TGM/DAM/RLP/7 0/46D
I fe
POSITION PAPER
INTRODUCTION
REDUCTION OF SMOKE NOXA
12 th March, 1987
Examination of any list of substances considered
to be noxious
agents in tobacco smoke (Table 1) reveals that
their likely sources in
the substrate are diverse. Views may varyion the
need to remove
individual items in this list, whilst at the same
time the priorities
attached to them could depend on short-term
commercial sensitivities as
much as on scientific logic. Therefore, if for no
other reason, for
new initiatives towards their reduction to be of
practical advantage,
they should seek tb offer substantial reductions
in individual components,
or across the board reductions in numbers of them.
Sources in tobacco of the so-called noxa range
from those where
the substance transfers direct to the smoke, ~.g.
metals such as cadmium
and polonium, through major defined sources of
smoke components on
combustion, e.g. nitrate giving rise to NOx and
volatile nitrosamines,
to the many assumed to have multiple origins, the
relative importance
of which individu,ally remains uncertain (see
Table 2).
It is against this background that we have
discussed approaches
to research on their reduction in mainstream
smoke, and the recommendations
made regarding projects.
Our brief at this stage did not include product
constraints such
as immediate commercial feasibility or subjective
consequences, although
we realise that these would ultimately need to be
integrated into any
successful outcome.
DISCUSSION
1. Potential Strageqies
The following routes by which modification might
be achieved have
been considered and are discussed briefly in turn:
a) Tobacco growing and curing,
b) Precursor removal
c) Combustion modification
d) Tobacco fractionation
e) Filtration
a) Tobacco Growing and Curinq
including variety selection
Explorations have been made, both within and
outside the
Group, of the options available for smoke
component control
via this route. Useful results have been achieved
for some
isolated components:
4 ~:.
C=)
CZ)
%.0
00
__J
r1 J
(-n
1.0
BATCo document for Province of BritiSh Columbia 28
October 1999


-2-
0 NOx and nitrosamines via fertilizer control and
variety selection (Burley) and curing modification
(Virginia).
ii) Use of high nicotine varieties to obtain
pro-rata
reductions in smoke constituents at constant
nicotine
delivery.
Unsuccessful attempts have been made to influence
CO by
cultural practice. I
Further work in this area has been discounted from
consideration at this stage because the benefits
so far
identified have been too limited to seek to
control farm
variables on the large scale required.
In the longer term, it is an area of potential,
particularly
through the medium of advanced variety
development. To do
this specific information on sources of noxa will
be
required. A summary has been made in Table 3 of
current
assumptions of means of control via this route.
Precursor Removal
The major obstacle to this remains the limited
number of
clear targets currently available. Nitrate removal
has
been developed by some companies, particularly for
application
to burley stein to modify delivery of oxides of
nitrogen.
On present information, this approach remains
limited to
modifying smoke nitrogen compounds and, possibly,
Ames
activity. Extension to others requires a clearer
understanding
of the tobacco precursors involved.
(c) Combustion Modification
Combustion mechanisms are obviously relevant to
the formation
of a number of the smoke compounds of possible
interest, so
that approaches to their control or modification
are
important. In practice, two broad areas have been
studied,
but generally in isolation:
i) The physical form of the tobacco, such as
strand size,
particle form and cigarette dimensions; and
ii) The use of additives applied either to t~e
tobacco or
paper - oxidants, retardants, modifiers, and also
in
studies of reconstitution such as PRT.
The prospects for combining some of these
approaches may
appear both uncertain and daunting. However, there
is a
case for examining the information and considering
its
implications. The earlier literature has been well
surveyed
(Report RD.1012-R), but it would seem timely to
update this C:)
in the current context. C:)
.10
co
0-
BATCO document for Province of BritiSh Columbia 28
October 1999


-3-
As a guide to the range of substances with a
history of use
as substantial additives, Table 4 lists those
permitted in
the UK at 3% or more. This demonstrates sorne
diversity of
those which have been us-ed rather than simply
submitted to
research study.
M Tobacco Fractionation
Extraction of tobacco has been examined
extensively with
particular reference to reconsbitution processes
and early
attempts at reducing the biological activity of
condensate.
The reverse of this has also been applied, in
endeavours to
isolate flavour fractions.
Comprehensive studies of tobacco fractionation
have been
rare and the few reported appear to have had
limited
objectives (e.g. that by Bush on tar contribution
and by
Bockelman directed to cell wall fractions). Apart
from
the PRT programme, in-house effort includes the
Cylproc
process which did achieve some selective change of
smoke
delivery.
A recurring theme in considering the other
approaches has
been the limited information on sources of the
noxa of
interest. Tobacco fractionation offers the
opportunity not
only to find possible routes to products with
reduced levels
of noxa, but also to provide the basic information
required
for knowledge-based effort via other routes.
An example of a tobacco fractionation scheme is as
follows.
Assume tobacco to provide three distinct elements:
i) Sources of pleasurable elements in smoke
ii) Carrier material
iii) Negative elements
The fractionation would be aimed at optimising
their
separation in order to increase the ratio of i) to
iii),
and to maximise the use of ii).
The argument advanced here is at the simplest
level, but
the advantage of the approach is that it allows a
diversity
of thought to its implementation and a variety of
technologies
could be brought to bear on improving the
separations and
modifying the fractions. Two schemes to
investigate this
subject are presented in the Appendix.
(e) Filtration
Numerous efforts have been made to obtain
reduction of
specific smoke components by filtration and some
successes C:)
have been recorded. On present evidence, it is
unlikely C:)
that this approach could be~ used to obtain M212r
reductions %D
in several components concomitantly. This is based
simply co
__4
BATCo document for Province of BritiSh Columbia 28
October 1999


