting of a 50% aqueous solution of a mixture of methyl dodecylbenzyl trimethyl ammonium chloride and methyl dodecyl xylene bis(trimethyl ammonium chloride)). The treated top, which contained 5.7% PEI 1000 and 5.7% Hyamine 2389 by weight, was dried, equilibrated, and fabricated into filters as described in Example I.

EXAMPLE V

Corona treated wool top as padded with an aqueous solution containing 4.5% w/w PEI 1000 and 4.3% w/w Chemiquat epsilon (a product of the Chemirad Corporation, U.S.A., consisting of an aqueous solution of quaternized poly(ethylenimine)) The treated top, which contained 4.7% PEI 1000 and 4.5% Chemiquat epsilon by weight was dried, equilibrated, and fabricated into filter plugs as described in Example I.

EXAMPLE VI

Corona treated wool top was padded with an aqueous solution containing 5% (w/w) modified PEI 1000 (pH reduced from 10.0 to 8.0 with hydrochloric acid). The treated top, which contained 5.6% by weight of PEI 1000, was dried, equilibrated, and fabricated into filters as described in Example I.

EXAMPLE VII

Corona treated wool top was padded with an aqueous solution containing 5% (w/w) PEI 1000 and 3% (w/w) of an aliphatic hydrocarbon sulphonic acid. The treated top, which contained 5.7% PEI 1000 and 3.4% alkyl sulphonic acid by weight, was dried, equilibrated, and fabricated into filters as described in Example I.

EXAMPLE VIII

Corona treated wool top was padded with an aqueous solution containing 5% (w/w) of modified PEI 1000 (partially quaternized with cetyl bromide) to give 4.2% by weight on the weight of wool. In a similar treatment omitting the cationic additive, PEI 1000 was applied to corona wool top to give 4.8 % by weight of PRI 1000. The treated tops were dried, equilibrated, and fabricated into filters by the methods previously described in Example I. Control filters were prepared from untreated corona wool top.

Filters prepared in Examples III to VIII are attached to standard cigarettes and the overall vapour phase ciliatoxicity determined by Paramecia bioassays (Example I type). Results of these comparative tests are recorded in Table III.

TABLE III ______________________________________ Filter material, treatment, and % (w/w) Paramecia treating material (LD 100) ______________________________________ Corona treated wool top with PEI 1000 (5.6%) and Vantoc N (5.6%) 29 Corona treated wool top with PEI 1000 (5.7%) and Hyamine 2389 (5.7%) 29 Corona treated wool top with PEI 1000 (4.7%) and Chemiquat E (4.5%) 32 Corona treated wool top with modified PEI 1000 (5.6%) (CHl as additive) 31 Corona treated wool top with modified PEI 1000 (5.7%) (RSO ₃ H as additive) 29 Corona treated wool top with modified PEI 1000 (4.2%) (treated with cetyl bromide) 27 Corona treated wool top with PEI 1000 (4.8%) 24 Corona treated wool top, no additives 19 ______________________________________

Results in Table III demonstrate the improved retention of ciliatoxic smoke constituents by filters treated with different cationic additives in combination with poly(ethylenimine) or with a poly(ethylenimine) modified in such a manner as to render it highly cationic in character.

EXAMPLE IX

Loose carbonized wool was carded into a fine open web which was overlapped several times. In a continuous process, the web was passed through a trough containing an aqueous solution of PEI 1000 (3% w/w) and Zephiran (3% w/w), squeezed at a pad mangle, and dried at 80°C for 5 minutes in a dryer. On emerging from the dryer the web was slit into 6 inch wide strips which, after equilibration at 65% RH and 21°C, had an end weight of 9.5 gm/metre. The bonded non-woven fabric contained 10% by weight of both PEI 1000 and Zephiran. Two other treatments were carried out in a similar manner. In the first, Zephiran was omitted from the treating solution and a non-woven fabric containing 10% by weight of PEI 1000 was produced. In the second, an aqueous dispersion containing 2% (w/w) of a polybutadiene-polyacrylate bonding agent was applied to the web to give a non-woven fabric containing 7% by weight of the binder.

Filter plugs were prepared from the above fabrics and comparative analytical tests carried out as described in Example I. The Paramecia bioassay was performed also as described in Example I except that the puff duration of the automatic smoking machine was altered to give a 17.5 cm ³ puff volume of 1 second duration at 60 second intervals. Results are recorded in Table IV.

TABLE IV ____________________________________________________________ ______________ Filter Treatments Constituents 7.0% acrylate 10% PEI 1000 10% PEI 1000 removed (%) bonder 10% Zephiran ____________________________________________________________ ______________ HCN 7 45 85 Acetaldehyde 0 17 32 Acrolein 0 9 29 Phenols (S.V.) 36 40 80 Paramecia (LD 100) 14 19 24 ____________________________________________________________ ______________

Results in Table IV show that filters prepared from a non-woven wool fabric bonded with poly(ethylenimine) containing a cationic additive such as Zephiran are superior in removing ciliatoxic constituents from the mainstream smoke of cigarettes.

