CA1317961C - Process for removal of basic materials from plant products using solvent extraction - Google Patents

Abstract

PROCESS FOR REMOVAL OF BASIC MATERIALS

Abstract of the Disclosure A process is provided for the selective removal of basic materials from plant products, in particular, for removing nicotine from tobacco with out materially affecting the content of the other components of the tobacco. Tobacco is traversed with a solvent in the supercritical state or in the liquid state wherein nicotine and the other components dissolve in the solvent. The solvent is then passed through an acid-containing trap where the solvent is essentially freed of nicotine. The solvent, depleted of nicotine and enriched in the other components, is then recycled to the tobacco to reextract nicotine.
In addition, the tobacco may be pretreated with a chemical base which does not substantially react with the tobacco components under ambient conditions but rather is activated under the conditions of the extraction process. Pretreatment of the tobacco with a chemical base neutralizes nicotine salts and permits the extraction process to be carried out under milder conditions.

Description

~ 3 ~ 7 ~ ~ ~ PM-1267 CIP

This invention relates to extraction pro-cedures and is particularly directed to a process for the selective removal of basic materials from plant products. The in~ention may be used for the selective extraction of particular substances from a great variety of plant products. However, it is particularly applicable and will therefore be described in connection with the extraction of nico-tine from tobacco.

Various processes have been proposed for the removal of nicotine from tobacco. Most of these processes, however, are not sufficiently selective for nicotine and remove other ingredients from the tobacco. Removal of these other ingredients adversely affects the desirable flavor and aroma properties of the tobacco. Other processes generally have been found to be limited in their scope and efXectiveness and are often complex and expensive to caxry out.
Nicotine has been extracted from tobacco using organic solvents with ~nd without neutralization of the nicotine salt. Representative of such pro-;

?- ~ 3 1 1 ~
cesses are those disclosed in U.S. Patent Nos. 678,3S2 ~Froehling), 1,196,184 (Villiers-Stuart), 2,048,624 (Roselius), 2,128,043 ~Garner), 2,227,863 (~hodes) and 3,096,773 (Prilly et al.). Nicotine has al~o been extracted from tobacco with aque~us solutions with and without neutralization of the nicotine salt.
Representative of such processes are those disclosed in U.S. Patent Mos. 2,~22,306 (Beuel et al.), 2,582,075 (Severi) and 3,874,392 (De Brunn et al.).
Nicotine h~s been separated from solutions of tobacc~
extracts and the nicotine depleted solvent xeturned to the tobacco. Representative of such processes are those disclosed in U.S. Patent Nos. 283,622 (Liebreich et al.), 802,437 5Wimmer~, 2,80S,667 (von Bethmann), 3,046,997 (Hind), and 3,139,435 (Staley et al.). Nicotine has also been separated from solu-tions of tobacco extracts and the depleted solvent saturated with the other components recycled to the extraction vessel. Representative of such processes are those disclosed in U.S. Patent Nos. 1,294,310 (Sayre et al.), 1,577,768 (Smith), 1,813,333 (Andrews), 3,396,735 (von Bethmann) and 3,612,066 (Jones).
Nicotine has been transferred from tobacco to a substrate without the use of a solvent by con-tacting the tobacco intimately with a substrate which has been treated with an acid as disclosed in United States Patent No. 4,215,706 (Larson). This process maybe carried out with or without heat.
United States Patent No. 4,153,063 (Roselius~
discloses a process for removin~ nicotine ~xom tobacco in which tobacco is contacted with an extraction solvent in a supercritical state. In a single step extraction process, moist tobacco is extracted with a solvent in a supercritical state. Thereafter thesolvent is evaporated. Because aroma components are also removed along with nicotine in this single step : :.

' l3- 13~7~
extraction process, a mu'lti-~tep process has been suggested. In the first ~tep, dry tobacco is extracted with a solvent in the supercxitical state to remoYe the aroma components. In the second step, the tobacco is moistened and again extracted with a solvent in the supercritical state to remove nicotine which is separated by either evaporating the solvent, contacting the solvent in a separate vessel with an acid, or adsorbing the nicotine on an active carbon column. In the third step, the stored aroma com-ponents from the first step are redissolved in a supercritical solvent and returned to the tobacco.
This multi-step extraction process is expensive and time consuming. In addition, the prolonged handling of the aroma components may adversely affect their properties.

This invention provides a process for removing nicotine from tobacco without also removing the desirable aroma generating components. Tobacco is extracted with a solvent either in the super-critical s~ate or in the liquid state. Thereafter, nicotine is selectively removed from the enriched solvent by passing the solvent through a trap con-taining a non-volatile acid which is not soluble in the extraction solvent. The trap may be contained on a support medium. The solvent, depleted of nicotine and enriched in the other components, is then recycled to the tobacco to extract nicotine again.
Alternatively, the tobacco may be pretreated with a chemically basic (alkaline) compound which does not significantly react with the tobacco components under ambient conditions but rather is effective in increasing the amount of extractable nicotine under the conditions of the extraction process. Pretreat-. " :
''' ~

1 3 ~

ment of the tobacco wi~h a basic compound neutrali~esnicotine salts and permits the extraction process to be carried out under milder conditions.
It is an object of this invention to provide a process for selectively xeducing the level of nico-tine in tobacco using a single stage extraction process with or without separate entrapment vessels.
It is another object of this invention to provide a process for the migration of nicotine from one tobacco substrate (leaf material or reconstituted leaf) to a second tobacco substrate (leaf material, reconstituted le~f matexial or tobacco stems) or to a non-tobacco substrate.
It is another object of this invention to provide a process for the migration of flavor and aroma components (with or without nicotine) from one tobacco substrate (leaf material or reconstituted leaf) to a second tobacco substrate (leaf material, reconstituted leaf material or tobacco stems~ or to a non-tobacco substrate.
It is another object of this invention to provide a process for the attenuation or removal of flavor or aroma substances.
It is a further object of this invention to provide a process using adsorption media (full flavor tobacco filler, reconstituted leaf materials, tobacco stems, cotton cloth, cellulose, carbon, cocoa shells, other plant by-products, porous ceramic, porous metal, etc.) to ~acilitate the selective removal of nicotine.
It is a further object of this invention to provide a process using aqueous absorption media (water, agueous ac~d, aqueous salt, etc.) to ~acili-tate the selective removal of nicotine.
It is a further object of this invention to provide a process for the extraction of nicotine from tobacco under relatively mild conditions.

