Product Design
Subject: Monthly Development Summary - Research Directorate
Abstract
Summarizes work on Phillip Morris research and development projects. Reports progress in the ART-STEM study, with the exposure of stems to citric acid and supercritical Co2. Evaluates organic acids and potassium salts for sidestream reduction. Mentions use of CORESTA smoking machine in the odor modification program. Summarizes TSNA precursor research. Studies sensory effects on taste and odor modification. Discusses varied cigarette paper design projects, including CaCo2 paper, Ecusta prepared magnesite, dolomite paper, Aromatek applied paper, and paper incorporating vanillin release compound. Notes that the initial results of Dr. Helena Chum's work with flax fiber cigarette paper are "extremely exciting because they suggest that eventually we may be able to systematically design cigarette paper parameters so as to optimize cigarette taste characteristics."
Fields
- Author
- Charles, James L., Ph.D. (PM, R&D VP, Pharmacologist, Industry Expert)Vice President of Research and a scientist for Philip Morris, Inc. Vice President of Research for Philip Morris, Inc. in 1986 and then again from 1992 to 1993.
- Recipient
- Houghton, Kenneth S., Ph.D. (Senior VP of R&D at Philip Morris)Defense
- Hypothesis
- Sidestream constituent yieldsModification of selected sidestream smoke constituents in response to health concerns.
- Use of additives
Modification of tobacco products through use of additives and measuring effects on dependence, behavior, and toxicity.- Use of filters, paper, and ventilation
Modification of tobacco products through use of filters, paper, and ventilation, and measuring effects on dependence, behavior, and toxicity.- Use of tobacco processing/ blends
Modification of tobacco products through changes in tobacco processing and use of blends, and measuring effects on dependence, behavior, and toxicity.- Sensory effects
Technologies used to measure, control, or alter sensory effects - Use of additives
- Keyword
- Flavor (Taste)
- Menthol delivery (Smoke menthol, menthol yield)
- Nicotine manipulation
- Pyrolysis
- Secondhand Smoke (Sidestream smoke, SS)
- Sensory response
- Sidestream visibility
- Tobacco taste (Attribute measure)
- Menthol delivery (Smoke menthol, menthol yield)
- Additive
- Ammoniasee also: Ammonium bicarbonate, Ammonium carbonate, Ammonium chloride, Ammonium hydroxide, Ammonium sulfide, Diammonium phosphate, and Urea
- Carbon dioxide
- Citric acid
- Magnesium carbonate
- Vanillin
- Ammonia
see also: Ammonium bicarbonate, Ammonium carbonate, Ammonium chloride, Ammonium hydroxide, Ammonium sulfide, Diammonium phosphate, and Urea- Menthol
- Carbon dioxide
- Smoke Constituent
- Acids
- Alkaloids
- ammonia
- Carbon dioxide
- Hydrogen cyanide (HCN)
- Menthol
- Nicotine
- Nitric oxides
- Nitrosamines (N-nitrosamines)
- Alkaloids
- Design Component
- Ash appearance
- Casing
- Cellulose acetate filter (CA filter, Conventional filter)
- Flax paper (Flax wrapper)
- Humectant
- Magnesium oxide paper
- Woodpulp wrapper
- Casing
- Operation/Project
- Project 1904
- Project ART (Denicotinized cigarette)
- Project OSIRIS
- Project ART (Denicotinized cigarette)
- Named Organization
- COMARK
- Coresta (Industry-affiliated Int'l scientific/research group)
An international organization whose objective is "to improve cooperation in scientific research and tobacco." Consists of 186 member companies/institute from 54 countries, including tobacco, paper and filter companies, and universities engaged in tobacco research.- DUPONT
- Eastman Kodak Co. (Kodak) (Cigarette filter mfg from 1950s to 1994.)
Manufacturers of quality control equipment for cigarette packaging- ECUSTA
- ELECTRONIC TECHNOLOGY IN IMAGING
- GENEBANK
- Kimberly-Clark Corp. (Specializes in the tobacco reconstitution process)
Specializes in the tobacco reconstitution process and in helping the tobacco companies control their nicotine- NELSON
- NY POLYTECHNIC
- PERKIN ELMER
- Pfizer Flavor & Fragrance Product Group (C.A.L.) (supplier of tobacco extracts in 1993)
A New Jersey supplier of tobacco extracts in 1993- R.J. Reynolds Tobacco Co. (Cigarette manufacturer (Camel, Winston, Doral))
Cigarette manufacturer (Camel, Winston, Doral)- RSA
- SOLAR ENERGY RESEARCH INST
- SUGARMAN
- Usda, U.S. Dept of Agriculture
- WATTENS
- CALLERY CHEMICAL
- Coresta (Industry-affiliated Int'l scientific/research group)
- Brand
- Ambrosia
- CABIN
- Marlboro (PM)
- CABIN
- Subject
- Filters (Design)
- Paper (Design)
- additives
- acids (additives)
- Humectants
- Menthol (Additives)
- nicotine technology
- Particle Size (Technology)
- Sensory Effects—Impact (Effects)
- Test/Smoke Machine (Testing)
- Secondhand Smoke/Constituents
- Paper (Design)
Document Images
PHILIP MORRIS U. S. A.
