RESTRICTED
GR&DC RESEARCH PROGRAMME
STATUS REVIEW NOTES
covering the period
January June, 1982
DISTRIBUTION:
Dr. L.C.F. Blackman
Dr. R.A. Sanford
Dr. P.J. Dunn
Herr E. Rittershaus
Dr. F. Seehofer
Dr. C.J.P. de Siqueira
Mr. P.M. Denton
Mr. R.C. Nicholls
Mr. K.L. Heard
Dr. C.I. Ayres
Dr. R. Binns
Mr. D.E. Conway
Dr. S.R. Evelyn
Dr. M.J. Hardwick
Dr. M. Oldman
Mr. P.H. Gannaway
00
cc
0 1982 British Amicrican Tobacco Co. Ltd. This
report must not be copied or shown to unauthorised
persons.
BATCo document for Province of British Columbia 5
November 1999


GR&DC RESEARCH PROGRAMME
.STATUS REVIEW NOTES
covering the period
January June, 1982
Contents:
Work Area 01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
Biological
Filters
Tipping and Paper
Cigarette Design
Combustion
Sidestream
Smoke Retention/Human Smoking
Psychology and Sensory Testing
Smoke Taste and Flavour Development
Leaf and Biotechnology
Tobacco Processing
Tobacco Expansion
Secondary Processing
Ventilated Cigarette Technology
Test Method Development
a 1982 British American Tob.2cco Co. Ltd. This
report must not be copied or shown to unauthoriscd
persons.
C---)
1-0
C;0
cc
-r-:.
C=)
BATCO document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 01 Biological
Author: R.E. Thornton
Period Covered: January - June, 1982
OBJECTIVE:
To assess, in quantitative terms, the biological
response to cigarette
smoke and its constituents and to develop
acceptable cigarettes with
minimum biological activity.
PROGRESS:
There are five areas of study:
1. Direct measurement of biological activity in
products and the assess-
ment of changes in product variables such as paper
permeability, etc.
2. Smoke Inhalation Studies.
3. Development and use of short-term tests related
to cancer.
4. Other short-term tests.
S. Bioassay of nicotine.
The work going on in these areas is as follows:
1. Products of reduced biological activity
(a) Empirical approach
The main activity of the year was quite unexpected
twelve months
ago. Product variables such as paper permeability,
tipping ventila-
tion, filter pressure drop and blend composition
had been investi-
gated in two skin-painting experiments (B14 and
B15). Unfortunately,
subsequent to an analysis of the results# errors
in the records
kept by the contract research laboratory carrying
out the study,
were detected. 53 animals in study B14 and 146
animals in B15 were
incorrectly recorded. These errors have now been
corrected and
re-analysis is proceeding as a matter of urgency,
and is scheduled
for completion in July.
(b) Fundamental studies
Studies into the role of tobacco leaf proteins as
precursors of
mutagerLic compounds of high activity are being
planned: assays
will be carried out by the Ames Test.
2. Smoke inhalation studies
(a) Investigation of product variables
A major project has been the adaption of
experimental protocols to
permit the examination of low tar products. For
many such products,
0 1982 British American Tobacco Co. LtcL This must
not be copied or sho- tc. unauthorised persorm
cc
CD
Qr
~Jl
BATCO document for Province of British Columbia 5
November 1999


changes in the concentration of smoke components
(such as nicotine
and carbon monoxide) relative to tar, precluded
examination by
earlier protocols. This work has been concluded
and a series of
low delivery cigarettes has been examined.
Cigarettes containing
expanded tobacco are being studied by a number of
protocols,
including a one day test developed by Battelle
(Geneva) for Gallaher
Ltd. The activity of sidestream smoke is also
under examination:
preliminary results suggests different sites in
the respiratory
tract are affected in comparison to mainstream
smoke.
(b) Development of improved techniques for
investigating the response
of rodents to inhaled cigarette smoke
This project has been abandoned. We came to the
view that an eight
week rodent (hamster) study was unlikely to give
useful information:
a two year study would be required. This was
considered to be an
ineffective way of utilising our limited
resources.
3. Short term test related to"cancer
(a) Development of mutagenicity assay
The Ames' Test using Salmonella typhimurium has
now been established
at Southampton. Different smoke condensates have
been examined in
a collaborative test involving the Hamburg,
Montreal and Southampton
laboratories: the results from the different group
were-in excellent
agreement.
In preliminary experiments, fresh sidestream
condensate has been
examined for mutagenic activity and has been found
to possess
similar activity to that observed for mainstream
smoke. However,
it is exvected that this relationship between
sidestream and main-
stream smoke may vary depending upon smoking
conditions, and this
is under investigation.
(b) Development of a chromsome damage assay
To fulfill the minimunt test battery to ensure an
optimum system for
detecting mutagens, an in vitro assay to
investigate the ability
of compounds under investigation to induce chr
omosome- damage in
lian cells is recommended. The in vitro human
peripheral
blood lymphocyte assay system is now under
development and will be
able to detect the ability of compounds to induce
chromosome muta-
tions.
(c) DNA-Bindi-ng
in view of the progress that has been made in
project (d) below,
work on DNA-binding has been given law priority
and this project
will be reviewed in the Autumn when a biochemist
(Dr. P.M.M. Godden)
is scheduled to join the staff.
(d) Biochemistry of chemical carcinogens
Studies involving the University of Surrey.
It has been demonstrated that the enzymp
ethoxyresorufin de-ethylase
(ERD) , a Cyto-P448 linked enzyme and part of the
microsomal detoxifi-
cation sYstem, is frequently involved in the
activation of chemical cc
compounds to potentially active carcinogens. The
general response CC)
0 1992 British Aincrican Tobacco Co. Ltd. This
must not be copied or shown to unauthorised
persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 3 -
of this enzyme following in vivo administration of
a spectrum of
structurally related known carcinogens or
non-carcinogens has been
studied at Surrey. The preliminary findings
indicate that the
induction of ERD is highly correlated with known
carcinogen exposure.
Collaborative studies between GR&DC and Surrey
have been concerned
with studying the changes in lung levels of ERD
together with changes
in its kinetic properties following smoke
exposure. Initial studies
indicate the following relatively short periods of
smoke exposure
ERD can be induced considerably (6-10 fold) with
changes in its
kinetic properties. These studies are continuing
to investigate
M the longer time course of smoke exposure (ii)
the effect of
exposure to different smoke dilution as wiBll as
different cigarette
types and (iii) changes in the lung ERD activity
following exposure
to known carcinogens.
4. Other short-term tests
(a) Test related to emphysema
Efforts have been concentrated on studying the
retention of nitrogen
dioxide, a compound that has been associated with
the development
of emphysema, work being carried out under
contract for the Verband
der Cigarettenindustrie. We have also kept in
informal contact with
similar work being carriedout at Papworth Hospital
by Dr. T. Higen-
bottam.
M Test related to bronchitis
This test was appraised by Dr. F.J.C. Roe in
September 1981. He
concluded that this technique had merit as a
method for quantifying
irritation, but was not necessarily related to
bronchitis. Work has,
therefore, continued and encouraging results
obtained. The time
course of the response has been established and
high sensitivity
(to smoke) established. Currently, the dose
response curve is being
established.
(c) Development of models for atherosclerosis
Work has progressed on the development of a
physico-chemical model
for modelling atherosclerosis, based on the notion
that athero-
sclerosis occurs in regions of low shear (contract
studies, Imperial
College, London).
S. Bioassay of nicotine
Demand for this popular assay continues to be
brisk. Currently we are
attempting to extend the assay to include
measurement of a second
nicotine metabolite, Nicotine N-Oxide.
6. ZORBONITE
The filter material nitrosobenzene, ('ZORBONITE'),
was examined by a
battery of tests chosen in consultation with the
Group's consultants.
Unfortunately, significant '~Aological activity
was found in certain
tests, notably skin sensitization, and the use of
this material was not
reca=ended. cc
cz~
0 1982 Bri6sh American Tobacco Co. Ltd. This must
not be copied or sho.n to unautborised pcrson3.
BATCo document for Province of BritiSh Columbia 5
November 1999


4
FUTURE WORK
Sidestream smoke will be studied in greater
detail, with attention being
given to t!ie development of better methods for
collecting sidestream
smoke. it is also likely that some further work on
expanded tobacco will
be required, if only to clarify differences in
results reported by
ourselves and Gallaher Limited.
New studies being considered include work aimed at
examining the effects
of smoking on weight gain and metabolic rate in
humans and collaborative
studies into the effects of smoking in animals and
humans on lung permea-
bility, the so-called "Leaky lungs".
CD
I'D
LO
CY-1
I
0 1982 BnmO American Tobacco Co. Ltd. This must
not be copied or shown to unauthorised pemons. CO
BATCo document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 02 Filter Research
Author: H.G. Horsewell/J.A. Luke
Period Covered: January - June, 1982
OBJECTIVE:
To develop novel filters and filter technology,
aimed at the development
of marketable lo-d-tar products, with particular
regard to human smoking
patterns.
PROGRESS:
The areas of study are as follows:
(1) Selective Filtration:
Filters which selectively remove nitrosamines,
nitric oxide or
hydrogen cyanide.
(2) Controlled Smoking Filters:
(a) Fundamental aspects of parameters which affect
smoke particle
capture;
(b) Concepts for producing filters which will
modify the puff-
by-puff profile;
(c) Concepts for producing filters which will help
in the design
of cigarettes having reduced deliveries of gas
phase consti-
tuents, particularly carbon monoxide.
(3) Filter Technology Development:
Investigate microwave drying technology in order
to achieve rapid
curing of bonding agents in filter manufacture.
(4) Advisory and Support:
(a) Development, with Hercules, of polypropylene
tow and filters;
(b) Evaluat-ion of techniques and materials from
external sources;
(c) Design and manufacture of special filter
samples for GR&DC
and Operating Companies.
1. Selective Filtration
(a) ?~'itrosamines
The main aim of this work is to discover filters
which will remove
the volatile non-tobacco specific nitrosamines.
The tobacco 00
specific nit-rosamines are substantially
non-volatile and would be Co
expected to reside in the smoke particles - hence
it is unlikely
CD
that these can be filtered selectively. (-M
0 1982 British Arnerk= Tobacco Co. Ltd. This must
not be copied or shown to unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 2 -
So far, a range of different fibrous filter
materials has been
examined which includes cellulose acetate
(control), polypropylene,
polyam.ide (nylon) and polyester. None of these
materials was as
effective as cellulose acetate but some were only
slightly less
efficient. There was a suggestion from the results
that the
addition of the hardener to polypropylene filters
increased the
filtration efficiencies. Work is in hand to
explore this further.
Additional future work will include a study of the
effect of
plasticising level (triacetin) and the effect of
changing the
filter pressure drop on the filtration of volatile
nitrosamines
and tobacco specific nitrosamines by cellulose
acetate filters.
A further aim of this study is to improve the
performance of poly-
propylene for nitrosamine removal as this material
is clearly less
effective than cellulose acetate. It is unlikely
that it will be
possible to chemically modify the nature of
polypropylene without
help from outside sources.
(b) Nitric oxide
After the decision to terminate the ZORBONITE
project, contract
work with Brunel University has commenced. This is
aimed at
understanding why activated carbon is unique for
the performance
of C-nitroso compounds. if this problem can be
solved then it
might be possible to find other supports for
additives to filters
which could remove nitric oxide.. So far,
techniques for studying
the form of the C-nitroso compound on carbon
surfaces are being
developed. It is anticipated that a further year
will be required
before the project will be completed.
A second approach which is being pursued is the
grafting of
suitable functional groups onto the fibres. In
this respect,
Dr. J-T. Guthrie of the University of Leeds has
been engaged as
a consultant. As yet, it has not been possible to
commence work
in this area, but very soon work in GR&DC will be
progressed
using cellulose acetate as the starting material.
(c) Hydrogen cyanide and aldehydes
No work on this has been carried out so far, but
the grafting
approach will be used to generate filters for
these smoke consti-
tuents.
2. Controlled Smoking Filters
(a) Fundamental work on smoke particle capture
This work has been in progress for the last year.
initially it
was necessary to develop new methods of measuring
the filtration
efficiencies when smoke is drawn through the
filter at different
velocities. A modified smoking machine with a very
large piston
has been obtained from Borgwaldt. The performance
of this machine
has been reported. A very sensitive
spectrophotometric method for
measuring very small quantities of smoke particles
in filter tips
has been developed. Preliminary studies of the
particle capture
mechanism, using cellulose acetate and paper
filters, have indica-
ted that the general smoke velocity/filtration
efficiency patterns
are similar to those found some years ago ustng
synthetic aerosols, cc
but the relative contributions of the various
mechanisms do not cc
agree with those found previously. One interesting
result is that
C__
0 1982 British Americ.n Tob.,,c. C3. Ltd. This
must not be copied or shown to unnuthorised
persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


