Product Design
[Low Tar/High Flavor Literature Review]
Abstract
Consists of review of product design literature related to flavor and impact aspects of low-tar cigarettes. Includes chapters: "Summary; I. Introduction; II. Smoke chemistry and subjectives; III. Tobaccos and tobacco chemistry; IV. Filtration and filter ventilation; V. Nicotine, smoke pH and smoke subjectives; VI. Conclusions; VII. Indicates substantial evidence suggests that "increased nicotine, and optimal combinations fo nicotine, volatile organic acids and/or total remaining sugars probably will yield major flavor improvements of [ultra-low tar and lowest tar] cigarettes". Recommendations; IIX. Acknowledgements; [and] IX. References". Recommends future studies in "1. Manipulating smoke pH; 2. Higher nicotine tobacco blend with added organic acid(s) and/or increased total reducing sugars; [and] 3. Utilization of the supercritical extraction process". Indicates accession number "90-047".
Fields
- Author
- Hale, Robert W.Defense
- Kroustalis, ChristopherDefense
- Lin, Sherman (Scientist)1995 USA
- Raymond, W.
- Spielberg, Howard L. (PM Research Professional)Research Professional
- Recipient
- Grantham, P.
- Hypothesis
- Free Nicotine
- Behavior TargetingCigarette's effect of enhancing/mitigating specific behaviors
- CompensationIncorporating knowledge of compensation and effects of human smoking behavior into cigarette design.
- Low-yield cigarettesModification of low yield products to assure that adequate levels of nicotine delivery are maintained, and effects of yield changes on toxicity and dependence.
- Nicotine transport, transfer, and uptakeDesign changes which alter nicotine delivery or effect how the product causes and maintains dependence, including transfer of nicotine from tobacco to smoke, and uptake into the body.
- Use of filters, paper, and ventilationModification of tobacco products through use of filters, paper, and ventilation, and measuring effects on dependence, behavior, and toxicity.
- Use of tobacco processing/ blendsModification of tobacco products through changes in tobacco processing and use of blends, and measuring effects on dependence, behavior, and toxicity.
- Sensory targetingTargeting of smokers through changes in sensory characteristics
- Sensory effectsTechnologies used to measure, control, or alter sensory effects
- Keyword
- Mouth feel (Mouthfull)
- Total particulate matter (TPM or Tar)
- Maillard reaction
- Additive
- Citrate
- Citric acid
- Cocoa (Chocolate) (Cocoa Shells, Extract, Distillate and Powder)Composed of nearly 400 identified chemical substances as of 1967
- Dicarboxylic acid
- Gluconic acid
- Honey
- Levulinic acid
- Licorice (Licorice Fluid Extract, Powder, and Root)
- Malic acid
- Monocarboxylic acids
- Nicotine citrate
- Nicotine levulinate
- Pyruvic acid
- Tartaric acid
- Tricarboxylic acid
- Smoke Constituent
- Amines
- ammonia
- Carbon monoxide
- Carbonyls
- Citric acid
- Cyclopentenenes
- Dextrose
- Diterpenoid alcohols
- Diterpenoids
- Ether extractables
- Furanones
- Furans
- Hydrogen cyanide (HCN)
- Indole
- Isovaleric acid
- Ketones
- Nicotine
- Oxalic acid
- Pectin
- Phenols
- Pyrazines
- Pyridines
- Pyrroles
- Terpenoid ketones
- Terpenoids
- Volatiles
- Design Component
- Oriental tobacco (Turkish)
- Flue-cured tobacco
- Burley tobacco
- Bright tobacco (Flue-cured tobacco)
- Turkish Blend (TB)
- Greek
- Casing
- Flavorant
- Expanded tobacco (Puffed tobacco, ET)
- Filter tow
- Filter efficiency (FE)
- Filter draft (Filter RTD)
- Filter channels
- Burley casing
- RTD
- Prototype
- DBC
- Designed Blend Component
- Operation/Project
- Project 202
- Project 605
- Project 6mg no-vent
- Project 2mg low-vent
- Project Best of Lowest
- Nicotine Tar Interaction Study
- Named Organization
- Federal Trade Commission (Enforcement agency for laws against deceptive advertising)Enforces laws against false and deceptive advertising, including ads for tobacco products. Ensures proper display of health warnings in ads and on tobacco products;collects and reports to Congress information concerning cigarette and smokeless tobacco advertising, sales expenditures, and the tar, nicotine, and carbon monoxide content of cigarettes.
