T H E J O U R N A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
98 2
IO,
No.
12
ORIGINAL PAPERS A STUDY OF THE CONDITIONS ESSENTIAL FOR THE COMMERCIAL MANUFACTURE OF CARVACROL By ARTHUR W. HXXSON AND RALPHH. MCKEE Received June 21, 1918
Carvacrol is closely related both chemically and physiologically t o thymol. The latter substance is used as a specific for hookworm disease a n d as t h e principal ingredient in many antiseptic preparations. Hookworm is probably t h e greatest handicap t o t h e full use and occupation of the tropics by t h e white races. On account of t h e general use of thymol in antiseptic manufacture and the widespread organized effort being made in many countries t o combat t h e enormous inroads of t h e hookworm disease, t h c demand for it has become so great t h a t the supply from present sources is entirely inadequate. Attempts have been made t o produce thymol synthetically, b u t up t o the present time no process of any promise commercially has been developed. Recent comparative tests' have shown carvacrol t o be practically equal t o and, in some cases, t o possess greater antiseptic value t h a n thymol. I t s importance as a substitute for this substance is sufficient t o warrant a n investigation of its production. Carvacrol is found as an ingredient in t h e essential oils of many labiate plants and particularly in those of t h e species Origanam. There are two kinds oE Origanum oil2 known commercially, namely, Trieste oil containing from 60 t o 8; per cent of carvacrol, and Smyrna oil containing from 2 5 t o 6 0 per cent. Both of these oils contain cymene. Carvacrol is also found in t h e oil of thyme from Thymus vulgaris in which it sometimes replaces all of t h e thymol.3 T h e quantity of carvacrol available from these natural sources, however, is very small and of practically no commercial importance. These facts indicate t h a t if carvacrol is t o be used in t h e place of thymol a process for its synthetic preparation on a commercial scale must be developed. T h e prospects were such as t o encourage a study of t h e synthetic preparation of carvacrol and t h e conditions essential for its commercial production, which is the object of this research. Carvacrol was first prepared synthetically b y Schweizer4 who found t h a t t h e same oil was obtained by treating caraway oil with potassium hydroxide, phosphoric acid, or iodine. C l a w 6 heated camphor with iodine and obtained a product which he called camphor-creosote which was identical with the product made by Schweizer. MtillerB while comparing cymene a n d thymol obtained from different sources, sul1 The average results of four viability tests using the organisms E typhosus, B communior, and slaphylococus Pyogenes a w e u s , furnished through courtesy of Dr A. R Balls, Department of Bacteriology, College of Physicians and Surgeons, Columbia University, and Dr. A. M. Buswell, Depart ment of Chemical Engineering, Columbia University, New York City.
U. S Dispensary, 19th Edition, 1905, 1432. I b i d , 19th Edition, 1905, 1571. 4 J . prakt. Chem , 24 (1841). 257. 5 I b i d , 25 (1842), 264. 6 B e ? , 2 (1869), 130 2
a
fonated pure cymene, made t h e sodium salt, fused i t with sodium hydroxide a n d obtained an oil which he identified as carvacrol. Kekul6 and Fleischer' treated carvone obtained from caraway oil with orthophosphoric acid and produced carvacrol. From cymene, obtained from camphor, P o t t 2 made potassium cymene sulfonate, which he fused with potassium hydroxide. H e poured t h e fusion mass into water, neutralized with sulfuric acid, and obtained a small amount of yellowish liquid which distilled a t 230' C. He recognized i t as an isomer of thymol. He also observed t h a t if a few drops of a n alcoholic solution of t h e oil were added t o a solution of ferric chloride, a characteristic green coloration would be produced. Reychler3 found t h a t when carvo-chlorhydrate is distilled, hydrochloric acid split off a n d t h e distillate contained carvacrol. E t a r d 4 treated monochlorcamphor with a I O per cent solution of zinc chloride and heated it. By distilling t h e mass and agitating the distillate with caustic soda he obtained carvacrol. Meads a n d Kremmers converted pinene into nitroso-pinene a n d by hydrolyzing this substance produced carvacrol. The yield was about 6 0 per cent. Wallachs made amidothymol from oxydihydrocarvoxime, treated i t with sulfuric acid, and found carvacrol t o be one of t h e products. Harries7 passed steam for a long time over hydrobromcarvone and produced a small amount of carvacrol. McKees has patented a process for t h e manufacture of carvacrol based upon the use of spruce turpentine as the source of cymene. A careful examination of these methods showed t h a t in each case, with t h e exception of t h a t of McKee, t h e raw materials used were of such a nature as t o make them impracticable for t h e production of carvacrol on a commercial scale. However, t h e discovery t h a t spruce turpentine consists mainly of cymene and t h e fact t h a t it is produced in large quantities as a by-product in the manufacture of wood pulp by the sulfite process indicated t h a t a method, along t h e line suggested by t h e experiments of Muller, Pott, and McKee might be capable of commercial development. For these reasons an investigation was made t o determine whether a method based upon the following reactions could be carried o u t on a commercial scale: I-Formation of cymene I-sulfonic acid by treating spruce turpentine with sulfuric acid. 11-Removal of t h e excess sulfuric acid used in (I) and formation of calcium cymene sulfonate by treatment with finely divided limestone. 111-Formation of sodium cymene sulfonate in solution and removal of calcium as carbonate by treating with soda ash. B e ? , 6 (1873), 1087, I b i d . , 9 (1876), 468. 8 Chem. Cent?., 63 (1892), 379. 4 Compt rend., 116 (1893), 1136 5 A m . Chem. J , 17 (1895), 607. 1 2
6
7 8
A ~ N .291 , (i896), 348. Ber., 34 (1901), 1924. U. S. Patent No. 1,265,800, M a y 14, 1918.
Dec., 1918
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
IV--Formation of sodium carvacrolate by fusing the sodium cymene sulfonate with caustic soda. V-Formation of carvacrol b y treating t h e fusion products of (IV) with sulfuric acid. A study of these basic reactions indicated t h a t t h e following fundamental operations would be necessary: I-Sulfonation of cymene. 2-Disposal of the sulfonation products. 3-Formation of calcium cymene sulfonate solution and precipitation of calcium sulfate. 4-Filtration and disposal of the filter cake. 5-Formation of sodium cymene sulfonate solution and precipitation of calcium carbonate. 6-Filtration and disposal of filter cake. 7-Evaporation of t h e sodium cymene sulfonate solution and disposal of t h e solid salt. of the solid sodium salt with caustic 8-Fusion alkali. 9-Disposal of t h e fusion products. Io-Neutralization of t h e fusion product solution and formation of carvacrol. of t h e carvacrol. 11-Separation I 2-Purification of t h e carvacrol. EXPERIMENTAL
The experimental work consisted of the determination of t h e relative importance of t h e preceding operations and the conditions under which they could be carried out with maximum efficiency. MATERIALS-A~~ of t h e materials used in t h e experimental work were of standard commercial purity, such as can always be obtained on the market without difficulty in normal times, with t h e exception of spruce turpentine, which u p t o this time, although produced in large quantities as a by-product, has had no commercial value. The spruce turpentine used was a steam-distilled product obtained from t h e New Process Gasolene Company, of Philadelphia, and was purchased in t h e crude form by t h a t company from t h e J. and J. Rogers Company, of Au Sable Forks, N. Y. This product was clear and nearly white. After remaining in the laboratory for several months a distinct yellow tinge appeared. The product was used in the "as received" condition without treatment of any kind. A fractional distillation of I . j liters gave t h e following results: Temperature Degrees C. Below 171.5 171.5-178.5............................ Above 178.5 ...........................