.
- - , :

~ 3 ~
..~
The6e and other objects and advantages of the invention may be seen in the following descrip-tion.
Brief Description of the Drawing FIG. 1 illustrates the apparatus for removing basic materials.
Detailed Description of thP Invention FIG. 1 illustrates one preferred apparatus for removing nicotine from tobacco and tobacco blends.
Extraction vessel 1 is filled with a sample of tobacco and entrapment vessels 2 and 3 are filled with an acid impregnated support medium or water, preferably containing some acid. An extraction solvent is supplied to vessel 1 which is connected to a pump 4 and a mass flow meter 5. The pressure in the vessel is controlled by means of a fill pump ~not shown) and the temperature is controlled by means of heat exchanger 6. The extraction solven~ enters the top of the extraction vessel, passes downwardly through the sample and leaves at the bottom of the vessel~
In passing through the sample, ~he extraction solvent becomes enriched with components of the sample. The solvent is then circulated through either trap 2 or trap 3, or both, again being introduced from the top, and then passing downwardly and exiting at the bottom. The entrapment time may be split between traps 2 and 3. In passing through the trap or traps, the nicotine in the solvent reacts wi~h the acid and becomes trapped ~herein while the remaining aroma components in the solvent pass throu~h the trap or traps fr~ely. The solvent, depleted of nicotine and enriched in the ot~er components, is then returned into the cycle by recirculating it to the extraction vessel.

-6- ~ 3 ~7 3 ~
The extraction vessel is preferably designed for radial flow or for axial flow of solvent. The entrapment vessels axe preferrably both designed for radial flow or axial flow but need not be of the same design as the extraction vessel. A radial flow of solvent will minimize compaction of solid material in a vessel and may allow for lower pressure drops within each vessel. Persons skilled in the art will recognize that many directions of flow will be effec-tive, e.g., flow from bottom to top, top to bottomor inward or outward radially in each vessel. Persons skilled in the art will also recognize that the pump can be placed on any of several lines in the system.
An especially pre~erred procedure makes it possible to contain both the tobacco sample and a relatively small volume of entrapment material in the same extraction vessel. In this procedure, the entrapment material is placed in the bottom portion of the extraction vessel r a porous plate is placed on top of the trap and the tobacco is loaded into the extraction vessel on top of and supported by the porous plate. The entrapment vessels 2 and 3 are removed from the flow system by valve adjustment.
The extraction of the tobacco sample is then carried out as befsre by introducing the solvent into the top of the extraction vessel and passing it downwardly through the sample until it exits at the bottom of the vessel.
One advantage of this process is that no additional vessel is necessary to contain a relatively large guantity of adsorption material to trap the nicotine. The ability to carry out the extraction in a single vessel,results in a more economical process for the above reasons and also becaus~ the solvent to tobacco ratio can be significantly lowered.
The ability to use less solvent also results in less `` ~3~7~