- Japan Tobacco Monopoly Corporation
- Philip Morris & Co. Ltd. (Cigarette manufacturer, incorporated in U.S. in 1902)Philip Morris & Co. Ltd.., was incorporated in New York in April of 1902; half the shares were held by the parent company in London, and the balance by its U.S. distributor and his American associate. Its overall sales in 1903, its first full year of U.S. operation, were a modest seven million cigarettes. Among the brand offered, besides Philip Morris, were Blues, Cambridge, Derby, and a ladies favorite name for the London street where the home companies factory was located - Marlborough.
- R.J. Reynolds Tobacco Co. (Cigarette manufacturer (Camel, Winston, Doral))Cigarette manufacturer (Camel, Winston, Doral)
- Technology/Method
- Richmond panel
- Supercritical Extraction Process
- SCE
- Brand
- HALF-NIC
- Kent III
- Lark
- Marlboro KS
- MARLBORO LIGHTS
- Merit (PM)
- MERIT ULTRA LIGHTS
- Next
- Winston KS
- WINSTON LIGHTS
- Subject
- additives
- aerosol (technology)
- Blends (Design)
- CNS/Brain (Effects)
- Effects—Smoking Behavior (Effects)
- Filters (Design)
- Levulinic Acid (Additives)reduces the harshness of cigarettes
- Low Yield Cigarettes (Products)
- Particle Size (Technology)
- pH Manipulation (Technology)
- Puff Count (Measures)
- Receptors (Effects)
- Sensory Effects—Impact (Effects)
- Sensory Effects—Taste (Effects)
- Smoke Constituents
- Smoke Delivery/Transport (Measures)
- Smoke Nicotine (Measures)
- Smoke pH (Measures)
- Smoothness/Harshness (Effects)
- T/N Ratios (Measures)
- Tar (Measures)
- Test/Consumer Preference (Testing)
- Test/Inhalation (Testing)
- Transfer to Smoke (Measures)
- Ventilation (Design)
Document Images
ACCESSION NUMBER: _ ?L)-n47
ISSUEI) TO:
ESEARCH CEN
, ..:.:~........_.
RICH1VI
I
PROJECT NUMBER: 2s 07
PROJECT ZZTLE: Basic Flavor Investigation
TYPE REPORT: (Q ANNUAL [Q SEMIANNUAL = COMPLETION
DATE: October, 1990 PERIOD COVERED:
REPORT TITLE: Low Tar/High Flavor Literature Review
WRITTEN BY:
SUPERVISED BY~
APPROVED BY:
0
SPECIAL
THIS REPORT IS CONFIDENTIAL TO THE BUSINESS OF THE COMPANY. IT
HAS BEEN ASSIGNED TO YOU. IT IS NOT TRANSFERABLE AND MUST NOT
BE PHOTOCOPIED.
IF THE REPORT HAS SERVED ITS PURPOSE AND IS NO LONGER NEEDED,
PLEASE RETURN IT TO THE,CENTRAL FILE AT THE RESI:ARCH CENTER
FOR RECORD KEEPING AND DESTRUCTION.
Smokers smoke cigarettes for the benefits of relaxation, concentration,
enjoyment and better management capacity. In the past decades, the low tar
cigarettes have gained substantial acceptance by the smokers, but not the
ultra-low tar nor the lowest tar categories.
The major problem for ultra-low tar and lowest tar cigarettes is their
failure in meeting smokers' expectation and needs. The cigarettes are
generally weaker in smoke impact and tobacco taste, and the smoke is thinner
and drier than the smoke with higher tar deliveries.
Among the thousands of compounds identified in cigarette smoke, the
single most abundant and important compound is nicotine. Nicotine contributes
to the smokers' physiological as well as organoleptical satisfaction. Nicotine
in tobacco is partially transferred directly to the smoke, and partially
decomposed to form amines and pyridines during smoking. Nicotine and its
decomposition products all contribute to the smoke's sensory properties,
particularly the smoke impact. Their sensory properties are influenced by the
smoke pH, which determines the ratio of free and protonated species of these
compounds. Higher smoke pH yields a harsher, hence a stronger impact. Lower
smoke pH yields a smoother, hence a weaker impact.
However, a stronger impact does not necessarily yield a more acceptable
cigarette. There seems to be an optimal impact for maximum acceptance at a
given tar level. It appears that some tar components have the balancing effect
on the sensory pruperties of nicotine and its decomposition products. The
sensory interaction between tar and nicotine seems to behave differently at
different smoke nicotine levels. when the smoke nicotine/tar ratio is at or
below half of the natural range, higher tar will yieldd stronger impact. But
when the ratio is at or above the natural range, higher tar will yield a
smoother and weaker impact.