...........................
Fraction
Cc. 50
1250 200
Per cent 3.33 83.34 13.37
Kerteszl found spruce turpentine t o contain 80 per cent of cymene, from I O t o 1 2 per cent of sesquiterpene, and t h e remainder diterpene. The fractionation results show t h a t nearly 80 per cent of t h e material came off a t about 175' C., t h e boiling point of cymene. SuLPOhTATIoN-The prime variables in this operation are ( a ) strength of acid, ( b ) temperature, (c) time, ( d ) proportional amount of acid, ( e ) amount of stirring, (f) type of sulfonating vessel. S T R E N G T H OF ACID-The adoption of 66' B6. SUIfuric acid as t h e most practical strength for t h e sulfonation of benzene in phenol manufacture led t o the 1
Chem. Ztg., 40 (1916), 945.
98 3
belief t h a t this strength would also be t h e most practical in the sulfonation of cymene. 400 cc. of commercial 66' BC. acid (checked b y titration with standard alkali) were placed in a liter Erlenmeyer flask with 2 0 0 cc. of spruce turpentine. This was placed in a water bath and heated t o 96' C. A two-blade glass propeller stirrer was placed in t h e vessel below t h e level of the acid and was run a t a speed of 7 0 0 r. p. m. in t h e direction t h a t would throw t h e acid toward t h e top of the flask. At the end of 3l/2 hrs. sulfonation was complete. This proved t h a t 66' BC. acid could be used. A similar experiment with 60' Be. acid showed t h a t sulfonation mas less t h a n 50 per cent complete a t the end of 1 2 hrs. Acids of greater strength were not tried although it was obvious t h a t the reaction period would be shortened somewhat by their use. The fact t h a t 66' BB. acid can normally be obtained a t less expense and trouble t h a n the stronger acids and t h a t it can be handled in a plant with less difficulty prompted its adoption for all of t h e sulfonation experiments. TEMPERATURE-The apparatus described in t h e preceding section was used. The sulfonation vessel was filled with zoo cc. of spruce turpentine and 400 cc. of acid. The stirrer was run at 7 0 0 r. p. m. Four hours was the standard time. At t h e end of the reaction period t h e stirrer was removed and acid allowed t o settle and separate from the cymene and the sulfonated portion. The upper layer was then siphoned off, shaken well, and 25 cc. removed b y means of a pipette. This was placed in a graduate and 7 5 cc. of water added, shaken well, and allowed t o stand for 1 2 hrs. The unsulfonated cymene formed a layer a t t h e top and its volume was read directly. The following table and curve, Fig. I show the results:
20
TABLE I Unsulfonated portion cc. 19.3
40
14.5
60 80 90 100
7.8 2.0 0.2 Trace
Temp. O
c.
Per cent sulfonated 22.80
42.00 68.80 92.00 99.20 99.5-1-
The rate of sulfonation varied almost directly with t h e increase of temperature u p t o goo. Between this temperature and 100' C. t h e rate was highest, indicating t h a t t h e temperature should be kept within this range for efficient sulfonation. This temperature being near the boiling point of water makes i t an easy one t o maintain in both laboratory and plant. Sulfur dioxide is evolved a t all temperatures. The amount was slight a t low temperature and increased as the temperature was raised. TINE-with the same apparatus and t h e same quantities of materials, time experiments were run. The d a t a in t h e preceding table indicate t h a t a temperature between goo and 100' would give t h e shortest time required for complete sulfonation. 96' C. was chosen for t h e reason t h a t it was convenient t o maintain. The extent t o which t h e reaction had proceeded was determined by the same means used in the preceding experiments, t h a t is, 2 5 cc. portions wese taken from t h e upper layer which had been separated
T H E J O U R N A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
984
Degrees C. FIG. 1
from t h e acid and, after diluting and standing for 1 2 hrs., t h e volumes of t h e unsulfonated portions were read. The results are given in Table 11. TABLEI1 Time Hrs. '/z 1 2 3 4 5
Unsulfonated portion cc. 18.4 12.4 4.3 1.2 0.1 Trace
Per cent sulfonated 26.40 50.40 82.80 95.20 99.60 99.64-
These d a t a show t h a t a 4-hr. period is sufficient for complete sulfonation at 96 O C. with efficient stirring. The results are shown graphically in Fig. 2. A M O U N T O B ACID REQUIRED-ZOO cc. charges of spruce turpentine were sulfonated a t 96' C. with quantities of acid varying from 400 cc. t o 150 cc. Sulfonation was complete with amounts down t o 2 0 0 cc. With amounts below this complete sulfonation could be obtained, b u t t h e time required was greatly increased. Many batches were r u n using equal volumes of acid a n d cymene and complete sulfonation was obtained in 4 hrs., with t h e temperature a t 96" C. These results show, contrary t o previous records, t h a t a volume of 68' BC. acid equal t o t h a t of t h e cymene is sufficient. It is t o be noted also t h a t t h e decrease in t h e amount of acid used t o t h e equal volume limit did not decrease t h e rate of reaction. A M O U N T O F sTIRRIxG-McKeel has shown t h a t t h e r a t e of sulfonation of hydrocarbons is distinctly dependent upon efficient stirring, other things being equal. By increasing t h e efficiency of his stirring device he was even able t o sulfonate kerosene with ease. No experiments were made t o determine t h e effect of different degrees of stirring upon t h e rate of sulfonation of cymene, but t h e t y p e of stirrer used, t h e speed a t which it was run, and t h e shape of t h e 1
Science, 36 (1912), 388.
IO,
No.