-4-
on the view that a multiplicity of mechanisms
would he
involved and the practicality of combining them
into a
single filter would be remote.
Despite this negative appraisal, the role of the
filter
will'obviously remain of the highest significance
and the
possibility remains that it may need to be
explored again.
One scenario would be success in reducing x
components by
another route allowing the opportunity to develop
a selective
filter to complete the j)icture with regard
possibly to a
remaining compound of importance.
A modified form of the filtration approach has
also been
proposed - the use of compounds added to the
tobacco.to act
as scavengers of the substances being formed and
transferred
to smoke. The prospects for success via this route
at
present remain unknown.
CONCLUSIO14S
Philosophically, it is possible to divide the
approaches to noxa
reduction into two groups.
1. Empirical, and
2. Knowledge-based
At present most strategies would have to be
defined as empirical
as we lack the knowledge to develop well-supported
and rational schemes
for the reduction of smoke components on a
selective basis. It therefore
seems appropriate that, in trying to make
progress, it should be done
in such a way that the information generated will
be useful in
understanding the problem that is being tackled
and will allow the
foundations to be laid for structured attempts to
produce 2nd or even
3 rd generations of smoking products.
We recommend that the following projects should be
considered which,
whilst different in scope, have complementary
features:
A. To evaluate current options for combustion
modification
The first stage would be to undertake a literature
review paying
particular attention to the prospects for
combining approaches
which individually have had limited success, and
which might have
possibilities of synergy. The procedures and
materials identified
would then be evaluated.
3. To fractionate tobacco in order to define the
contribution to
noxa from different sources and to achieve their
separation
Two approaches to this have been set out in the
Appendix. "A"
comprises a general scheme which would not only
identify sources
of noxa, but would also indentify potential
approaches via genetic P~6
improvement and combustion modification. "B" is
more limited and C)
sets out to achieve reductions in a number of
components. C:)
ca
__J
BATCo document for Province of British Columbia 28
October 1999


Beyond this, it is envisaged that this programme
would also help
to define the future desirable combinations of
tobacco - identical
or related materials for 2nd generation smoking
products. An
example of the latter can be proposed now and may
be considered
as a speculative project for support -
polysaccharide threads of
controlled diameter impregnated with alkaloid and
combustion
modi fi ers. Alternatively a less extreme version
might involve
the knowledge of combustion modification or
extraction to derive
altered smoke deliveries using the DEER technology
or other
appropriate process route.
CX3
BATCo document for Province of BritiSh Columbia 28
October 1999


-6-
APPENDIX
A. OUTLINE OF A GENERAL SCHEME FOR TOBACCO
FRACTIONATION
The following is by way of example only. It is
envisaged that it
woul: be conducted separately on both flue-cured
and burley tobaccos,
usin:: blends of lamina only initially.
PROC:_-URE
1. Prepare 500 kg base blend in strip form
2. Sample for analysis. Store for fractionation
programme
3. Laboratory scale fractionation to optimise
initial approach
4. ?ilot scale 'to establish schedules
5. Treatment of bulk sample to produce fractions
sufficient for
cigarette manufacture
6. Analysis of fractions and cigarettes
7. 'Development of modifications (chemical,
physical, enzymatic'
Exam:
FRACTIONATION BASED ON SEQUENTIAL EXTRACTION
POLAR -- NON-POLAR OR VICE-VERSA
SOLVENTS1
a - b - c
e. a w pentane
b = acetone
c - water
SDL~,SLE
FRAC71ONS Analyse RESIDUE .......... What
inorganics r--nain?
a, c How much N left?
RecombineI D_
cigarettes
mo,-;fly Modify ........... Can we remove residual
inorganics, secondary
Recombine thickening polysac:.harides,
nitrogen?
C=)
C)3
-j
I%j
01-
Jt~-
BATCo document for Province of British Columbia 28
October 1999