The volatile organic acids in the smoke seem to modify the smoke pH, and
he:nce the sensory properties. The acids can be derived either from a direct
transfer from tobacco, or from precursors present naturally in tobacco or in
added flavorants.
The Maillard reaction during curing, primary processing and smoking all
seem to contribute to the smoke flavor. There seems to be a close relationship
between the ratio of total reducing sugars/total alkaloids and the smoke's
impact strength as well as acceptability.
Raising the smoke impact by increasing smoke nicotine, and
simultaneously, smoothing the harshness by adding a suitable organic acid
and/or by adjusting the ratio of total reducing sugars/total alkaloids to an
optimal level, may be able to improve the quality of ultra-low and lowest tar
cigarettes substantially.
TABLE OF CONTENTS
Pace No.
I. SL,mmary ---------------------------------------------------
Introduction ---------------------------------------------- i
1
II. Smoke Chemistry and Subjectives --------------------------- 4
1. Smoke Aerosol ------------------------------------------ 4
2. Smoke Components --------------------------------------- 5
III. Tobaccos and Tobacco Chemistry ---------------------------- 7
IV. Filtration and Filter Ventilation ------------------------- 10
1. Filters ------------------------------------------------- 10
2. Filter Ventilation ------------------------------------- 10
3. Ventilation Effect on Smoke Deliveries and Subjectives - 11
V. Nicotine, Smoke pH and Smoke Subjectives ------------------- 16
1. U.S. Patent - Nicotine and Levulinic Acid --------------- 16
2. PM Studies on Low Delivery Cigarettes ------------------- 19
VI. Conclusions ------------------------------------------------ 35
VII. Recommendations ------------------------------------------- 37
IIX. Acknowledgments ------------------------------------------- 39
IX. References ------------------------------------------------ 40
I. Introduction
Cigarettes are customarily classified into or perceived as four different
categories based on their FTC tar levels: Full-flavor (16mg or higher); Low tar
(15mg or less); Ultra-low tar (6mg or less) and Lowest tar (2mg or less) [4]. In
the last 30 years, the U.S. cigarette market has undergone some dramatic changes
with regard to cigarettes' FTC tar delivery. The industry sales weighted average
tar decreased from about 35mg in the 1950's to about 14mg in 1980 [1]. Since 1983,
it has stabilized around 12-13mg [2).
The mechanism of this tar lowering trend is two-fold:, Lowering of the tar
level of the existing brands and the introduction of new lower tar brands. This
trend of lowering cigarette delivery is best demonstrated by Fig. 1, by comparing
the numbers of brands vs tar delivery levels from the CI reports of August 1973
and November 1989 [3, 4]. Three interesting points are observed: 1) The tar level
for full-flavor category in 1973 was greater than 18mg, but in 1989, the majority
of the full-flavor category has shifted to 14-18mg range, most commonly around
15mg; 2) In 1973, the number of brands progressively decreased as the tar
decreased, but in 1989, the distribution curve is very much like a standard
Gaussian curve, with the highest point at 8-14mg range; 3) The total number of
brands has doubled in the last 16 years, from 130 in 1973, to 262 in 1989.
Another important trend of the U.S. cigarette market is the continuous
dec:.ine' of total sales in recent years. The total industry sales had enjoyed a
steady growth from 1940 to 1980, with the exceptions of 1955 and 1965 [5].
However, since 1980, the total industry sales have declined steadily at the rate
of about 2 percent annually [2].
These cigarette market trends can be attributed to the negative social and
political mood about smoking. This market environment has become a strong driving
force for the cigarette industry to continue lowering the cigarette tar delivery.
However, the growth rate of ultra-low tar cigarettes in the past 18 years is much
slower than that of the low tar category [1]. This may suggest that the current
ultra-low tar brands fail to gain popular acceptance from the smokers.
Facing today's political and social anti-smoking environment and the sales
potential for acceptable ultra-low tar and lowest tar cigarettes, it is of great
interest to Philip Morris to develop the capability of making cigarettes with
lower tars than current low tar cigarettes, and yet still with a flavor and taste
which satisfy smokers' needs and palates.
There are several active developmental projects, i.e., Project 202, 605, 6mg
no-vent, 2mg low-vent and Best of Lowest, with the objectives of achieving the
goals mentioned above. In search of new or additional insights and/or ideas about
the subject, a literature review was initiated. This review was conducted by
focusing specifically on the following questions: 1) What are the major flavor
deficiencies of lower tar cigarettes? 2) What are the major causes of these flavor
problems? 3) How is each of the current technologies for achieving lower tar
delivery, i.e., tobacco blending, cigarette constructionr filtration and filter
c'ig. 1 Brand distribution . ar DeliveriPs
105=
100 i
95-
90i
85i
N 80I
o
$ 75,
O
f
10
FTC Tar mg/cig.