12
reaction chamber insured a n extremely intimate contact of reacting materials. TYPE OF SULFONATING VESSEL-TO determine whether a cast iron or steel sulfonating vessel could b e used, one was made b y screwing a cast iron cap on t h e lower end of a 6 in. length of 4 in. pipe. A similar cap provided with stuffing-box openings for a stirrer and t h e thermometer was used for a cover. A two-blade stirrer of t h e propeller t y p e was used. The blades were set at such a n angle t h a t when run a t speeds above 500 r. p. m. t h e liquid was thrown against t h e cover of t h e vessel. T o t h e stem a series of pulleys of different diameters was fastened in order t o use different speeds. The stirrer was driven with an electric motor. A thermometer was placed in t h e vessel a t such a depth as t o be well in t h e liquid, a n d it was held in place by a stuffing-box similar t o t h a t used for t h e stirrer. When t h e cover was screwed on well t h e vessel was gastight. T h e vessel was set into a water bath t o such a depth t h a t t h e surface of t h e water came t o t h e lower edge of t h e cover. The water b a t h was heated b y an ordinary Bunsen burner and t h e temperature was controlled within two degrees without difficulty. With this apparatus many runs using 300 cc. of spruce turpentine and 300 cc. of acid were made. T h e temperature in all cases was from 96' t o 98". With t h e stirrer running 500 t o 600 r. p. m. sulfonation was complete in from 3l/2 t o 4 hrs. A larger amount
Dec., 1c)18
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY
tion was less t h a n one degree below t h a t of t h e bath. When t h e sulfonation products were poured into a glass vessel and allowed t o cool, a white solid settled t o t h e bottom of t h e acid layer. This was found t o be ferric sulfate. The iron had reacted with an excess of concentrated sulfuric acid forming ferric sulfate and sulfur dioxide. This accounted for t h e increased amount of sulfur dioxide observed during t h e reaction period. The amount of ferric sulfate formed varied from 3.55 to 5.6 g., representing a loss of iron of from 0.73 t o 1.57 g. The weight of t h e sulfonator was 3500 g. The loss was quite small and undoubtedly in a larger vessel of special cast iron would be still less on account of t h e relatively smaller contact area a n d refractory skin on t h e surface of t h e vessel. The presence of this iron was not objectionable. If for a n y reason its removal might be desired this could be done a t practically no increase of cost b y t h e addition of a small quantity of lime just after t h e precipitation of t h e calcium sulfate in t h e next operation. SULFONATIOP; PRoDucrs--'CVhen s h o n a t i o n was complete and t h e mixture allowed t o stand for a short time, two distinct layers formed. The lower, lighter colored layer contained the greater portion of t h e excess sulfuric acid and about 20 per cent of t h e total amount of t h e cymene sulfonic acid. The upper, darker colored layer contained t h e greater part of t h e cymene sulfonic acid, some sulfuric acid, and materials resulting from t h e action of t h e acid on t h e impurities in t h e spruce turpentine. The formation of t h e layers took place rapidly when t h e materials were hot. When cold, t h e upper layer became very thick and viscous. When allowed t o stand for a few days a t z o o C. t h e cymene sulfonic acid began t o crystallize and finally t h e whole layer became solid. With t h e temperature below 10' C. t h e upper layer solidified very rapidly. Colorless, transparent crystals of cymene sulfonic acid, isolated from t h e upper part of t h e lower layer, melted at 50' t o 51' C. This was t h e melting point found b y Spica' and later by Eaton and McKee2 for a cymene sulfonic acid of t h e composition C L O H I ~ S OzHzO, ~ H . which t h e latter two made from spruce turpentine. All of t h e constituents of t h e upper layer were found t o be soluble in water with t h e exception of a small amount of a very finely divided white substance which settled out after standing for a number of days. Examination of this white precipitate showed i t t o be sulfur. Evidently it came from t h e complete reduction of a small portion of t h e sulfuric acid. D I S P O S A L O F S U L F O N A T I O N PRoDucTs-l'he formation of two distinct layers which could easily be separated suggested t h a t a recovery of t h e unused sulfuric acid might be possible. The problem was t o recover t h e sulfuric acid without losing t h e cymene sulfonic acid which was present in considerable quantity in t h e lower layer. Inasmuch as it was necessary t o add water in t h e next operation, experiments were made t o determine whether t h e distribution of t h e substances in t h e layers was affected by dilution. 1
B e y . , 14 (1881). 653.
* Unpublished thesis, University
of Maine, 1911.
98 5
For these experiments 1 5 0 cc. of spruce turpentine a n d 1 5 0 ec. of 66' BB. acid were used in t h e sulfonation. When sulfonation was complete t h e hot products were poured into a 5 0 0 cc. graduate and allowed t o stand until t h e volumes of t h e layers became constant. After reading t h e volumes a definite quantity of water was added and t h e mass shaken until t h e mixing was complete. The mixture was t h e n allowed t o stand until t h e volumes became constant again. I O cc. samples were taken with a pipette and analyzed. The total acid content was determined by titrating with standard sodium hydroxide. The free sulfuric acid was determined b y precipitation with BaC12.
CC.
W a f e r Added FIG.
3
The difference between these gave t h e sulfonic acid content which was calculated as sulfuric acid. The results are given in Table I11 a n d are shown graphically in Fig. 3. TABLEI11
Water added cc. None 30 60 90 120 150 180
Free HzS04 lower layer
G.
108.16 124.88 123.60 121.03 116.43 99.39 79.06
Free HzSOa upper layer
G.
49.22 40.73 36.57 39.87 45.38 58.68 82.70
Combined HzSOa lower layer
Combined HzSO4 upper layer
G.
G.
19.94 11.89 6.54 7.59 7.62 7.63 6.50
85.19 94.24 97.30 98.15 96.96 97.93 98.64
These results show t h a t b y t h e addition of a quantity of water equivalent t o one-fifth of t h e total volume (approximately 300 cc.) t h e combined sulfuric acid, which represents t h e cymene sulfonic acid, dropped from 19.94 g. t o 6 . j 4 g. in t h e lower layer and increased from 85.19 g. t o 98.15 g. in t h e upper layer. After this t h e values remained practically constant with further dilution. The values given in t h e above table for t h e combined sulfuric acid represent approximately t h e percentages of cymene sulfonic acid in t h e two layers. The dilution experiments were carried out t o t h e point where t h e two layers merged; analyses a t these
986
T H E J O U R N A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
dilutions were not made for t h e reason t h a t t h e acid was not worth recovering. Although i t was possible t o reduce t h e cymene sulfonic acid content in t h e lower layer 60 per cent, the amount which still remained was such t h a t its loss probably more t h a n balanced t h e value of t h e acid recovered and t h e amount of ground limestone saved. Market and plant conditions will be t h e deciding factors. If t h e acid is recovered t h e conditions t h a t will give t h e lowest cymene sulfonic acid loss should be used. At t h a t dilution t h e acid would have a concentration of about 60' B6. and could be used t o neutralize t h e fusion mixture in a later operation. The separation of t h e layers could be made in t h e sulfonation kettle after the products were coo:ed, if the kettle was made with a bottom discharge as is usually t h e case. N E U T R A L I Z A T I O N O F THE S U L F O N A T I O N PRODUCTS-
This is a standard operation in many processes and the conditions controlling it are well understood. The neutralizing agents may be a good grade of finely divided limestone or lime. Limestone is t h e cheaper and is efficient, although t h e operation does not go quite as smoothly as with lime on account of t h e evolution of a large amount of carbon dioxide. A lead-lined t a n k with a stirrer and foam breaker should be used unless care is taken t o discharge t h e sulfonation products into a slurry of limestone in which case t h e lead lining is not necessary. The calcium sulfate formed may be removed without difficulty by using a filter press. The filter cake has no value. F O R M A T I O N O F S O D I U M C Y M E N E SULFONATE-sodium carbonate or sodium hydroxide may be used. Under normal conditions soda ash would be t h e material t o use on account of its cheapness and t h e greater ease of filtering t h e precipitated calcium carbonate as compared t o calcium hydroxide. A standard, mechanically stirred wooden t a n k should be used.
SIMULTANEOUS NEUTRALIZATION O F SULFONATION F O R M A T I O N O F S O D I U M C Y M E N E SUL-
PRODUCTS A N D
F O N A T E soLuTIos-This may be done by neutralizing partially or completely t h e sulfonation products with limestone and then adding t h e requisite amount of soda ash. By this method the calcium sulfate and calcium carbonate can be removed by a single filtration. This procedure requires closer chemical control t h a n when t h e two operations are separated for t h e reason t h a t i t is much more difficult t o tell when t h e reaction with t h e soda ash is complete. Unless close watch is kept on this operation under plant conditions an excess of soda ash will often be used by t h e workmen. It is doubtful if this combined procedure will work out as well as t h e former in plant practice as t h e resulting saving in limestone and labor will be small. EVAPORATION
OF
THE
SODIUM
CYMENE
SULFONATE
is obvious t h a t the use of as little water as possible in t h e preceding steps will save time and expense in the production of dry sodium cymene sulfonate. This salt is very soluble in water and its water solutions are difficult t o evaporate t o dryness a t atmospheric pressure. The presence of a small amount of water causes t h e salt t o form a thick pasty SOLUTIOK-It
IO,
No.
