WO2005115182A2

WO2005115182A2 - Tobacco smoke filter

Info

Publication number: WO2005115182A2
Application number: PCT/GB2005/002045
Authority: WO
Inventors: Anthony Denis Mccormack; Francisco Rodriguez Reinoso
Original assignee: Filtrona International Limited
Priority date: 2004-05-25
Filing date: 2005-05-24
Publication date: 2005-12-08
Also published as: WO2005115182A3; GB0411690D0

Abstract

A tobacco smoke filter containing activated carbon and a volatile flavourant in an amount of between 5% and 35% (w/w) of the activated carbon, which activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of less than 200 cm3/g at a relative pressure (P/Po) value of 0.1, and one or more of an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm3/g; and an increase in the volume of nitrogen adsorbed. between relative pressure (P/Po) 10 values of 0.75 and 0.99 of greater than 200 cm3/g.

Description

TOBACCO SMOKE FILTER

The present invention relates to tobacco smoke filters containing particulate sorbent.

The use of sorbent particles to remove vapour phase (VP) components from tobacco smoke is well known. Cigarettes containing volatile flavourant (e.g. menthol) are also well known. Many previous attempts to use both volatile flavourant and particulate sorbent in a filter cigarette have been unsuccessful, it having proved difficult to provide a satisfactory level of flavour delivery whilst maintaining a satisfactory level of VP constituent removal by the particulate sorbent.

Our International patent application No. PCT/GB2003/005151 describes a tobacco smoke filter that is capable of delivering satisfactory levels of flavour (notably menthol) when used in a cigarette whilst maintaining a satisfactory level of VP constituent removal. This performance was achieved through the use of an activated carbon in the filter, the carbon having a micropore volume of at most 0.3 cm³/g, a mesopore volume of at least 0.25 cm³/g and a volume of larger mesopores of at least 0.12 cm³/g. The definitions of micropore (less than 2 nm diameter) and mesopore (2-50 nm diameter) were in accordance with the accepted IUPAC terminology and definition. Larger mesopores were defined as those greater than 7 nm as measured by mercury porosimetry. The micropore and mesopore volumes were calculated from measurements of the nitrogen adsorption/desorption isotherms. The amount of gas or vapour adsorbed by a solid is dependent on the pressure of the gas or vapour above the surface. A graph of the amount of gas or vapour adsorbed versus pressure at a fixed temperature forms the basis of the measurement of an adsorption isotherm. A nitrogen adsorption/desorption isotherm measures the quantity of nitrogen gas adsorbed at cryogenic temperatures and is the most widely used analytical technique to derive information on pore size distribution. This technique is used to obtain values for micropore and mesopore volume, but suffers from the limitation that the numerical treatment of the isotherms to obtain pore size volumes is open to a degree of interpretation. The nominal definitions of micro- and meso- pores do not adequately define pore size distribution for the purposes of the present application.

The historic work of Brunauer classified adsorption isotherms into five types. Type I isotherms characterise microporous adsorbents, types II and III describe adsorption on macroporous adsorbents with strong and weak adsorbate-adsorbent interactions respectively, whilst types IV and V are similar to types II and III except that they exhibit hysteresis and are associated with materials containing fairly large pores. Subsequent to the work described in PCT/GB2003/005151, we have carried further investigations into the porous structure of carbons that are able to simultaneously deliver flavour and remove volatile compounds when used in a cigarette filter. We have found improved activated carbons which have have a pore size distribution which gives rise to a characteristic nitrogen adsorption isotherm shape. Such isotherms cannot simply be defined in terms of type I to type V. Furthermore, carbons having this optimum pore size distribution (as defined by this optimum isotherm shape) are not currently commercially available. The activated carbons we have developed have a particularly preferred range of pore size distribution that gives improved performance when used in filter cigarette with flavour. This preferred range has been demonstrated by the use of nitrogen adsorption isotherms. We have also found that the amounts of carbon and menthol in the cigarette filter (or cigarette) as well as the ratio of carbon to menthol in the cigarette filter (or cigarette) are of great importance in determining the performance of the final cigarette product in terms of menthol yield and the efficiency of vapour phase removal. It is desirable to have the capability to deliver as high a menthol yield as possible, whilst still retaining as high an efficiency as possible for the removal of vapour phase compounds from cigarette smoke. At a given weight of carbon in the filter, as the ratio of menthol:carbon increases, so the menthol yield per cigarette increases and the efficiency of vapour phase removal decreases. The previous work simply allowed the activated carbon to adsorb menthol for a defined period of time and tests were then carried out on the resultant mentholated carbon filter. However, it then becomes difficult to identify the optimum characteristics of the activated carbon due to the comparative results being affected by the menthol:carbon ratio. In addition, in any final cigarette product it is necessary to have reproducible performance, which would necessitate the use of predetermined quantities of carbon and menthol. For these reasons, it is necessary to use defined ratios of menthol:carbon. We have found that there is an optimum range of mentho carbon ratio: if this ratio is too low, menthol will not be released; if the ratio is too high then there will be insufficient removal of vapour phase compounds on smoking. Furthermore, in this latter case, the cost of the filter and/or cigarette starts to become prohibitively expensive due to the high cost of the menthol present.