INTER-OFFICE CORRESPONDENCE
Richmond, Virginia
CONTI D EN I i P`U;
r
To: Dr. K. S. Houghton Date: June 28, 1990
From: J. L. Charles C90-03771
Subject: Monthly Development Summary - Research Directorate
ART - STEM STUDY' - Progress was made in efforts to identify the major chemical changes
in stems resulting from exposure to supercritical carbon dioxide, to supercritical CO, sol'ubles
of filler, and the addition of citric acid.
The samples analyzed were uncased CRS, citric acid treated CRS, citric acid treated CRS
extracted at BHPP with no filler present, citric acid treated CRS extracted~ at BHPP with filler
present, and CRS without citric acid extracted at BHPP' with no fill'er present. Gas
chromatographic analysis, evolved gas analysis, NMR analysis,and mass spectrometric
analysis were performed on these samples. The citric acid treated CRS before and after BHPP
extraction with no filler showed differences attributable to selective loss of some compounds
resulting from the extraction: These differences are the same in the SFE/GC analysis of CRS
and the GC analysis of the hexane extracts. The interpretation of the changes being observed
awaits subjective data which will be obtained in the near future.
Sidestream Reduction - Numerous combinations of organic acids and their corresponding
potassium salts were evaluated for sidestream (SS) reduction. In general, the addition of acid
enhanced SS reduction, but often at the expense of a longer static burn time (SBT). It appears
that we should be able to utilize judicious combinations of salts and acids in order to optimize
SS reduction and SBT.
A regression analysis was completed of data from a matrix of malonic acid/KH2 PO; sizing
solutions. A small synergistic effect between KH2 PO, and malonic acid was found. The
contribution of the malonic acid for SS reduction was greater than that of the ICI4, PO, .
Comparisons of several dibasic and monobasic potassium salts of organic acids (e.g.
K, malonate vs. KHmalonate) showed the monobasic forms usually to be better SS reducers.
KHfumarate, however, was surprisingly poor. Low levels of Na, fumarate were found to be
ineffective.
The effects on SS reduction of a variety of sizings were evaluated on a high basis weight
CaCO, paper. Potassium gluconate performed well, with its performance proportional to the
K+ level. Potassium galacturonate, however, performed poorly. The addition of fructose to
KH, PO, decreased SBT relative to KK PO, alone and decreased SS reduction: Disodium
fumarate was shown to give SS reduction similar to KH, PO, with a lower weight add-on to the
paper, while dipotassium furnarate at a comparable level of cation addition was clearly inferior.
The mixed sizing of KH, PO, and K, H2 P= O, was examined'. This combination worked well and
offers the potential of reducing SBT compared to KHz PO, alone. Higher total levels of sizing
need to be examined.
Preliminary results have been obtained on research being conducted under contract with Dr.
Helena Chum at the Solar Energy Research Institute. Clear dependencies on pH (of applied
solution), temperature, and additive concentration (as % K+) have been seen for both pyrolysis
and combustion products obtained from the flax fiber in cigarette paper. For combustion, four
dominant classes of products were identified: levoglucosan derivatives, furan derivatives,
"lignin" derivatives, and CO; and related compounds.. These initial results are extremely
exciting because they suggest that eventually we may be able to systematically design cigarette
paper parameters so as to optimize cigarette taste characteristics.
Selected Ecusta-prepared magnesite (Baymag A) handsheets from a matrix study were used to
prepare handmade cigarettes for SS analyses. The best SS reductions were obtained for the
least porous papers.
Evaluation of a newly processed sample of mined magnesite (Baymag C) was conducted. The
Baymag C sample was found to be of equal or greater purity than the Baymag A sample which
previously had been examined. Analyses showed good SS reduction for both medium and high
basis weight papers. Also, the ash appearance was irnproved compared to previous magnesite
samples. In addition, subjective screening yielded quite favorable responses, suggesting that
this new material may be a good candidate for preparation of a mill run of paper by Ecusta. At
this point; the only concern for Baymag C is that it is somewhat low in brightness, which might
necessitate its being used as part of a mixed filler system along with CaCO, .
Previous results had shown that SS reduction of magnesite papers is not dependent on K+ level,,
with essentially equal performance from a wide range of additive levels (-2 - 10% K, citrate or
K, succinate). The lower sizing levels gave longer SBT's. However, when unsized magnesite
papers were examined it was found' that the resultant handmade cigarettes would not burn. This
finding suggests that the primary role of the sizing agents examined to date on magnesite
papers is to function as a burn control compound. A preliminary examination of low levels of
acid sizing (without K+) on magnesite papers has shown that the papers will burn, but with
very long SBT's. Further investigation of other acids and monobasicpotassiurn salts of organic
acids on magnesite papers is in progress.
