206
T H E J O U R N A L OF IiVDUSTRIAL A N D ENGINEERIYG CHIIMISTRY
TABLE 11-PRODUCTS
O B T A I W E D B Y FLDXING GR.4HAMITE W I T H RESlDUtJM
MEXICAN
--
PERCENT
MELTING POINT O F .
MATERIALS 7 h _ , CHARACTER AND Fixed R i n g and Ball Barrett OF MIXTURES Carbon Ash Water Glycerine Bir PRODUCT M Mexican 17.26 tr. 104.7 residuum 117.3 Too soft Semi-fluid Too R 485” F: 20 21.22 0 . 0 4 hard 191.5 179.0 Tough hr. residue Per cent 70loss on M. Gr. fluxing .,. . . 130.3 A 95 5 145.4 133 . O ~&ogeneous B 90 10 181.4 162.0 Homogeneous 20.34 0 . 0 8 ) c 85 15 2 : i s 212.0 187.0 Homogeneous 2i:?2 0:24 Ton 238 3 229.0 Homogeneous D 80 20 1 . 9 0 E 75 25 1 . 1 5 . . hard 257 .O 241.0 Homogeneous ,. Fairly homo26.62 0 . 2 4 j F 70 30 1.25 Brittle
..
Vol. 7 ) NO. 3
material or on t h e penetration a t one specified temperature. The flattening out of t h e temperaturepenetration curves far below t h e so-called melting points emphasizes this point. A series of determinations plotted as in Fig. 2 may be of assistance in many
{
.
I
~
3 / 8 in. in diameter a n d weighing 3l/2 g. is placed on top of t h e sample which is then suspended in a 600 cc. beaker containing 400 cc. of glycerine, t h e bottom of t h e cylinder being I in. from t h e bottom of t h e beaker. A thermometer is suspended in t h e bath with the bottom of t h e bulb flush with t h e bottom of t h e cylinder. The b a t h is heated at t h e rate of 7’ C. per minute with constant stirring. The temperature registered at t h e time the sample touches t h e bottom of t h e beaker is recorded as the melting point of t h e material. T h e rate of heating exercises a marked effect on t h e melting point and must be watched closely t o obtain concordant results. The reflux solubilities and t h e air melting points were made as described by S. R.Church.’ The various other tests noted were made according t o t h e methods
cases in predetermining and comparing t h e values of new “mixes.” KoTE-The writer is indebted to &Ira S. Drucker for t h e preparation of t h e graphs which accompany this article. TECIIWICAL DEPARTMENT, MORRISA K D COMPANY UNIOXSTOCK BARDS, CHICAGO
1NVESTIGATIONS ON THE OIL OF EUCALYPTUS GLOBULUS OF CALIFORNIA By CHARLESE. BURKEA N D CHARLESC. SCALIONE Received December 21, 1914
recommended by Prevost Hubbard2 and by Clifford Ri~hardson.~ CONCLUSIONS
I-Fig. I shows a fairly regular increase of melting points of t h e fluxed products with increasing percentages of t h e asphaltite. 11-It is of interest t o note in Fig. 2 , t h a t t h e temperature-penetration curve of t h e residue from a 485’ F.-20 hour evaporation test of t h e original residuum follows very closely t h e curve given by t h e sample containing I; per cent of t h e asphaltite. 111-The writer believes t h a t too much emphasis is often placed on the melting point of a n asphaltic 1 2
3
THrs
JOURNAL, 3 (1911). 227, 6 (2913). 195. “Dust Preventives a n d Road Binders.” “ T h e Modern Asphalt Pavement.”