~
~
1973
®
1989
ventilation related to the flavor problems of lower delivei:y cigarettes? and
Can we derive some new concepts from this review for the £lavor improvement
these lower tar cigarettes?
4)
of
To facilitate the understanding of the basic nature of this flavor problem,
cigarette smoke chemistry and subjective attributes, tobacco chemistry, filtration
and filter ventilation were reviewed. Several books, review articles and seminar
printouts were used for this general review. Internal technical repnrts, external
publications and patents related to the subject were examined. Those judged to
contain relevant and valuable information were reviewed in greater detail. The
pertinent data were re-examined from different perspectives in search of new
meanings and insights.
Extensive interviews with many technical staff members of various
disciplines within the Philip Morris R&D organization were conducted. Discussions
with certain individuals were constantly held during the course of this review and
the writing of this report. Many statements, ideas and concepts presented in this
report represent the general conviction of those staff members interviewed and my
own personal interpretation.
3
II. Smoke Chemistry and Subjectives
1. Smoke Aerosol
Cigarettes are unique among all the products consumed by hu:nans through the
mouth. All the other products, foods, drinks and medicines, are consumed in solid
and/or liquid forms. But cigarettes are consumed by first burning the cigarette
rod and then inhaling the smoke generated from the burning of tobacco.
Cigarette smoke is an aerosol, a state of colloidal particles dispersed in a
gas medium. The smoke aerosol consists of two parts: one which passes through a
glass-fiber filter, and another one which is retained on a glass-fiber filter. The
one which passes through glass-fiber filter can be further divided into gas phase
containing gaseous components at room temperatures, and vapor phase containing low-
boiling compounds. The particulate phase is further divided into semivolatiles
containing medium and hig:i boiling compounds, and nonvolatiles.
According to the calculation of Okada et al. [6], thel$verage particle size
in mainstream smoke is 0.18 u, with af9ncentraon of 3x10 particles/cc, which
is equivalent to approximately 35x3x10 U~ 1x10 particles per normal puff.
Keith and Tesh [7] reported that the cigarette smoke was composed of about
88% by weight of gaseous components of nitrogen,' oxygen, carbon monoxide and
carbon dioxide, 8% total particulates and 3% vapor phase.
Since cigarette smoke is an aerosol, the aerosol properties besides its
chemical composition may have an effect on the sensory perception of the smoke.
The size and size distribution of the particulates, the number of particulates per
puff and the surface activities of the particulates may affect the "presentation"
of the smoke to the mouth.
They may affect smoke sensation in two way3: One is the physical feel, the
feeling of fullness or thinness in the mouth, and another is the transfer
efficiency of smoke components from aerosol to the sensory receptors. The effect
of aerosol properties on smoke flavor is very much like the effect of food texture
on food flavor. Texture is not normally considered as a flavor attribute, but it
has an important impact on the perceived flavor quality of foods. Unfortunately,
very little is known about the correlation between aerosol properties and smoke
flavor quality.
According to discussions with Hale and Keritsis [8], they have observed in
the past that the smoke aerosol "presentation" can be altered by modifa.cation of
the filter or filter construction. Since a substantial dectree of filter
ventilation is normally applied to achieve ultra-low ter, and`_lo.west tar
deliveries, a better understanding on this subject could be very' heapfu; ..in the
total consideration of improving the flavor of lower tar cigarettes.
2. Smoke Components
Cigarette smoke components are derived from the incomplete combustion of
tobacco during puffing. The smoke components are derived from three different
mechanisms: 1) Direct thermal distillation of the volatile and semi-volatile
components of tobacco and added flavorants; 2) Thermal decomposition (pyrolysis)
products of tobacco components and added flavorants, and 3) Pyrosynthetic products
- pyrolytic products recombined to form new molecular species [9),.
Ishiguro and Sugawara 1103 of The Japan Tobacco Monopoly Corporation wrote a
comprehensive review on this subject in 1979. The translated version of this
review is available from the PM R&D ?.ibrary. The total number of smoke components
identified by 1979 was well over 2,500 [10). Although many of them could be
artifacts formed during smoke isolation and analysis processes, the total number
of smoke components is still very large. Ishiguro and Sugawara [107 listed a total
of 1,889 smoke components with the nomenclatures and structures in their review.