The performance of a paper made with a precipitated derivative of dolomite (Pfizer) was shown
to depend strongly on its sizing agent. Potassium succinate was much more effective than
i
-2-
K, citrate, which is similar to previous observations of their relative effectiveness on
hydromagnesite papers.
Sherry Baldwin was sent on a trouble-shooting mission when problems were encountered by
Kimberly-Clark in the application of Aromatek-245 to the 091 paper with 12% KK PC?, . Both
their method of application and their method of analysis were inappropriate. The "wet pick-up"
method that they usedd for analysis was unacceptable. The result was that pick-up was
underestimated. Compounding this problem was the fact that the application method used an
open pan beneath the roller to hold the coating solution, in which ethanol was the solvent.
Evaporation during the 2-hour coating period also resulted in high levels as well as
nonuniformity. A suggestion was made by Sherry to use a closed reservoir to contain the
solution and apply in a fashion similar to size-press coating. K-C concurred, having
independently reached the same conclusion. An uv spectrophotometrio method also was
developed by K-C to analyze the paper for the aromatek. That method gave results in excellent
correlation with our HPLC procedure, although with a slight bias on the low side. This
procedure is fast and accurate and'is well suited for a QA method. After some trial and error to
obtain required coating solution concentrations, the target specifications were achieved.
A series of machine-made cigarettes known to give a wide range of SS reductions was analyzed
on both the laser single-port SS visibility instrument and the 8-port instrument. The same
relative rankings of extinction coefficients and SBT's were obtained on both instruments.
In a continuing examination of handmade cigarette production using a small roller-maker,
subjective screening of roller-made cigarettes indicated minimal differences from handmade
cigarettes produced in the typical manner., SS analyses on additional models suggested that
roller-made cigarettes may tend to give slightly greater SS reductions than the usual handmade
cigarettes. This result might be due to the slightly smaller circumference of the roller-made
cigarettes, which in turn may produce a slightly greater filler density. These roller-made
cigarettes may actually be better predictors for machine-made cigarettes. Further comparisons
are in progress.
For the month 71 handmade cigarette models were prepared for the reduced SS program. A set
of handmade cigarettes made from cigarette paper handsheets incorporating a vanillin-release
compound (CR-2936) also was prepared. SS visibility analyses were conducted on 74 models.
Thermall analysis support is being provided for the evaluation of inorganic fillers which are O
being prepared as part of the paper technology effort. ~
~
ThermogravimetriclDifferential Thermal Analysis (TG/DTA) was used to determine the CA
composition of calcium carbonate mixtures. All samples were heated from ambient ~
temperature to 1000 ° C at rates of 10 ° or 20 ° per minute. The samples were found' to be N
~
-3-
mixtures of calcium carbonate and/or magnesium hydroxide, calcium hydroxide and water.
The percentage of each component was reported.
Several analytical studies were reported in association with a series of reduced sidestream
cigarettes sent out for a POL study. These studies included selected mainstream gas phase
components and also the toltiene/benzene ratio.
The compounds determined were hydrogen cyanide (HCN), acetaldehyde (CH, CHO), carbon
monoxide (CO), nitric oxide (NO), benzene, and toluene. The cigarettes were made with the
same blend but with differing filters, papers and paper additives. The test cigarette was a 100-
mm cigarette with 45-g/m2 paper, added 12% KH2 PO, and a charcoal filter. The control
cigarette was a 100-mm Marlboro Lights model with a 25-g/m2 paper, a CA filter and no paper
additives. The HCN, CH, CHO, CO, and NO were determined by the PTIR. The benzene and
toluene were determined GC analysis.Compared' to the control, the test cigarette had lower
HCN and toluene deliveries and increased CO: deliveries. In addition, the toluene/benzene MS
delivery ratios were lower for the cigarettes with the 45-gm/m2 paper and 12% I{H~ PO, .
The increased level of CO and the decreased toluene\benzene ratio both suggest that the the
cigarette with the high basis wt. paper had a higher pyrolytic temperature than the control. The
decreased cyanide delivery is due to the use of a carbon filter on the test model.
Sidestrearn Smoke - Multiple smoke runs were made on the prototype CORESTA smoking
machine in support of the sidestream odor modification program. Mainstream and sidestream
smoke was collected on TPM pads for analysis by the Analytical Research Division. Samples
included cigarettes prepared with Aromatek 100 and Aromatek 150 in adhesive, cigarettes
prepared by coating of Aromatek 150 on the wrapper, Chelsea and Horizon (RJR commercial
products), and models with Aromatek 245 coated on the wrapper.
Videotapes have been made of various cigarettes utilizing the visibility setup previously
described. Significant differences have been observed from cigarette model to model and even
within an individual pack from the same model. Observations about the ash and interesting
views of smoke emanating from the unburned wrapper should prove important to future plans
in the sidestream program.
In conjunction with CAD, the pc-based Perkin-Elmer Nelson Turbochrom II chromatography
data system has been put; into operation. This is being evaluated as a replacement for the PE-
LIMS system currently in use. A Hewlett-Packard Chemstation will be installed for
comparison with the PE-Nelson system.