the filtration efficiency for nicotine decreases
markedly as the
velocity through the filter is increased. A
detailed interpreta-
tion of the results is premature.
(b) Filters for modification of smoke profiles
The purpose of this work is to devise methods of
controlling puff-
by-puff delivery profiles of cigarettes without
affecting adversely
the "Smoking Mechanics". The study has been
directed to evaluating
basic principles rather than specific commercial
designs.
Working on the hypothesis that the high pressure
drop in the HEE
filter was the result of immobile tar deposited in
the constriction
of the filter, several filters were examined
designed to ease the
migration of tar in the filter and hence relieve
the pressure drop.
Results obtained suggested that the mechanism by
which these filters
produced level delivery profiles wai inescapably
linked to increases
in pressure drop. This, combined with the
variation in puff volumes
and puff velocities ' observed in populations of
human smokers,
suggests that controlling deliveries by a ve loci
ty-dependent
mechanism is impracticable. It is therefore
necessary to re-think
the approach.
Cigarette deliveries are affected by tobacco
blend, tobacco column
construction, filtration -and ventilation. To
control delivery
profiles, these parameters would need to change
during smoking.
The initial work included controlling tip
ventilation as a means of
regulating the delivery profile. Normally,
ventilation is a function
of hole size, number of holes and the balance of
draw resistance
pre- and post-ventil zone. To circumvent the
predominantly linear
pressure drop versus flowrate relationship of
fixed hole size
ventilation, filters have been examined with flaps
to provide
variable hole size and high efficiency pre-ventil
filter sections
to increase pre-ventil draw resistance during
smoking. The results
from some of these designs are encouraging.
Various factors which can change during smoking
and hence regulate
puff-by-puff deliveries have been examined. The
important parameters
are length, cross-sectional area, structure,
pressure drop, filament
denier, tow denier, packing fraction, total
surface area and specific
surface area.
An initial prototype being examined causes the
smoke path through
the filter to change and hence alters the
effective length of the
filter during smoking. The design depends upon the
use of small-
circumference filters wrapped in porous plugwraps
and surrounded by
an annular channel down which the smoke passes
before entering the
filter. Alternative pathways controlled by the
changing relative
pressure drops as the cigarette is smoked
essentially change the
effective length' of the filter. The design can
also incorporate
:everal different filter sections in a composite
filter.
mall-circumference filters have been obtained for
evaluation of
he variable path length filter, both as single
filters and as
composite filters. CZ)
1 ~0
(c) High pressure drop/low efficiency filters C)O
CIO
About two to three years ago, GR&DC was asked by
BAT (Hamburg) to _V__
investigate the possibilities of producing filters
which had a C::)
CT1
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 4 -
pressure drop of 50-100 mm WG and an efficiency of
20% or less.
The filters must be capable of being ventilated.
The purpose
was to provide draw resistance for highly
ventilated cigarettes
for reducing carbon monoxide deliveries. Other
companies (B&W,
Sphweitzer and the Celanese Corporation, USA) had
realised the
attractive potential of this idea.
Two somewhat different approaches were adopted to
meet this
objective. Attention was given to the possibility
of using
large-diameter fibres. The main difficulty was
obtaining such
fibres. Bridon supplied 14 and 32 dpf
polypropylene, which was
difficult to process into filters and the
resulting products
had filter efficiencies which were too high. ._
Filters were then prepared using very large nylon
filament.
It was found that the specification could be met
but the
filament denier required was well outside the
normal range of
filter tows and the filter tips were much too
heavy.
meanwhile, attention was directed towards a filter
having a
small-bore tube sealed into it to provide a smoke
passage. A
tube about 0.8 - 1.0 mm diameter gave the desired
performance
and met the specification. Patent coverage has
been applied
for. Samples have been sent to BAT (Hamburg) and
comments are
awaited. At present, filters of this type can only
be made in
the laboratory by hand. If there is a requirement
to produce
these in quantity, consideration can be given to
sealing the
tube into cellulose acetate filters by spin
moulding. Alterna-
tively, it should be possible to produce filters
in which the
tube is embeddecf in closed cell foam, preferably
cellulose
acetate, or polypropylene.
It is not intended to carry out further work on
this type of
filter until there are definite indications that
these filters
are of commercial interest.
Spin moulding of filters was the second approach
adopted and
the code given to this type of filter was CSF11.
The CSF/1
filter has an orifice restriction to the smoke
flow and this
restriction is created by the spin moulding
operation. Rela-
tively high pressure drop is generated by the
orifice and low
filtration efficiency is achieved by using high
denier per
filament tow. For good formation of the orifice it
is necessary
to use a thermoplastic plugwrap.
Simple tests showed that the CSF/I filter could
significantly
reduce the carbon monoxide to tar ratio for low
delivery
products (ri5mg) whilst retaining good draw
resistance. At
that stage it was decided to explore the technical
and commer-
cial viability of the CSF/I filter concept at the
5mg delivery
level. The objectives were to evaluate the filter
manufacturing
process (spin moulding) and the assembly of
cigarettes, to
examine machine smoking performance (including
variability in
delivery) and to carry out a subjective
evaluation. Encouraging
results have been obtained for each of these
objectives.
Recent effort aimed at comparing the subjective
qualities of
the CSF/I product with a LOCO product (a
"conventional" design
stretched to the limit for producing a low carbon
monoxide to
tar ratio) and a control (conventional) product
showed that, cc
with a small group of confirmed low tar consumers:
CD
0 1982 British American Tobacco Co. Lid. This must
not be copied or shown to umuthorised persons.
BATCO document for Province of BritiSh COIUMbia 5
Novernber 1999


- 5 -
- CSF/1 and control compared favourably;
- CSF/1 has a subjective advantage over the LOCO
product.
All of the products under comparison were designed
to give 5mg
of tar (PWdNF); both the CSF/1 and the LOCO
products gave a
carbon monoxide to tar ratio of 0.5 : 1, whereas
the conven-
tional control product gave a ratio of 0.9 : 1.
Currently, work is under way to confirm these
initial findings
by carrying out a much wider subjective assessment
of the
three products through conducting a pilot scale
consumer test.
To support the work on the CSF11 filter and to
provide opportu-
nities for creating novel filters or products,
effort has been
directed towards the further development of
thermoplastic plug-
wraps and the spin moulding technology.
The highly porous thermoplastic plugwrap produced
for us by
Schweitzer has been developed to a point of
commercial readi-
ness. In parallel a zero-permeability plugwrap has
been
developed with Wiggins-Teape. Work is continuing
on producing
further materials to cover the permeability range
in between
these. The machine technology centres on a
pre-production
machine which will now produce CSF type filters (a
relatively
complicated spin moulding design) at 750 rods per
minute in a
single operation. Since the Wiggins Teape plugwrap
allows
the elimination of sealant from this filter,
speeds up to
1,000 r.p.m. should be realised soon.
The flexibility and scope of the process is being
explored by
the creation of a series of novel moulded product
ideas which
can be produced as a result of having mouldable
plugwrap and
mouldable tipping papers. These designs are being
carefully
screened by a variety of techniques, including
subjective
assessment. It is not intended to go beyond the
feasibility
stage for these products.
3. Filter Technology Development
Microwave Drying
Following the closure of the Lucas Aerospace
Microwave group,
attempts were made to establish working
relationships with
both Philips and Thompson to permit joint
development of
systems which would meet our objectives.
Contractural negotia-
tions were not a success and were brought to an
end. Currently,
discussions are being held with Raytheon and
Westinghouse.
4. Advisory and Support
(a) Polypropylene tow
Discussions on the continuation of the working
relationship
between GR&DC, Souza Cruz and Hercules have taken
place
throughout the period but have not resulted in a
clear defini-
tion of GR&DC's role. Emphasis has remained on the
develop-
ment of a 4.SY/35000 tow for Souza Cruz which is
being test Co
marketed. The development of 2.2Y/28000 tow has
not progressed 0__
as intended because of these uncertainties,
resulting in a
cp'?
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 6 -
moratorium an joint work with Hercules, so the
direction of
GR&DC work has been altered towards more
fundamental issues
and alternative methods of processing fibres.
Major studies
have iacluded O/W`P and selective filtration,
subjective assess-
ment of polypropylene filter and hardener
properties, initial
work on an alternative hardener, measurement of
on-machine
tension forces generated in tow, and tensile
testing. A con-
tract has been placed with Shirley Institute for
the develop-
ment of the scientific understanding of filter
hardness.
This model is intended to identify whether
polypropylene
fibres are likely to exhibit different mechanical
characteris-
tics to cellulose acetate fibres, because of
differences in
modulus, surface friction coefficient, etc. Gear
crimping
may have some value for crimping polypropylene
tows and initial
studies have been undertaken.
(b) Cellulose Acetate Tow Development:
Several types of experimental cellulose acetate
tow are being
examined including tows with mixed sized fibres,
different
crimp forms or large dpf. Main interest lies in
obtaining
better yield and hardness in rods through enhanced
blooming
and better crimp forms. This is especially true
for lower dpf
tows (<2pdf) - The quality of crimp and the
matching of the
crimp and fibre characteristics to the machine
capability have
been of particular interest.
(c) Sample Filter Manufacture:
A wide range and large number of filter samples
have been
produced for GR&DC laboratories, product
development work and
Operating Companies to support research and
product projects.
This activity continues to provide a base of
current filter
making capability and know-how from. which new
ideas can be
evolved.
FVrURE WORK:
The current work on the filtration of nitrosamines
will be extended
to include a study of changing the plasticiser
level in cellulose
acetate filters. It is also intended to explore
the effect of
changing the level of hardener in polypropylene
filters. In
addition, the effects of changing the filter
pressure drop will be
studied.
Future work on the filtration of nitric oxide and
hydrogen cyanide
will include a feasibility study of grafting
suitable functional
groups onto cellulose acetate and, in due course,
advantage will be
taken of any finding as a result of the work at
Brunel University.
The fundamental work on smoke particle capture
will be extended to
a study of different filter types, including those
with different
fibre sizes, the overall aim being to discover the
important factors
which influence the particle capture mechanisms.
In the future, it
will be necessary to combine this work with
particle size measure- CD
ments as the puff velocity could cause a change in
particle size
which can change the particle capture mechanism.
cc
Co
CD
C7 N
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unauthorised persons.
BATCO document for Province of BritiSh Columbia 5
November 1999


- 7 -
Work on filters for the modification of smoke
profiles will concen-
trate for the time being on filter designs in
which the effective
length of the filter is increased during smoking.
Obviously,
alternative designs based on this theme will be
examined.
The spin moulding technique, together with the
associated thermo-
plastic plug-draps, are likely to figure strongly
in the design of
novel filters or products, for modifying puff
profiles, reducing CO
deliveries or imparting other subjectively
beneficial characteris-
tics to low delivery cigarettes. Thus most
emphasis will be on
fully exploiting the potential of the process and
extending the
range of per-meabilities for the plupwraps.
Support for CSF11
filters will continue.
The exploitation of polypropylene tow remains a
major goal. It is
anticipated that a research programme will be
formulated with
Hercules to cover areas of knowledge essential to
the satisfactory
world-wide use of polypropylene tow but which will
not be dealt
with by the Souza Cruz/Hercules Joint Venture.
Areas proposed
include selective filtration, hardener and thermal
bonding. Some
work on alternative texturing may be pursued.
New process techniques or filter materials will be
examined for
their potential value in filter operations as they
become available.
00
cc
Xli-
CD
0 1982 British Ancrican Tobacco Co. Ltd. This must
not be copied or shown to unnuthorised persons.
BATCO document for Province of BritiSh Columbia 5
November 1999