Those listed are:
InorcJanic gases --------------------- 11
Nitrogen-containing compounds ------ 573
Sulfur-contair.ing compounds --------- 46
.vdrocarbons ----------------------- 432
Or.ygenated compounds.--------------- 732
Halogenated compoundc --------------- 17
inorganic elements ------------------- 76
Others ------------------------------- 2
The smoke components of sensory interest and their potential sources in
tcbacco are summarized as follows:
GompOnents of Sensory_Interest Potential Sources in ^'obacco
1. Ammonia, non-aromatic amines Nicotine, free amino acids
2. Pyrroles Sugars
3. Ma;llard reaction products Sugars, ammonia, amino acids
4. Pyridines Nicotine, amino acids, proteins
5. Pyrazines Maillard reaction
6. Nicotine Direct transfer from tobacco
7. Polycyclic N-compounds Amino acids, proteins
8. A.mino acids Direct transfer from tobacco
9. Sulfur compounds S-amino acids, S-proteins
10. Terpenoids Diterpenoids, diterpenoid alcohols
11. Furans Sugars
12. Carbonyls;
Terpenoid ketones Eosential oils - direct transfer
Furanones, cyclopentenones Sugars
13. Volatile organic acids Direct, Sugars
Isovaleric acid Direct (Oriental tobacco)
14. Phenols Same as mentioned above
15. Brown pigments Same as mentioned above
5
=n= more detailed information, please refer to the review article by
Ishiguro and Sugaware. (10].
The asse~sment of the contribution to the cigarette smoke flavor from
different compounds or different classes of compounds is very difficult if not
impossible. Because of its comp.lexity, most statements about flavor contribution
of smoke components in the literature are speculative in nature. However, there
seems to be sufficient evidence to indicate that smoke pH and smoke nicotine
content aze closely related to smoke impact strength and smoke quality. The effect
of smoke pH and nicotine are reviewed in greater detail in section V.
:~ iS~
III. Tobaccos and Tobacco Chemistry
;-re three major types of tobaccos used in typica.L American blended
cigarettes. They are flue-cured (Bright), Burley and oriental (Turkish or Grec:k).
These tobaccc,s contain similar components but in different concentrztion ratios.
Therefore, their smokes possess different sensory characteristics.
Blending of these tobaccos in certain ratios will provide a richer, more
balanced and better tasting cigarette than a cigarette of using any single type of
tobacco. tio4:ever, blending of various tobaccos alone may not bc' able to provide
the ultimate richness of flavor. Therefore, various flavorants are also added to
the tobaccos before they are made into cigarettes.
Flavorants, i.e., volatile aromatic materials and casings rich in sugars,
are applied to tobacco during primary processing. The casings, which may contain
sugars, licorice extract, honey and cocoa, are designed to compensate or makeup
the desirable components deficient in certain tobaccos, i.e., sugar deficiency in
Burley. The volatile components and other non-sugar, non-volatile components in
the casings will also contribute to the total flavor of the smoke. The volatile
aromatic flavors, i.e., the aftercut flavoru, are designed mainly to balance the
aroma of the finished blend, the cigarette rod and its smoke.
The main objective of adding flavors to tobacco is to supplement or enrich
the tobacco characteristics of cigarette smoke: If the added flavors yield
predominating non-tobacco i,otes and become distinctive, they av~ not desirable for
American cigarettes. The only ex::eption is mentholated cigarettes.
As disctissed in section II, the tobacco components which have major effects
on cigarette smoke qualities are ni::otine, total volatile bases, amino acids and
proteinu, organic acids, total reducing sugars, pectins & cell wall subscances,
essential (vc,lati'-e) oils, and alc.ohol s,'.uble resins. Each group of substances
contributes differently to the cigarette smoke subjective attributes. Therefore,
each type of tobacco has its characteristic aensory properties. Those sensory
differences can be associated with the differences in their chemical compositions.
There are other major factors influencing tobacco chemical composition and
sensory properties. They are the lAaf's stalk position, the soil and climate
conditions of production location, the type of fertilizers applied and the
conditions of curing after the tobacco leaves are harvested. Therefore, there are
many difi'eren;. grades of tobacco within the same type.
In order to have a more uniform tobacco for assuring cigarette quality
consistency, saving cost and smoothing production and inventory control, PM is
committed to using one standard DBC (Designed Blend Component) for each type of
tobacco used in making most of the PM cigarettes. For instance, a DBC flu-cured is
a blent of flue-cured tobacco from the whole stalk of the flue-cured tobacco
plant.
7