degradation and loss of the aroma producing components and consequently in an improved tobacco product.
In another embodiment of this invention, the tobacco may be pretreated with a chemic~lly basic compound which does not significantly react with the tobacco components under ambient conditions but rather is activated under the conditions of the extraction process~ Nicotine is thus liberated from its salts and immediately taken up in the extraction solvent before other base induced chemistry is initiated within the tobacco. Since the solubility of the nicotine free base is generally higher than that of the nicotine salts in the extraction solvents of choice, the extraction process may be carried out under milder conditions. Pretreatment can include spraying or soaking the tobacco with the basic com-pound or a suitable solution thereof.
Persons skilled in the art will recognize that the said chemically ~asic compounds include those which are not necessarily affected ~y the process conditions but are nonetheless effective in increasing the amount of nicotine which can be extracted by the process of this invention.
A number of extraction solvents having solvent capacity for nicotine in both their liquid and gaseous phases can be employed to reduce the nicotine content of tobacco. Carbon dioxide in the supercritical state is the preferred solvent in this invention. Other useful solvents include, for example, halogenated hydrocarbons including up to about 4 carbon atoms such as CF4, CHF3CClF3, CBrF3, CF2=CH2, CF3-CF2CF3, CHClF2, CC12F~, CHC12F, CC13F, C~rF3, CFCl=CF2, CH3-CF3, octafluorocyclobutane and hydrocarbons including up to about 5 carbon atoms ` 35 such as propane, butane, pentane; o~h~r useful solvents include N20, SF6 and argon. Mixtures of solvents or additives or co-solvents may be used to ` ~3.l7~
-8,~
obtain improved 601vent characteristics. In addition, water, ammonia, or aqueous ammonia can be mixed with the extraction solvent to obtain improved solvent characteristics. Carbon dioxide is the preferred solvent because it is a naturally occurring compound and leaves no non-tobacco residue in the extracted tobacco.
A solvent in the supercritical state is a solvent in the gas phase at a sufficiently high lQ temperature so th t it cànnot be liquified by an increase in pressure. A solvent in the subcritical state is a solvent which can be liquified by an increase in pressure.
Supercritical carbon dioxide is carbon dioxide which is above its critical temperature, i.e., above 31.3~C, and above its critical pressure, i.e., above about 70 atmospheres. Extraction with carbon dioxide in the supercritical state is carried ~ut at a pressure in the range of from about 70 to about 1500 atmospheres and at a temperature in the range of from just above the critical temperature to about 1~0~C. The range of temperature and pressure for the supercritical state of other useful solvents are of generally the same order of magni~ude.
The preferred acids for use in this inven-tion are acids which are non-volatile and non-soluble in the extraction solvent under the conditions of the extraction. Useful acids are sulfuric, phosphoric and nitric. Other useful acids include polycarboxylic acids such as tartaric, citric, malic, malonic, succinic and glutamic. Monovalent salts, such as the alkali metal salts, of the above acids are generally preferred because these salts are less volatile and less soluble in the solven~. ~ preferred salt of an acid is monopotassium citric acid. Mono-ammonium and diammonium salts of the above acids may also be used. Polyvalent salts of the above acids ~ l 7 ~ ~ ~
are also useful but are less efficient in trapping nicotine.
The acid in the trap i~ preferably, though not necessarily, in contact with a support medium, which does not impede the flow of ~he solvent. The acid may be impregnated on, deposited on, or otherwise in contact with the support medium. Useful support media are carbon, tobacco filler, reconstituted leaf materials, tobacco stems, cotton cloth, cellulose, cocoa shells, other plant by-products, porous ceramic, porous metal and the like. The tobacco stems may be long stems, cut and rolled, shredded, expanded, treated or untreated. Especially preferred support media are shredded ~obacco stems and cocoa shells.
The support medium for the acid may even be water, as in the case where the solvent is bubbled through an aqueous solution of the acid. A preferred trap material is an aqueous solution of citric acid.
An especially preferred trap material is an aqueous solution of monopotassium citrate.
The ratio of acid to nicotine may range from about 10:1 to about 1:1 and preferably from about 4.5:1 to 1.5:1.
The concentration of the acid in the support medium is not critical. ~he concentration should be hi~h enough to minimize the volume o support medium required within the vessel but low enough not to impede the flow charac~eristics of the solvent through the support medium. The concentration of acid in the support medium may vary but in the case of tobacco stems the concentration generally ranges from a~out 5% by weight to about 40% by weight and preferably is about 15% by weight. Highex acid concentra~ions, i.e., sa~urated or crystalline forms are not excluded.
The chemical bases for use in this inven-tion for pretreating tobacco are bases which do not significan~ly react with the tobacco components under ~ o ~ 3 ~
ambient conditions. Chemical bases which do ~ignifi-cantly react with the tobacco components under ambient conditions may initiate base catalyzed chemistry in the tobacco, blackening the tobacco and otherwise adversely affecting its smoking characteristics.
Rather, the preferred base is one which i5 effectiYe in increasing the amount of extracted nicotine wi~hout excessively increasing the extraction of flavor con-stituents. Such a base may react with the components in the tobacco under the conditions of the extraction process. In this way, nicotine is liberated from its salts and immediately taken up in the extraction solvent before other base induced chemistry is initi-ated within the tobacco. Since the solubility of the nicotine free base is higher than that of the nicotine salts in the extraction solvent, one may carry out the extraction process under milder condi-tions. In this way the ~uality of the subjective smoking characteristics is preserved. Chemical bases ~hich do not significantly react under ambient condi-tions but are activated under the conditions of the extraction process include ammonium bicarbonate, sodium or potassium carbonate or bicarbonate, glyco-sylamines, N-glycosides of aldoses, N-glycosides of ketoses and the like.
Other chemical bases are useful in practic-ing this invention, including ammonia, aqueous ammonia, trimethyl amine and triethyl amine, which can be effective in increasing the amount of extract-able nicotine under the extraction conditions,althou~h some such compounds in some concentrations may adversely affect the subjec~ive characteristics of the tobacco. In general, bases which liberate nicotine from its acid salts are effective in practic-ing this invention, especially bases with a pK~gr~ater than about 7.2 and less than about 10. Com--' ~3~73~L

binati~ns of ~uitable ba~es are ~lso within the scope of the invention.
A glycosylamine useful in the present invention is illustrated by the following formula:
H
c NH
I HO---C H
o HO---C H
I HO---C- H
C - H
R
where ~ is a hydrogen, methyl or methylol substituent.
Illustrative of the invention glycosylclmine compounds is l-amino-l-deoxyglucose (R=CH2OH). Other glycosylamine compounds include l-amino-l-deoxymannose (mannosylamine), l-amino l-deoxyribose (ribosylamine), l-amino-l-deoxygalactose (galactosylamîne), 1-amino-1-deoxyrhamnose (rhamnosylamine), 1-amino-1-deoxyfucose ~5 (fucosylamine), l-amino-l-deoxyxylose (xylosylamine), 1 amino-l-deoxyar~binose (arabinosylamine), 1-amino-1-deoxylyxose (lyxosylamine), and the like.
The glycosylamine compounds included in this invention are, in pure form, stable and odorless compounds at ambient temperatures. In addition, the glycosylamine compounds decompose at a relatively low pyrolysis temperature (e.g., 200~-300~C.) to release amm~nia, pyrazine and related compounds.
Ammonia-derived glycosylamines with an unsubstituted amino-group ~-NH2) are more stable than glycosy~amines in which the amino group is sub-stituted (-NHR or ~R2~. If the amino group is an amino acid structure, then there is a self-catalyzed Amadori Rearrangement, i.e., a conversion of N glyco~
side of aldose sugar to an amine derivative of the :
,~- ', ' - ~ :