RESTRICTEO
STATUS REVIEW NOTE
Work Area: 03 Papers and Tippings
Author: H.G. Horsewell
Period Covered: January - June, 1982
OBJECTIVE:
To create opportunities for advancing the design
of cigarettes by:
(1) Identifying recent developments in Uppings and
cigarette paper;
(2) Modification of cigarette papers with chemical
additives and
different fillers.
PROGRESS:
All the work carried out in this Work Area has
been connected with the
Sidestream Projects (work Area 06). "Tippings and
Papers" will not
appear in the 1983-1985 Research Programme.
C)
1 ~0
co
cc
0 1992 British A-6-n Tob,2cco Co. Ltd. This must
not be copied or shown to unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NCTE
Work Area: 04 Cigarette Design
Author: T. Hirji
Period Covered; January - June, 1982
OBJECTIVE:
To optimise the exploitation throughout the Group
of new Cigarette Design
Concepts and raw materials.
PROGRESS:
There are three broad areas of current study:
I .
1,2. The development, evaluation a nd application
of techniques which are
more effective for the characterisation of tobacco
and smoke;
3. To identify the optimum use of expanded tobacco
in Virginia and US
blended products;
4. To understand the characteristics of acceptable
ultra-low delivery
cigarettes.
1. Grade Matching
Over the past six months, the Oil/Water Partition
(O/WP) data collected by
I.T.L. (Canada) for their 1979 crop have been
validated on the Cyber. From
this data, Ontario grade H2 (medium mature
unblemished cutter) has been
matched using other Ontario grades. The model used
to derive the match
depends upon the fact that grades which are
acknowledged to have different
smoking qualities by the blender have different
O/WP profiles. Thus, the
grade matching model combines the expertise of the
blender with the expertise
of the chemist and this has been possible largely
due to the co-operation and
support from Mr. Andy Schaeffero the Imperial
blender.
cigarettes based on the computer solutions have
been manufactured and are
currently being tested in Canada. The samples have
also been assessed by the
GR&DC DELTA panel and the results are encouraging.
-
2. Tobacco and Smoke Characterisation by Fourier
Transform Infra-Red (FTIR)
Spectroscopy
Considerable effort has been spent on:
(i) Perfecting the diffuse reflectance technique
(sample preparation
and minimising within sample variance);
(ii) Setting up photo-acoustic spectroscopy
facility, which is also
available in the FTIR instrument;
(iii) Establishing the use of FTIR as a df--'ector
to characterise the
fraction from a gas chromatographic examination of
smoke extracts, C::'.
e.g. O/WP solution. %1C
C-C
cc
I*,-
0 1982 British American Tobacco Cm Ltd. This must
not be copied or shown to unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 2 -
The work on smoke fractions is in the early
stages, but it has been shown that
the instrument can characterise tobaccos very
quickly and very accurately. The
instrument provides a continuous scan over the
wave number range 400cm-1 to
4000cm-1, which means that a lot of detailed data
is generated. Consequently
the FTIR has been linked to the Cyber computer so
that the information can be
available quickly and in a convenient form for
further analysis.
FTIR has also been used to analyse liquid samples
(for Flavour Group) and for
surface studies related to filter tows and
carbons.
3. Expanded Tobacco
The study of the effects of DIET expansion on a US
blended product (POET II)
is now completed and the report is in preparation.
As expected, there were
losses of humectants and casings during expansion
- 10 to 20% loss of tracer
compounds for casing materials and 5 to 20% losses
of humectants have been
observed. The wide range found in the case of
humectants will necessitate
further analysis.
Detailed chemical and subjective assessments of
cigarettes containing 33% of
different cortions of the blend in the expanded
form have been carried out.
Multivariate mapping techniques are being applied
to obtain qualitative/
quantitative description of the chemical changes
and attempts are being
made to link these with the subjective assessment.
The delivery reductions from POET II samples are
much in line with expecta-
tions. However# the work carried out and reported
in Project RUGBY (an MPDC
project done jointly with Product Development
Services - PDS) suggests that,
in the case of 100% expanded tobacco cigarettes,
the deliveries may reduce
with increasing density. The reason for this is
not clear at present but, if
this is a real effect, then it may be possible to
exploit it in cigarette
design. The same project also shows that, with
increasing levels of expanded
tobacco, it will become more and more difficult to
make cigarettes which
match the current cigarettes in all physical
properties. Some compromise
will therefore have to be reached and it is no.
becoming essential that we
know what physical properties are considered to be
most important by the
consumers.
Practical blending strategies applicable to
ultra-low delivery Virginia
products have also been investigated and reported.
The experimental products
(containing 40% expanded tobacco) with deliveries
of 1-2mg were found to
have smoking properties equivalent to the 4-5mg
gommercial products used in
the test.
4. Smoking Dynamics
it has been recognised for some time now that low
delivery products of
equivalent machine deliveries can produce very
different human sensations,
particularly during puffing. These sensations are
used by consumers to
discriminate between the low delivery products.
Considerable effort has been and is being spent in
identifying what sensations
are involved and how to maximise these desirable
sensations from cigarettes
delivering 5mg or less to the human smoker.
Various concepts have been
developed and presented to groups within and
outside GR&DC with encouraging
response.
Currently, an exploratory project (with PDS) for
developing models for ranking
cigarettes in smoking sensation terms using
laboratory data is in progress.
0 1982 British America,, Tobacco Co. Ltd. This mwt
not be copied or shown to onauthorised persom.
cc
CZ)
ON
CZ~
BATCo document for Province of British Columbia 5
November 1999


- 3 -
The concept of "reward for effort" forms the basis
of the present investiga-
tions which involve M determination of deliveries
(reward), flows, pressure
drops, etc., (effort) under non-standard
conditions and (ii) subjective
assessment of pairs of cigarettes whilst recording
the smoking behaviour of
the subjects. We believe this is the first time
that such a technique has
been used within the Group.
FUTURE WORK:
For practical applications, it is essential that
the grade matching model
performs consistently and that the data required
can be collected easily on a
routine basis. To satisfy the former criteria, a
series of matches based on
increasingly restricted sets of candidate grades
have been planned with I.T.L.
To satisfy the second criteria, two different
approaches are under investiga-
tion. Firstly, the development of a crop year
model such that only a few
grades need to be analysed each year - this will
be of general interest to
many Operating Companies who rely on domestic
tobaccos. Secondly, identifi-
cation and evaluation of alternative techniques
which are more practical and
convenient for analysing a large number of samples
in a short time - e.g.
FTIR, X-Ray fluorescence, near Infra-Red scan.
A direct comparison will be made between the
capabilities of the various
scanning techniques using Canadian flue-cured, US
Burley, Turkish Oriental
and US flue-cured tobaccos. All except the US
flue-cured tobaccos are on
site and data collection using FTIR and X-Ray
flourescence has commenced. On
receipt of the US flue-cured tobaccos near the end
of thi year, a full battery
of analysis, including O/WP and neutral volatile
flavour scan, will be applied
so that chromatographic and spectroscopic
techniques may be compared. Some
consideration will also be given to the potential
application of a pyroprobe-
GC (flash pyrolysing a small amount of tobacco in
controlled atmosphere and
analysing the pyrolysis products by GC) and laser
spectroscopy.
At the Technical Exchange Meeting, great interest
was shown in the use of
multivariate mapping techniques for describing
normal tobaccos and specially
processed tobaccos (e.g. with non-standard
conditions, expansion, denitration,
etc.). It is planned to develop suitable models in
the next period.
No more work on expanded tobacco has been planned
within this work. area.
The smoking dynamics project mentioned above
represents an empirical approach
and a much more fundamental approach for
understanding the science behind the
sensations is being planned as a major GR&DC
activity involving various
specialist Groups for'the conLing year. .
Besides smoking dynamics, work on
self-extinquishing cigarettes has been
identified as an area of continued interest. In
the last six months, this
subject has been discussed with Public Affairs
(Woking, Millbank, Brazil,
Mexico), personnel from B&W` and Courtaulds (UK).
In collaboration with the
Combustion Group, plans for the next phase are
under discussion and experi-
mental work should start as soon as these have
been finalised.
C=)
co
CC
-r-
0 1982 British Amcric.. Tb.cc. Co. Lid. This must
not be copied or shown le- unauaborised persons.
BATCo document for Province of British Columbia 5
November 1999


RLSTRICTED
STATUS REVIEW NOTE
Work Area: 05 Combustion
Author: R.R. Baker
Period Covered: January - June, 1982
OBJECTIVE:
To have available a comprehensive understanding of
the processes occurring
in a burning cigarette in order to provide a sound
basis for designing and
developing new products.
PROGRESS:
There are three broad areas of current study:
(a) the processes which contribute to the
formation/re lease of substances
in the burning zone of the cigarette;
(b) the factors which contribute *t*o whether the
products enter mainstream
or sidestream smoke;
(c) the processes which affect the final
concentrations as the smoke flows
through the tobacco rod and filter.
Each of these areas is being studied both
experimentally and by computer
simulation.
1. Release of nicotine and semi-volatile material
Reports on a general survey of the semi-volatile
components of mainstream
smoke (with boiling points between 70 and 300*C)
'and their formation and
distribution between the mainstream and sidestream
have been issued. This
work was done using a VG 707OF mass spectrometer
linked to a capillary column
gas chromatograph and data system.
About seventy identified substances in the
semi-volatile fraction of smoke
have been estimated quantitatively from cigarettes
containing various tobac-
cos. Important differences in the levels of the
semi-volatile components
in the different smokes were seen, suggesting that
the method could be used
to characterise the smoke from different
cigarettes. Pyrolysis studies have
shown that most of the substances are formed from
tobacco at temperatures
below 600*C and a literature survey has shown that
the majority are formed by
the thermal degradation of many tobacco
components. 80% of the semi-volatile
components monitored were delivered preferentially
to the sidestream, although
the sidestream/mainstream ratios varied from I for
4-methylphenol, through 4
for nicotine, to 50 for trimethylbenzene.
Probe studies in the cigarette burning zone have
shown that most of the
nicotine is released in the pyrolysis/distillation
region of the cigarette,
just downstream from the main combustion zone. The
exact release temperature
depends on the form of the nicotine in the tobacco
and pyrolysis studies have
shown that it ranges from 50-220*C for free
nicotine to 150-370*C for nicotine
hydrochloride.
2. Formation of ;idestream
A variety of experimental techniques have been
used to investigate the mecha-
nism by which paper with "active" magnesium oxide
filler produces low visible co
sidestream. To date, no satisfactory explanation
of the mechanism has been cc
found. Possibilities include a catalytic effect on
the sidestream precursor
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unauthorised pcrso.s.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 2
vapour as it diffuses through the partially
degraded paper, an effect of water
vapour released from the endothermic decomposition
of magnesium hydroxide on
sidestream aerosol formation, or an effect of the
elongated coal which is
normally associated with the magnesium oxide
paper. All of these possibili-
ties vill be investigated.
In order to obtain fundamental information of
sidestream formation, detailed
velocity distributions have been made in the plume
using a laser velocimeter
on loan from the Atomic Energy Authority at
Winfrith. The plume accelerates
from 400 mm/s at 10 mm above the cigarette to
about 600 mm/s at the 50 mm
level, due to the influx of air. The hydrodynamic
characteristics of the
plume are largely determined by the natural
convective flow of air across the
combustion coal.
Various computer simulations of sidestream
formation are being developed with
the help of external consultants. A model which
treats the coal as a point.
source is available and can make reasonable
predictions of the temperature
and velocity distributions in the plume. . This is
being extended to include
the occurrence of chemical reactions in the plume.
An existing general
hydrodynamic program at Imperial College, London,
is also being adapted to
describe the coal and plume. I ..
3. Puff-by-puff delivery model and empirical
parameters
A co=uter program for a semi-theoretical model of
a burning cigarette is now
available on the GR&DC computer. The program Is
simple to use and interactive
with the user. It calculates the-puff-by-puff
deliveries of carbon monoxide,
TPH, nicotine, water and PMWNF, as well as the
puff-by-puff Ventilation and
pressure drop of the cigarette. The model is a
mathematical analogue of many
of the physical processes occurring in the burning
cigarette and is based on
studies at BAT and elsewhere over the last twenty
years. It also includes
many experimental relationships concerning gas
diffusion, flow, filtration
and combustion. The present version can predict
trends for a wide variety of
cigarette designs, including expanded tobacco
rods, ventilated filters, dual
filters, unconventional filter materials and
diffusion filters. It can also
accommodate any machine or human smoking
parameters.
Comparing predictions of the model to experimental
results indicates some
discrepancies in predicted values but not in
predicted trends. Some of the
empirical parameters in the model need
re-determining and other physical
processes which occur in real cigarettes need to
be incorporated - for example
the variation of filtration coefficient with flow.
The effects of incorporating expanded tobacco in
cigarettes on many of the
combustion parameters that influence product
deliveries have been determined
using a variety of experimental techniques -such
as pyrolysis, temperature
determinations, carbon monoxide formation in the
burning zone and filtration
of particulate material. The results constitute a
bank of basic mechanistic
data vith which to rationalise the observed trends
of incorporating expanded
tobacco.
4. Computer simulation of combustion processes
The combustion processes occurring in the
cigarette include chemical reac-
tions, heat generation and transfer, phase changes
and mass transfer. These
processes occur in three phases and three
dimensions and vary with time.
The ultimate test of our understanding of the
interaction of these processes
- and the means by which the understanding can be
applied to cigarette design CD
- is through computer simulation. Several
simulation programs are being 110
developed which are mutually complementary. 00
CO
A model of the chemical combustion and pyrolysis
reactions has been formulated
and the Mathematics group at GRP.DC are applying
existing numerical packages
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shou-n to unauthoriscd prsons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 3 -
available on . the computer to solve the
differential equations. This model
Will define the chemical interactions occurring
and will provide essential
input for the other models. Because of other
commitments in the Mathematics
group, this model is still at a very preliminary
stage.
Pressure and flow aspects of a simplified single
phase model have been solved
numerically using an iterative finite difference
scheme. Good agreement has
been obtained between the numerical solution and
the fully analytical solution,
giving confidence in the numerical scheme chosen.
The same iterative scheme
was then used to solve the energy equation to give
predictions of the gas
phase temperature distributions. Graph plotting
routines have been developed
which allow isobar, velocity vector and
temperature contours to be plotted
interactively via a computer terminal.
A program has been written, based on empirical
correlations in the literature,
which predicts the variation of physical
parameters of gases with temperature.
This program will be used as a subroutine in the
other combustion model programs.
A finite difference numerical solution of a
simplified set of equations
describing the burning cigarette is nearing
completion. The model is of gas
and solid phases in one dimension 'with time as a
variable and a simplified
set of pyrolysis and combustion reactions
generating carbon monoxide and
carbon dioxide. This work has been undertaken by a
consultant in Oxford and
the computer program will be transferred to GR&DC
in the next few weeks.
This will provide us with the first working model
of the combustion processes
and will be used to determine which are the
important parameters in defining
the overall combustion characteristics of the
cigarette.
A more detailed model in two dimensions is being
solved by a group of consul-
tants in Southampton using a finite element
numerical scheme and should be
completed by the end of 1982. As well as providing
a more comprehensive
simulation of the cigarette, the model will enable
a direct comparison to be
made of the two numerical methods available for
solving sets of differential
equations - finite difference and finite element.
FUTURE WORK:
More details of nicotine release will be
investigated - the effects of heating
rate, gas velocity, tobacco particle size and
oxygen levels. A similar study
to determine the factors that are important in
nitric oxide formation will
also be undertaken.
Work will continue on the mechanism of the
magnesium oxide paper, as indicated
above. However, the mechanism is apparently
complex and the most obvious
aspects have already been investigated.
Consequently, the project is likely
to continue beyond 1982.
The sidestream velocity measurements, together
with equivalent temperature and
gas distributions, will be used as basic data in
testing the predictions of the
sidestream computer models which are under
development.
The combustion parameters determined for expanded
(and unexpanded) tobacco will
be incorporated into the puff-by-puff delivery
model. This will enable a wide
range of cigarette designs to be explored by
computer simulation.
In the last six months the basic combustion
modelling has largely been at a
development state, in which work on individual
programs of specific sets of C:)
the combustion processes has been- initiated. In
the next six months these '-0
programs will be completed and used predictiVely.
Subsequently, the programs co
will be joined together in order to obtain a more
complete model of the cc
burning cigarette, which should be a major input
in future cigarette design. _~:-
C=p
_1 J
'IQ
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unauthodscd persons.
BATCo document for Province of British Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 06 Sidestream
Author: H.G. Horsewell
Period Covered: January - June, 1982
OBJECTIVE:
To design cigarettes which deliver lower levels of
those components of
sidestream smoke which are thought to contribute
to nuisance effect or
alleged hazards.
PROGRESS:
There are four broad areas of current study:
(1) Development of analytical procedures for
measuring volatile and
tobacco-specific nitrosamines, ammonia, smoke pH,
vapour phase con-
stituents and other constituents which may
contribute to undesirable
sensory effects and irritation.
(2) Modification of cigarette paper by the use of
alternative fillers
and additives to reduce sidestream generally or to
reduce specific
smoke components.
(3) The design of prototype cigarettes using
upto-date knowledge and
materials as these become available.
(4) The development of techniques for monitoring
levels of smoke consti-
tuents in ambient atmospheres in rooms, etc.
1. Development of Analytical Methods for
Constituents of Sidestream
(a) Nitrosamines
This is part of a general programme of work on the
study of nitros-
amines aimed at discovering means of reducing
these hazardous
materials through gaining an understanding of how
these materials
are formed and released into the smoke.
During the last six months reliable and reasonably
rapid procedures
have been developed for measuring the levels of
tobacco-specific and
non tobacco-specific nitrosamines in mainstream
smoke. This method
also enables tobacco-specific nitrosamines; to be
measured in side-
stream smoke. A method for tobacco-specific
nitrosamines in tobacco
has also been completed. A method for the non
tobacco-specific
nitrosamines (volatile nitrosamines) in
sidestrearn smoke is still
required and work is in hand.
(b) Ammonia
C=)
A method for measuring ammonia in sidestream has
been investigated. 110
Although this is not suitable for absolute
results, the procedure C:0
does give comparative results which can be used
for the time being. cc
The main problem is that some of the ammonia is
trapped on the walls _C1__
CD
0 1982 British Arnerican Tobacco Co. Ltd. This
must not be. copicd or sho-n to unauthorised
persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