For many years considerable interest has been taken in California, in t h e Eucalyptus globulzts (Blue gum), a tree which, while a native of Australia, has been transplanted and thrives well on many parts of t h e Pacific Coast. The fact t h a t t h e tree grows very fast, and yields a hard wood makes it a tree particularly desirable in these localities which are almost destitute of hard woods. Many difficulties have been met in seasoning t h e wood which has a great tendency t o warp a n d check, b u t these have been largely overcome by experienced lumbermen and i t is now easily possible with certain precautions t o prepare from t h e Eucalyptus a good clean lumber, which for hardness and tensile strength compares very favorably with oak or hickory. I n Australia one of t h e valuable products of t h e tree has been the oil which is obtained by the distillation of t h e leaves; this oil finds extensive use in the arts a n d in medicine and is imported in large quantities annually from Australia into t h e United States. The United States Pharmacopeia describes the oil as: “A colorless or pale yellow oil, having a characteristic or aromatic and somewhat camphoraceous odor, and a pungent, spicy, and cooling taste. Specific gravity, 0.905 t o 0 . 9 2 5 a t 2 5 ’ C. Soluble in all proportions in alcohol; also soluble in three volumes of 70 per cent alcohol. I t s alcoholic solution should be neutraI t o litmus. It is dextrogyrate. the anglc of rotation being not more t h a n + I O ’ in a I O O mm. t u b e
Mar., 1915
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
a t a temperature of 25' C. If 2 cc. of t h e oil be mixed with 4 cc. of glacial acetic acid a n d 3 cc. of a saturated solution of sodium nitrite be gradually added, t h e mixt u r e when gently stirred should not form crystals of phellandrene nitrite. It should become semi-solid on being stirred, when cold, with a third or a half of its volume of phosphoric acid of commerce of specific gravity 1.750 (presence of a due proportion of cineol)."
these leaves gave a total yield of oil corresponding t o 0.8 per cent of t h e total weight of t h e green leaves. Baker a n d Smith have obtained yields as high as 0.918per cent; this of course varies somewhat with t h e season of t h e year. For t h e rest of our work, since we wished t o make our analysis as valuable as possible commercially, we have used oil obtained from a commercial still in North Berkeley. This oil was obtained from a grove of almost pure Globulus b u t containing a sprinkling of other varieties a n d would more fairly represent the average Californian bil t h a n a n absolutely pure Globulus oil. This oil was fractionally distilled a n d t h e following table shows t h e fractions obtained with t h e constants of each. Fraction No.
Temperature Percentage oil O C. in each Dzo 0.866 85-100 3.1 100-155 0.885 0.8 0.887 155-165 6.5 165-170 0.896 35.6 30.0 0.909 170-180 0.921 11.1 180-190 0,980 190 up 12.5
............. ............ ............. ............. ............. ............. ...........
1
2. 3 4 5 6. 7.
FIG. I-EUCALYPTUSGROVEO N THE CAMPUS OF THE UNIVERSITY THIRTYYEARSOLD CALIFORNIA.TREESAPPROXIMATELY
OF
The Australian oil fulfils these specifications b u t t h e Californian oil does not, and therefore t h e latter has not been utilized t o any great extent. I t is this fact t h a t has led us t o a n examination of t h e Californian oil, t o determine whether there is a n y difference i n t h e crude oil from t h e same species of Eucalyptus grown in Australia a n d California, or whether t h e difference is simply in t h e method of refining. Baker and Smith' after a very extensive research on t h e Eucalyptus oils of Australia state t h a t "the constituents of t h e oil of Eucalyptus globulus are practically constant wherever t h e tree is grown; a sample of oil distilled from trees in New South Wales differed in no respect from t h e product of trees grown in Tasmania or Victoria. The amounts of t h e constituents may vary b u t t h e constituents found are always t h e same." Our work on t h e Californian oil has fully corroborated this statement, since we have found exactly t h e same constituents in t h e Californian oil t h a t Baker a n d Smith found in t h e Australian oils. The amounts of t h e constituents vary so greatly, however, as t o give t h e crude oils entirely different properties. This may be seen from t h e following comparison of t h e Australian oil a n d Californian oil.
207
N : 1 .4485 1.4606 1.4635 1.4635 1.4645 1.4675
..
[CY], 20
15.25 20.11 19.17 16.00 8.80
3.81
...
From t h e curves i n Fig. I1 it will be seen t h a t t h e r e are at least four constituents in t h e oil: t h e first distils between 85" and 155') t h e second between 155' a n d 170', t h e third between 170 a n d 180, a n d t h e fourth above 180. BUTYRIC, ISOVALERIC,
AND CAPRONIC ALDEHYDES-
I n t h e fraction boiling below 155' C. t h e strong irri-
CALIFORNIAN OIL AUSTRALIAN^ OIL 0.913 . . . . . . . . . . . . . . . . . 14' 42 90 2
D20... . . . . . . . . . . . . . . . . . 0.9052 [ a l20 -r;..
.N :
...................
1.46053
?
Insoluble in Soluble in 1.5 volumes 70 per cent alcohol of 70 per cent alcohol Free acid.. .............. 0 . 7 3 1.1 Saponification No. 2.5 1 .o Acetylation No. 56.0 ?