Sidestream Smoke Chamber - Several additional chamber runs were carried out in support of
the Biochemical Research Division nitrosamine work. Collection methods for unaged
sidestrearn smoke were evaluated this month. Preliminary indications are that artifactual
-4-
nitrosamine formation is occurring in, the collection procedure currently used, and that bubbler
trap RTD may have a sibnilficant impact on this phenomenon.
The instrument to monitor mainstream CO and CO, in conjunction with the normal chamber
sidestream operation has been reconfigured by the manufacturer to allow CO, determination at
the 0'-20%Ieve1. It was determined that the origina10-10°To range for CO, was insufficient for
the actual mainstream levels found. Actual smoke runs using IM1 3 cigarettes yielded CO and
CO= values consistent with those expected based on calculation and CI data. This will continue
to be monitored to determine consistency of operation and will become a routine part of the
normal sidestream chamber operation.
A sampling system for the LASX particle size monitor has been constructed and installed in the
chamber. This system incorporates a filter/capillary tube dilution device at the inlet of a 1/2
inch sampling tube inserted into the chamber. The dilution device reduces the particle number
concentration in the sampling,system by drawing most of the flow through a Cambridge filter.
A small amountof the flow passes through a capillary tube which does not remove the aerosol
particles. In effect, the aerosol sampled from the chamber is diluted with its own gas phase.
This system is driven by an air pump set to pull 1.6 I/rnin. The LASX inlet tube is placed~ at the
center of the 1/2 inch sampling tube through~ a'I'-fitting and draws the diluted aerosol into the
counter at a flow rate of 1 cm' /sec. The system will be fully tested as soon as the
modifications to the chamber are completed.
Installation of the Honeywell alarm/control modifications has begun. Additional and upgraded
software will be included in these modifications.
An investigation to determine if vanillin could be collected from chamber sidestream smoke in
a similar manner to that used for other aldehydes has been completed. It was found that
vanillin swept onto a DNPH treated Seppak and the Seppak eluted and analyzed by HPLC
yielded a peak corresponding to the same retention time as a peak from vanillin dissolved
directly into: DNPH/acetonitrile solution. The peak elutes in the same region as acetone and
propionaldehyde. All work to date has been qualitative. Quantitation will follow if chamber
runs on vanillin release materials are required.
In collaboration with personnel of ARD, IML3 chamber runs have been carried out to compare
scrubber traps and silica gel tubes for ammonia collection. Acid extracts of the silica gel tubes
and scrubber solutions have been cross checked by ion selective electrode and ion
chromatography. All results will be summarized upon completion of additional IM13 runs in
the ARD chamber.
Modifications have been made to the transmissometer to facilitate test point access and~ ease of
calibrations. The instrument is designed to operate unattended and in extreme weather
-5-
conditions. The main: electronics are located inside a sealed protective cover which: must be
removed for test and calibration. An access window has been provided with a lexan window.
The test points and adjusting pots are made accessible by holes drilled through the lexan
window. These routine procedures can now be carried out with minimal effort.
Tobacco Specific Nitrosamines - Precursor Studies: Alkaloid- and preformed-TSNA-depleted
filler prepared by ethanol extraction and preformed-TSNA-depleted filler prepared by the ART
process were subjected to ageing at room temperature and at - 30° C. After 14 weeks the
ethanol extracted filler has shown no increase in filler TSNA levels for either storage
temperature. No changes in filler TSNA content was noted at either storage temperature for the
ART filler after 5 weeks..
Washing DBC burley filler with a lanthanum chloride solution did result in reduction of water
insoluble calcium by -50%: The effect of this calcium reduction on MS TSNA has not been
measured.
A new approach to analysis of fillier for pseudooxynicotine (PsON) is being investigated. The
method includes extraction of filler alkaloids (including, presumably, PsON), reduction of
PsON to deuteronicotine with sodium cyanoborodeuteride, and analysis for nicotine and
deuteronicotine by gc/ms. Preliminary results are promising.
PEI: Preparation of partialiy-purified PMT extracts from hydroponically grown Burley 21
tobacco plants continued. A method' for elution of phenylethanolamine transferase (PET) from
an agarose isoelectric focusing (IEF) gel was developed. PMT samples eluted from IEF gels
were examined by SDS-PAGE. Most of the material present in these PMT enriched samples
had a molecular weight of 60 kD. Fractions with PMT activity were eluted from native
polyacrylamide gel and with the ELFE apparatus.
In the experiments tolabel PMT and PET with' H-SAM in the absence of photolysis, several
major (in the molecular weight area of interest) as well as minor (not in the molecular weight
area of interest) protein bands were labeled. This suggests a non-specific binding effect under
the reaction conditions examined.
A portion of a PET preparation was given to members of Project 1904 for isoelectric focusing
gel electrophoresis separation. Following separation, portions of the gel containing PET
protein were subject to PAL following by 1D-SDS-PAGE and fluorographic analyses. The IEF
separation PET sample contained radiolabeled proteins at PET's reported molecular weight and
the labeling pattern was very similar to an untreated PET control. This indicated that the IEF
procedure can be used to separate proteins which can then be photoaffinity labeled!.
