Guildford Misc
54-page document: WITH COMP TS ' BR[TISH-Ai~ERICAN TOBACCO COMPANY L~D
Fields
Document Images
WITH COMP TS '
BR[TISH-Ai~ERICAN TOBACCO COMPANY L~D
Westminst'sr House 7 MS/bank London SWI - Te]~ephone Abbey 1222
X=
CD
P,0
Ox
CO
,.O
".O
r,O
.'/ ....~ .... .. .......... ° ....... j_
Confidential
CaWoon copy No. 0
Photo copy No.13
Eesearc~ and Development B.A.T. Germany/Eo-mburg
FE~eport No.IS
• ...
I
!
(
I
l
t
Title:
Active Charcoal as a Constituent of
Cigarette Filters
Authors: H. Borowski and M. Jacob±
Coauthor: E- .Dep~e
~,,~I~,.,~ b7: F. Seehofsr
S e 'i ...... " P
Distribute~ by:
• M'r. S.J. Green
11
Herr E. Soring
Herr _~. Sottorf
He_~r W. Se!bmann
.
~-. .p- .~[o. l-S
-U~. 9
[~'c. !d
" :" .:'0 "l!
\
0
",.o
.i°°
i
Conten~s
!
!
!_
i. Introduction
I.I General notes on active charcoal
1.2 Terminology for filtration and sorption processes
1.5 Factors fo~ evaluation.
2. Test results
2.1 }Tature of samples
2.Z Preseiectlon
2.5 Activity towards test chemicals.
3. Tests on c lgarette filters with active charcoal
5.1 Activity towards individual smoke constituent substances
5.2 Comparison with competitors' products
~.3 Taste evaluation.
4. Discussion
4.1 Summary of results
4.2 Discussion of procedures and filter types
@.5 Discussion of Patent and legal situation.
5. Recommendations (in accordance with present state of knowledge)
6. Plan for further work
Append ix, I
I~! Figures and tables of data (see special index)
i/2 Description of test methods
1/2.1 Test methods in use
i/2.2 0ther known test methods
I/3 List of charcoal types under test
Appendi~ II
II/l Patent Specifications
II~l.! List of patent specifications evaluated to date
II/1.2 Other patent specifications with short statement
of contents
II/2 Literature references.
~h-_ .
JCD
Special index for i/!.
List of Figures
Relating to 2.3: Activity towards test chemicals.
2.51 Survey of all types under test~ chlorobenzene
2.32 Survey of selected types~ adsorption
2.33 Phenol adsorption
2.54 Hydrogen cyanide adsorption
Relating to 3.1: Activity towards cigarette smoke
3.11 Amount of aldehydes in main smoke
3.12 Volatile organic acids in main smoke
3.13 Phenol content in main smoke
3.14 Ciliate test on particle phase
3.15 Cilia test on gas phase
i
i
I
List of Tables
3.16 Summarising chart of the most important
analytical parameters.
Relating to 5.1: Table of data referrin~ to the Figures 5.Ii Phehol adsorption (Fig.2j33)
5.12 Hydrogen cyanide adsorption (Fig.2o34)
5.13 Amount of aldehydes in main smoke (Fig.3.11)
5.14 Volatile organic acids in main smoke (Fig.3.12)
5.15 Phenol content in main smoke:
Ciliate test on particle passe (Fig.3.13/3.14)
and full smoke analysis
5.16 Cilia test on gas phase (Fig.3.15)
5.3 List of charcoal tToes under test (Fig.2.3~2.32)
O
r~O
Q~
co
~D
Ln
I. Introduction
I.I General notes on active charcoal
A charcoal rich in pores with a more or les~ extensive
internal surface area is formed during a suitable
carbonisation of plants. Such types of charcoal possess
a high adsorption capacity, and they are therefore known
as "active charcoal". The production of a gooa active
charcoal is not altogether simple. It has been found
that not all plants and plant parts can be considered
suitable for the preparation of an active charcoal. The
most suitable materials for this purpose are wood, peat,
coconut shells, bitumen residues and similar substances.
Normal heating of these materials admittedly only produces
a charcoal with a low adsorption capacity because the tar
formed at the same time blocks up the fine pores of the
charcoal so that these cannot adsorb vapours and gases.
go,ever, the active charcoal industry can nowadays produce
charcoals with an extraordinarily high activity. The
excellent adsorption qualities of active charcoals have
therefore for some time been exploited in many fields of
chemistry and of chemical technology for the purification,
decolourlsatiou and detoxicatlon of liquids, gases, aerosols
and smoke gases. It is therefore obvious that consider-
ation should be given to the activity of active charcoals
in relation to the filtration of cigarette smoke.
Charcoal adsorbs all volatile materials and really has no
specific selectivity. The adsorption capacity is
inversely proportional to the saturation vapour pressure,
i.e. in general terms, subs=shoes with a higher boiling
point are adsorbed more strongly than those with a lower
boiling point.