....... .........
These data were obtainedfrom leaves of t h e Eucalyptus globulus gathered on t h e University Campus a n d distilled with steam a t a pressure of four pounds; 1 "A Research on Eucalyptus with Particular Reference to Their Essential Oils," by Baker and Smith. Published by the Australian Government.
FIG. 11-BOILING POINT,SPECIFIC GRAVITYAND INDEX OF CURVESFOR
REFRACTION
EUCALYPTUS OIL^
tating odor suggested t h e presence of aldehydes; on redistillation this p a r t was easily divided into three fractions, t h e first distilling a t 75-85') t h e second a t 85-100', a n d t h e third a t 100-140'. The first of these fractions readily formed a n addition product with ammonia which after recrystallization from alcohol melted at 26-27' C. T h e second one also formed a n 1
See Burke and Scalione, THISJOURNAL, 6 (1914). 804.
T H E J O U R N A L O F I i V D U S T R I A L AATD E N G I N E E R I N G C H E M I S T R Y
208
addition product with ammonia which after recrystalThe first of these fractions lization melted at 52-53'" was undoubtedly butyric aldehyde, a n d t h e second isovaleric aldehyde since t h e ammonia addition products of these two aldehydes melt a t 29', a n d 56', respectively. The third fraction formed no addition products with ammonia b u t t h e boiling point and other physical properties indicated capronic aldehyde. A quantitative analysis of t h e fresh oil, by t h e sodium sulfite method of Sadtler a n d Burgess,' showed the total aldehyde content of t h e oil t o be 6 per cent of the total volume. PINENE-After t h e removal of t h e aldehydes t h e remainder of fractions I , 2 a n d 3 was combined with iraction 4 a n d redistilled. The fraction distilling a t 156-1 j7 a constituted approximately 2 0 per cent of t h e original volume of t h e oil and t h e following constants would seem t o indicate t h a t i t was practically pure pinene. Distillate 156-157' D28 ........................ 0.878 ..................... 340 Boiling point 156-1 57 O Melting point . . 102-103°
C.
PUREPINESE 0.8767 2+49' 1 56-1 5 i," 102-1 03
rag..
A quantitative analyQis for pinene was also made on t h e crude oil as follows: a weighed portion of t h e oil was fractionated a n d t h e fraction boiling a t 13-176' retained; 5 cc. of t h e oil in this fraction were placed in a cassia flask and j o per cent resorcinol solution was added t o take up all t h e cineol and traces of aldehydes present; after t h e solution became clear t h e unabsorbed oil amounted t o 21-22 per cent of t h e whole. This oil had t h e same properties as t h e pinene above which had been purified by distillation. CIsEoL-The remainder of t h e oil from t h e first fractions was added t o fractions j and 6, a n d redistilled; after several fractionations a fraction was obtained which boiled a t 176". A comparison of t h e constants of this fraction with those of pure cineol shows t h a t this fraction was cineol. ........... ..........
e.. .................................. Melting point iodol (after recrystallization).
,
.
Distillate 176-177' 176-177O 0.930 1.457 111
PURE CINEOL 176 0.9267 1.4559
1
Burgess, Analyst, 169 (1904),i 8 .