-6-
Singlestranded (ss) DNA was prepared from selected'phagemids. The ssDNA from four
clones was sequenced. A recently acquired computer package is being used to search for any
homology of DNA sequences generated at PM with DNA sequences in the Genebank database.
Prior to further differential hybridization experiments, the total RNA was isolated from the
roots of hydroponically grown tobacco and tomato plants.
Lowered Biological Activitv - Bioassay Development: Two modified - protocol soft agar
assays testing the ability of 2R1 CSC to transform JB6 cells were completed. The modifica-
tions were 20% serum in the agar medium and pretreatment of the cells with the 2R1 CSC.
Neither modification produced a significant number of transformants over that observed with
the control cell cultures.
Model Evaluation: S/M assay work continues to be done to support the reduced'sidestream
smoke program.
Model Development: Preliminary indications are that application of tobacco solubles to base
web by soaking is as efficient as spraying. Condensate has been prepared from cigarettes
which included tobacco solubles modified by treatment with hydrogen peroxide.
Glutathione Depletion Assak: The V79 mutation assay was conducted using 2R1 CSC and
DEM pretreatments to deplete the cells of GSH. Non pretreated cells were used as controls.
All the cells were then treated with MNNG to induce mutagenicity. There were no significant
differences in the number of mutants per survivor between non-pretieated or pretreated cells.
These results indicated that GSH depletion by CSC or DEM (which did occur at the expected
levels), had no effect on MNNG mutagenicity..
Entomolqgy - i~B oloav: As a first step in method development, tobacco hornworm purge fluid
has been obtained and subjected to ultrafiltration to isolate peptides which may be of value in
cigarette beetle control.
Low-oxygen modified atmospheres similar to those generated by Isolcell(D machines were
shown to kill CBs. However, Isolcellt9 - type fumigations would require at least 11 days to
complete and this does not fulfill the original objective of a 5-day treatment.
A memo was issued documenting this result.
A memo was issued which documented a 1 year warehouse study which examined temperature
as a possible means of CBB control. The results tended to show that the warehouse moderated
the outside temperature changes; higher hogsheads in a stack had a higher temperature than
lower stacked hogsheads; and more CBs were caught in the highest trap position. Since so few
CBs were caught in all the traps for the duration of the study, the initial objective was not
accomplished.
-7-
Service to Others: Advice on infestation control practices was given to PME and PMI's
Mexican affiliates and licensees. Methoprene residue results on a variety of tobacco samples
were communicated to PMI. Collaborative studies with Engineering personnel revealed that
zone 3 in an ordering cylinder is the optimal site for applying Kabatt9 to ET.
Cell/Tissue Culture Research - Tobacco Microbiology: Samples of bright casing (water
based) stored at 4, 27, 37.5 and 50 ° C have been analyzed for changes in microbial populations
after 0, 24 hr, 48 hr, l week, 2 weeks, 4 weeks and 8 weeks of storage. To date, no detectable
microbial populations have been observed in any of the samples.
Samples of cased burley stored at 27 ° C and uncut blend and cut filler stored at 4 and 27 ° C
were analyzed after 0, 4, 24 and 48 hr of storage. Bacterial counts for all samples were similar
to those observed for the laboratory standard. ''
Expanded stem samples from different points in the process at Louisville were analyzed for
bacteria, yeast and mold. There were small decreases in the microbial counts after conditioning
and expansion, but not at a significant level. Microbial counts taken throughout the process did
not exceed the feed sample mean count.
The FTR procedure for the determination of ergosterol was evaluated' as an alternative indicator
of mold growth on tobacco. Bright tobacco was analyzed for ergosterol after it was allowed to
visibly moldd under jungle room conditions (90°F, 85% RH). The ergosterol increased from
0.9 µg/g of tobacco at zero time to an average value of 10 µg/g after 9 days of incubation.
Mold counts at time zero or day 9(fust visible mold observed) using the existing SOP did not
correiate with ergosterol results. Interestingly enough in this same study, bright tobacco
samples that started out with a low mold spore count (-42/g) versus bright samples with a high
zero time mold spore count (-3500/g) had visible mold on them after only 9 days of incubation.
The high zero time mold spore count samples did not mold until day 13!
Hand-made cigarettes (without filters) containing bright tobacco with high mold spore counts .
(-3500/g) were dry puffed or smoked under normal smoking conditions. The "particulate
matter" was collected on a Cambridge pad and analyzed for mold spores. None were found.
Alternate Preservative ProLRram: Caffeic acid, gallic acid', sebacic acid, phytic acid, 2,6-
dimethoxyphenyl and (3-carotene were tested for their antimicrobial properties in the agar
inclusion assay. None of the compounds displayed efficacy equal to that of propylparaben.
Electrophvsiolol;icat Studies - PREP testing to compare the electrophysiological and
subjective effects of ART-blend cigarettes made from varying proportions of extracted and
unextracted fillers has been completed. Latency and amplitude measurements are currently
-8-
being obtained for all of the PREP waveforms. Following the completion of these measure-
ments, both the subjective and the electrophysiological data will be subjected to various
statistical analyses.