-7-
B. PROPOSAL FOR A FRACTIONATION STUDY BASED ON
WATER EXTRACTIONS ONLY
Speci-ic targets: Reduction of heavy metals,
Cyanide and Ames activity,
Carbonyl compounds,
NOx and nitrosamines (Burley only)
HYPOTHESIS:
1. Heavy metals, specifically cadmium, will be
mainly present in
insol;ble residue after water extraction.
2. Significant cyanide and Ames activity may be
generated from
nitrcien in insoluble fraction. -
3. Carbonyl compounds may be reduced by
modification of residue.
4. NOx End nitrosamine may be reduced by removal
of nitrate etc.
from -sater solubles.
EXPERIMENT;7ION:
1. Water extraction of tobacco to give 2
fractions. Analyse for
meta'.s and nitrogen in each.
2. Trea-. residue:
a) with chelating compounds to modify metal
content
b) use enzymes to solubilise nitrogen and possibly
modify
polymers;
c) consider chemical or physical treatments to
modify form
of residue;
3. Tre-x-. extract (Burley) to remove nitrate etc.
CD
k_n
BATCO document for Province of BritiSh COIUMbia 28
October 1999


-8-
TABLE I
"Tar"
CO
Carbonyl compounds
Aromatic and polyaroinatic compounds
Cyanide Listed by
Nitrosamines Isc
Oxides of nitrogen
Metals including Cd, P0210
Free radicals
Ethylene Additional components
Mutagenic Nitrogen heterocycles we have considered
"Ames' activity
ISC : Independent Scientific Committee on Smoking
and Health (UK)
CD
CD
1.0
OD
BATCo document for Province of BritiSh Columbia 28
October 1999


-9-
TABLE 2
SMOKE COMPONENTS AND THEIR ORIGINS
MAINSTREAM SS/MS
COMPONENT PYROLYSIS/PYROSYNTHESIS DELIVERY RATIO
(ug/CIG)
NO NITRATES 60 - 300 3 - 6
HCN AMINO ACIDS, PROTEIN 100 - 300 0.006
VOLATILE ALKALOIDS, VOLATILE 0.005 10 - 150
NITROSAMINES BASES, NITRATES
TOBACCO SPECIFIC DISTILLATION (FORMED DURING 0.03
- 0.9 1 - 25
NITROSAMINES CURING): NICOTINE AND
TOBACCO ALKALOIDS
ACETALDEHYDE DISTILLATION; CELLULOSE, 200 - 1000 1
- 2
STARCH, LIGNIN, SUGARS
BENZO[A]PYRENE SOLANESOL. SUGARS, AMINO ACIDS,
0.01 3.4
NICOTINE, TERPENE, LIPIDS, ...
PHENOL DISTILLATION; POLYSACCHARIDES, 200 - 400
2.6
LIGNIN, PROTEINS, AMINO ACIDS...
Cd TRANSFER FROM TOBACCO 0.007 - 0.4 3 - 7
FREE RADICALS ? ? ?
SS - Sidestream; MS Mainstream
c0
BATCo document for Province of British Columbia 28
October 1999


-10-
TABLE 3
PROSPECTS FOR MODIFYING LEVELS OF SMOKE
CONSTITUENTS BY GROWING PRACTICES
Smoke Direct (1) Indirect (2)
Component Consequence Consequence
Tar Genetic(?)
co Genetic(?)
Carbonyl Compounds Genetic(?)
Aromatic and Poly- Genetic Genetic
aroinatic Compounds Curing
Cyanide Genetic
Curing
Oxides of Nitrogen Agronomic
Genetic
Nitrosamines Agronomic Genetic
Curing
Free Radicals Agronomic(?)
Genetic(?)
Curing(?)
Ethylene Genetic(?)
Metals Agronomic Genetic(?)
Ames Activity Genetic Genetic
Curing(?)
Note
(1) A Direct Consequence indicates where selective
change in the area
indicated may be expected to lead to change in the
component indicated.
(2) An Indirect Consequence indicates a potential
change. For those
with (?), this is an assumption rather than based
on existing evidence.
CZ)
CD
1 ~0
co
-4
00
BATCo document for Province of British Columbia 28
October 1999


-11-
TABLE 4
MAJOR PERMITTED ADDITIVES TO CIGARETTES FROM THE
ISC LIST I
Maximum Allowed
Ammonium Carbonate 4
Sulphate 4
DiAmmonium hydrogen phosphate 4
Bentonite 5
Butane-1,3,-diol 3
Calcium carbonate 5
Calcium chloride 5
Cellulose 5
Citric Acid 4
Dextrin 4
Diatomaceous earth 5
Glucose 10
Glycerol 10
Invert sugar 10
Liquorice 4
Magnesium chloride 5
Methylcellulose 5
Potassium carbonate 5
Propylene glycol 6
Sodium carboxymethyl cellulose 5
Sodium chloride 5
Sodium hydrogensulphate 5
Sorbitol 8
Sucrose 10
Use of the above is also restricted to total
amount of these and other
'List I' materials not exceeding 15% bi-weight in
the tobacco blend.
4 :h
C)
C=)
-10
c0
--4
BATCo document for Province of BritiSh Columbia 28
October 1999