The present invention provides a tobacco smoke filter containing activated carbon and a volatile flavourant in an amount of between 5% and 35% (w/w) of the activated carbon, which activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of less than 200 cm³/g at a relative pressure (P/Po) value of 0.1 , and which shows one or more of an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g; and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 200 cm³/g. Preferably the volatile flavourant is menthol. Preferably the flavourant (menthol) is present in an amount (w/w) of between 10 and 30%, more preferably between 15 and 25%.

Our work has shown that a flavourant (menthol) loading of less than around 5% w/w on the carbon does not give rise to any release of flavourant (menthol) on smoking, whilst loadings greater than around 35% w/w are inefficient. The optimum loading of flavourant (menthol) will depend on the performance required in the final cigarette and, importantly, the pore size distribution characteristics of the activated carbon.

Herein, unless specified otherwise, a volume (or volume adsorbed, or volume of nitrogen adsorbed) expressed in cm³/g means said volume as measured by nitrogen porosimetry, using a Coulter Omnisorp 610 Instrument to measure the nitrogen adsorption isotherm over a range of relative pressure (P/Po) values at a temperature of 77K using the protocol described below. All measurements are taken from the adsorption part of the hysteresis loop, rather than the desorption part of the loop. The wording "the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm" refers to the nitrogen adsorption isotherm of the activated carbon measured using the specified nitrogen porosimetry protocol (in the absence of flavourant). Activated carbons which, when tested using the specified nitrogen porosimetry protocol, show the characteristic nitogen adsorption isotherms defined in the claims may be suitable for use in tobacco smoke filters according to the invention.

Preferably, the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g, and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 200 cm³/g. Preferably, the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 180 cm³/g.

Preferably, the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 280 cm³/g, more preferably greater than 300 cm³/g. Particularly preferred are activated carbons which have a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 550 cm³/g.

Preferably, the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of between 110 cm³/g and 200 cm³/g at a relative pressure (P/Po) value of 0.1, more preferably between 140 cm³/g and 200 cm³/g, even more preferably between 160 cm³/g and 180 cm³/g.

Preferably the activated carbon has a pore size distribution defined by an adsorption isotherm substantially similar to that of Example F4 described herein (see Fig. 2). Thus, preferred activated carbons for use with the invention have a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of between 150 cm³/g and 200 cm³/g at a relative pressure (P/Po) value of 0.1, an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 300 cm³/g and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 550 cm³/g. Preferably the activated carbon has a pore size distribution defined by an adsorption isotherm substantially similar to that of Example F3 described herein (see Fig. 3). Thus, preferred activated carbons for use with the invention have a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of between 110 cm³/g and 200 cm³/g at a relative pressure (P/Po) value of 0.1 , an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 180 cm³/g and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 350 cm³/g.

Preferably the activated carbon has a pore size distribution defined by an adsorption isotherm substantially similar to that of Example F1 described herein (see Fig. 4). Thus, preferred activated carbons for use with the invention have a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of between 150 cm³/g and 200 cm³/g at a relative pressure (P/Po) value of 0.1 , an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 280 cm³/g.

We have found that activated carbon loaded with the specified amount of volatile flavourant and having the defined isotherm shape(s) (and pore size distribution(s)) shows a satisfactory level of adsorption of volatile flavourants such as menthol; and provides excellent release (yield) of the flavourant under smoking conditions to deliver satisfactory taste without any requirement for heavy loading of (expensive) flavourant; together with a good adsorportion of VP components from tobacco smoke with maintainance of satisfactory level of VP removal in the presence of volatile flavourant. According to the present invention in a further aspect there is provided a tobacco smoke filter containing activated carbon and a volatile flavourant in an amount of between 5% and 35% (w/w), preferably between 10% and 30% (w/w), of the activated carbon, which activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm substantially similar to that shown in any of the attached Figs 2, 3 and 4.