:--~2- ~3~73~
corre~ponding ketose, in addition to ~ther side-reac-tions which occur at room temperature.
Another preferred form of the invention includes pretreating tobacco by ~praying it with an agueous solution of ammonium bicarbonate. Ammonium bicarbonate îs useful if applied at about 1% of the dry weight of tobacco (mass/mass) and can be app~ied at up to 3% of the dry weight of tobacco. The resulting tobacco, after extraction, is generally less harsh than non-pretreated tobacco.
Neutralization of nicotine salts may also be carried out by contemporaneously adding a source of base during the supercritical extraction such that neutralization and extraction of nicotine occur simul-taneously and no significant base-induced chemistry within the tobacco is initiated.
In a typical example of the process of this invention, an acid-containing trap is prepared by impregnating an aqueous solution of an acid such as monopotassium citric acid into a support such as expanded tobacco stems. The entrapment material is then placed in an extraction vessel, a porous plate placed on top of the acid-containlng trap material and the tobacco to be extracted is loaded on top of the plate. The vessel is charged with solvent, the pressure is brought to 260 atmospheres and the temperatur~ is brought to 70 C.
The tobacco to be extracted is adjusted to contain an oven volatile conten~ of about 25%. The percentage of oven volatiles (%OV) in the tobacco is a measure of the moisture content plus a minor frac-tion of other components and is determined as follows:
%OV = weiqht losg of sample after 3 hrs at 100C
sample welght The tobacco is then traversed with an extraction solvent such as carbon dioxide in the ~3-supercritical state, and nicotine and other component~
are dissolved in the extraction solvent. The enriched supercritical solvent is then passed ~hrough the acid containing trap wherein the solvent is freed of nicotine. The supercritical solvent, depleted of nicotine and enriched in other components, is then recycled to the tohacco. The aroma-generating com-ponents are extracted from the tobacco only during the initial stages of the cycle since the solvent rapidly becomes saturated with these components.
Because nicotine is continuously removed from the solvent, upon recycling the solvent is able to extract additional amounts of nicotine from the tobacco.
The process is carried out until the desired level of nicotine reduction in the tobacco is achieved.
Usually 30 to 60 minutes is sufficient. An advantage of this process is that principally the nicotine is removed from the supercritical solvent and the aroma- !
generating components are substantially preserved.
Valve and instrumentation hardware may be arranged to allow (a) by-pass of the Co2 flow to all vessels, (b) CO2 flow from an extraction vessel to any entrapment vessel, (c) CO2 flow both upflow an~
downflow in any vessel, (d) faster turn-around time, (e) use of both large and small C02 fill pumps and (f) pressure drop instrumentation for up and down flow differential pressure measurement in the entrap-ment vessels. The flow of C02 is in the opposite direction from the force of gravity in the upflow mode and in the same direction in the downflow mode.
For bett~r xetention of tobacco subjective smoking characteristics, the extraction vessel may ~e by-passed during ~he C02 fill and heat period, and the C02 flow directed only through the entrapment vessels. Once extraction process conditions are reached, the CO~ flow is then directed through the extraction vessel or vessels. Extraction process 7 ~ ~ ~
conditions are reached quickly (4-8 minutes). In an experiment using this process, the beginning of each extraction run was counted when process conditions were reached (temperature, pressure) and the end of each run was counted when the re~uixed amount of C2 mass (m/m, mass of extraction solvent/mass of tobacco) had traversed the tobacco.
Table 1 illustrates the results of extrac-tion runs carried out using carbon as an adsorbent support.
Full flavor American blend tobacco at 25 OV (oven volatiles) was extracted over a period of 30 minutes using activated carbon as an adsorbent (Run-16, Table 1). The nicotine content of the tobacco was reduced 97.2%. Tobacco flavor was im-proved, as judged by subjective tests, in comparison to tobacco subjected to longer extraction periods.
When activated carbon saturated with potas-sium citrate was used as the adsorbent support, the carbon to tobacco ratio was significantly reduced, from 2:1 for carbon alone, to 0.25:1. The level of nicotine extracted was slightly lower due to break-through of nicotine through the entrapment columns.
The tobacco subjectives (aroma, flavor and other smoking characteristics) were very poor ~Runs-17,-18, Table 1).
In an attempt to extract subjectives only, tobacco, without being premoistened, was extracted with supercritical C02 under control conditions (260 atmospheres, 70C, 12% OV, 30 min, 150 m/m). Contrary to the published patent literature (United States Patent No. 4,153,063, Roselius~ 94.4% of the nicotine was removed from the tobacco (Run-21, Table 1).
Table ~ lllustrates the results from extrac-tion runs carried out using potassium citrate txeated stems as the adsorbent support.

~7~
-~5-The best subjectives overall in ~he experi-ments summarized in Table ~ were obtained when shredded stems were used as the adsorbent support a~ a stem to tobacco ratio of about 1:1 by weight and a potas~
sium citrate to nicotine molar ratio of 8:1 (Run-41, Table 2). The entrapment material was divided equally between two entrapment columns. Subjective quality approached that of the unextracted contxol and 93.7%
of the nicotine was removed from the tobacco. A one hour extraction period was used at a lower CO2 flow rate in order to minimize compaction of the stems in the entrapment columns.
Table 2 also shows the following results:
Use of dual entrapment columns gives higher levels of nicotine removal than use of a single entrapment column containing the same amount of entrapment material.
An entrapment time split of 15 and 45 minutes or 20 and 40 minutes, in entrapment vessels 2 and 3, respectively, is preferred over a time split of 30 and 30 minutes ~based on nicotine breakthrough profiles).
A high stems to tobacco ratio, e.g., in excess of about 2:1 gives a "stemmy" character to the extracted tohacco.
Predrying of the stems to maximi~e potassium citrate loading results in a toasted note. Use of stems, without predrying, gives a more acceptable product.
Maximum nicotine removal from full flavor tobacco requires (1) a high potassium citrate level on the stems, (2) a low level of background nicotine in the adsorbent support, and (3) use of dual entxap-ment columns.
Shredded stems are the preferred stem type for an adsorbent support. Use of cut and rolled stems probably results in a pressure drop problem ~nd use of long stems results in poor nicotine extraction.

.. .

, " ~

:

7 ~

Table 3 shows the results from extraction runs using potassium citrate treated full flavor American blend tobacco as the adsorbent support.
Subjectives were judged good ~ut nicotine removal was low, in the range of 83~ to 88% (Table 3) due to the high level of nicotine already present on the adsorbent support. Higher nicotine removal levels probably require significantly higher potassium citrate loading.
High levels of full flavor tobacco subjec-tives were present in the cO2 at the end of the extraction as evidenced by the waxy coating on the metal surfaces of the extraction/entrapment vessels.
Table 4 shows the results from extxaction runs using non-tobacco adsorbent supports.
When potassium citrate treated pure 100%
cotton fabric was used to remove nicotine from tobacco, subjectives were judged not as good, thin with slight mouthcoating, as with potassium citrate treated stems (Run-27, Table 4).
Use of potassium citrate treated pure alpha-cellulose gave a low removal of nicotine from tobacco, probably due to the non-uniform distribution of potassium citrate on the cellulose support. Subjec~
tives were judged unacceptable, bitter~reen, dry, and astringent (Run-32, Table 4).
Use of cocoa shells as an adsorbent support gave a nicotine removal of 96.~% using a single entrapment column. Subjectives were characterized as acceptable, tobacco like, slightly burnt and sweet with no mouthcoating ~Run-30, Table 4).
Use of cocoa shells as the adsorbent support also permits the transfer of desirable flavor attri-butes from one natural substrate (cocoa shells) to another ttobacco). One may also transfer the desirable flavor subjectives from Burley tobacco to an expanded stem support.