Work on the nicotine-menthol interaction study continues. The objective of the study is to
compare the electrophysiological and subjective effects of ART-blend cigarettes containing
factorial combinations of four levels of menthol (0.00, 0.41, 0.85, and' 1.95 mg/cigt) and three
levels of nicotine (0.08, 0.41 and 0.91 mg/cigt). PREPs are being recorded before and after the
smoking of each of the twelve experimental cigarettes. Additionally, at the end of each
experimental session, subjects are required to complete a ballot evaluating the cigarette that
was smoked. To date, the study is 30% completed.
Ambrosia - Solubility problems associated with different crystal mophology for aromatek 150
samples prompted a study of several batches of this material by analytical microscopy.
Samples with smaller particle size had better water solubility. Alcohol reprecipitation tended to
be associated with filamentous crystals which cause some difficulties when mixed with glue for
sideseam adhesive application.
Extensive hplc support was provided for CR compounds and for aromatek 100, 150, and 245
samples this month. In particular, a series of experiments were performed on MS and SS
smoke collected on TPM pads. The objective of these studies was to quantitate the transfer of
intact release material to smoke as well as the expected breakdown products. In addition to the
hplc work on release compounds in paper and sideseam adhesives, the Chemical Analysis
section developed a rapid UV procedure for Aromatek 245 coated on paper., Each study was
communicated to appropriate New Products staff on a pre arranged schedule in order to:
facilitate meeting target dates for these products.
©11tical Proces.sing - The search for a business partner to develop a commercial machine
vision system incorporating the OSIRIS inspection methodology has continued. A series of
meetings have been held with representatives from several major machine vision companies.
Significant interest has been expressed in the OSIRIS inspection method. Proposals are
expected from each of the vendors by the first week of July.
Changes have been made to the design of the OSIRIS display screens to accommodate the
needs expressed during last month's advisory group meeting. Problems encountered with the
Matrox imaging hardware have been communicated to Matrox. Solutions to these problems are
to be incorporated in the next software release which is due July 1. The fourteen slot PC
backplane was found to have an incompatible slot spacing. A ten slot backplane will be used
until October when COMARK delivers a new fourteen slot backplane with appropriate
spacings.
-9-
A meeting was held with the president of Electronic Technology in Imaging (ETI) to discuss
the application of their inspection hardware to the print inspection problem. ETI has a strong
background in web inspection and is licensed to market Dupont's inspection products. Their
product is a strong candidate for providing a cost effective solution to print inspection. A
proposal from ETT has been solicited.
New Expansion Processes - High speed Spin Physics cameras were used, with the assistance
of Larry Marshall of Industrial Engineering, to photograph expansion and collapse of CO=
impregnated'bright and burley tobacco. Expansion and collapse occurred at table top
conditions under two 6000 watt lights. The objectives were to establish: 1) whether expansion
occurs in both the x and y directions in the plane of the leaf as well as 'ut the z direction
perpendicular to it, 2) whether quantitative measurements of the changes in each direction
could be obtained, 3) the rate at which the tobaccos expand and collapse, and 4) -the maximum
expansion achieved and net gain after collapse.
The results indicate that expansion does occur along each of the three axes. Quantifying the
change in each direction proved impossible due to sample handling difficulties and buckling.
Growth in the x-y plane was affected by veins and defects. By averaging the changes in these
two directions, estimated maximum expansion in this plane was 30%, and this occurred in
about 8 sec. Subsequent collapse in this plane occurred over an approximate 20 to 30 sec.
period leaving a net gain of about 10%. The resolution and depth of field of the cameras were
insufficient to permit measurements of changes in the z direction. An estimate of expansion in
this direction could be made from specific volume(SV) measurements of the overall changes in
volume which were about 65%: With an approximate increase of 10% in each the x and y
directions, this would yield about 36% expansion in the z direction.
In order to determine the brittle point of DIET feedstock acoustic emission tests were
performed. The filler was cooled to the appropriate temperature and then subjected to
compression while acoustic emissions were being monitored. The onset of brittleness was
determined from the amplitude and total number of events. The brittle point was found to be a
strong function of filler OV. For filler at 23.5% OV the brittle point was -20 t5° C, that for
16.3% OV was -5 t5° C, and for 11.5% OV was 0 t5° C: Due to experimental limitations
these numbers should be considered estimates only.
The test grid described last month is still being pursued. One major observation to date is that
large differences in stability exist between the gaseous impregnation conditions of -18 ° C and
+10° C. After 30 minutes better than 80 0 of the impregnated CO~ , depressurized isothermally
from 230 psig, was retained in filler chilled to -18° C while less than 15% was retained on
filler chilled to +10° C even though the initial CO2 uptake was approximately the same.
IJncased tobacco, JOQG, at 12.5% OV was used in both cases. Further work will focus on the
effect of OV on the amount of CO, retention and stability.