The filter according to the invention may be of any design previously proposed for particulate sorbent - containing tobacco smoke filters. For example, the activated carbon may be dispersed through a filter plug, carried on the tow or fibres or sheet material which is gathered to form the plug; it may instead be adhered to one or more threads which extend through the matrix of the filter plug or be adhered to the inner face of a wrapper around the filter plug; or it may form a bed sandwiched between a pair of plugs (e.g. of cellulose acetate tow) in a common wrapper. The activated carbon may be treated with the flavourant prior to filter production so that it acts as a carrier for the flavourant and minimises migration of the flavourant during storage.

The invention also provides a tobacco smoke filter according to the invention incorporated in a filter cigarette.

According to the present invention in a further aspect there is provided a filter cigarette comprising: a tobacco smoke filter containing activated carbon having a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of less than 200 cm³/g at a relative pressure (P/Po) value of 0.1, and which shows one or more of an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g, and/or an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 200 cm³/g; the cigarette containing a volatile flavourant in an amount of between 5% and 35% (w/w), preferably between 10% and 30% (w/w), of the activated carbon. The volatile flavourant, which may be menthol, may be within the tobacco smoke filter, for example applied to the activated carbon. The volatile flavourant, which may be menthol, may be included in the tobacco. Filters according to the invention may additionally include one or more particulate sorbents other than the activated carbon required by the invention (e.g. silica gel, or a different carbon), mixed with the carbon required by the invention and/or separate from this.

The present invention is now described in more detail with reference to the attached drawings, in which:

Fig 1 shows nitrogen adsorption isotherms for examples of activated carbons for use within filters according to the invention (activated carbons F1 , F3 and F4) and comparative examples, obtained using the nitrogen porosimetry protocol described herein; Fig 2 shows the nitrogen adsorption isotherm for F4;

Fig 3 shows the nitrogen adsorption isotherm for F3;

Fig 4 shows the nitrogen adsorption isotherm for F1 ;

Fig 5 shows the nitrogen adsorption isotherm for the comparative examples; and Figs. 6 and 7 respectively are schematic sectional side elevation views, not to scale, of an individual filter and filter cigarette according to one embodiment of the invention.

In the following, examples with example numbers prefixed with F (F3, F4 etc.) are according to the invention; the remainder (A, B, C6 etc.) are comparisons. Examples of preferred carbons (F1 , F3 and F4) for use in cigarette filters according to the invention were prepared by the following laboratory procedure of chemical impregnation, followed by activation.

Impregnation

60 grams of dried olive stones (previously washed with a dilute solution of sulphuric acid to remove any inorganic compounds present) were weighed. A 7wt% anhydrous CaCI₂ solution was prepared (8.4g CaCI₂ and 111.6g water for 60g olive stones) and put in a beaker. The beaker was placed in a water bath and the olive stones slowly added into the beaker with stirring. The beaker was then covered using a watch glass (full of water to avoid evaporation) and the water bath heated to a temperature of 85°C. Impregnation started when the temperature of the water bath reached 85°C; impregnation was continued (that is, this temperature was maintained) for 7 hours. After 7 hours, the watch glass was removed and the water bath heated to boiling, allowing slow evaporation of the calcium chloride solution. The beaker was heated until the level of the calcium solution and the level of the olive stones were the same. The beaker was removed from the boiling water bath, the solution filtered to leave the impregnated olive stones (the sample). The sample was held in an air-circulating oven at 110°C until completely dried.

Activation The sample was put in an horizontal furnace, and heated according to the following protocols (the numbers F1 , F3 and F4 are used subsequently). The flow of CO₂ gas was started immediately and heating commenced; the time given was the duration of heating once the final temperature (Oven T) was attained:

After heating for the specified time, the furnace was allowed to cool to room temperature. The sample (F1 , F3 or F4) was then heated in a vacuum oven at 50°C for 4 hours (to remove CO₂ adsorbed in the horizontal furnace).

Washing The sample was washed using a hot 5% (wt.) HCI solution, until no further gas was evolved; washed with distilled water (to completely remove the acid); and then held in an air circulating oven until completely dry. The sample was then weighed. The yields obtained using the above method were F3 - 14.5%, F4 - 11.9% and F1 - 7.7%. Other samples of carbons for use in filters or cigarettes within the scope of the invention may be prepared by this or similar methods.