~7~1 -~7-~able 5 illustrates the results from extrac-tion runs carried out using unwashed coarse shredded s~ems, unwashed fine shredded ~tems and washed fine shredded stems as the adsorbent support.
Combinations of stem type ~nd shredded stem particle size were tested for extraction efficiency and product guality. Best extraction efficiency was obtained with washed fine shredded Bright stems.
Table 6 illustrates the results of a number of extraction runs carried out using washed fine shredded Bri~ht stems.
Extraction of more than 96% of nicotine was achieved yet the product was subjectively rated as average in smoking characteristics.
Table 7 illustxates the results from extraction runs carried out utilizing 2% ammonium ~icarbonate pretreatment of the tobacco filler.
The following variables were tested:
Solvent to tobacco ratio was reduced from 150 to 113 m/m, extraction -time was shortened from 60 min to 45 min., and the temperature of the extrac-tion process was lowered from 70 to 55C. Subjec-tives were judged very good and nicotine extraction was high, over 96%.
Extraction of 2% ammonium bicarbonate-pretreated tobacco at 70C, 75 m/m and for 30 minutes gave 95.2% nicotine removal (Run 51, Table S).
However, subjectives were judged not as good compared to product obtained at lower extraction temperatures.
Further reduction in extraction time (30 minutes), and further reduction of the solvent to tobacco ratio (75 m/m) at 55C gave reduced extraction efficiency, 94% nicotine removal (Run 57, Table 5).
Table 8 illustrates the results from extrac-tion runs carried out using a single vessel design in the upflow and downflo~ modes.

' , . :
-.
.
' '' ' ' ' - ~3~7~
il8-In the upflow mode, the flow of c~rbon dioxide is in the opposite direckion as the force of gravity and in the downflow mode, the flow of carbon dioxide is in the same direction as the force of gravity. The incipient fluidization velocity is about 1.1 cm/sec and the pressure drop does not become significant until up to about 1.6 cm/sec. The tobacco extraction data obtained at about 0.9 cm/sec, up~low, compared favorably with the downflow control (Run 41B
vs. Run 49B, Table 8).
Table 8 also illustrates the relationship between m/m and extraction time.
The impoxtance of the solvent to tobacco ratio was established in two experiments by varying the extraction time (from 30 min to 75 min) at a fixed mjm. The results show that a minimum solvent to tobacco ratio is needed (about 113 m/m) to achieve in excess of 96% extraction of nicotine. The time of extraction is not considered important within the range tested (Run 45-30 min.; Run 66-45 min.;
Run 67-75 min.~.
Tobacco materials in solution in CO2, without nicotine, were collected in two runs from full flavor American blend filler (Runs 42, 43, Table 8). The tobacco was pretreated with potassium citrate to convert nicotine to its salt. The tobacco solubles were extracted wi~h supercritical C02 at 260 atmospheres and 55C.
The process disclosed was also used to deposit tobacco materials in solution in C02, obtained from the dry ice expanded tobacco process, as disclosed in United States patents Re 32,013 and 3~,014 onto the tobacco filler (Run 45, Table 8).
Extensive C02 sampling during the extraction runs showed that nicotine concentration in C02 follows a first order ra$e of extraction.

-~-9-~bles _ beneral Notes ~ 11 runs were ~arried out at about 260 atmospheres, 70C, 25% OV (oven volatiles).
Subjective rating was based on a value of "zero" for unextracted ull flavor American blend tobacco. A positive value indicates an improved rating.
All runs were carried out without extraction during the heat-up via bypass o~ the entrapment io vessels during the fill and heat-up period.
The oven volatiles of the tobacco filler in the extraction vessel decreased when the oven volatiles of the stems were 25% and 35%. when the oven volatiles of the stems were 45% and 55%, the results were inconclusive.
Pre-drying of stems before potassium citrate treatment was not done for the following Runs (10, 25, 26, 28, 29, 31, 35 thru 39, 41 thru 47) -, .
- . - ., . , : .
,. - , : ~ -, ~ ' ' ~ , , ~3~ 7~ ~

KEY TO THE TABLES

AB - 2~ ammonium bicarbonate ALPHA - alpha cellulose CARBON - activated carbon 5 COCOA - cocoa shells with fines removed via screening COT - cotton CRS - cut and rolled stems ETOH - ethyl alcohol 10 FF - full flavor American blend tobacco filler KC - mono~potassium citrate (+n.mKC = n.m moles sprayed onto the entrapment support) LS - long stems 15 LTAR - low tar blend tobacco M/M - mass of carbon dioxide/mass of tobacco PG - propylene glycol SS - shredded stems STP - strip blend tobacco 20 UCSS - unwashed coarse shredded stems UFSS - unwashed ~ine shredded stems WFSS - washed fine shredded stems ~3~7~

~ ~ o o U~
U~
~ o E~ ~
Z ~ ~ o ~ ~ ~ E

5 1 ,,, . o ~i ~ z I U ~ ;~
Z; O ,C ~-W ~ O ~ o E
m ~
~: ~ P. rl O ._ G ~ ~ 3 C~ N

O 3 ~ ~ w ~ 8, E ,,~ C

~`I C
X ._ o c~^ ~ ~ ~ +
~ ~' E a~ O

~ E ~

' I o ~ ~ i ,. .

.