2
(c) pH
A preliminary study of the pil of sidestream has
been completed which
shows that differences between blends can be
detected.
(d) Vapour phase
Preliminary work has been carried out on
methodology for measuring
various vapour phase constituents. This involves
absorption of the
vapour phase in a packed tube. The compounds are
desorbed and
measured by gas chromatography. This work has
already provided
useful information on the mechanism of sidestream
reduction by the
Ecusta magnesium oxide pdper. The work indicated
that the visible
sidestreant is converted to invisible vapour
phase. A report on
these findings will be issued shortly.
2. Modification of Cigarette Paper
The main aim of this work is to discover ways of
reducing sidestream by
-changing the properties of the cigarette paper,
thus providing papers
which are exclusive to the BAT group.
Some time ago, it was discovered that the use.of
cigarette papers having
very low air permeabilities resulted in a marked
reduction in sidestream.
Patent protection has been sought.
A study of the effect of _bermeability on
sidestream levels has been
completed and reported. This work showed that
cigarettes wrapped in very
low permeability papers have a drastically reduced
sidestream. As the
permeability is increased the sidestream emission
increases sharply.
A preliminary study has been carried out which was
directed towards an
understanding of the advantages of incorporating
magnesium oxide in
cigarette paper for sidestream, reduction. As a
result of evaluating
cigarettes wrapped in papers containing different
grades of magnesium
oxide, it is now possible to identify the grades
of this filler which
will reduce sidestream. This is particularly
useful when discussing
magnesium oxide papers with Ecusta. Further, a
pilot scale batch of
paper with the most suitable grade of magnesium
oxide was made in GR&DC
which gave a similar reduction in sidestream to
the Ecusta paper.
3. Low Sidestream. Prototype Cigarettes
Progress Reports on development of low sidestream
cigarettes are issued
at two-monthly intervals and this section is
intended to be a brief
summary of the activities. There are four main
areas - cigarettes based
on papers containing magnesium oxide supplied by
Ecusta, papers supplied
by Schweitzerr low permeability/slow burning paper
made by Schweitzer at
the request of GR&DC and consumer tests on low
sidestream cigarettes in
the UK.
(a) Ecusta paper
r_igarettes have been made using the latest batch
of 509-M-2 magne-
sium oxide paper and a paper containing 25% chalk
and 11% magnesium CO
oxide (code-named "Velvet") which Ecusta expect to
be the basis of cc
_S~:-
a commercial specification. The sidestream
reduction is satisfactory.
0 1982 British Amtrictn Tobncco Co. Ltd. This must
not be copied or shown tct unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 3 -
The ash characteristics of the "Velvet" paper show
an improvemefit
over the other types of Ecusta paper. Howevere the
burn rate is
slower and the mainstream carbon monoxide delivery
is about 75%
higher than that from a conventional cigarette
with a similar
permeability paper. Ecusta have been asked to
produce a "Velvet"
paper giving a faster burn rate and a lower
mainstream carbon monoxide
delivery.
(b) Schweitzer papers
Two samples of high basis weight chalk filler
paper have been tested.
The higher weight paper (44g.M-2) reduced the
sidestream particulate
matter by 50%. The ash characteristics were
satisfactory. This
paper is a possible competitor for the Ecusta
"Velvet" paper.
(c) Low permeability/slow burning papei~
Cigarettes made with this paper give a high
reduction in sidestream.
However, because of the slow burn rate the
mainstream deliveries and
puff numbers are very high. This can be overcome
by reducing the
tobacco column length and the addition of a long
filter length to
produce a normal length of cigarette.. Products of
this type can be
viewed as "poor value for money". Some work is in
progress to
attempt to lengthen the tobacco column using
expanded tobacco.
(d) Consumer testing
Cigarettes using the earlier versions of the
Ecusta magnesium oxide
paper and short tobacco column/long filter
cigarette with the low
permeability paper have been tested in the UK. The
reduced level
of sidestream was favourably received. The short
tobacco column
cigarette was perceived as poor value for money
and had a tendency
to cease burning. The cigarette with the Ecusta
magnesium oxide
paper was fairly well received but some comments
were made concerning
fast burn rate, harshness and off-taste.
Cigarettes with the later
version of Ecusta paper, including the "Velvet"
paper, were not
included in these tests.
4. Sidestreant in Ambient Atmosphere
The objective of this part of the programme is to
study various aspects
of the presence of cigarette smoke in the
atmospheres in rooms.
The status at present is that it has been proposed
that a pair of similar
rooms be built at GR&DC. Visits have been made to
two Universities which
have special rooms for carrying out the assessment
of the quality of air
in rooms. one facility is at Yale University in
the USA and the other at
the Institut fUr Hygiene & Arbeitsphysiologie in
ZUrich, Switzerland.
This has provided very useful information on the
important features of
room design for this work. So far, discussions
have taken place with
contractors on the design of the rooms and
proposals will be put forward
by the end of'July.
Meanwhile, more attention has been given to the
instrumentation required CD
for monitoring the composition of the atmospheres
in these rooms. It is C)O
proposed to monitor levels of carbon monoxide,
carbon dioxide, nitric cc
oxide and smoke particle concentration with
standard instrumentation. It -.r--
is also intended to monitor other smoke
constituents, such as vapour phase CD
components, using the type of technique referred
to - under Item 1.
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unnuthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


4
FuTURE woRK:
It'is intended to develop procedures for the
measurement of volatile
tobacco non-specific nitrosamines as soon as
possible. The procedure for
measuring sidestream vapour' phase will be
upgraded by the purchase of
special equipment for this purpose# which will
also provide a means of
collecting and measuring vapour phase in
atmospheres in rooms. There is
a requirement for a means of measuring nitrogen
dioxide independently of
nitric oxide; the feasibility of accomplishing
this is being investigated.
Work on the modification of cigarette papers will
continue. The low
permeability papers have a lower burn rate, which
gives high mainstream
deliveries; work will be carried out to add
chemicals to these papers
to correct this deficiency. In addition, it is
intended to commence a
programme of work aimed at studying a range of
chemical additives to
paper for sidestream reduction.
Preliminary arrangements have been made to set up
a joint development
programme with De Mauduit/Schweitzer, aimed at
producing new cigarette
papers which will reduce sidestream. A meeting to
decide the work
programme will be held shortly.
Prototype cigarettes will continue to be designed
aimed at reducing
sidestream, taking advantage of any new cigarette
papers or tobacco blend
modifications which are developed. Further work to
extend the short
cigarette with the low permeability paper, using
expanded tobacco, will
be undertaken. Some attention will be given to an
attempt to reduce the
mainstream carbon monoxide to tar ratios for
cigarettes wrapped in the
Ecusta magnesium oxide paper.
Work on ambient smoke in rooms will not commence
until late 1982 or early
1983 as this is the earliest date for completion
of the rooms. Meanwhile,
information on monitoring equipment will be
obtained and some analytical
procedures will be developed. A programme of work
is being formulated in
conjunction with the Psychology section, which
will be aimed towards the
reaction of subjects to sidestream from low
sidestream cigarettes, the
effects of changing ambient conditions
(temperature and humidity) on the
quality of sidestream, the effects of time after
smoking on the odour of
the room and possibly the effect of deposited
smoke on furnishings.
cc
cc
0 1982 Briti-sh American Tobacco Co. Ltd. This
must not be copied or shown to unnuthcrised
persons.
BATCo document for Province of British Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 07 Human Smoking Studies
Author: G. Read
Period Covered: January - June, 1982
OBJEC-,IVE:
To provide a scientific understanding of the
processes involved in the
interaction of smoke with the smoker that occur
during the complete
Smoking event and to identify the consequences of
cigarette design for
the smoker.
PROGRESS:
There are two major areas of current study:
1. The continuation of external and internal
smoking behaviour
studies, designed to identify and characterise
human smoking
behaviour patterns in relation to mouth-level
intake and to
determine the effect of product modification in
terms of both
design and delivery on smoking behaviour;
2. (a) The characterisation of smoke aerosols in
terms of both
physical and, to some extent, chemical properties
that
influence the potential of these particles to
deposit within
the respiratory tract and
(b) The development of techniques to quantify
retention of
particulate material.
Also, we have developed a respiratory physiology
facility to provide
fundamental information, initially on respiratory
patterns, necessary
to relate mouth-level intake to potential
respiratory tract deposition.
1. Smoking behaviour in relation to product
development
over the six month period, two external smoking
behaviour studies have
been in progress. The smoking behaviour component
of the Cambridge
study (MS07) has been completed and is currently
undergoing analysis.
This study will provide information on the effect
of long-term switching
of middle- and low-tar smokers to alternative
lower delivery products.
In addition to this information, estimates of
daily nicotine intake by
these smokers are being determined. The first
phase of the London
University study has been completed and the second
phase is on target to
begin in the autumn.
The decoding/duplication of the smoking behaviour
studies continues to
be the rate-limiting step in this area. In an
effort to improve
progress in the future, the possibility of
operating the existing a-id
expected duplicator facilities in parallel is
being considered.
The data from a wide range of smoking behaviour
studies is being collated co
to form the foundation of a smoking behaviour
database. The database cc
is being used to identify the expected ranges for
smoking behaviour _r~t.
0 1982 British Amcrican Tobacco Co. Ltd. This must
not be copied or shO-11 to un;tuthorised persons.
BATCo document for Province of British Columbia 5
November 1999