Measurement of nitrogen adsorption/desorption isotherms

Numerous activated carbon samples were prepared and their nitrogen adsorption/desorption isotherms measured. During this current work, the isotherms were measured on a different piece of apparatus to that described in PCT/GB2003/005151. Details of the experimental procedure used for the present application are as follows.

Sample preparation for adsorption tests A weight of over 0.1 grams of the activated carbon specimen

(Sample F3, F4 etc.) is placed in a sample holder of a Coulter Omnisorp 610 instrument (made by the Coulter Corporation, now Beckman Coulter, Inc. of Fullerton, CA 92834-3100 USA), which is then outgassed at a vacuum of 10^"4 torr and heated to a temperature of up to 250°C for 4 hours. The sample is then allowed to cool down to room temperature and the sample holder filled with helium (to a pressure slightly higher than atmospheric) and the outgassed weight measured. The sample is then taken down to a vacuum of approximately 2 x 10^"5 torr prior to measurement of the isotherm.

Isotherm Measurement

The Coulter Omnisorp 610 puts into the manifold a 100 torr dose of nitrogen and the gas in the manifold expands over the sample (which is maintained at 77 K). After 60 seconds equilibrium time, the equipment measures the pressure (which is expected to be lower than that measured 60 seconds previously due to the adsorption capability of the sample). The equipment then waits another 60 seconds and re-measures the pressure, comparing this value with the previous one. Only when it finds five points with the same value, does the equipment put another 100 torr dose of nitrogen into the manifold.

Both adsorption (increasing nitrogen pressure) and desorption (decreasing nitrogen pressure) loops of the isotherm are measured by the equipment using this procedure. A cut-off point of 0.3 is used as the P/Po value at which desorption is regarded as finishing. The adsorbed volume is obtained by the equipment by means of pressure difference measurements. As soon as the isotherm measurement has finished, the equipment generates a plot of P/Po vs Vads (cm³/g).

The above techniques and procedures are well known and readily understood by those skilled in the art. Figure 1 shows the nitrogen isotherms for a number of activated carbon samples. These are:

A, a coconut-based activated carbon as typically used in cigarette filters.

A2 is sample A that has been further activated to enhance its pore volume (particularly in the mesopore range).

B is a carbon similar to that described as sample B of PCT/GB2003/005151.

B2 is sample B that has been further activated to enhance its pore volume.

C is a carbon similar to that described as sample C of PCT/GB2003/005151.

C6 is sample C that has been further activated to enhance its pore volume.

Samples F1 , F3 and F4 are samples according to the present invention. The preparation of F1, F3 and F4 is described above. For each of these carbons, a sample was dried and mixed with menthol in the ratio of one part menthol to five parts carbon (except where stated otherwise). The resultant mix was stored in a sealed container for four days at a temperature of 55°C. 'Triple granular' cigarette filters (which are well known in the art) were then assembled, each containing 100 mg of the mentholated carbon in a packed bed between two cellulose acetate filter segments. The filter cigarettes were smoked under ISO conditions (a 35cm³ puff of 2 second duration taken once per minute) and the menthol yields from the cigarettes were measured. The vapour phase (VP) of cigarette smoke was also collected and the percentage reduction of a selected number of VP compounds measured; the mean reduction in these VP compounds, and the reduction obtained from an equivalent filter with 100 mg of the same carbon prior to exposure to menthol, were measured relative to an equivalent filter cigarette with no carbon.

The results are summarised in table 1.

Table 1

* At a mentho carbon ratio of 1:7

It should be noted that the menthol yield data in the above experiment is comparative within this set of samples; the actual values obtained cannot and . should not be directly compared with the corresponding data from the previous patent application

PCT/GB2003/005151.

It can be seen that samples F1, F3 and F4 exhibit the best performance: that is, they deliver high menthol yield whilst simultaneously removing a high quantity of VP compounds. A comparison of the nitrogen isotherms in the figures with the data in table 1 shows that carbons F1 , F3 and F4 are particularly characterised by a steep increase in the volume of nitrogen adsorbed at relative pressure (P/Po) values of greater than around 0.75. In addition, the volume adsorbed at lower P/Po values is not too great. The isotherms for samples F4, F3 and F1 are shown separately in Figs 2 to 4, and those for all comparative examples are shown in Fig 5. Some data points corresponding to the plots in Figs 1 to 5 are interpolated in table 2. Data is given for both the adsorption and desorption loops of the isotherm (Ads is adsorption loop of isotherm; Des = desorption loop). Table 2

Carbons according to this invention that are particularly effective in delivering menthol whilst simultaneously removing VP compounds are those which have pore size distributions which give nitrogen adsorption isotherms (when tested using the above protocol) which show a volume adsorbed of less than around 200 cm³/g at a P/Po value of 0.1 , and which show an increase in the volume of nitrogen adsorbed between P/Po values of 0.75 and 0.95 of greater than around 150 cm³/g, and/or an increase in the volume of nitrogen adsorbed between P/Po values of 0.75 and 0.99 of greater than around 200 cm³/g.