. ` ~3~7 l , ~
O O Ql O N O 5~ O It~
c~ ~ ~ .a ~ c~ ~ ~ o ~q O O
Z ~ ~ ~ Cl` O ~D CO

:~1 o o o o o o o ~ U~ O

U~
^l o o E~ ~ ~ o o o o o o o E~ ~ ~D
IY ~
~ o ~

~ L~ o :1 ~ E-l ^ 3 ~
O ~ C~
C~ ~ S~~C ~ y CJ C~ C~ D O
Om ,_ ~ ~+ '1 E-~ ~ Z U~
~ tLl C~ Z Z; Z Z oo u~ o _~
~'7 V~
E~ c~
_ Z ~
X ~C ~ ~ ~ ~ CJ
Z; ~ ~-- ~ C~ O
H ~1 ~ + 1- + ,~ . + t~

a U
u~ ,1 E
z 5!; _ E

~ C~
~ ~o~ ~ ~ ~ o o I ~ o 3 ~
~ z~
~ ~ ~ o ~ U~
~ ~ o o o o o " ~ ~ ~ o~

~1 --I ~ ~ ~ ~ ~ ~-1 r l z zl ~ ~ ~ ~ ~
E~ ~ I_ o o o o o o o ~~ E~ c~
" ~:
:~ ~ 3 .~ V~ ~ ~
~ ~ 0~ ~ oo oo ~J E~ c~
_ ~:

~ OE~ c~
~: C ¢u~ ~ 6 X ~ ~C X
C~ r~ ~ ~ ~r) r~
~o 3 Z ~ + + + + + +

;~: E~
:~ ~ X X X ~ X
Z ~ u~ _ ~ ~ ~ ~
~ ¢ ~ ~ ~ + +
o z tn u~

~ Z~
¢ u~
~ X
t~ ~ a~
't Z

', ' -, ' :

~ 7~&~
hZ
U'~
~ o o ' -~ o o o O O
Z ~ I~
C~ o ~ _ o Cr~

~1 ~ ~ ~ xl 0~ 0 ~ k ~
O ~ Z~ I + +~ +

X E~
X
~ ~ ~ + +

~n :Z
X
~ ~.~ J~ 't Z

t 3 ~ 7 ~3 ~ ~

~ u~
~ , ,~ j , , U~ . .. . .

E~
o o . ~ ~ , ~ o ~1 0 1 o o Z ~ Z~
~ ~ ~ o , o uq o e fi ~ E~ I 'u :z ~ ~ ~ O E
1~ 2~ ~ ~
W , Ei ~Z~

Z ~ i ~ :
~ ~ O In r~
~; ~

~ , . . .. .

:

~7~:3~ ~

~Z
,.
ua I ~

~1 o ~
Z; ~
a~ oo oO

~ O
--I ,~ o Q~

o ~ zl `J 3 u7 E~ ~
a 0~ 0~
O C~ C~
c~ 3 oo 0~ 3 O ~ . . _ ~ ~ 0~ ~ v v ~ ~ ' ~ E~ Q~
E~
o ~ ~n ~ ~
~C ~ ~ ~ ¢ + ~ .
E~ ~ ~
~ ~o :~; E~ ~
Z; ~ ~ ~
~ Z~ oo~ +
~ 6 c~ ,n~
FL Z; O
;~ Z o ~4 o W
~q l P~
Z ..
Z
:~ ~ C`J O
;~: ,~

~ 3 ~

i~
u~
~ ~ o ,~ . U'~ ...
;, ~ o ~ o E~
o o ~, ~
E ~

:~ I o o o ~ o ~ _ ,~ ~ i ~
~ ~ Zl ~ ~
X ~ ~ ~ o o o o o E~ r~ E~ ~
~ ~, O 3 ~1 L~ ~ ~
Z
O
X
C ¢ U~
E~ ~ ~ ~ ~ ¢ 't:
O Q k~ ; Z z;
E~ _ ~ ~C Y
Z P~ rn, '1: U~ ~ ,~ , E~
Z C~ O ;~
~ Z
ez v~ i~ O
E~ ~ u~
~ ~ Fs~
1~ ~1 C~ ~ ~ ~ + ~ ~ ~ ~ ~ +
,~
O
~ U~ o o ~L 3 ~ 7 ~ ~ ~

z oo In u~ ~ ~ ~
, o o o o o o , , , , , o, ~ O I ~ ~ c~ ~ ~ ~ E~
Z ~ o~ ~ ~ ~ Z

o o o o o o o u~ E~
C~i ,, ~ o o o o V~ ~
~ ~ ~~ ,~ . ~
E~ ~
3Zl E~
O 0~ o Z~
U~C~
X Zo Z~ 3 v~^ r~ ~ ~ X
o u~
~1 ~ ~ E-- ~~ t /~
I
E~ t~1 o ~ Z ~ ~
~ ~ tJ~ ~
_~ ~ ~ ~ tJ) ~!; ~ ;1~ ~ ~ ;Y; J~l X U? t O
~ ~r tt ~t~ tl~
a z ~c~ o ~ ~ ~ C q t~ L'l i~ O
~:; Z -~ Z 2 t ~ ~ J~ z ~0 H ~ W~ ~/$t.q ~IC ~ ;Y ~n U3 Y U~ Y t~) ~ ~U~ 3 c~ 3 ~
0~ t ~ t ~ ~ ~
Z~ ~ ~ O
~D ~ + ~ ~ ~ + ~ ~r ~ ~ + ~ +
O R::
tJ~ ~1 0 t.~l ii5 ~1 g! ~o ~ ~ tJ~

~ ~ ~ ~ o ~
X ~
.~ . .

o ~1 ~ O O O O 0 1~ ) Z

~ 3 ~

~ Zl u~ x u~ , ~ ~ ~ x ~
~ ~1 o o o T ,o o ol o o ~:
Z "~
a~

~n ~ ~1 '~
~1 ~ ,~
, ~ ~ o o o o o o o o C/~
~ a u~
a Z 3 z ~ ~ ~ L~ L~ L~ L~ L~ L~ L~ L'~
C`i V~ Z 0~

a . , Z E~C ~J
L~ L~ L~ Lr~ Lr~ Lr~ L~ L
E~ r .~ L~) Lr) L'l L'') 0 I_ ~ r-l rt ~
E~ t~
E~