12
mass which becomes liquid above 7 0 ' C. This last portion of solvent may be removed, in plant practice, b y either drying in vacuum or by use of a film dryer such as t h e drum dryers (atmospheric pressure or vacuum) which have lately come into such wide use in drying concentrated or pasty substances in chemical plants. F U S I O N O F T H E S O D I U M C Y M E N E SULFONATE-The problem was t o determine ( a ) the best fusion reagent, ( b ) t h e proper fusion temperature, ( c ) the most suitable type of fusion kettle, ( d ) t h e time required for completion of the reactions, and ( e ) t h e minimum amount of fusion reagent for maximum yield. The apparatus used for t h e preliminary fusion experiments consisted of a cylindrical steel vessel 41/4 in. in diameter and 5 in. deep. The steel was 3/1e of a n inch thick. T h e cover was a steel plate with openings for a stirrer shaft and thermometer and could be closed tightly b y means of stove bolts and winged nuts. It was necessary t o have t h e stirrer work through t h e whole mass of t h e liquid in order t o break up surface crusts and prevent the material from sticking t o t h e bottom and sides of t h e kettle. The vessel was heated with a Fletcher burner. After making a number of fusions with this apparatus i t was obvious t h a t it was not possible t o control t h e temperatures closely enough. T o overcome this difficulty the fusion vessel, equipped as described, was placed in an insulated b a t h containing about 2 0 lbs. of a eutectic mixture of sodium and potassium nitrates. This b a t h , provided with a propeller type stirrer, was heated with a Fletcher burner. With this arrangement there was no difficulty in keeping the temperature constant within one degree. The difference between t h e temperature of the bath and t h a t inside of the fusion chamber was less t h a n one-half degree when the stirrers were running. FUSION REAGENT-"YtOfOre, those who have prepared carvacrol by a fusion method have used potassium hydroxide in large excess. Although scientific literature favors t h e use of this reagent for t h e fusion of sodium cymene sulfonate a n d similar salts, such as sodium benzene sulfonate, modern commercial practice on t h e latter has demonstrated t h a t caustic soda can be used with equal efficiency and a t much less expense. For this reason a good grade of commercial caustic soda was used in all of t h e fusion experiments. FUSION TEMPERATURE-TO determine what effect temperature has upon t h e efficiency of t h e fusion operation, fusions were made at different temperatures. The charges consisted of 1 5 0 g. of dry sodium cymene sulfonate and 450 g. of caustic soda. T h e fusion period was 6 hrs. The caustic soda which melted a t 319' C. was fused first. T o this the sodium cymene sulfonate (in granular form) was slowly added. A t t h e end of t h e fusion period t h e products were poured gradually into 2 liters of cold water, forming a strongly alkaline solution which was neutralized b y adding dilute sulfuric acid (40' Be.). The carvacrol set free in this operation was extracted with benzene. After separation of t h e benzene solution from t h e neutral
T H E JOURNAL OF INDUSTRIAL A N D ENGINEERING CHEMISTRY
Dec., 1918
liquors, the benzene was distilled off a n d recovered. The residue was distilled yielding carvacrol and a t a r r y residue. T h e carvacrol was weighed, a n d the yield t h u s obtained was used as t h e criterion by which the efficiency of t h e fusion operation was determined. As t h e sodium cymene sulfonate dissolved in the molten caustic t h e temperature a t which t h e mass remained. molten rapidly fell. T h e average time required t o get a smooth fusion was one-half hour. It was possible t o lower t h e fusion temperature t o 255' C. a n d still have t h e mass molten enough t o stir well. If a quantity of water equal t o I O per cent of t h e weight of the caustic was added, t h e fusion took place much more smoothly a n d the mixture was kept molten a t a still lower temperature. However, t h e addition of water was of no advantage for the reason t h a t its rapid evolution as steam a t higher temperatures made i t difficult t o keep t h e molten material in the fusion chamber until it was all given off. When 2 8 0 ' C.
I
I
I
> 30
275
285
295
305
315
323
335
345
355
965
975
985
Deqrees 6, FIG.4.
was reached (fusion vessel uncovered) a bluish white fume was observed. This increased in amount rapidly with t h e rise of temperature until a t 3 2 5 ' C. i t was quite dense. T h e mass thickened under these conditions a n d a t t h e end of t h e fifth hour i t was granular a n d would not pour. If t h e temperature was kept below J O O O C.. although some fume was evolved, t h e material remained liquid until t h e end of the fusion period a n d poured well. At 280' C. when a flame was held in t h e upper p a r t of t h e fusion chamber a flash was observed. As t h e temperature was raised t h e flash became more pronounced and a t 300° C. t h e gas burned for a number of seconds. A t 3 2 5 O C. it burned longer. With t h e fusion chamber open a t temperatures from 2 7 5 " t o 300' C. t h e yield of carvacrol varied from 8.2j t o 43.19 g. Above 300' C. the yields varied from 35.30 t o 45.27 g. and, though higher t h a n for temperatures below 300' C., were very erratic and difficult t o duplicate. These facts are revealed b y the d a t a in Table I V .
987
TABLE IV (Fusion Chamber Uncovered)
fig,
m No. G.
.Y
u F G. Hrs.
g C.
J
3 k
267 275 275 285
G. Per cent 15.75 16.53 15.10 15.80 8.66 8.25 23.45 24.62
6 5
295 300 300 325
35.00 43.19 45.27 35.30
36.74 45.34 48.57 37.06
3.75
345
36.15
37.85
1 2 3 4
150 150 150 150
450 450 450
4.50
6
5 6 7 8
150 150 150 150
450 4.50 450 450
6 6
9
150 450
6 6
6
Remarks Nofume. No fume. Inflammable gas. No fume. Inflammable gas. Slirrht fume. Small amount Gf inflammable gas. Morefume. Moregas. More fume. More gas. More fume. More gas. Rapid evolution of fume. Solidified before end of period. Rapid evolution of fume. Solidified before end of period.