The adsorption capacity of a charcoal depends on its active
surface area, i.e. the sum of the surface areas of all
pores and channels.
Since it is possible for 1 g of active charcoal to have a~
active surface area of about 1200-1400 sq.m. and more, it
will be evident that the pore diameter can only be of the
order of a few molecules. A reduction In the particle
size (external surface area) only increases the total
active surface area of a charcoal by a smail amount which I
can be neglected.
!
A minimum requirement of i0 (mete!c) tons of active char- !
coal is required for the meekly production of a brand at '
the rate of i00 million cigarettes/mouth (for I00 mg/tip), i
!
CD
~O
t~
- 2 -
1.2 Terminolo~ for filtration and sorption processes.
"Sorption" is the inclusive term for the physical and ohemica!
individual processes durin~ a filtration operation. In
general, this refers to the take-up of a material (sorbant)
by a material (sorbent). However, if the material is already
present in a sorbed state in or on the material, it is said
to be sorbed or aasorbed:
The inclusive term "sorptlon" mainly comprises four
subsidiary Seres:
!. Adsorption: This implies the concentration of a vaoour or
gas on ~he surface of a solid material without the formation
of solid solutions or chemical reactions
(Example: gas adsorption through active charcoal, nicotine
adsorption in an acetate Xilter).
2. Capillary condensation: This is a special case of adsorption
and occurs when a condensation (liquefaction) of the ad-
sorbant is ea~sed by an increase in its concentration within
the capillaries of porous materials.
(Example: benzene vapour condensation in active charcoal)
3. Absorotion: A solution process occurs during absorption,
and during this process the absorbant is dissolved in the
absorbent.•
(~mple: hydrogen absorption by palladium or platinum)
4. Chemosorption: When the sorbant is chemically combined
w'ith the sorbent, chemosorption occurs.
(Example: chemosorption by CO2 by caustic alkali or water).
When the overall efficiezcy of a cigarette filter is consideredI
without a detailed separate investigation of special processes,!
it is usual to refgr to a "filter effect" which may be good or i
poor, or may be high or low. When it is intended to
characterise the efficiency of a filter in relation to the
whole smoke or individual comoonents in smoke by means of a i
quantitative expression (say in terms of percentages, as usual)i
it is best to employ the term "retention" (filter retention)
originating from the Anglo-American usage. ,.
!
!
The term "filter adsorption" or "filter absorption" as often
emoloyed is definitely wrong. We can only refer to
"nicotine adsorption" or "phenol adsorption", etc.
When certain components in smoke are preferentially retained
by a fil~er, this has a "selective" activity and exhibits a
degree of seleeti#ity.
CD
~o
Cr~
~0
-3-
According to a more recent definition, it is possible to
refer to a cigarette filter having a ~'positive selectivity"
when the difference (or the quotient) of the retention
activity towards a substance x and the dry main smoke
MS(dr) condensate is greater than I
or > I
- 1 SCdr. ) R ,mCdr)
A better reference basis than MS(drJ would be a smoke
constituent without a vapour pressure which can readily
be analysed "(e.g. solane sol, scopoletine, etc.).
g
T
O
O~
(Do
kO
kO
(3o
- 4-
i
1.5 Factors for evaluation.
It has already been pointed out that active charcoal does not
adsorb selectively, and it is therefore relatively difficult
to decide from which angle an evaluation of a charcoal type
for its usefulness as a cigarette filter should be carried out.
The following properties are inter alia desirable from the
point of view of its satisfact-~-~--pr--~ical appllcatio~.
Retention of aldehydes, acrolein, phenol, hydrogen cyanide,
volatile organic acids, volatile organic bases, favourable
behaviour during biological tests, the least possible influence
on taste, optimum processing characteristics, etc.
Since our laboratory is of course not equipped with the
apparatus required for the methods of investigation normally
employed in the active charcoal industry, it was necessary to
develop or apply more practical methods for the evaluation of
active charcoal from our point of view.
Since the object of the work of investigation in the field of
active charcoal was ultimately directed towards the discovery
of a type of charcoal whose activity in a biological test should
be at least equivalent to that of the material in a Lark filter,
the decisive quality looked for related to this factor during
the final selection. For this purpose, two independent
biological tests, the ciliate test on paramecia {CAT) and the
cilia test on mussels (C~T), were employed for evaluation.
The CAT establishes the activity of the particle phase and the
C~ the activity of the gas phase. During the final selection,
the degree of phenol retention was employed as a further
useful parameter.
2. Test results.
2.1 Nature of samples
The active charcoal types which are commercially available
were obtained through the normal trade channels. For this
purpose, an approach was made to the normal chemical suppliers
as well as to the special suppliers and to the active charcoal
industry in general. Closer contact was established with
two firms in order to characterise samples of special
charcoals:
Ringsdorff-Werke, of ~ehlem, and
~I " 1 I1
Drager-Wsr~, of L~beck.