_~ ~ u~
Z
C~ ~ ~ + ~ ~ ~ + ~ + ~ ~ ~ +
X Z-~

CZ~ ~ ~ u~ X cn o ~
Z;
W ~ _,L ~ ~ ~ + ~ + ~ + ~ + ~r + ~ +
W
~ oz ~
~ v~
~ t~ ~ ~
X ~

O
Z '.
t~
1~ 0 0 C ~ O ~1 ~

~ ~7~

Z -~ ~ ~ ~ ~r) ~ ~n u~ 1~ ~ u~
o o o o o ., o o o o o , ~ , ~ + ~ + , ~ + + + +

o ¢
D
~ O ,_ ~_ I~ r~ r~ l_ ` ~ oo 1` 1`
X 2 ` ~ C~` ` O` CJ' o~ c~ ~ cr~ c7~

al ~1 0 0 o o O O O O O O

` O V~

~; O O Z
E~ L ~ ~:
O X
~ ~ t_~ .

E~
~:: z ~J ~ ~r _ ~ ~ `~ ~
o ~Z

:4 ~ U~ X U~ X U~ X U~ X
~ Z ~ ~ ~ ~ ~ ~ ~ ~ ~
_~ ~ ~ + ~J ~ ~ + ~ ~ ~ + `$ + ~ + ~ + ~ + ~ ~ ~ +
Y
_ ~_-:z ~ ~a v~ ~ cn X u~ Y ~ ~ u~ ~ ~,q ~ v~ ~ c~
c: ¢ u~ u~
o ~ 3 ~ ~ ~ + ~ I ~ + ~ ~ ~ +
v~
E~ ~ ~ ~ m ~ ~
c~ u~ ~ ¢
+ + + + + +
X ~
.
ol ~ o U~ o ~ ~ ~ ~ ~

` ~3~7~

$ + $ ~ ~ + o o ,~
O ~ I~ u~ V ~ ~ ~ 0 z ,' C~
Q ~ ~I t~
i!; "I: ~, E~ O u~
Z; ~ :Z 3 z c~ ~ ~ ~ ~ ~ u~
~ 1_~ ~
.~ !~ o~ c~
~_:1 ~
~ ~;
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I
_~

C~ :Z
~_ ~ ~r + ~ + ~ + ~ + ~ + ~ * ~ ~ ~ + ~ + ~ + ~ +
X E~
Z ~
~ ~ ~ v~ X u~ ~ U7 X v~ X u~ ~ cn Y
O r, ~ D t;~ `D ~ ~ ~ `D
~Z;
kl ~ + ~:r + ~ + ;r ~ ~ + ~ ~ ~ + ~ + ~3 +
t~
U~ Z
~ O r ~ ~ ~ S
.t, ~

Zl '.
o~ .

1 ~ 7~ ~
~ Z ~ ~ ~ ~
,~ $ + + +

o ~
t, o , ,~,~ ,~
~i ~ ~ o~
~e Il U~
~1 Z
E- O ~n O ~:; O O Z
5 ~ ~ ~ U~
L~
t~ ~ o~ 8 C Z
E~ ~

~ ~ ~ U~
~E~ -~;~
;~ U~ X U~
~ ~ g _, ~ ~ +-:r + ~ +
x E~
Z ~ ~ ~
~ a~ ~ u~ X u~
q ¢ U~ U7 U~ ~q O ~ g ~
~_1 ;~ t +
C~ Z
E~ ~ ~ ~ m ~U~
X~ t4 Z

r :

z ~3 l7 ~1 U ~ P

E~ k~
O O Z ~ Z ~

Z ~ o ~, i~ æ ~ ~ z ~ ~
D E-~ Z co ~ Z ~ .
~ _ ~ ~

~ ~ ~n ~ ~ u~
~ E~ E~ ~ w - E~ ~.1 ~ ~ ~ ~ S ~ ~ ~ +
Y
z ~ ~ u~ ~ o ~zl~ ~ ~ ~
~ ~ L~ v~ :~ a ~ u~ ~ ~
~: ~ ~ ~ U~
z ~ :zg ~$ ~+
f~ o~ ~! z h ~ ~ ' ~ ~ :2~ ;y;
.. X~ ~ ~ ~
~ ~ a~ + ~ +

~ Zl :ZI

' ' ' ' o o l l ~ ~

o ~ ¦ o z ~¦ ~ ~ ~ ~ ~ ~æ

~o I U~
~ o `I ~ ~ ~ - - - a h ~ O
:~ o~ ~ ~ ` ~ ~ --E~
O o o o o U~ U~ ~
Z 3Zl ~

~3 o x~ o z ~ ~ t~ ~

O 2 ~ o o c o o o ~ _~ ~
~4 Z
X ~
3 ~ ~ ~

~ _ ~ ~+ + ~+ ~ ~
~ Y ~ Q

30 ~ ~
~ ~ ~+~ + ~
a w zi ~1 ~ ~ ~
~ ~ Y o ~
k~ ~ ~ + ~ ~
~ ~` o o u~
~ Zl , 2 Z ~ R~
~ ~ ~ ,-.

, ~ 5_ ~17~
The following examples are illustrative.
Ex~mple 1 Each of the entrapment vessels 2 and 3 in Figure l was loaded with 2.2 kg of cotton cloth which were impregnated with 2 liters of an aqueous solution containing 920.8g monopotassium citrate. 12 Kg of full flavor American blend cigarette filler t~bacco (25% oV) was placed in extraction vessel l.
With the valves to the entrapment vessels closed, the supercritical C02 was brought to 70C and 260 bar. Then the valve to the first trap was opened and the stream of supercritical C02 was circulated through the extractor and the first trap for 15 minutes. At the end of this time, the valve to the second trap was opened and the valve to the first trap was closed. The flow of supercritical C02 was circulated to the second trap for an additional period of lS min. At the completion of 30 minutes total extraction time, the circulation was stopped and the C02 was removed from the system. Tobacco blend and entrapment materials were removed from the system and submitted for nicotine analysis. A reduc tion in nicotine ~ontent of the tobacco blend of 77.4% was obtained. The tobacco blend retained a strong characteristic aroma which was not different from the unextracted blend.
Upon smoking, simiIar tobacco impact was obtained compared with unextracted tobacco.
Exampl 2 4.4 Kg of expanded bright tobacco stems were impregnated with 1.78 1 of a~ueous monopotassium ci~rate to yield a moistened stem support containing 409.8g of monopotassium citrate. This entrapment support was placed in the bottom portion of extraction vesse~ l. A porous plate was placed on top of the ~. .