From I jo g. of sodium cymene sulfonate a yield of 9j.25 g. of carvacrol should have been produced. The percentage yields in this table and those following were calculated on this basis. Above 300' C. there was less variation in the yields. This and the fact t h a t t h e yields were higher led t o t h e belief t h a t still higher temperatures might give better results. I t was also noted t h a t when t h e fusion chamber was covered no fume was evolved, b u t as soon as the cover was removed it appeared, showing t h a t there was a reaction with the oxygen of t h e air. T o determine whether this reaction was in a n y way responsible for t h e great variation in yields, a series of fusions was made a t different temperatures with t h e fusion chamber closed. The cover with a n opening and connections for a condenser was screwed down tightly after each charge was inserted. A Liebig condenser provided with a n air-tight receiver was attached. From one opening in t h e receiver a tube r a n t o a gas holder. With t h e exception of the cover a n d the accessories mentioned, t h e conditions for this series were the same as for t h e previous one. The d a t a and results are given in Table V and shown graphically in Fig. 4. TABLEV (Fusion Chamber Covered) Charges consisted of 150 g. sodium cymene sulfonate and 450 g. caustic soda. Fusion period, 6 hrs. CarvaTemp. crol Yield No. C. G. Per cent Remarks 1 275 29.65 31.13 6.5 cc. Hz0 and trace of yellow oil distilled off. Small amount of inflammable gas. 2 300 35.71 37.41 6 cc. HzO. 15 CC. amber colored oil. Oil fluorescent. 3 325 37.16 39.01 6.5 cc. Hz0, 21 CC. amber colored oil. Oil fluorescent. About 3 liters gas. 4 325 38.20 40.10 8 cc. water, 22 cc. amber colored oil. Fluorescent. Passed gas through bromine. No reaction. 5 340 43.57 45.74 9 cc. water, 25 cc. amber colored oil. About 4 liters gas. Oil less fluorescent. 6 342 44.81 47.04 7.5 cc. water, 23.5 cc. oil. Oil amber colored and fluorescent. 7 355 49.36 51.71 Lost distillate. 8 360 48.33 50.14 9 cc. water, 22 CC. amber colored oil. Fluorescent. 9 360 49.91 52.3 1 9.5 cc. water, 23 cc. oil darker colored and more fluorescent. 10 360 49.14 51 . G O 11 cc. water, 22 cc. amber colored oil. Fluorescent. When cover was removed, mass flashed. 11 370 48.26 50.65 11 cc. water. 22 cc. lizhter colored oil. N o t quite as fluorescent. 12 375 47.92 50.31 14 cc. water, 22 cc. amber oil. Oil fluorescent. When cover was removed, mass ignited. 13 375 46.97 49.31 11 cc. water, 22.5 cc. amber colored oil. Fluorescent. When cover was removed, mass ignited. 14 385 46.10 48.39 10.5 cc. water, 22 cc. amber colored oil. Oil fluorescent. Mass did not ignite when cover was removed.
-
988
T H E J O U R N A L O F I N D U S T R I A L A X D E N G I L V E E R I N G C H E M I S T R Y Vol.
The d a t a of Table V show t h a t t h e yields of carvacrol increase and become more uniform with t h e rise of temperature. Also t h a t t h e range for maximum uniform yield is from 350' t o 370' C. Schorgerl states t h a t t h e fusion temperatures should not be above 300' C. T h a t this is not correct is demonstrated b y t h e results of these experiments. Neither could he get uniform yields a t temperatures below 300' C. Above 3 7 0 ° C. decomposition becomes noticeable and t h e yields decreased. Between 360' C. and 370' C. t h e fusion mass had a tendency t o ignite when exposed t o t h e air. This was much more marked at higher temperatures. A comparison of t h e d a t a obtained from this series of fusions and t h a t of t h e previous one shows plainly t h a t it is necessary t o use a covered fusion vessel. Without a cover t h e fusion mass thickens, due t o reaction with oxygen of t h e air, and t h e volatile oil which distils off is lost. It was noted t h a t in all of the fusions a n amber colored, fluorescent oil came off. The quantity distilled from t h e different fusions was quite constant. It varied somewhat, in color and in t h e degree of fluorescence with different fusions. As a rule it became darker on standing. I n some instances it became more fluorescent, and in others less, when exposed t o t h e air for some time. A quantity of this oil was carefully fractionated; 7 5 per cent of it boiled between 1 7 2 ' C. and 178' C. This fraction was water-white and h a d t h e odor and characteristics of cymene. T h e higher boiling fraction varied in color from light straw t o very dark brown and was relatively small in amount. All of t h e fluorescent material boiled above 2 1 0 " C. T o verify t h e belief t h a t t h e oil was principally cymene, t h e fraction boiling between 1 7 2 ' C. and 178" C. was sulfonated in t h e usual manner. The sodium salt was made a n d fused with caustic soda. Fifty grams of t h e sodium yielded I O g. of carvacrol a n d 3.5 cc. of a n amber colored, fluorescent oil similar t o t h a t described. This proved t h a t t h e oil which distilled from t h e cymene sulfonate was essentially cymene. Inspection of t h e results of this series of fusions shows t h a t t h e cymene recovered in t h e distillate from t h e fusions represents a decomposition of from 18 t o 2 0 per cent of t h e sodium cymene sulfonate fused. Experiments showed t h a t this cymene could be easily recovered and re-used. On a factory scale its recovery would be profitable. T h e presence of sodium sulfate in considerable quantity as one of t h e fusion products along with cymene seemed t o indicate t h a t two reactions took place between the sodium benzene sulfonate and t h e caustic soda, one of which formed sodium carvacrolate and sodium sulfite, t h e other cymene and sodium sulfate. However, t h e evolution of hydrogen and methane and t h e formation of a considerable quantity of t a r r y matter indicated t h a t other reactions took place. T h e gas evolved during t h e fusions varied in quantity from 2 t o 4 . 5 liters. Samples from two fusions were analyzed and were found t o contain hydrogen and methane. There were no traces of carbon monoxide, oxygen, or unsaturated hydrocarbons. 1
THISJOURNAL, 10 (1918). 259.
IO,
KO. 1 2
TABLEV I Gas from Fusion So. (Table V) 3 6
Methane Per cent
Hydrogen Per cent 78.16 80.00
by volume
To get some idea of t h e stability of sodium cymene sulfonate Ijo g. (containing 0 . 5 7 per cent moisture) were placed. in t h e fusion kettle alone and heated. Between 345' and 350' C. it melted and showed n o signs of decomposition. T h e temperature was gradually raised t o 3 7 5 " C. a n d a t this temperature a distillate consisting of 1 3 ~ / 2 cc. water and 7 cc. of a d a r k oil came over during t h e first hour. The heating was continued for 3 hrs. During t h e entire period hydrogen sulfide came off in large quantities. T h e mass gradually thickened and was sticky and black when poured. The oil from this salt was somewhat similar t o t h a t obtained from t h e fusions with caustic soda except t h a t it was smaller in amount, very much darker in color, and was saturated with hydrogen sulfide. T h e hydrogen sulfide formed showed t h a t t h e reaction without caustic soda was not t h e same as t h a t with it. If it were, sodium sulfide would have formed during t h e fusion. This, in t u r n , would have reacted with t h e sulfuric acid in t h e neutralization operation and hydrogen sulfide would have been evolved. Such was no t h e case. 5t
41
31
21
Approximofe+7ean FusionSer/od v/C/d Curvq /d
1
2
9
4
5
6
8
IO
I2
15
20
24
Hours FIG.5 TIME REQUIRED FOR FUSION was determined b y making a number of fusions with t h e temperature and t h e composition of t h e charges constant using t h e quantities of carvacrol as t h e criteria. T h e results appear in Table VI1 and Fig. 5. Fusion periods of from 4 t o 6 hrs. gave t h e best results. With longer periods t h e yields gradually fell off and were more or less erratic. The same was true for t h e shorter periods. The curve in Fig. 5 shows t h e relation between t h e length of fusion period a n d t h e yield of carvacrol. T h e products of t h e 15, 2 0 , and 2 4 hr. periods were dry, granular masses
Dec., 1918
T H E JOlrRiVAL O F I L f T D U S T R I A LALVD E N G I N E E R I N G C H E M I S T R Y
when removed from t h e fusion chamber. On exposure t o air they gradually became hard and stony. I n t h e cases of t h e 2 0 and 24 hr. fusions t h e products were liquid up t o within z hrs. of the end of t h e periods. T h e amount of distillate was practically t h e same for all of t h e fusions of more t h a n 2 hrs. duration, showing t h a t t h e consistency of t h e products a t the end of t h e period was in no way related t o it. Temperature, 360' C. soda.