- 2 -
Parameters and to develop means of classifying
smokers according to
smoking behaviour. An initial report of these
findings is in the course
Of Preparation.
A preliminary smoking behaviour study has been
initiated in which the
interaction of cigarette design features and
smoking behaviour is being
investigated.
2(a) Aerosol research; particulate deposition
within the respiratory
tract
Smoke aerosol research is being conducted at both
GR&DC and Bristol
University. The work at Bristol has involved the
investigation of the
properties of aerosols that influence their
stability, coalescence and
interaction with aqueous surfaces. One aspect of
the work investigating
the mechanism of aerosol emulsion formation has
been completed and the
Study will be issued as a technical report..
Research into the genera-
tion of aerosols of mixed aqueous and organic
phase composition and
structure is continuing. Initial. experiments have
indicated that it
is feasible to generate, under controlled
conditions, aerosols of this
type.
Earlier research at GR&DC indicated the need for
additional equipment
to study the particle size distribution of smoke
aerosols over a range
of physical conditions. Equipment for this purpose
has been purchased
and is currently undergoing validation. This
facility will allow
smoke aerosols in the particle size range 0. 1 -
6.0 li'm and 0.5 - 20 pm
to be studied. The former size range will be
suitable for the deter-
mination of the particle size
distribution/concentration of ambient
smoke, whilst the latter will be used to study the
growth potential of
smoke aerosols in both model lung and human
smoking studies.
A study indicating the growth potential of smoke
aerosols generated
from a range of Swiss brands, largely confirming
earlier observations
vith respect to tobacco type, has been reported. A
human retention
study designed to identify the relationship
between aerosol growth
potential and particulate retention has been
completed and the results
are currently undergoing detailed analysis. The
initial findings of
this study tend to confirm the expected trend of
reduced respiratory
retention with aerosols with low growth potential.
in conjunction with work in this area, preliminary
studies have been
instigated to assess the quantitative significance
of smoke lost from
the mouth prior to,inhalation - mouth-spill. If
this is a widespread
occurrence it may be an important contribution to
the overall assessment
of the maximum smoke dose available for
inhalation.
2(b) Identification of the relationship between
mouth-level intake and
inhalation pattern; the potential for particulate
retention
one possible consequence of modifying cigarette
deliveries and hence
mouth-level intake is a change in a smoker's
inhalation pattern. The
consequence of this would be to affect the
distribution and mean
residence time of smoke within the respiratory
tract - important
considerations when assessing the potential for
particle growth and %10
subsequent deposition. Preliminary studies with
gas phase markers c0
(Co and NO) have indicated that these are not
indicators of depth of cc
inhalation but rather reflect the difference
between inhaling and
non-inhaling. C:)
__1!
c0
Q 1982 British Arterica., Tobacco Co. Ltd. This
must not be copied or shown to mmuthoriscd
pcrsons.
BATCo document for Province of British Columbia 5
November 1999


- 3 -
The initial task of the newly-appointed
respiratory physiologist is to
implement techniques to characterise the
inhalation patterns of subjects
in relation to their smoking behaviour, in an
essentially non-invasive
Manner.
FUTURE wr--uc:
To maximise progress and to develop thinking in
the area of human
smoking smudies, the continued integration of the
expertise within this
group willy be continued and, where appropriate,
overlap with other work
areas engaged on related topics.
It is intended that during the remainder of the
year, the smoking
behaviour of smokers (including respiratory
monitoring) engaged in sub-
jective product assessment, will be instigated.
These studies will be
used to indicate the relationship between
objective measurements and
subjective assessments.
The need for long-term, fundamental studies in the
area of smoke aerosol
research has been recognised. The future work
within this section will
include the character i sation of particle size
distributions of smoke
from both mainstream and sidestream. These
measurements will be used
to relate real-time particle growth to the
potential for particulate
retention within the respiratory tract.
In recognition of a BAT group requirement for
specialist knowledge in
the area of nicotine pharmacology/sensory receptor
interaction, a
research scientist in this field has been
appointed. A work programme
will be ',itiated in the latter half of 182 to
begin to investigate the
role and mechanism of nicotine in mediating a
pharmacological response
as part of the smoking motivation process.
Expertise in this area will
be directed toward developing lower delivery
products with enhanced
satisfact~ion/acceptability.
CO
CO
2 ~:-
CD
0 1992 Duish American Tobacco Co. Ltd. This must
nut be copied or shown to unauthorised pcrsom
BATCo document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 08 Psychology and Sensory Testing
Author: R.P. Ferris
Period Covered: January - June 1982
OBJECTIVE:
To advance the f-andamental understanding of the
needs and behaviour of
smokers, and to develop and apply methods for the
sensory assessment of
tobacco smoke.
PROGRESS:
The current programme is divided into four major
project areas;
1. Consumer Studies, which includes attitudinal
and motivational
research with particular reference to smoking and
health and
social disapprobation issues, the development of
market segmenta-
tion approaches and the validation of sensory
testing techniques
using large samples of low delivery cigarette
smokers.
2. Subj ec tive/P sycho logical Methods, including
product communication
research, the role of smoking in social situations
and arousal
modulation, and the development of brand image
measurement tech-
niques.
3. Preference Research, including the formation of
preference judge-
ments, particularly in relation to unfamiliar
cigarettes/novel
flavours.
4. Sensory Evaluation Methods, which includes the
development of tech-
niques for low delivery evaluation# flavour
characterisationt side-
stream visibility/irritancy/aroma evaluation and
subjective varia-
bility est4--tion.
1. Consumer Studies
Project NILE, a large scale consumer validation of
DELTA has been com-
pleted in the U.K. and is currently at reporting
stage. The results
demonstrate the utility of scaling up in-house
panel methods to product
test applicaticas, showing acceptable reliability
in the field and
cost effectiveness. Furthermore, this 'appiFoach
offers subsequent
access to reliable small subsets of the overall
test popUlation which
may be defined according to a range of
sociodemographic, psychographic
and smoking his-tory criteria.
A second report in the ARIES series (c-Lncerning
social disapprobation
of smoking) has been completed and concerns
perception of smoking as a
social nuisance as a function of particular
situations.. Coniensus
between smoker/aon-smoker groups is apparent. in
terms of those' situa-
tions generating most and least annoyance.
However, never smokers
appear to apply more extreme and less
differentiating judgements.
0 1982 British A-nerican Tobacco.Co. Ltd. This
must not be copied or sho,.n to unauthorised
persons.
CC)
00
cD
C)o
C:>
BATCo document for Province of BritiSh Columbia 5
November 1999


An immediate future commitment to rationalising
the volume of informa-
tion gathered under Projects LIBRA (smoking and
health) and ARIES
(Social disapprobation) is under way and has two
aims for completion
within 1982:
M to develop a short form monitoring questionnaire
for tracking
changes in smoking and health/social
disapprobation attitudes
in the marketplace;
(ii) to develop a model for market segmentation
based on motiva-
tional/attitudinal stances which could be taken to
be predic-
tively related to quitting/switching behaviour.
2. Subjective/Psychological Methods .
A contract has been initiated with the University
of Southampton to
develop an explanatory model of 'the relationship
between gestural use
of the cigarette in social coping situations and
arousal level mainte-
nance. Prime elements of the model are observed
relationships between
certain smoking gestures (e.g. puff taking) and
averting of eye gaze,
which in turn are related through the latter to
central nervous system
activity level. The basic hypothesis of this
(initially one-year)
contract is that smoking allows a more refined
control of social coping
and central arousal.
A battery of product communication researches have
been developed.
Priorities are largely derived from the 1979
Jesteberg Guidelines on
the future of marketing under conditions of
limited access to communi-
cation media.
Research approaches have been formulated for pack
design, pack within
point-of-sale, and poster prominence or 'stand
out', all based on
measures of (a) stimulus duration time required
for recognition of
salient features, and (b) recordings of eye gaze
tracking behaviours
across complex visual fields. The requirement to
overcome the "famili-
arity element" of current pack studies is being
met by a "crosss-over"
study using UK and Finnish pack samples in their
respective countries.
outline proposals have also been developed for
researching the effec-
tiveness of differing combinations of media in
communicating novel/
innovative product features and for extending the
'recently completed
NILE method to brand image measurement; both
researches will be commis-
sioned within 1982.
project FUTURE, a qualitative study of the Finnish
(total ban) market
concerned with social/cultural bases of product
information flow has
been completed and is approaching report stage. It
has proved difficult
to isolate truly post ban phenomena with this type
of research since
most pre/post ban effects are apparently of degree
rather than type.
principle observations include evidence of an
increasing demand for
technically innovative produ,:ts, a developing
reaction against pack
homogeneity and increased importance of access at
the point-of-sale CD
for visual/tactile examination of packs.
cc
cc
41.
C=)
CC)
0 1982 Britisla American Tobacco Co. Ltd. This
must not be copied or shoum to unauthorised
persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


3
3. Preference Research
A three-year contract with the University of
Birmingham reaches comple-
tiOn in the second half of 1982. The aim of the
work is to apply
learning theory to an understanding of the
development of preference
for novel flavoured cigarettes in those smokers
seeking low tar cigar-
ettes with taste and "satisfaction". The
conclusions are likely to
support parallel in-house findings that the
development of preference
is most significantly mediated by accompanying
brand message material
in the short term and nicotine deliveries in the
longer term.
4. Sensory Evaluation Methods
(a) Project BROMLEY
This development of an alternative standards-based
approach to low
delivery evaluation has progressed to analysis of
a final U.K. quanti-
tative validation designed to establish the panel
vocabulary and a
model which may be used to extend this to product
test applications.
The factor structure of consumer vocabularies
appears similar to DELTA,
with the notable addition of a 'visual-tactile'
factor. The results
indicate feasibility of using consumer samples to
model acceptability
by using previously suggested measures of
'baseline expectancy' and
'tolerance bandwidths'. Priority training of a
BROMLEY panel is inclu-
ded in the current research year.
(b), (c) Flavour Panel Development
Two reported screening exercises have so far
refined a large population
in terms of flavour sensitivities (specific
anosmias) and flavour
recognition. Multidimensional scaling, cluster
analysis and factor
analysis treatments of subsequent subjective
evaluations of a wide range
of flavours has yielded the structure of the
flavour 'perceptual space'
used by these subjects. This structure is very
similar to professional
flavourist systems, including floral, musk,
balsamic/vanillic, fruity,
minty, smoky and spicy categories. Method
development will proceed to
leaf type familiarisation (lit and unlit) and
specific flavour/leaf com-
binations.
(e) Temperature/humidity variability
This research, at data analysis stage, considers
the effect of varia-
tions in temperature and humidity conditions on
subjective evaluations
of a flue cured and a U.S. blended type of
cigarette. These subjective
results are to be related statistically to
parallel findings of objec-
tively measurable changes in chemical composition
of these cigarettes
found to take place under the same environmental
conditions.
Highly significant changes have been recorded at
both objective and
subjective level. However, since not all trends
are unidirectional a
statistical model is being developed to consider
this interrelationship.
During the year, work has been carried out on
pressure drop/tip ventila-
tion variability. This research, at report stage,
considered the co
effect of variations in ventilation and pressure
drop on subjective cc
evaluations of a single brand of cigarette deawn
from a normal produc- _4~_
C:)
tion run. Results show the effects of pressure
drop and ventilation cc
Q 1982 British American Tobacco Co. Ltd. This must
not be copied or shown tc, unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 4 -
to be additive and linear. Increases in either
variable are associated
with significantly increased perception of amount
of draw effort;
decreased impact, mouthfull, mouth/ throat/nose
feel, taste; and no
association with aroma or acceptability.
FUTURE WORK:
Activity in the subjective evaluation area should
be significantly
extended by developments of methods to evaluate
sidestreant visibility,
irritancy and aroma. Currently, a cubicle based
approach has been
pilotted and proposals have been generated for a
room based approach
to ambient evaluations.
Motivational/attitudinal surveys of smoking and
health/social disappro-
bation will now revert from a basic information
gathering need to
a more time/cost effective emphasis on 'monitoring
and segmentation
issues.
Finally, planning of joint studies with Work Area
07 is now advanced
in a number of areas.
cc
cc
4-1.
CD
cc
0 1982 British Amcricitn T-b-cco C.. Ltd. This
must not be copicd or shown to unauthoriscd
p-sons.
BATCo document for Province of British Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 09 Smoke Taste and Flavour Improvement
Author: R.A. Crellin
Period Covered: January - June, 1982
OBJECTIVE:
To identify and implement routes to the
augmentation of smoke sensory
characteristics (especially taste and flavour) of
low delivery products.
PROGRESS:
The work area can be divided into the five
following aspects for the
purposes of discussion:
1. The examination of green leaf flavour
precursors and factors affec-
ting their formation and accumulation;
2. The transformation of flavour precursor
materials into compounds and
mixtures with flavour potential;
3. The behaviour of flavour materials added to
cigarettes, particularly
their transfer into the smoke stream;
4. The development of an appropriate flavour
methodology, to facilitate
identification and monitoring of important flavour
constituents and
their precursors;
5. The formulation of new compounded flavours from
new and traditional
ingredients to meet specific low tar product
requirements.
The work area was described in detail as part - of
a Flavour Conference
held by BAT in Hamburg in Februaryl 1982.
1. Flavour precursors studies
The Importance of genetic (choice of tobacco
variety) and environmental
(agronomic conditions) factors in the green leaf
accumulation of diterpe-
noid flavour precursors has been extensively
studied. Recent results
relate to Virginia, Burley and Maryland varieties,
high nicotine varie-
ties and the influences of harvesting time.
Generally, it appears that
varieties capable of synthesising both classes of
diterpenoids (cembra-
noids and labdanoids), e.g. Oriental and Brazilian
varieties, produce
the highest levels of leaf surface waxes and
diterpenoids. Additionally,
harvesting at normal maturity (rather than
under-ripe) resulted in the
maximum diterpenoid accumulation.
A BAT-sponsored Post-doctoral Fellow at Nottingham
University has con-
tinued work on the synthesis of ( )-cembrene,
believed to be the precursor c::)
of the cembran e diol (CBD) which is the major
cetabranoid diterpene of
green tobacco. The synthesis is close to
completion. Experiments have cc
also begun at Nottingham designed.to convert CBD
(supplied by BAT) into 00
a radio-labelled cembrene (probably tritiated);
success will enable
detailed itudy of flavour precursor synthesis by
tobacco plants and co
plant cel s.
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unnuthoriscd persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