However, after it had become apparent in the pertinent
industrial circles that the cigarette industry has become
interested in active charcoal, samples and offers of active
charcoal were also received without request from other sources.
2.2 Preselection.
In view of the large commercial supply of active charcoal, it
is appropriate to carry out a general prsselection operation
orior to the start of soecial test procedures. In technology,
charcoals not intended for any special field of auplication areO
characterised according to their adsorption capacity for a
benzene vapour/air mixture. The increase in weight up to the ~
attainment of the degree of saturation is termed the
Cry.
co
~O
@
-5-
maximum charge capacity. We have therefore prese!ected the
charcoal types available in commerce according to their
maximum charge capacity, except that we employed chlorobenzene
instead of benzene3 owing to its higher ~ensity and equally
good lipophilic properties.
The most active charcoal types established according to this
method were subjected to an activity test towards test
chemicals. Thereafter, their behaviour towards cigarette
smoke was investigated.
2.5 Activit~ towards test chemicals.
2.32 Chlcrobenzene adsorption (cf. Fig. 2.34 and Table 3.16~.
The top performance group with maximum adsorption
qualities consists of the two types Ringsdorff and
Lurgi (Desorex 21).
2.83 Phenol adsorotion (cf. Fig. 2.33 and Table 5.11)o
It is worth noting the relatively poor adsorption
efficiency of the type "Fluka, granulated" compared
with the top performance group.
2.34 Hydrogen cTanide adsorption (cf. Fig. 2.34 and Table 8.12).
Separ~bion of the charcoal types into three different
activity classes:
1. Top performance type (Merck 1.5 mm, granulated)
2. Intermediate group (Ringsdorff suecial with Cu and Fe;
Riedel de H~"en, pure, granulated;
Ringsdorff ~0r', activated 1
3. Poor type (Fluka, granulated ).
3. Tests on cigarette filters with active charcoal.
3.1. Activity towards individual smoke constltuen~ substances.
8.11 Aldehydes (cf. Fig. 8.11 and Table 5.!3)
Separation of the charcoal types into three different
activity classes :
I. Top performance type (Rin~sdorff "0", activated)
2.
!ntermediate group (Riedel de H~en, pure, granulated;
Merck l.S me. granulated)
3. Poor type (Dr~ger K 238)
3.12 Volatile organic acids (cf. Fi~.$.12 and Table 8.14)
Separation into two different actlvity classes.
1. Top performa=ce type (Ringsdorff 0 , activated)
2. Eauivalent intermediate group [Merck 1.8 m~., granulated;
" Riedel de Haen, pure,
granulated; .
Dr~ger K 285)
5.13 Phenols (cf. Fig. 3.13 and Table 5.15)
Superiority of the chamber filters compared with the
Filtrona-Bonded Filters. The constant order
of ~he activity of the charcoal types for a given
O
O~
~O
:CD
~CID
....... ~ (ID
- 6 -
type of filter on comparison with a different type is
particularly marked.
Top performance type: Ringsdorff "0", activated.
5.14 Ciliate test on particle phase (of. Fig. 3.14 and
Table 5.15)
The top performance type in the chamber filter arrangement
is far superior to the Lark filter and the Filtroua-
Bonded Filters.
3.15 Cilia test ion ~as phase (cf. Fig. 3.15 and Table 5.16)
The distinct superiority of the top performance type
"Ringsdorff "0" activated" in relation to the Filtrona
and Lark filters is especially marked in this test.
5.2 Comparison with competitors,' products
The following summary provides a survey of the physical date
for the charcoal filters of cigarette brands commercially
available.
3.3 A taste evaluation is extremely important during the development
of a cigarette wi~h a charcoal filter.
Attempts were therefore first of all made to discover whether
certain tobacco varieties are particularly suitable for applicat- i
ion in cigarettes with charcoal filters. The following smoking
evaluations by experts indicate that all tobaccos which have been
air dried or naturally processed (in the widest sense) do indeed
provide better smoking results than Virginia varieties. These
experiments were carried out with the same charcoal filter
(Filtrona FF 561).
After the selection of active charcoals according to chemico-
biological criteria had been completed, a smoke taste evaluation
was carried out on a homogeneous tobacco mixture. The smoking
evaluation by experts during this test which is also given below
has indicated the distinct superiority of the Lurgi sample.
_This smoking evaluation is of interest since it makes possible
some fundamental conclusions regarding the problem of the cause
!
of
the..
"strange
taste due to charcoal".
I
!
It follows that an increase in the "charcoal taste" is not i
necessarily related to an increase in the adsorption efficiency i
of a charcoal type (otherwise, a different order of preference
would have been established). It appears in fact that the result i
!
of a smoking test deteriorates as the level of the dust fraction
in the sample rises. Since this recognition will allow us to
redtme the charcoal taste, though it cannot be wholly eliminated,
coupled with the fact that exhaled smoke filtered through
charcoal has an unpleasant smell (see below), we may propose the
following hypothesis :
O
~O
~O
CD
CD
_~--~