7~

entrapment material and 3.52 kg of burley tobacco (25% ov) was loaded into the extractor vessel. The entrapment vessels 2 and 3 in Figure l were removed from the flow system by valve adjustment. A stream of supercritical C02 was circulated through the extractor with the temperature and pressure brought to 70C and 260 atmospheres, respectively. After 30 minutes extraction time, the circulation was stopped and the CO2 was removed from the system.
Tobacco product (burley) and entrapment material (expanded stems) were removed and submitted for nico-tine analysis. The burley tobacco product had a reduction in nicotine content of ~2.4%. Analysis of the expanded stem entrapment material yielded a corresponding increase in nicotine content. The burley tobacco product retained a strong character-istic aroma which was not different subjectively from the unextracted burley tobacco.
Upon smoking the extracted burley tobacco, similar tobacco impact was obtained compared with the tobacco impact from unextracted tobacco.

Claims (30)

1. A method for the selective removal of basic components from plant products containing a plurality of substances including acid salt forms of said basic components, said method comprising first contacting a plant product with a chemical base which does not substantially react under ambient conditions with the acid salt forms of said basic components or with other plant components but can react with the said acid salts under the conditions of an extraction process, then contacting the plant product with an extraction solvent, in the super-critical or liquid state, under conditions such that the chemical base does react with the acid salt forms of the basic components in the plant product.

2. The method according to claim 1 wherein the basic component in the plant product is nicotine.

3. The method according to claim 1 wherein the extraction solvent is selected from the group consisting of carbon dioxide, argon, SF6, N2O, lower halogenated hydrocarbons and lower hydrocarbons.

4. The method according to claim 3 wherein the extraction solvent is carbon dioxide.

5. The method according to claim 1 wherein the chemical base has a pKa greater than about 7.2 and less than about 10.

6. The method according to claim 1 wherein the chemical base is selected from the group consisting of ammonium bicarbonate, glycosylamines, N-glycosides of aldoses and N-glycosides of ketoses.

7. The method according to claim 1 wherein the chemical base is selected from the group consisting of sodium carbonate, sodium bicarbonate, potassium carbonate, potassium bicarbonate, ammonia, aqueous ammonia, triethylamine and trimethylamine.

8. The method according to claim 1 wherein the extraction solvent is further contacted with a non-volatile acid which is not soluble in the extraction solvent.

9. The method according to claim 8 wherein the acid is selected from the group consisting of polycarboxylic acids and the monovalent salts thereof.

10. The method according to claim 9 wherein the acid is monopotassium citric acid.

11. The method according to claim 8 wherein the non-volatile acid is dissolved or suspended in water, acid, aqueous acid or aqueous salt solution.

12. The method according to claim 11 wherein the non-volatile acid is monopotassium citrate.

13. The method according to claim 8 wherein the acid is contained on a support medium.

14. The method according to claim 13 wherein the support medium is selected from the group consisting of cotton cloth, tobacco stems, carbon, cellulose, and tobacco filler.

15. The method according to claim 14 wherein the support medium is tobacco stems which are long stems, cut and rolled, shredded, expanded, treated or untreated stems.

16. In a method for the selective extraction of nicotine from tobacco, preserving aroma-producing components, using normally gaseous solvents, said method characterized in that the tobacco is extracted with the solvent in the presence of at least 10% by weight of moisture based on the weight of the tobacco, said solvent being either in the liquid state at a pressure of from about 50 to 300 atmospheres or in the supercritical state at a temperature from the critical temperature of the solvent to about 120°C at a pressure of from about 70 to 1500 atmospheres, the improvement which comprises pretreating the tobacco with a chemical base which does not significantly react with acid salts of nicotine under ambient conditions but does react with the nicotine salts under the extraction conditions.

17. The method according to claim 16 wherein the moisture content of the tobacco is up to about 30% by weight.

18. The method according to claim 16 wherein the extraction solvent is selected from the group consisting of carbon dioxide, argon, SF6, NO2, lower halogenated hydrocarbons and lower hydrocarbons.

19. The method according to claim 18 wherein the extraction solvent is carbon dioxide.

20. The method according to claim 16 wherein the chemical base has a pKa greater than about 7.2 and less than about 10.

21. The method according to claim 16 wherein the chemical base is selected from the groups consisting of ammonium bicarbonate, glycosylamines, N-glycosides of aldoses and N-glycosides of ketoses.

22. The method according to claim 16 wherein the chemical base is selected from the group consisting of sodium carbonate, sodium bicarbonate, potassium carbonate, potassium bicarbonate, ammonia, aqueous ammonia, triethylamine and trimethylamine.

23. The method according to claim 16 wherein the extraction solvent is further contacted with a non-volatile acid which is not soluble in the extraction solvent.

24. The method according to claim 23 wherein the acid is selected from the group consisting of polycarboxylic acids and the monovalent salts thereof.

25. The method according to claim 24 wherein the acid is monopotassium citric acid.

26. The method according to claim 23 wherein the said non-volatile acid is dissolved or suspended in water, acid, aqueous acid or aqueous salt solution.

27. The method according to claim 26 wherein the non volatile acid is monopotassium citrate.

28. The method according to claim 23 wherein the acid is contained on a support medium.

29. The method according to claim 28 wherein the support medium is selected from the group consisting of cotton cloth, tobacco stems, carbon, cellulose, and tobacco filler.

30. The method according to claim 29 wherein the support medium is tobacco stems which can be long stems, cut and rolled, shredded, expanded, treated or untreated stems.