-10-
h%(enthol Studies - The modified tow mentholator is running well. Calibration runs were
completed to determine tow residence times. Samples of tow were mentholated with residence
times varying from 15 to 130 seconds. The mentholation rate was determined to be 0.041 mg
menthol/g tow - sec residence time. Jay Pierotti is currently working on developing a method
for estimating the quantity of inentholl on the tow.surface versus menthol inside the fiber.
Continued research on application methods for the GMC'menthol release compound prompted
a number of studies in support of providing quantitative application levels and in support of
qualifying the latest large scale preparative batch of GMC from Lee laboratories.
The amount of GMC (Sugarman) on filler was determined in nine samples of tobacco after
spray application in the Semi-works batch cylinder. The average recovery of the GMC was
99% with respect to the target concentration, with a negligible amount found on the cylinder
walls after spraying.
Filler samples submitted by L. Haws were analyzed for ethanol to model equipment to be used
for ethanol removal after GMC spray application.
Methods were developed and samples analyzed for the determination of benzene, menthyl
chloride, and pyridine in Sugarman.
In addition, selective ion electrode fluoride determination were conducted on sugarman, no
fluoride was detected.
Operations Support - Pellets of sludge from reconstitution processes were analyzed to
determine the concentration of eight elements. These samples were also analyzed for residual
nicotine.
Two charcoal samples from Cabin Milds filters were found to be plant-based charcoal after
comparison with a coal-based charcoal and a coconut-based charcoal. The particle size of the
charcoal within the inner filter was about 0.2mm in diameter while the charcoal within the
middle filter space was about 1.8mm in diameter.
A foil-paper laminate was examined to determine the depth of penetration into the paper of the
Na silicate-based glue. The glue was located'close to the foil-paper interface, but it was not
uniformly thick. The glue was not detected on the outer side of the paper.
German certification continues for the PMI export materials not covered by the first round of
testing. A certificate of analysis was issued for each of the materials used in both the export
and domestic formulations.
-11-
Specifications were completed for a fourth vendor and a meeting with that vendor was held on
site discussing the contents of the specifications as regards their materials. The written
specifications and associated methodology was very favorably received and some additional
samples/methods/protocol exchanges are planned. Work continues on the FRG Direct
Materials and the Dry Flavors.
A method has been developed to monitor low levels of ammonia in environments were ETS is
present. The method offers ease of use and relative high sensitivity and precision. A special
report is in preparation. This method is currently being used to pressure the ammonia levels for
Project PACT.
Commerciali7ation of GMC and Polymic - Analysis of GMC batches 003R and 004 is in
progress. Using an alternate HPLC method, additional peaks were observed. These are a- and
(!-anomers of the known isomers of GMC as well as other minor isomers of GMC not
previously observed. All of this data has been reviewed and work is in progress to better define
the GMC anal'ysis by HPLC.
Callery Chemical has developed a promising synthesis of the PoIyMIC monomer. Full details
were received in early June. Their process uses potassium hexamethyldisilazide to form the
potassium enolate of acetone which is isolated as a solid and in turn is reacted with menthyl
chloroformate to form the desired monomer in good yield. Callery willl be carrying out some
additional research in June and issuing a final status report about the 1st of July.
A final review has been completed with Eastman Kodak on their work directed to finding an
improved synthesis of the PoIyMIC monomer. Kodak was unable to find any process which
they feel would be suitable. Effective mid-June they have stopped all work on this project.
Flavo Release TechnologX - Aromatek 150 (ethyl vanillin glucoside) was recrystallized from
aqueous methanol and aqueous ethanoL The solids obtained from these recrystallizations were
examined by Scanning Electron Microscopy. The samples were forwarded to Flavor
Development for evaluation.
3,4,6-Tri-O-acetyl-D-glucal was reacted with vanillin to give 4'-formyl-2'-methoxyphenyl 4,6-
di-O-acetyl-2,3-didehydro-2,3-dideoxy-D-erythro-hexoside. The purification and conversion
of this material to the unacetylated hexoside are in progress.
The reactions of 2-oxo-(3-ionol with di-t-butyl dicarbonate, methyl oxalyl chloride and oxalyl
chloride to obtain t-butyl-3-oxo-R-ionyl carbonate, methyl 3-oxo-p-ionyl oxalate and di-3-oxo-
(3-ionyl oxalate respectively have been carried out. All of these oxoionyl derivatives are
designed to: release megastigmatrienone on pyrolysis. Purification of these materials for
pyrolytic and subjective evaluation is currently in progress.
-12-
R.S.A. Company has been contracted to devel'op: a laboratory scale synthesis of vanillin
chloroformate which can be used to prepare a large variety of vanillin release compounds. The
research was begun on 6/13 and initial trials have been encouraging. An interim status report
will be sent on 6/25.
All analytical work has been completed for Vanil-3 (the large scale production of CR-2910 at
Aldrich) and the material has been released for internal evaluation.