Example F11

In a specific example of a filter and filter cigarette according to the invention as described with reference to Figs.6 and 7, the filter is 27 mm long and about 25 mm in circumference. The buccal end plug 2 is a 10 mm long non-wrapped acetate (NWA) plug - i.e. a preformed non-wrapped plug of plasticised cellulose acetate filaments gathered and bonded together such as is well known in the art.

The upstream end plug 3 is a 10 mm long wrapped acetate (WA) plug - i.e. a preformed wrapped plug of plasticised cellulose acetate filaments. The filter wrapper is 27 mm long to give a cavity 6, which is 7 mm long, extending between plugs 2 and 3. The cavity 6 is filled with 100mg of granules 17 of mentholated activated carbon, according to the invention. In this example, the activated carbon is equivalent to sample F1 (prepared as described above with reference to F1) which has been dried and mixed with menthol in the ratio of one part menthol to five parts carbon, the resultant mix having been stored in a sealed container for four days at a temperature of 55°C. The filter rod is attached by a ventilating tipping overwrap 12 to a commercial wrapped tobacco rod 10, 11. It will be appreciated that Example 11 is similar in construction to a known triple granular filter but includes activated carbon according to the invention.

It will be appreciated that the filter according to the invention may be of any design previously proposed for particulate adsorbent - containing tobacco smoke filters with the substitution of the known particulate adsorbent with the activated carbon of the invention.

Claims

1. A tobacco smoke filter containing activated carbon and a volatile flavourant in an amount of between 5% and 35% (w/w) of the activated carbon, which activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of less than 200 cm³/g at a relative pressure (P/Po) value of 0.1 , and one or more of an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g; and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 200 cm³/g.

2. A filter according to claim 1 wherein the volatile flavourant is menthol.

3. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g, and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 200 cm³/g.

4. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 180 cm³/g.

5. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 280 cm³/g.

6. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 550 cm³/g.

7. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of between 110 cm³/g and 200 cm³/g at a relative pressure (P/Po) value of 0.1.

8. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 300 cm³/g and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 550 cm³/g.

9. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 180 cm³/g and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 350 cm³/g.

10. A filter according to any preceding claim wherein the activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 280 cm³/g.

11. A tobacco smoke filter containing activated carbon and a volatile flavourant in an amount of between 5% and 35% (w/w) of the activated carbon, which activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm substantially as hereinbefore described with reference to any of appended Figs 2, 3 and 4.

12. A filter cigarette including a filter according to any preceding claim.

13. A filter cigarette comprising: a tobacco envelope and a tobacco smoke filter; wherein the filter contains activated carbon having a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of less than 200 cm³/g at a relative pressure (P/Po) value of 0.1 , and which shows one or more of an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g; and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 200 cm³/g; and wherein the filter cigarette includes a volatile flavourant in an amount of between 5% and 35% (w/w) of the activated carbon.

14. A cigarette according to claim 13 wherein the volatile flavourant is within the tobacco smoke filter.

15. A cigarette according to claim 13 or 14 wherein the volatile flavourant is applied to the activated carbon.

16. A cigarette according to claim 13 wherein the volatile flavourant is included in the tobacco envelope.

17. A tobacco smoke filter substantially as hereinbefore described with reference to the description and examples.

18. A tobacco smoke filter substantially as hereinbefore described with reference to the attached Figures 6 and 7.

19. A filter cigarette including a filter according to any of claims 17 or 18.

20. A tobacco smoke filter containing activated carbon, which activated carbon has a pore size distribution defined by a nitrogen adsorption isotherm which shows a volume adsorbed of less than 200 cm³/g at a relative pressure (P/Po) value of 0.1, and one or more of an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.95 of greater than 150 cm³/g; and an increase in the volume of nitrogen adsorbed between relative pressure (P/Po) values of 0.75 and 0.99 of greater than 200 cm³/g.