Growths of tobacco in the GR&DC greenhouse in 1981
and 1982 are being
used to evaluate the influences of variety, plant
maturity and leaf
Position on the production and accumulation in
leaf of carotenoid
terpenes, also considered significant green leaf
precursors of tobacco
flavour. 'The concentration and distribution
during leaf growth of
individual carotenoids is being determined.
2. Transformation of flavour precursors
The collaborative project with Svenska Tobaks, has
progressed to the point
at which leaf surface gums, extracted in Brazil
and USA from a number of
Virginia, Burley, Maryland, Brazilian and Oriental
varieties, have been
treated in Stockholm by electronically-excited
states Of oxygen (to mimic
chemically the changes occurring during curing and
ageing) and assessed
sensorily on cigarettes in Southampton and MalmF)
(Svenska Tobaks Product
Development Department). Subjective reaction has
been very positive
(particularly from Maimb). This result has
prompted a preliminary
appraisal of process economics; .. this in turn
has emphasised the need to
switch attention from process investigation to
process optimisation. The
change entails maximisation of gunt/diterpene
yield, maximisation of
treated gum potency and cost reduction. The Swedes
are evaluating
variations in treatment conditions (reaction time
with singlet excited
oxygen and choice of acid rearrangement
conclitions) to optimise sensory
activity.
Collaborative work has also been progressed with
Biotechnology (see Work
Area 10) in microbial routes to flavour. Taking
the crisomer of CBD as
the precursor, microbial conversion resulted in
the isolation and identi-
fication of two products. - One of these was a
known constituent of tobacco,
the other was an entirely novel compound. Attempts
to produce compounds
from O-CBD by this route have not succeeded so
far; products are complex
mixtures. The situation using ambroxide as
precursor is very similar and
the saarch for better process conditions is under
way. lonones have also
been examined; cr-ionone was shown to yield
optically active crionol (100%
pure) by the microbial route. Further products
from crionone are being
investigated. 0-ionone has already been examined
in this way by Japan
Tobacco and Salt Corporation, but studies in
Southampton have produced
different compounds microbially and these have
been shown to have interes-
ting and useful flavour properties, particularly
on Virginia-style
products.
Amino acids and sugars are precursors to nitrogen
heterocycle flavour
compounds, including pyrazines. Four grades of
poor-quality flue-cured
tobacco from Canada have been subjected to heat
treatments in the presence
of amino acid/sugar casings. Smoking trials
revealed considerable flavour
improvement and reaction to submitted samples is
awaited from Montreal.
A Product Development Review brochure entitled
"Taste improvement of low
grade tobaccos by casing and heat treatment" has
been widely circulated.
3. Migration and transfer of flavour
A successful tobacco flavour compound must not
only have useful flavour
properties per it must also show minimal migration
away from the
c::>
product into the pack during storage and efficient
transfer (without 110
degradation) into the smoke. Studies carried out
with radiolabelled co
tetramethyl pyrazine, cournarin, cinnamyl alcohol,
13--phenyl ethanol and 0C
amyl salicylate showed a range of migration and
transfer characteristics.
(Z)
c0 -
0 1982 British American Tobacco Co. Ltd. This must
not be copied or shown to unauthoriscd persons
BATCo document for Province of British Columbia 5
November 1999


- 3 -
However, in all cases, addition of the flavour
compound to tobacco resulted*
in higher smoke transfer than filter addition and
the inclusion of
cigarette ventilation only exaggerated the
advantage of tobacco addition.
Transfer to smoke reduced after product ageing;
this is partially due to
migration. of flavour to the filter. Filter
migration of polar flavour
cOrnpounds might be minimised by using a non-polar
filterf e.g. polypropy-
lene. Attempts have been made to compare
polypropylene with cellulose
acetate using acetyl pyrazine, ambroxide,
ambrettolide and sclareolide
in unlabelled form, but results suggest that
radiolabelling will be
necessary to resolve flavour transferred from
smoke background.
Migration away from the product (or into areas of
the product, e.g. the
filter, from which smoke transfer is poor) might
also be minimised by
some form of aLicroencapsulation. Discussion with
Wiggins-Teape resulted
in proposals to trap menthol in cigarette seam
adhesive by emulsification.
Preliminary observations imply that migration is
still occurring but,
nevertheless, efficient transfer to the mainstream
results.
4. Flavour methodology
A significant increase in the mass spectrometry
library of spectra coupled
with experience gained in linked capillary gc/rns
systems has resulted in
a marked increase in success rate in identifying
constituents of tobacco
extracts. Tbis system is now used routinely- in a
wide range of flavour
studies; currently headspace samples from a range
of tobacco types and a
series of UK brands are being evaluated in this
way.
A scan of the neutral volatile constituents of
tobacco and smoke has been
developed and reported; . this is attracting the
attention of flavour
scientists t~Lroughout the Company. The scan
detected a series of ketones
and lactones derived from cembranoid and
carotenoid terpenes and also
detected the diterpenoid precursors. The technique
is of interest,
particularly in process research. Another,
relatedp technique which
enables continuous extraction of tobacco and
condensate steam-volatiles
is under investigation. The technique results in
organoleptically intense
and characteristic extracts and the sensory
effects of these extracts on
add-back will be assessed in addition to
analytical method development.
In conjunction with Computing Services, a software
system has been
developed wbLich enables spectral comparisons to
be made between known
carotenoids and the unknown constituents eluting
from the HPLC column.
Attempts are being made to develop a quantitative
capillary g.c. scan of
minor cembranoids in tobacco, for analysis in
greenhous e- grown tobacco of
flavour precursor accumulation patterns during
plant growth and in
different plant positions.
In conjunction with Psychology Group (Work Area
08), a further stage in
the selection of a "high calibre" sensory panel
has been completed;
selected candidates will now be educated in leaf
and smoke flavour
terminology.
Collaboration with Biological Group (Work Area 01)
has prompted the
assessment bv Ames test of a number of new single
and formulated flavours.
Equallyl Sve-nska Tobaks have confirmed through
the Ames test that their
flavour fractions from green leaf gums; are Ames
negative, i.e. should
not influence condensate mutagenicity. cc
CC
_rt::b
0 1982 Brkish American Tobacco Co. Ltd. This must
not be copied or shown to unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


4
5- Flavour formulation
A Flavour Brochure has been finalised, assembled
and circulated to 43
Operating Companies. This describes new single
flavour compounds and
formulated flavours resulting from the research
described in the foregoing
sections. Initial comments from Brochure
recipients have been very
Positive and a number of requests for samples have
been made.
Following the creation of a number of
Kretek-alternative flavours and
subsequent assessment of these by BAT Indonesia, a
flavour has been
accepted by Indonesia for use on a new brand
launch.
Sidestream modifiers from commercial Flavour
Houses have been evaluated
by the sidestream panel; two of the submissions
produced a noticeable
improvement in sidestream smoke character. Recent
work has involved the
testing of several open-space odourisers' from a
Fragrance Housel these
show a positive aromatising effect on aged
sidestream smoke.
FUTURE WORK:
The objective of the work area has now been
broadened to encompass all
smoke sensory characteristics considered deficient
in low delivery
products. Consequently, much future work will be
centred around the body
and mouthfull impression of smoke. Attention will
be directed towards a
chemical understanding of this attribute and the
discovery of routes by
which it can be augmented. Already, some fatty
acid esters have shm-n.
promise as mouthfull enhancers. Additionallyr the
degree to which varioas
sensory attributes (impact, flavour, irritation,
mouthfull) interact with
each other (augmentation, synergism, potentiation,
masking) will be
investigated.
Further leaf experimentation in the USA (conducted
under the aegis of t-he
Leaf Steering Committee) is aimed at evaluating
the diterpenoid potential
of some new varieties (the so-called T.I.'s -
Tobacco Introductions) and
the carotenoids potential of high and low
carotenoid lines developed at
North Carolina State University. Leaf experiments
are also planned :n
Brazil; these will evaluate high density growing
of Oriental and Brazilian
tobaccos as a high yield source of green leaf gum
for the Svenska Tobaks
collaboration. A study of growing and extraction
costs will be recorded
as part of this work.
Microbiological studies will include the use of
tobacco plant cells as
transforming agents (this also eliminates the
requirement for careful
isolation of products prior to sensory work) and
the use of carotenoids
as substrates will be considered.
Migration and transfer studies will in future
include the technique of
gas proportional counting. This will enable smoke
transfer of flavour to
be analysed into direct distillation (the desired
effect), degradation
during transfer and reaction with smoke
constituents. The effects of
emulsification and microencapsulation on
distillation transfer will be
assessed.
The Flavour Brochure will be kept up to date by
the circulation of CD
additional pages when useful new compounds are
discovered and new formula- 110
tions are developed. 00
c0
CC)
Q 1982 British Americ.n Tobcco Co. Ltd. This must
not be copied or shown to umuthorised pcrsons.
BATCo document for Province of British Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 10 Leaf and Biotechnology
Author: T.G. Mitchell
Period Covered: January - June, 1982
OBJECTIVE:
To obtain improvements in products or processes by
the alteration of
leaf properties prior to primary manufacture or by
the application of
biological methods.
PROGRESS:
Current projects are in the following broad areas:
1. The growing and treatment of tobacco to meet
specific product
requirements.
2. The production of smoke flavourants by
microbial and plant cell
transformation.
3. The study of new biotechnology techniques.
4. the development of new approaches to
processing.
1. The production of tobacco from growing to
post-curing
(a) Growing: The main activity in the review
period has been to plan
and agree field experiments to be carried out in
the USA and Brazil
this year. The objectives of the trials which have
been initiated
comprise optimising the levels of flavour
precursors in the tobacco
before harvest, the production of plant gum as a
source of flavour
enhancers and the identification of the agronomic
factors involved
in nitrosamine formation. A report on earlier
trials on the latter
subject has also been issued. In additiorv, a new
area of investi-
gation is to compare individual varieties and
types of tobacco in
respect of physical properties and, ultimately,
their value in
cigarette manufacture.
(b) Curing: In some countries a proportion of the
Virginia tobacco
crop is air-cured. Apart from possible energy
saving aspects, it
was considered that there could also be
differences in leaf
chemistry relevant to smoke flavour. A pilot
experiment to examine
this was mounted in 1981 using tobacco grown in
Honduras. Some of
the differences found in the tobaccos were
expected, the sugar
content of the air-cured being half that of the
flue-cured sample.
Differences possibly affecting smoke flavour have
also been recor-
ded, including some increases in flavour compounds
and lower resi- C=)
dual carotenoids in the air-cured tobacco sample.
The final stage '_~o
of this experiment comprises the panel testing of
low delivery co
cigarettes containing a proportion o~ the test
tobaccos in the cc
blend.
00
0 1982 British AmeriC2n Tobacco Co. Ltd. This must
not be copied or shown tv unauthoriscd pcrsons.
13 ATCo document for Province of British Columbia
5 November 1999


- 2 -
(C) Aging and Fermentation: Numerous individual
studies have been
made in the past of treatments to accelerate aging
or to ferment
tobaccos. The purpose of such studies has varied
so that un-
certainty remains over the effects obtained an
various aspects of
leaf and smoke chemistry as well as acceptability.
A limited
programme of investigation was devised to study
the effects of
selected forced aging treatments on flue-cured and
Burley tobaccos.
Treatment of flue-cured tobacco at 50*Co 20%
moLsture for 14 days
caused minimal changes in smoke deliveries,
alterations in the
neutral volatile scan and enhanced tobacco aroma.
A more extreme
treatment at 70*C and 30% moisture caused
unacceptable physical
changes and a marked increase in cyanide delivery.
2. - Microbial and plant cell transformations of
flavour precursors
Many changes in the chemistry of tobacco leaf
during maturation and
curing are initiated by enzyme action. The
application of microbial or
plant cells containing enzymes is one approach to
converting isolated
flavour precursors to compounds with activity as
smoke flavourants,
analogous to the natural process.
Following the earlier demonstration of the
apparent utility of this
method, work has concentrated on obtaining
sufficient quantities of
selected transformation products to enable their
isolation and assess-
ment on cigarettes. Emphasis has been placed on
ionones, in view of
claims made by Japanese Tobacco & Salt Monopoly,
cembranes as tobacco
specific compounds and ambroxide as a smoke
flavour compound of parti-
cular interest. The status of products
derived.from these is discussed
in the review of Flavour Research.
Two specific difficulties have emerged from
studying these compounds.
In moving from initial screening using shake
flasks to fermenters of 3
or 10 litres in order to increase the quantities
produced, toxicity
problems have occurred when adding the substrate
in a single unit to
the rapidly growing mould. Incremental addition of
substrate and
monitoring of C02 output have been used to
overcome this. Yields of
products have been disappointingly low with these
substrates when
compared with earlier studies using sclareol.
Plant cells could offer advantages over microbial
cultures as a
transforming system, including their identity with
tobacco and feasi-
bility of using the total reaction mix as an
additive. Three cell
lines from tobacco were established in callus and
suspension culture
via a contract with Sheffield University and a
first screening on
selected flavour precursors was completed.
Transformation of a cembra-
trienediol was obtained with two of the three cell
lines.
In theory, isolated enzyme systems would allow
greater control of
transformation reactions. In view of the great
variety of compounds
claimed to be affected by the enzyme system known
as the cytochrome
P448/P450 systemt its application to flavour
precursors is being
studied at Surrey University using sclareol as a
model substrate.
3. Assessment of new biotechnology techniques
A programme of work to start in 1982 has been
proposed to investigate
plant cell culture as a source of cigarette
filler. An outline costing
of such a process, based on productivity figures
published from Japan,
was extremely unfavourable. This view was
confirmed in the course of a
visit to the Japanese Monopoly (LCFB/ALH). As, in
addition, the subjec-
tive character of smoke from such a product is
also said to be extremely
poor, the decision was taken not to proceed with
this project.
0 1982 British American Tobacco Co. Ltd. This must
not be copied or sho"In to unauthoriscd persons.
CZ)
cc
cc
C)
C-11
__O
BATCo document for Province of BritiSh Columbia 5
November 1999