Carvacrol No. G. 15.15 1 13.71 2 17.21 3 2 31.15 4 2 31.43 5 3 39.78 6 3 41.14 7 4 47.12 8 4 49.36 9 5 49.81 10 11 6 49.91 12 6 49.33 13 6 49.14 8 46.31 14 I10 35.41 15 15 15.41 16 20 20.01 17 24 20.10 18 1 Solid a t end of period. Time .Hrs. 1 1 1.5
Percentage Yield of Carvacrol Liquid Liquid Liquid Liquid Liauid
Remarks when poured. when poured. when poured. when poured. when Doured.
Liquid Liquid Liquid Liquid Liauid
when when when when when
poured. poured. poured. poured. Doured.
Remelted after 5 hours.1
TABLEVI11 Temperature, 360' C. 150 g. sodium cymene sulfonate. Fusion period, 6 hrs. Percentage Caustic CarYield Soda vacrol Carvacrol No. G. G. G. Remarks 450 52.31 49.91 48.12 150 50.52 100 51.39 48.95 50.92 48.50 75 51.96 49.50 50 Distillate 11 cc. water 22 cc. oil. 19.91 18.97 25 Distillate 7 cc. water. '17.5 cc. oil. Distillate 13.5 cc. water, 7.5 cc. black oil. 0 00.00 00.00
It is interesting t o note t h a t t h e quantity of t h e fusion reagents could be reduced almost t o t h e theoretical amount required without much effect upon t h e carvacrol yields. With less t h a n I O O g. t h e mass could not be poured from t h e kettle. With smaller amounts t h e products were of a pasty consistency and had t o be scraped out. It is quite necessary t h a t the contents of t h e kettle be discharged rapidly in order t o prevent excessive loss due t o reactions which take place on exposure t o t h e air. These reactions were so rapid t h a t ignition took place on two occasions. This was especially true when t h e fusion products were semisolid. The results of t h e fusion in which no caustic soda was used have been discussed previously. FUSION O F CALCIUM CYMENE SULFOKATE-It is possible t o fuse calcium cymene sulfonate with caustic alkali and obtain carvacrol. Schorgerl obtained his highest yield b y using this salt. Although one filtration and t h e soda ash required for t h e making of t h e sodium salt, would be saved, more caustic would be necessary for t h e fusion; and mechanical difficulties, resulting from t h e insoluble calcium sulfite formed Lac.
Cit.
during fusion, would be experienced, thus over-balancing t h e advantage of t h e process.
TABLEV I 1 150 g. sodium cymene sulfonate, 450 g. caustic
Q U A K T I T Y O F F U S I O S R E A G E N T REQUIRED-In 8.11 of t h e fusions made in t h e previous experiments a large excess of caustic soda was used. T o ascertain t h e minimum amount required for t h e highest carvacrol yields and t h e best working conditions, fusions were made with different quantities. The results are given in Table VI11 and Fig. 6..
1
989
Two fusions using calcium cymene sulfonate were made with t h e following results. The charge consisted of 80 g. of caustic soda and 80 g. of calcium cymene sulfonate. The temperature was 360' C. Per cent of Carvacrol Theoretical G. Yield Remarks 14.82 28.80 16.5 cc. of fluorescent oil and 11 cc. of water distilled during fusion. Fusion was thick when poured. 2 17.70 34.41 14 cc. of fluorescent oil and 13 cc. of water came off during fusion. Fusion quite thick when poured
No. 1
The oil which distilled from t h e fusion was identical with t h a t obtained from t h e fusions of t h e sodium salt. Although t h e two fusions were made under t h e same conditions, there was quite a perceptible difference in t h e yield of carvacrol. NEUTRALIZATION O F T H E F U S I O N PRODUCTS-The fusions were poured into 2 liters of water and allowed t o dissolve. T o this solution dilute sulfuric acid (40' Be.) was added in sufficient quantity t o neutralize t h e excess caustic soda and free t h e carvacrol from t h e sodium carvacrolate. When t h e acid was added in excess it reacted with t h e sodium sulfite formed during t h e fusion, and sulfur dioxide was evolved. This served as a means of determining when t h e neutralization was complete. It was necessary t o add the acid t o t h e solution by leading it through a tube t o the bottom of the vessel. If this was not done a n excess of acid on t h e surface reacted with the sodium sulfite in t h a t part of t h e solution before neutralization throughout was complete. The appearance of sulfur dioxide under such a condition was not evidence t h a t neutralization was complete. If the excess sulfuric acid were recovered a t t h e end of t h e sulfonation
990
T H E J O U R N A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
operation it could be diluted and used for this purpose. If concentrated acid is used, the time required for neutralization will be greatly lengthened, due t o the great amount of heat produced. If concentrated acid is added rapidly, t h e heat evolved will cause foaming and loss of much carvacrol. S E P A R A T I O N O F THE CARVACROL-The Specific gravity of carvacrol lies so near t o t h a t of water t h a t it does not separate readily from water and dilute water solutions of salts. While some of it collects on the surface and forms an oily layer a great deal of it stays in t h e body of the solution in the form of an emulsion. For this reason it was necessary t o remove the carvacrol either by means of a solvent or by steam distillation. When extracted by means of a solvent, onehalf liter of benzene was used. It was well shaken with the neutral fusion liquors until it had dissolved all of t h e carvacrol. The benzene layer was then removed by means of a separatory funnel. The removal of the carvacrol from the fusion liquors by steam distillation had t h e advantage t h a t a considerable portion of the solid and suspended tarry matter was left behind. A great deal of this went into t h e benzene layer and caused trouble when t h e solvent was applied directly. The yields of carvacrol were t h e same by both methods. I t was necessary to use a solvent t o completely remove the carvacrol from t h e steam distillate. T h e benzene was recovered in all cases and it was found t h a t a n average of 4 per cent was lost in the extraction and distillation. When a n excess of acid was used in neutralization a large amount of sulfur dioxide formed was taken u p b y t h e benzene. After the removal of t h e benzene the residue containing the carvacrol was distilled. The carvacrol came over in the form of a clear, light yellow oil between 2 2 7 ' C. and 245' C., leaving a tarry residue which averaged 0 . 2 5 g. for each gram of carvacrol. It was noted t h a t when the yields of carvacrol were low the quantity of t a r extracted was also low, The amount of the residue depended upon t h e method used for removing the carvacrol from the fusion liquors. When t h e steam distillation method was used t h e quantity of t a r was much smaller. If the carvacrol is extracted directly b y means of a solvent, a still with a bottom discharge should be used for t h e distillation of the extract. This would provide for t h e removal of the tarry residue when it was hot. If allowed t o cool it formed a hard, brittle mass. P U R I F I C A T I O N O F T H E CARVACROL-This was done b y redistilling the product obtained from the fusion liquor by either of t h e two methods mentioned. No difficulty was experienced in getting a product with a fairly constant boiling point. I
LARGE SCALE EXPERIMENTS
Having determined t h e optimum conditions for t h e several operations involved in the production of carvacrol from spruce turpentine on a laboratory scale, it was desirable t o test them b y using larger quantities of materials. Accordingly this was done with apparatus of semi-commercial size. suLFoNAT1or;-Fifty-three pounds of spruce turpen-
IO,
No.