EXAMPLE X

A polyalkylene polyamine resin was synthesised from ethylenediamine (0.67 mole) and epichlorhydrin (0.2 mole), and low molecular weight material removed by steam distillation. An aqueous solution containing 5% (w/w) Zephiran was padded on wool top which had been pretreated with dry chlorine gas (1% w/w uptake). The treated top contained 7.1% polyamine resin and 7.1% Zephiran by weight. A treatment omitting Zephiran was applied to the chlorinated top in a similar manner to give 8.8% by weight of the polyamine resin. The treated tops were dried, equilibrated and fabricated into filter plugs by the methods described in Example I. Comparative tests using the Paramecia bioassay (Example I type) were performed on the treated filters and on filters prepared from untreated chlorinated wool top Results are shown in Table V.

TABLE V ______________________________________ Filter Material, Treatment and % (w/w) Paramecia treating material (LD 100) ______________________________________ Chlorinated wool top treated with epichlorhydrin-ethylene diamine resin (7.1%) and Zephiran (7.1% 37 Chlorinated wool top treated with epichlorhydrin-ethylene diamine resin (8.5%) 25 Chlorinated wool top -- no further treatment 19 ______________________________________

Results in Table V demonstrate that a polyalkylene polyamine resin applied to chlorinated wool top in combination with a cationic additive such as Zephiran results in a filter considerably more effective in reducing overall smoke ciliatoxicity than one treated only with the polyalkylene polyamine component.

EXAMPLE XI

A strip (6 inches wide, end weight 10.0 g/metre) of bonded (polyacrylate-polybutadiene binder) non-woven wool fabric was padded with an aqueous solution containing 2% (w/w) PEI 1000 and 2% (w/w) Zephiran. The treated fabric contained 6% PEI 1000 and 6% Zephiran by weight. A second treatment omitting the Zephiran was carried out in a similar manner to give 9.4% by weight of PEI 1000 on the bonded fabric. The treated fabrics were dried, equilibrated, and fabricated into filter plugs by the methods described in Example I. Control filters were prepared from the untreated fabric.

Comparative analytical tests for hydrogen cyanide and steam volatile phenols together with the Paramecia bioassay were performed as described in Example I. In addition, analyses were performed on the vapour phase for total volatile aldehydes and on whole smoke for total volatile acids. The method of smoking for the bioassay was modified as described in Example IX. Results are recorded in Table VI.

TABLE VI ____________________________________________________________ ______________ Filter material, % Removal of smoke components treatment, and Hydrogen Volatile S. volatile S. volatile Para- % (w/w) treating cyanide aldehydes phenols acids mecia material (LD100) ____________________________________________________________ ______________ Prebonded non-woven fabric + PEI 1000 (6%) + Zephiran (6%) 91 35 89 62 22 Prebonded non-woven fabric + PEI 1000 (9.4%) 43 11 66 52 17 Bonded non-woven fabric untreated 13 0 56 41 14 ____________________________________________________________ ______________

The results in Table VI show that filters prepared from prebonded wool non-woven fabric treated with a combination of poly(ethylenimine) and a cationic compound such as Zephiran are superior in removing ciliatoxic constituents from the mainstream smoke of cigarettes.

EXAMPLES XII TO XVIII

Samples of wool top, which had been treated with a corona discharge, were padded with aqueous solutions containing different concentrations of PEI 1000 and Zephiran to give the treatments shown in Table VII. The treated tops were dried, equilibrated, and fabricated into filter plugs by the methods described in Example I. Comparative tests were made on these filters and on filters prepared from the untreated corona wool top using the Paramecia bioassay (Example I type). The results are recorded in Table VII.

TABLE VII ____________________________________________________________ ______________ Example Soln. conc. (% w/w) Wght. on wool (% w/w) Paramecia No. PEI 1000 Zephiran PEI 1000 Zephiran (LD 100) ____________________________________________________________ ______________ XII 5.0 5.0 5.8 5.8 30 XIII 10.0 5.0 11.0 5.5 27 XIV 10.0 2.0 10.8 2.0 25 XV 5.0 8.5 5.4 9.1 30 XVI 3.0 9.1 3.2 9.8 30 XVII 5.0 0 4.8 0 24 XVIII 0 7.0 0 8.4 20 Untreated -- -- -- -- 19 ____________________________________________________________ ______________

The results shown in Table VII demonstrate that wool filters treated with PEI 1000 and Zephiran as a combination are more effective in reducing the overall ciliatoxicity of cigarette smoke than filters treated only with PEI 1000 or Zephiran. In addition, the results show that the ratio of polyamine to cationic additive is important for optimum results.

It is believed evident from the foregoing detailed description that we have shown an improved tobacco smoke filter for selectively reducing the ciliatoxicity of tobacco smoke. While specific embodiments of this invention have been described for clearness of understanding, it is apparent that many variations can be employed without departing from the spirit and scope of the present invention.