- 3 -
Somatic hybridisation is a technique with
applications in plant
breeding. A research group in Ottawa, Canada
studying somatic hybrids
of Nicotiana offered to make available leaves from
plants produced in
this way. Extracts of a series of plants
representing hybrids of N.
tabacum and N. sylvestris were prepared in the
Montreal laboratories
of Imperial Tobacco and have since been examined
at Southampton. The
conclusion reached is that the somatic hybrids do
show quantitative
differences in diterpene content. As the N.
tabacum cell line used in
this work was not from a commercial cigarette
variety, the hybrids
obtained are judged to be of no immediate
practical interest.
4. Alternative processing methods
A relatively simple procedure for fractionating
tobacco constituents
has been proposed. It involves pressing harvested
tobacco leaves which
have been yellowed, but not dried, to obtain an
extract of cell sap and
a leaf residue which can be used directly. The
name Cylproc has been
used to refer to it. A quantity of Virginia
tobacco was processed in
Canada in 1981 using this procedure.
From the data obtained so far, it is clear that
all targets were not
achieved on the first samples - the yellowing
phase was not completely
effective so that residual starch concentrations
are high and liquid
extracted was only half of that anticipated.
Cigarettes are being
manufactured to complete the assessment of this
first trial.
FUTURE WORK:
Field studies involving growing of tobacco will be
the subject of
analysis during the latter half of the year (green
leaf extracts from
the USA) and into 1983. Further work on
comparisons of air- and flue-
curing has been initiated in the USA this year and
will be studied both
in relation to smoke flavour chemistry and
nitrosamine formation.
Studies of forced aging and fermentation will
concentrate on aspects of
smoke flavour and the prospects for selective
control of changes via
the treatment conditions.
Work on flavour precursor transformations is to
continue with the aim
of increasing the yield of products obtained.
Further work on plant
cells will aim at producing sufficient quantities
of transformation
products for evaluation.
It is proposed to produce further quantities of
tobacco by the Cylproc
technique this year, modifying the process to take
account of current
defects.
on new biotechnology techniques, a programme is
being developed on
selective gene transfer into tobacco via treated
pollen at Birmingham
University and a programme of work on enzyme
applications to cured
tobacco is being considered.
Data has been published from Japan indicating that
protein in tobacco
is the source of a high proportion of the
mutagenic activity .'n cigarette
smoke. A programme of work is proposed to validate
the claim and to
devise methods for reducing the protein content of
cured tobacco.
co
cc
-C72-
CD
0 1982 British American Toliacco, Co. Ltd. This
must not bi copicd or shown to unauthorised
Persons
BATCo document for Province of British Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 11 Tobacco Processing
Author: R.G. Hook
Period Covered: January - June, 1982
OBJECTIVES:
I. To establish the fundamental principles of how
tobacco process
conditions can:
(a) change the chemistry and physics of tobacco
(b) affect the properties of the final product
2. To enable Operating Companies to control the
properties of the
blend to permit final product specifications to be
achieved.
I .
3. To develop new process technol ogy, drawing on
inputs from (1)
and (2) and from other sources.
PROGRESS:
1. Fundamental Principles
(a) A study has been made of the relative
importance of the different
mechanical and geometrical properties of tobacco
particles in influencing
the density of cigarettes at a given firmness. The
main inputs have been:
- a review of literature, particularly a series of
papers by Japanese
workers which have recently been obtained in
English translation.
- a series of simple non-standard compression
tests on tobacco in
bulk.
- simple mathematical and physical modelling.
This will be reported in the third quarter of
1982.
..The conclusions at this stage are only partly
quantified. It is clear
that although tobacco particle density is the most
important single factor,
it is far from dominant. There.is a group of
additional factors which
combine to determine the size of the air spaces
between the tobacco
.particles (i.e. the void fraction). When an
assembly of tobacco particles
is compressed over the range of bulk densities
which occur in cigarettes
the tobacco particles themselves are not
compressed to any significant
extent. The spaces between the particles are
compressed by distortion
of the shapes of the particles and slippage
between them. The elastic
component of the overall bulk compression can be
attributed to the
elastic distortion of the particles, whereas the
non-elastic component
is caused partly by non-elastic distortion of the
particles and partly by
irreversible slippage between particles. The
quantification of the
relative importance of these effects will require
the application of measure-
ment methods which are mostly novel to the tobacco
industry, and further
more detailed modelling work. CIC
0 ~
0 1982 B.'Ai3h American Tobacco Co. Ltd. This must
not be copied or sho-n tc, unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


-2-
The next stages are:
- detailed measurements of the geometrical
structure of cigarettes
and tobacco particles (image analysis).
- measurements of key mechanical properties of
tobacco particles
(elastic moduli and coefficients of friction).
- more formal mathematical modelling. A plan for
this work has
been drawn up. It will be carried out mainly under
contract
by-Brunel University.
(b) A simple steady state model of the operation
of a rotary dryer has
been developed and tested experimentally using
on line measurements of process settings, input
and output tobacco
moisture, tobacco surface temperature and process
air temperature
and humidity. A report has been issued on the
latter technique
(RD 1847).
equilibrium water vapour pressure measurements
over tobacco at
different temperatures and tobacco moisture
contents.
laboratory drying experiments.
The agreement between experimental results and
predictions of the model
is moderately good bearing in mind the simplicity
of the model. Some
further physical effects will need to be
incorporated into the model to
give a better overall fit to the main trends of
the experimental data.
The objective at this stage is to use the minimum
of physical
assumptions to match the overall pattern of the
data rather than to
obtain precise numerical agreement. In this way we
will confirm that
we have a: correct understanding of what are the
really important
physical effects. This limited number of key
effects is likely to
apply to all types of wetting and drying
equipment.
A report is being prepared on the development and
testing of the mathe-
matical model.
A modified version of the air sampling probe is
being made to permit
measurements to be made on the process atmosphere
inside the small scale
Dickenson (ITH type) dryer.
( C) The development of the method of estimating
neutral volatile flavour
compounds in tobacco has been completed and it has
been applied to
representative tobacco types and grades (Report
T.78). The preliminary
application of the method to smoke was reported at
the Flavour Conference
in Hamburg.
The method was also used to obtain a preliminary
view of the effects of
warm, moist (fermentation) processing and hot, dry
(oven) processing on
the compounds being measured. The form, r
increased and the latter
decreased the level of breakdown products, in the
widest sense.
C:0
cc
4 ~-
C=>
0 1982 Rritish American Tobacco Co. Ltd. This
-nust not be copied or shown to unRuthorised
persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


-3-
The next stage is to establish the method as a
routine service and to
apply it to process experimentation in more
detail.
(d) As part of the study of the pre-primary
process a review of air
curing was issued (RD 1879).
The experimental programme in collaboration with
ITL Canada tracing
physical properties through all stages of
processing has progressed as
planned. The next stage will be to bring together
results from
Canada and GR&DC and begin interpretation and
planning of further
experiments on the 1982 crop.
A general investigation of size reduction
processes has begun, the
first stage being a study of novel variations on
lamina cutting. A
report is being prepared incorporating an overall
review and the results
of the first stage.
The experiments explored the effects of cutting
with serrated blades and
of crimping the tobacco prior to cutting or
shredding. These processes
were compared with conventional cutting over a
wide range of cuts per
inch and the interactions with subsequent DIET
expansion were also
investigated. The trends in filling value for the
control samples
were as expected from previous work but the two
experimental processes
gave unexpectedly low filling values. Measurements
of tobacco particle
density and shape were carried out to understand
the mechanisms involved.
In these experiments unexpected trends were
observed even with the control
samples. The initial indications are that the
tobacco particles were
compressed during cutting to a surprising extent
and that this effect
was reversed during DIET expansion. one
possibility is that the
cutting blades used in this small scale experiment
were not as sharp
as those used in normal large scale production.
Further experiments
are required for clarification.
Apparatus is being assembled for laboratory scale
investigation of
the effects of extreme conditions of wetting,
drying, heating and
cooling on the basic properties of tobacco
particles. The experiments
themselves will begin in the second half of the
year.
2. Reduction of Blend Variability
(b) Near infra-red analysis has been the subject
of bi-monthly progress
reports. The main points are summarised below;
Collaborative experiments have been carried out
with Export Leaf, who have
acquired their own Infra Alyzer for routine use on
the 1982 cropt and with
Brazil, Indonesia, Malaysia and Switzerland.
The collaborative studies have involved:
- calibration of the NIR method against wet
chemical methods for the CZ)
tobacco types and crop years used by the different
companies. 110
(DO
Oc
0 1982 British A-crican Tobacco Co. Ltd. This must
not be copied or sho-n to un;tuthoriscd persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