12
tine were treated with 114 lbs. of 6 6 " BB. sulfuric acid for 6 hrs. a t 98' C. in a cast iron sulfonation kettle. The time was longer t h a n would have been necessary had t h e kettle been equipped with a thoroughly efficient stirring apparatus. R E M O V A L O F T H E EXCESS S U L F U R I C ACID-NO attempt was made t o recover the excess acid. The sulfonation products were slowly poured into a 150 gal. wooden tank containing a slurry of limestone (95 per cent passed IOO mesh). On the basis of the spruce turpentine containing 80 per cent cymene the calculated amount of limestone required was 95.2 lbs. The quantity needed for complete neutralization was 1 0 2 lbs. The neutral solution was filtered with a 1 2 in., 12-plate Sperry press. P R O D U C T I O N OF SODIUM C Y M E N E SULPOBATE-TO the filtrate from t h e liming operation 16.5 lbs. of 58 per cent soda ash ( 5 8 per cent NaaO) were added. T h e calculated amount was 17.1 lbs. The calcium carbonate formed was removed b y filtration and t h e clear solution was evaporated in a 50 gal., steam jacketed, open iron kettle t o a thick, sticky consistency. The salt was dried in a steam-jacketed shelf vacuum dryer t o a moisture content of 0.7 per cent. The yield was 70.5 lbs. of dry sodium cymene sulfonate. The calculated yield was 74.8 lbs. I n factory practice t h e sodium cymene sulfonate solution should be evaporated t o saturation with a vacuum and then finished with a film d r u m dryer. F U S I O N O F THE SODIUI$ SALT-Forty pounds Of 7 6 per cent caustic soda (76 per cent NasO) were fused in a 30 in. cast iron fusion kettle heated with gas. T o the fused caustic 70.5 lbs. (calculated t o d r y basis) of sodium cymene sulfonate were slowly added. The kettle was tightly covered, the condenser connected, and the temperature gradually raised t o 350' C. T h e temperature was kept between 350' and 360' C. during t h e 6 hr. fusion period. At 2 7 0 ' C. the fluorescent oil began t o distill. The rate a t which i t came over increased as the temperature was raised. The fusion went smoothly and poured readily. Less caustic could have been used without t h e substance solidifying during fusion. The salt was poured into a n iron t a n k containing 30 gal. of water. N E U T R A L I Z A T I O N O F T H E F U S I O N LIQUOR-Dilute sulfuric acid (40' Be.) was slowly added until t h e fusion liquor was neutral. 67.5 lbs. of acid were required. The calculated amount was 70.32 lbs. T h e operation was carried out in a steel tank. S E P A R A T I O N O F THE CARVACROL-Forty-five pounds Of benzene were thoroughly agitated with the neutral fusion liquors. The benzene solution was separated b y drawing off the water solution from below. T h e benzene (43.75 lbs.) was recovered b y distillation with a steam-jacketed still. The benzene loss was 2 . 7 per cent. There being no direct heated still of sufficient size available, I liter of the extract was distilled in a distilling flask. The product obtained was slightly fluorescent a n d contained a little finely divided carbon which came from t h e cracking of the tarry substance.
Dec., 1918
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
When redistilled a clear, yellowish oil t h a t boiled a t 232' C. was obtained. From t h e quantity of carvacrol obtained t h e total yield was calculated. The d a t a a n d results of t h e large scale experiments are given in Table I X . TABLBIX-DATA
RESULTSOF LARGESCALEEXPERIMENTS Theoretical Quan- Per Quantity cent Quan- tity tity ReYield Recov- ReUsed quired Yield Per ered covMATERIALS Lbs. -Lbs. Lbs. cent Lbs. ered AND PRODUCTS Spruce Turpentine.. 53 Sulfuric Acid (66O B8.). 114 32.2 Limestone 102 95.2 16.5 17.1 Soda Ash Sodium Cymene Sulfonate. 70: 5 94: 25 40.0 23.9 Caustic S o d a . . Sulfuric Acid (40° Be.) 67.5 70.32 45 4 i : ? 5 9;:s Benzene 7.3 Fluorescent Oil (recovered for re-use) Sp. Gr. = 0.8890 Carvacrol.. 25.3 5 i : i Tar 3.8 AND
.............. . . . .. .. .. .. .. .. .. .. .. .. ........... ....................... .......... ........................ .. .............. :. : . . .. .................. . . . . . . . ............ . . . .. .. .. .. .. ......................... .. .. .. .. .. .. .. .. .. .. .. . . . . . ................................... ........ D I S C U S S I O N O F RESULTS A N D O B S E R V A T I O N S
The most favorable conditions for t h e various operations as determined in t h e laboratory when applied on a larger scale gave similar results. With t h e greater quantities of materials t h e conditions were more easily controlled. This was especially t r u e of t h e fusion operation which was t h e most difficult one in the process t o handle. T h e quantity of t h e cymenebearing oil obtained from t h e fusion, per unit of sodium cymene sulfonate fused, was smaller t h a n t h a t obtained from laboratory experiments. From the latter it amounted t o about 2 0 per cent of t h e original cymene used a n d from t h e former t o 13.7 per cent. ,The apparatus required t o collect this oil is simple and inexpensive and t h e quantity of t h e oil given off is such as t o make its recovery imperative. When the caustic soda was fused first a n d t h e sodium salt added afterwards, trouble was experienced with foaming unless t h e sait was added very slowly. However, this procedure may be used if t h e cover of the kettle is such t h a t i t can be opened and closed quickly so as t o prevent oxidation and loss of t h e distillate. When t h e sodium salt a n d the caustic were well mixed together previous t o charging, no trouble with boiling over during t h e fusion was experienced. T h e stability of t h e sodium cymene sulfonate permitted this t o be done. T h e fusion kettle, however, should be of ample size t o t a k e care of t h e temporary swelling of t h e fusion mass. When d r y sodium cymene sulfonate was exposed t o t h e air i t took u p moisture rapidly a n d became sticky. This property would prevent i t from being stored in a n open bin. T h e time required for each operation was such t h a t i t could be completed within 8 hrs., t h e ordinary working day. This would be a n important factor in plant operation. The yield of carvacrol from t h e operations with t h e larger quantities of materials was about 5 per cent greater t h a n with t h e quantities used in t h e small scale experiments. If t h e cymene recovered from t h e fusion operation is taken into consideration, as should be done, t h e carvacrol yield will be increased. The difference this makes is shown in t h e following table. The cymene content of t h e original spruce turpentine
991
a n d of t h e oil recovered was taken as So per cent in each case. The cymene content of t h e recovered oil was subtracted from t h a t originally taken. From this t h e theoretical yield and t h e percentage yields were calculated Per cent Yield Not Per cent Yield Taking Recov- Taking Recovered Oil into ered Oil into Increase NO. Consideration Consideration Per cent (Table V) 63.94 12.23 7 51.71 62.60 12.46 8 50.14 64.65 12.34 9 ................ 52.31 63.65 12.05 10 51.60 62.51 11.86 1 1 . . .............. 50.65 Large scale experiment.. 57.2 66.4 9.2
................ ................ ................
....