Flavor Chemistry - Work continues on the recovery of organic components from the ART
process water column effluent. A total of 35g of organic neutrals, obtained from Stage I water
column effluent, was fractionated on i kilogram of silica gel to provide 8 fractions of
increasing polarity. Of these, 7 have been characterized by odor profiling and subjective
smoking as follows:
Fraction 2 (1.6g)
odor: fruity (non-citrus), honey, spicy, tea, green-herbal.
flavor tea-sweet, bright topnotes, good mouth feel.
Fraction 3 (7.3g)
odor woody, hay, fresh tobacco smoke.
flavor. sweet, flue-cured roasted, peppery, increased impact.
Fraction 4 (11.3g)
odor tea, sweet, hay, honey-floral.
flavor. full, rounded with sweet roasted note, peppery.
Fraction 5 (6.8g)
odor: tea/hay, honey, musty, pungent.
fliavor. full, blended sweet with salivation, good mouth feel.
Fraction 6 (3.3g)
odar: tea/hay, honey, sweet.
flavor. peppery tongue-bite, salivating, sour, lbts of mouth feel.
Fraction 7 (3.4g)
odor fruity (non-citrus), honey, sweet, hay, nutty.
flavor high amplitude, full-rounded, bitter.
Fraction 8 (0.9g)
odor. burnt-smoky, sweet, honey.
flavor peppery, musty, lots of mouth feel.
-13-
The subjective smoking information was obtained using nicotine-free cigarettes injectedd
with 3SOppm of each fraction.
Inorganics as Novel Tobacco lkiaterials Additives - Smoking studies have been completed on
a number of "mag carbonates", and the results are extremely encouraging with many giving
excellent sidestream reduction and a few with extremely high subjective ratings. One of the
best leads to date has been a magnesite, MgCO, , produced by the hydrothermal treatment of a
Reheis Mg(OH), paste in a CO, atmosphere. High-basis weight (63 g/m' ) handsheets of this
material have been prepared and are pending smoking d'ata. The preparation of this material
was reproduced and the sample submitted for papermaking. An additional sample containing
47% magnesite and 53% Mg(OH)2 was also prepared by the hydrothermal treatment of Reheis
Mg(OH)2 paste in a COt atmosphere and submitted for papermaking.
A high surface area "mag carbonate" generated from the pyrolysis of an ammonium magnesium
carbonate gave significant sidestream reduction with acceptable ash. Work with this system is
continuing.
A number of inorganic samples have been generated and either have been or will be submitted
for paperntaking. These include; 1) a sample from the large scale synthesis (11 liter reaction)
of the material from the magnesium methyl carbonate matrix study which gave the most
promising smoking results, 2) a freeze-dried sample of one of the samples from the same
magnesium methyl carbonate matrix study that was found unsuitable for handsheet preparation
due to excessive drainage times, 3) a 1:1 mixture of "mag carbonate" and calcium carbonate
prepared by reacting a suspension of calcium carbonate in magnesium methyl carbonate with
water and potassium hydroxide, 4) a 19:1 mixture of hydromagnesite and Mg(OM prepared by
reacting aqueous KOH with hydromagnesite, 5) a potassium magnesium carbonate generated
from the pyrolysis of a potassium ammonium magnesium carbonate, 6) two sol-gel CaCO,
samples, one with a surface area of 312 m= /g at 200 ° C, and 7) two "mag carbonate" samples,
one generated at N.Y. Polytechnic and the other in-house by an aqueous sol-gel route.
As a possible replacement for monopotassium phosphate, the hexasodium and hexapotassium
salts of phytic acid [myo-inositol-hexakis-(dihydrogen phosphate)] were prepared in solution.
The aim is to mimic the contribution of the soluble phosphate additives without consumption
by the formation of inert insoluble phosphate phases.
Cigarette Design - Last month we reported on a series of Kimberly-Clark cigarette wrappers
made from wood pulp. Cigarettes were fabricated with these wrappers and were found to have
equivalent puff counts to controls made with conventional flax wrappers. During this period,
we repeated this test with wood-pulp wrappers supplied by Wattens. Again we found no
significant differences in puff counts between the two wrappers. Additionally, we examined
the puff counts reported in the CI database for our competitors' brands made with wood pulp.
-14-
Cigarette design modeling showed no significant differences between their puff counts and
comparable cigarettes fabricated from flax wrappers.
Filter Research and Development - In addition to developing methods of acetylating paper
webs, we are also acetylating pulped wood fibers and making hand sheets from these. This has
been successfully accomplished on a laboratory scale. We are also seeking outside vendors
who can provide acetylated pulp and produce continuous sheets of web with it. Recent
discussions with USDA proved promising, and we are exploring the possibility of obtaining a
large-scale sample of acetylated pulp from them for producing a small bobbin of web material
for large-scale filter making and testing.
Nicotine Antibody: Several experiments were conducted to determine the cause of the varia-
tion observed in the ELISA to measure nicotine. One ELISA plate was shaken during each of
the incubation periods and one was not shaken. The plate that was shaken exhibited higher
absorbance readings than the stationary plate; however, the shaken plate had more variation in
the data.
/df
PC: Directors
Research Fellow