-4-
investigations of sample preparation procedures.
- scanning NIR studies using the Technicon 500 at
Aberdeen
University, to determine the best filters to use
for a
given analysis.
Appropriate advice has been given to the companies
concerned.
(d) A prototype device for simultaneous testing of
filling value
and impedance has been developed based on earlier
research in ITL
Canada and GR&DC (Work Area 15). A report is in
preparation.
Experiments to test the feasibility of applying
tests of this type
to tobacco at process moisture and temperature
(without conditioning)
will be carried out in the second half of 1982.
A computer system has been designed for
implementation on the
PDP 11 computer, for monitoring and logging of
pilot plant operations.
The system provides, within one overall system and
data base design, for
the recording of:
routine clerical details of All experiments.
- routine on-line logging of process settings and
conditions.
- special purpose experimental data logging as
required.
The detailed systems specification is nearing
completion and some pro-
graming has been carried out. The full
implementation will be carried
out in stages over the next six to twelve months.
3. Process Innovation
A systematic review and evaluation of ideas for
process innovations has
been carried out. The next step is to draw up a
plan for the further
investigation of the more promising options.
CO
cc
-9 ::.
0 1982 British American Tobacco Co. Ltd. This must
not be copied or sho,,M to un;tutho6scd persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 12 Tobacco Expansion
Author: R.W. Hedge
Period Covered: January - June 1982
OBJECTIVE:
To identify and develop "second generation"
expansion processes and other
techniques to reduce tobacco usage.
PROGRESS:
Three lines of study have been evolved as a
consequence of a review of the
features of existing technology and the limited
understanding of expansion
and weight saving techniques in general. These
studies, in part, interact
with and depend on research into -particle physics
(Work Area 11) and are:
(1) to demonstrate whether there is a continuum of
expansion processes
and whether or not this relates to the expansion
process/agent;
(2) to study the chemical and physical
relationship between tobacco and
expansion agents and the expansion achieved;
(3) to explore techniques for weight saving by
stiffening of tobacco
particles, in combination or not with expansion.
1. Is there a continuum of expansion processes?
As tobacco particle density is the most important
factor in weight saving
(Work Area 11), especially for lamina, the first
approach in this programme
has been to use lamina thickness as a direct
measure of expansion. In the
case of stem it is intended to use cell
"circularity", if expected progress
in particle physics is made. These measures are
being combined with routine
cylinder volume measures and results from a new
technique, which simulta-
neously measures the flow impedance of a bed being
compressed dynamically
in a cylinder and the bed volume.
Samples have been obtained, over the period of
this review, from a wide
range of commercial processes based on
water/heat/equipment configuration
(e.g. ITH dryer, HHD, HAD, TAMAG/COMAS, WTS),
along- with processes using
other expansion agents (e.g., DIET, G-13, G-13C,
Reemtsma (nitrogen)).
The procedures adopted are relatively
time-consuming, though appropriate
to all materials and small quantities of sample.
it is also possible to
define potential limits, such as, "full expansion
is equivalent to freeze-
dried fully-swoLlen treated particles" - to be
distinguished from the
potential in an uncured leaf before any
processing.
Qualitative observation of the samples and results
from the materials
received earliest in the period suggest that
lamina and stem (and their
related tzeatments) have to be treated as two
separate exercises. In CD
addition, the belief that currently water
expansion of lamina falls well 110
short of other lamina treatments is confirmed. In
this context, freeze- cc
drying of water in a fully swollen cell is a
different mechanism from the cc
remaining processes.
CD
Measurement continues.
0 1982 Britisb American Tob2CCO Co. Ltd. This must
not be copied or sho%n to unauthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 2 -
2. Chemical and physical characteristics of the
expansion agent
Studies of several potential mechanisms have been
considered. Among these
are the contribution of cell wall dynauLics (glass
transition temperature,
which'relates to flexural properties, is a
function of temperature, mois-
ture, rate of strain and exact chemical make-up),
transfer of energy to
the expansion agent, the transport properties of
the agent (including
interactions with the cell material) and the
energy properties of the
agent in contact with the cell material
(volatility, latent heat, bonding
energies etc.).
An initial experiment has been proposed using a 50
mm. diameter all-steam
'expansion tube, of a design allied to the
DIET/G-13 expander. It is proposed
to compare the expansion behaviour of several
homologous series of compounds
covering a range of volatilities, bonding energies
(polarities), latent
heats, molecular sizes etc. It is intended to use
watert Freon-11 and
several related fluoro-chloro and chloro-carbons,
along with some lower
alkanes and alkanols. The expander system has
recently been modified and
calculations are being checked, to, minimise any
hazard from the use of
alkanes/alkanols.
3. Tobacco stiffeninq
Work has not yet started on this concept, except
for some preliminary
literature surveys.
FUTURE WORK:
Emphasis will be placed on completing and
reporting the first two studies
outlined above. However, additional effort has
just been transferred to
this programme and further refinement of the
concepts of expansion will
be pursued. Experimental programmes to evaluate
the potential of item 3
will be considered.
CZ)
N-0
cc
cc
_rtt:.
CD
0 1982 British American Tobacco Co. LtdL T-his
must not be copied or sho-n to un.-tuthoriscd
persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 13 Secondary Processing
Author: W. Lumsden
Period Covered: January - June, 1982
OBJECTIVE:
To assess the needs and problems, both long-term
and short-term, of secondary
manufacturing and to formulate a programme of
research activities.
PROGRESS.
Study reports, surveys and plans issued by GRGDC
and Millbank Production
Departments during the last few years have been
examined. This preliminary
examination has been aimed at the identification
of possible new work areas
and has been supported by numerous informal
discussions with the appropriate
personnel. A presentation of the results of this
examination has been made
to a wide group of senior GR&DC staff and this has
been followed by a number
of committee discussions within the Secondary
Process R&D Department to reduce
and categorise the emerging activities.
The work areas that are emerging can be divided
between
(a) the activities which are capable of being
defined with some
detail now
and (b) activities which require a further period
of assessment and
definition.
The former category will become the basis of
programme proposals and will
probably include studies on the problems
associated with low density cigar-
ette manufacture, high speed packaging aspects of
carton materials and novel
cigarette constructions.
00
cc
_t!-
CD
0 1992 Britisb American Tobacco Co. Ltd. This must
not be copied or shorn to unnuthorised persons.
BATCO document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 14 Ventilated Cigarette Technology
Author: J.K. Milner
Period Covered: January - June, 1982
OBJECTIVE:
To develop methods of controlling the ventilation
level of cigarettes during
production in order to reduce product variability.
PROGRESS:
Our work falls into three clear areas, each
associated with one of the three
main machinery suppliers:- Hauni, Molins and
SASIB.
1. HAUNI
a) Laser Perforator
Following the production test of the 50 watt laser
last year and preliminary
.testing of the 100 watt laser which superseded
it, we conducted a production
trial of the 100 watt unit early this year. During
this trial, 20 million
cigarettes were manufactured for BAT(UK&E) on the
Mk9-5/MAX S-L in GR&DC at a
speed of 5,000 cpm. The first 5 million cigarettes
were made to the prescribed
tip ventilation level of 40% but, after smoking
machine tests had shown that
deliveries where higher than desired, the
ventilation level was increased to 43%
for the remainder of the production run. Analysis
of the results of this trial
confirmed the expected improvement In ventilation
performance provided by the
100 watt laser. At a mean tip ventilation of 43%.
the standard deviation was
2.4% compared with 4.4% for the 50 watt trial.
This standard of individual
cigarette ventilation performance provides a good
base for the application of an
automatic, mean ventilation control system. In the
meantime, we recommend that
the long-term performance of the laser perforator
should be established as soon
as possible using manual control in genuine
factory production conditions.
b) Ventilation Monitor
Last year, we investigated the Hauni on-machine
cigarette ventilation monitor
and concluded that its ability to quantify changes
in mean cigarette ventilation
level was poor. Moreover, we reckoned that
substantial re-design would be
necessary before it could be made to perform
acceptably. Hauni agreed with our
findings and said that they had improvements in
hand.
During the production trial of the 100 watt laser
this year, we took the oppor-
tunity of checking the on-line monitor against the
BAT Ventilation Tester. The
limits we set for control action were + 3%, i.e.,
when the mean ventilation
level reached 40 or 46%, the laser should be
re-adjusted. There were 15 occasions
when the Hauni monitor failed to indicate that
action was necessary and 24 whez.-
it indicated that action was necessary when in
fact action was not necessary4--)
There was never an occasion when it gave an
indication which would have led vmo
correct control action. co
cc
We have informed Hauni of our continuing concern
about the performance of thei
monitor and await their reply to see if they can
re-assure. us about the progres
they are making. k0
c0
(D 1982 British America- Tobacco Co. Ltd. This
must not be copied or sho" to unauthorised
persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


-2-
Report No. RD. 1878, 5.5-82, covers the
performance of the laser and the monitor
during the recent production trial.
2. MOLINS
a) On-Line Perforation
Molins' approach to the on-line creation of
ventilation is to perforate mechani-
cally an the PTA filters wrapped in non-porous
plugwrap and to perforate the
tipping either electro-statically or mechanically.
We have conducted tests in GR&DC with expe
'rimental filter perforating equipment
provided by Molins but have not, as yet, been
offered any tipping perforating
equipment for test. So far, the on-machine
perforation of filters has only
proved capable of matching relatively low porosity
plucrdrap suitable for tip
ventilation levels of up to, say, 20%.
Progress is dependent on Holins, who"do not seem
to be putting great effort into
the project.
b) On-Line Ventilation InsDection
Our work on the CID6 has just been concluded with
a semi-production test. T'he
results were encouraging in that the unit gave an
adequate indication of mean
ventilation over a wide range of ventilation
levels and cigarette constructior-s
and rejected individual cigarettes with faulty
ventilation. These tests are
reported in 'RD. 1885 with the recommendation that
the CID6 should be given a fuu
factory evaluation.
3. SASIB
a) Electro-static Perforator
Tests in GR&DC last year on an experimental
electro-static perforator showed
that tipping could be perforated to match the
Benkert zone perforated Z5/120
material, which is suitable for cigarettes with a
tip ventilation just over 30%.
A prototype unit, incorporating the lessons learnt
during last year's experiment-s,
has just been provided by SASIB for testing in
GR&DC. This is currently under-
going off-line proving trials before incorporation
on-line with the SIGMA/Ho.
b) On-line Ventilation Inspection
The co-operative programme of work undertaken by
SASIB in Bologna and ourselves
in GR&DC has reached the stage where a concept for
a ventilation information and
control system has been validated on low speed
machinery. A system based arx
this concept would have the ability to measure not
only ventilation but also
other pneumatic properties of the cigarette, such
as the balance between tip and
cigarette rod ventilation. The work to validate
the concept has been reported
in RD. 1867, 21.4.82, and patents have been
applied for in appropriate countries.
SASIB have installed recently in GR&DC a SIGRA/Ho
capable of operating at speeds
up to 6,500 cpm in order that a practical on-line
ventilation system may be
developed for use with nodern, high speed
machinery. Work has already star--ed
on experiments aimed at obtaining satisfactory
pressure signals from the inspec- CDC
tion drum. CDC
J~!
0 1982 British Americ2n Tob=o Co. Ltd. This must
not be corwied or sho-n to unauthoriscd persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


3
FUTURE WORK:
Until such time as Hauni are able to offer a
substantially re-designed ventila-
tion monitor, there would seem to be no need for
us to carry out any further
testing of their equipment in GR&DC. Molins have
declined our offer of testing
the CID6 in conjunction with the Hauni MAX S-L but
SASIB have offered to supply
an experimental ventilation monitor for test with
the MAX later this year. For
the next few months, our efforts will be
concentrated on our co-operative exer-
cise with SASIB to develop an on-machine
perforation system with automatic
control of ventilation. The technology developed
for this system will be capable
Of application to machinery from all suppliers.
(ZD
`10
co
co
0 1982 British American Tobacco Co. Lt(L T'jis
must not be copied or shoun LO un-lUthorised
persons,
BATCo document for Province of BritiSh Columbia 5
November 1999


RESTRICTED
STATUS REVIEW NOTE
Work Area: 15 Test-method Development
Author: H.F. Dymond/A.W. Cronshaw
Period covered: January-June# 1982
OBJECTIVE:
TO ensure that suitable test methods are developed
to meet both the needs
Of GR&DC projects and of Operating Companies,
including those dictated by
external requirements and government regulations.
PROGRESS:
There are six broad divisions of this work area:
1. Liaison with regulatory bodies to establish
approved analytical methods.
2. To provide a back up function for all research
projects by providing
chemical test methods as necessary.
3,4. The development of analytical/testing
procedures which are relevant to
current anticipated Group needs.
5. Air flow measurements and their use in
instruments - for example on-line
measurement of the air permeability of tipping
paper, off-line semi-
automatic measurement of air permeability of
paper, puff profile and
puff volume in smoking machines and the
measurement of pressure drop.
6. General test method development and the
application of physical measure-
ment for on-line and off-line work in association
with other projects -
for example, filters, ventilated cigarette
technology and tobacco pro-
cessing.
1. Liaison with TAC
Methods for determining nitric oxide and acrolein
(plus otherc(,P-unsaturated
compounds) in cigarette smoke have been submitted
to TAC and will be included
in the manual of standard methods (RP11). A
method.for determining hydrogen
sulphide in cigarette smoke, based on continuous
flow, has not proved reliable
and is being further investigated.
2. Analyses Related to Research Projects
Analytical support has been given to almost all
projects in the GR&DC pro-
grammes. Much of this work has involved the
application of standard methods
to a wide variety of samples, but specific
analyses and modification of
methods have been required in many instances.
Work in the environmental room to examine the
influence of temperatu-e and
humidity on objectively measurable parameters of
low delivery has been
successfully completed. Preliminary analysis of
the data indicates that the
present study will show similar effects to those
reported in an early study co
of high delivery products. The constituents under
investigation are PMWNF, 00
nicotine, CO, NO, acetaldehyde and steam volatile
phenols. r--
cz:>
0 1982 British American Tobacco Co. Ltd. This must
rKx be copied or shown to un-luthorised persons.
BATCo document for Province of BritiSh Columbia 5
November 1999


- 2 -
Considerable support has been given to the Filter
Research and Sidestream Work
Areas- Development of a method to determine
hurnectants in cigarette smoke is
in progress. A newly installed amino-acid analyser
has provided comparative
results on leaf study/flavour projects but
quantitative aspects require
improvement.
3. Development of a Method for the Measurement of
Ultra-Low Delivery
Products
The existing Cambridge filter method of
determining PMWNF, recognised as
inadequate for cigarettes delivering less than
ca.5 mg per cigarette, has
been modified to use Cambridge filter holders made
from non-hygroscopic
materials. The measurement of PMWNF is now
considerably more reliable (RD
1865).
In addition, an in-situ extraction procedure,
injecting specially dried
isopropanol into sealed Cambridge filter holders,
gives figures for water in
TPM which are lower than those obtained by the
conventional extraction proce-
dure. Combining the two modifications yields more
realistic PMWNF/nicotine
and water/PMVINF ratios; no negative values have
been obtained for ultra-low
products. Using the new procedurel PMWNF values
from sub 1 mg cigarettes
are generally higher than those values from
conventional methods.
As part of this project, a coulometric Karl
Fischer moisture meter was
investigated. This instrument requires no
calibration and gives a read out
directly in micrograms of water in less than 30
seconds. Fliminating the
need to calibrate represents a significant
simplification of the procedure
for the determination of water from U.L. delivery
products.
4. Development of a Rapid Automated Gas
Chromatographic Method for the
Simultaneous Determination of Nicotine and Water
in Cigarette Smoke
A report (RD 1845) has been issued which describes
the development of this
method. A total analysis time of four minutes per
injection with a throughput
of 100 samples plus standards in 7 hours is
possible. The system is fully
automated and requires little operator
intervention.
The method has been subjected to exhaustive trials
and a comparison between
this method and the continuous flow method for the
determination of nicotine
has been performed. No interfering substances have
been found and the method
har, been successfully applied at very low levels
of nicotine and water.
Initial examination of the data shows extremely
'good agreement between the
G.C. and continuous flow methods.
S. Air Flow Measurements
The- development of the off-line semi-automatic
air permeability meter was
completed. A report was issued. The work on the
off-line instrument was a
necessary part of the development of the on-line
permeability meter.
The development of the on-line permeability meter
continued. Three styles of
test head were designed, built and tested on a
laboratory test-rig. Three
measurement arrangements were tested with the
third head:
(a) The original method where air was blown at the
paper, a large C=)
amount of this air was allowed to leak to
atmosphere and the 110
air flow into the chamber on the lee-side of the
paper was used cc
as an indicator of paper permeability; cc
X".
C)
0 1982 British American Tob--cco Co. Ltd. This
must not be copied or shown to un.,uthoritcd
persons.
BATCO document for Province of BritiSh Columbia 5
November 1999


- 3 -
(b) A modification of method (a), where the
pressure in the chamber
on the lee-side of the paper was used as an
indication of paper
permeability;
(c) A method where constant air flow was drawn
through the paper and
the pressure in the chamber on the lee-side of the
paper was used
as air indication of paper permeability.
All three measurement methods showed adequate
resolution in the results.
Method (b) is the preferred arrangement.
The puff volume and puff profile work was halted
because the manufacturer of
the pressure transducer was developing a new
transducer that should overcome
some of the problems with the existing transducer.
The new transducer has.
been delivered and built into