Carvacrol can be produced by t h e process outlined with t h e same equipment as t h a t used in a plant for t h e manufacture of phenol or beta-naphthol, with but few changes. A fusion kettle equipped with a close fitting, easily opened cover a n d a water-cooled coil condenser would be necessary. I n addition t o this a direct heated still for t h e distillation of t h e carvacrol would be essential. Otherwise the phenol or betanaphthol plant could be used as it is. Inasmuch as practically t h e same plant can be used, the cost of production of carvacrol can be compared with t h a t of phenol. T h e United States Government has fixed t h e price of phenol a t 38 cents per lb. This is commonly known t o give t h e manufacturer a n average net profit of 7 cents per lb., t h u s making t h e total average cost 31 cents per lb. The total material cost per pound of carvacrol produced on t h e basis of 6 0 per cent yield would be about 35 cents per lb. The labor and overhead costs would be higher t h a n those of phenol due t o the lower yield obtained. T h e overhead cost would also be somewhat higher on account of t h e extra equipment required. The total cost per pound of carvacrol produced would be close t o 60 cents, a cost well within t h e limits of commercial possibility. SUMMARY
A process for the manufacture of carvacrol from cymene has been outlined a n d studied in detail. T h e process depends upon the use of spruce turpentine as the source of cymene. T h e optimum conditions for the necessary operations have been determined. Briefly stated, t h e process consists of: I-Making cymene sulfonic acid b y thoroughly agitating spruce turpentine with a n equal volume of 66' BB. sulfuric acid a t a temperature of go' t o 100' C. for 4 hrs. in a cast iron sulfonating kettle. a-Neutralization of t h e excess sulfuric acid, formation of calcium cymene sulfonate in solution b y adding ground limestone t o t h e sulfonation products a n d removal of t h e calcium sulfate formed by filtration. 3-Formation of sodium cymene sulfonate by adding soda ash t o t h e hot calcium cymene sulfonate solution and removal of t h e precipitated calcium carbonate by filtration. 4-Concentration of the sodium cymene sulfonate solution in a vacuum evaporator t o the point of saturation. 5-Precipitation a n d drying of t h e calcium cymene sulfonate by means of a rotary steam heated film dryer, or other suitable means.
99 2
T H E J O U R N A L O F I N D U S T R I A L A N D EA’GINEERING
6-Fusion of t h e dry sodium cymene sulfonate with approximately one-half of its weight of 76 per cent caustic soda in a cast iron or steel fusion kettle provided with a cover and water-cooled condenser a t a temperature of 3 j o ” t o 370” C. for 6 hrs. 7-Pouring t h e fusion products into a minimum amount of cold water and neutralization of t h e solution so formed b y adding just enough 40’ BB.sulfuric acid t o neutralize t h e excess caustic soda a n d set free t h e carvacrol from t h e sodium carvacrolate. 8-Separation of t h e carvacrol from t h e neutral fusion liquid b y steam distillation or b y agitating with a solvent such as benzene. 9-Recovering t h e solvent b y distillation. Io-Distillation of t h e carvacrol from t h e benzene extract with a direct heated still. I I-Purification of t h e carvacrol b y redistillation with t h e same still. This process was tested on a large scale which gave even better results t h a n those obtained with t h e smaller quantities. CHEMICAL ENGINEERING LABORATORY COLUMBIA UNIVERSITY N E W YORK C I T Y
THE SEEDING METHOD O F GRAINING SUGAR B y H. E. Z I T K O W S K ~ ~ Received June 1 7 , 1918
There is a disposition in some quarters t o deny t o t h e sugar industry its claim as a member of t h e chemical industrial family. T h a t t h e beet sugar industry, t h e direct descendant of scientific research and probably t h e oldest member of magnitude of t h e chemical industry family, should find i t necessary t o establish any claim in this direction is anomalous. Someone, sometime, as a labor of love, will bring this out as a matter of record. Here I desire merely t o state t h a t nowhere else in industry has technical accounting been carried t o t h e point t h a t it has in t h e beet sugar industry. The beet sugar industry has taken laboratory manipulations or processes such as dialysis or diffusion, precipitation, filtration, evaporation, and crystallization and adopted t h e m t o factory scale, handling millions of pounds of material daily, and with a refinement which taxes tha ingenuity of the most expert manipulator t o now duplicate on a laboratory scale. It is even held t h a t t h e beet sugar industry, which established itself in Europe during t h e Napoleonic wars, deserves t o a very large dzgree the credit for t h e rapid development of t h e chemical industry of Germany. It was t h e beet sugar industry which furnished t h e technically trained and experienced men, capable of transferring laboratory reactions and processes t o a factory‘scale and keep t h e commercial requirements in mind, when t h e modern chemical industry sprang into being. Men go so f a r as t o state t h a t it was the beet sugar industry of Germany which made possible t h e terrible war t h a t Germany is waging, not only because i t was t h e foundation stone for t h e chemical industry b u t also because t h e cultivation of t h e beet brought with 1 Paper read before the American Institute of Chemical Engineers, Berlin, N. H., June 19, 1918.
C H E M I S T R Y l7o1. I O , No.
12
i t scientific agriculture which doubled t h e agricultural yields, thereby making Germany largely self-sustaining and iliminating t h e threat of being starved t o submission b y blockade. There is much t h a t can be said in defense of such a view-point. However, a t this time here it is desired t o discuss briefly t h e large scale practical application of t h e well known (‘seeding” method of inducing crystallization. The oldest, and for many years the only method of producing sugar crystals was t o concentrate t h e properly purified sugar-bearing syrups t o t h e required density or supersaturation and set them away. I n t h e course of days, or weeks, or even months, as t h e solution cooled, sugar would crystallize out. Even after t h e introduction of t h e vacuum pan method of “boiling” sugar, for many years this was the only method and was known as “boiling blanks.” Sometime during t h e fifties of t h e last century the a r t o r rather t h e “trick of t h e trade” of “graining” sugar while yet in t h e vacuum pan was acquired, though this was n o t generally adoptzd till 2 0 years later, and even u p t o this day frequently, for reasons which need n o t be discussed here, blanks are boiled. The general procedure a t present is as follows: A quantity of the properly prepared sugar-bearing syrup with a water content of from 30 t o 40 per cent is introdubed into a vacuum evaporator or “pan” and is concentrated till saturated. At this point t h e boiling mass will be at a temperature from 70” t o 80’ C., and under a vacuum of from 2 0 t o 2 5 in. Under certain conditions aqueous sugar solutions have t h e property of forming supersaturated solutions and in t h e presence of t h e non-sugars or impurities, such as occur even in purified juices, this tendencyis greatly increased, so t h a t in factory practice it is always necessary t o carry t h e concentration t o some degree of supersaturation before crystallization occurs. Now it is not t o be inferred t h a t in all cases simple supersaturation will bring about Crystallization, for, if the content of non-sugars or impurities in t h e solution is great enough, crystallization will not occur even though evaporation be carried t o t h e point of dryness. Under t h e normal conditions of sugar manufacture, however, t h a t degree of supersaturation is finally reached at which crystal formation begins. Sometimes a sudden shock applied t o t h e boiling, supersaturated mass is resorted t o in order t o induce crystallization, such as a sudden raising of t h e vacuum bringing with it violent ebullition, or t h e introduction of a hot syrup of a lower density which has t h e same effect, or t h e injection of steam or air into the mass. No matter how produced, a t t h e moment of their formation t h e crystals are infinitely small and some time is required t o attain a visible size, though this may be only a few moments. Eventually t h e crystals formed do become visible and t h e n t h e critical moment of t h e “boiling” of t h e [‘pan” arrives. It becomes t h e attendant’s business t o allow t h e formation of crystals t o proceed till, in his judgment, t h e proper number of nuclei for t h e apparatus in ques-