Tobias,' namely, a bicyclic alcohol with a n unsaturated grouping, t h e theoretical molecular index of refraction would be 68.069; t h e observed molecular index of refraction was 68.001. The acetate prepared by a mixture of acetic anhydride a n d sodium acetate gave constants identical with those obtained by Tobias and Semmler. This a n d t h e general properties of the body seemed t o indicate t h a t we were dealing with a n alcohol rather t h a n a n oxide as suggested by Baker and Smith. In t h e higher boiling portions a body corresponding t o globuol was separated. Semmler a n d Tobias consider t h a t this body bears t h e same relation t o eudesmol t h a t borneol bears to isoborneol. Considering t h e formulas of the two alcohols to be C I ~ H Q t~hO e acetylation number would indicate t h a t t h e alcohols form 23.22 per cent of the total weight of t h e oil, The complete analysis shows t h a t t h e oil contains t h e following constituents in t h e given percentages: ALDEHYDES(butyric, isovaleric, and capronic) . . . , , 6 PINENE ......................................... CINGOL ......................................... ALCOHOLS(eudesmol, globuol, with traces of free acids and esters) .................................... ~
21-22 47
23
In order t o determine whether or not t h e oil from t h e burrs, which are necessarily gathered in large quantities with t h e leaves, h a d any detrimental effect upon t h e oil as a whole; a quantity of the burrs were
4G
35 30
25
1120
The iodol formed from this distillate after recrystallization melted a t 111'. Cineol iodol melts a t 112'. A quantitative analysis by t h e U, S. P. phosphoric acid method gave 47 per cent cineol by volume. E U D E S N O L A N D GLosuoL-The fraction boiling at 190' and up was placed in a shallow dish a n d allowed t o stand for several days until crystals began t o separate; these were removed, dried on a porcelain plate, dissolved in alcohol and after adding enough water t o produce turbidity allowed t o evaporate spontaneously until crystals separated. After another recrystallization soft white flaky crystals were obtained which gave t h e following constants: B. p. = 255' M. p. = 80' DZo= 0.988 [ a]? = 3 1 . 2 I ' in I 2 per cent chloroform solution. Assuming the formula proposed by Semmler a n d
+
Vole 7 ? NO. 3
2.0
15
LO
Per cent Pinene by W u m e 5 /o /5 20 F I G . 111-EFFECT
2,
OF P I N E N E UPON THE SOLUBILITY OF CINEOL
collected, ground a n d steam-distilled. The yield of oil corresponded t o 0 . 2 per cent of t h e weight of t h e burrs. The oil thus obtained was of a lighter specific gravity a n d had a greater rotation t h a n t h e leaf oil; i t dissolved in 18 volumes of alcohol (70 per cent b y weight) while t h e leaf oil dissolves in 1 5 volumes. This would seem t o indicate t h a t t h e burrs have a f
Semmler and Tobias, Be.,43 (Z), (1913), 2026.
Mar., 1915
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
detrimental effect upon'the oil and should be removed as far as possible. T h e above analysis would seem t o indicate t h a t t h e chief difference between the Californian a n d Australian oils is i n t h e exceedingly high content of pinene found in t h e Californian oil; this would account for t h e high rotation and low specific gravity of this oil. I n order t o determine t h e effect of t h e pinene upon t h e solubility in alcohol, pure pinene and cineol were mixed in definite proportions a n d t h e solubility in alcohol determined. Fig. I11 shows t h e effect of varying concentrations of pinene upon t h e solubility of cineol. It is readily seen t h a t above certain percentages of pinene t h e mixture would become practically insoluble in 70 per cent alcohol. SUMMARY
The above investigation has brought out t h e following facts: I-The oil from t h e Ewatyptus globulus, from trees grown i n California, has t h e same constituents as t h e oil of t h e Australian trees b u t in different proportions. 11-The reason why t h e Californian oil does not fulfil t h e U. S. P. requirements is probably due in great part t o t h e exceptionally high pinene content. Since Eucalyptus oil could be easily produced in large quantities in various parts of California and other Pacific coast states, a method of refining t h e oil so as t o bring i t up t o t h e U. S. P. specifications would be particularly valuable. The investigation is being continued b y t h e present authors with this idea in view. CHEMICAL LABORATORY UNIVERSITY OF CALIFORNIA
BERKELEY
THE SEPARATION OF GASES BY FRACTIONAL DISTILLATION IN A VACUUM AT LOW TEMPERATURES By G. A. BURRRLL AND I. W. ROBERTSON Received November 25, 1914
The authors have separated many gaseous mixtures by means of fractional distillation in a vacuum at low temperatures, a n d have found t h e method useful, accurate a n d applicable t o a wide range of mixtures. Some of t h e latter are impossible t o separate b y other methods and some are exceedingly difficult. Hence there is presented in this paper. a summary of t h e results so far obtained, with a discussion of t h e application of t h e method t o mixtures other t h a n those experimented with. T o date three papers on t h e subject have been presented b y t h e Bureau, as follows : I-Gas analysis by fractional distillation a t low t e m p e r a t u r e s 2 I n this paper was shown the separation of t h e paraffin hydrocarbons in natural gas. 2-The separation of t h e illuminants of mixed coal a n d water gas.3 I n this paper was shown t h e method of separating t h e paraffin hydrocarbons, olefine hydrocarbons and benzene in mixed coal and water gas. 3-The determination of gasoline vapor in air.' Published by permission of the Director of the Bureau of Mines. 1537. 8 THISJOURNAL, 7 (1915), 17. 4 Ibid.. 7 (1915), 112. 1
* J . A . C. S..S6 (1914),
209
I n this paper was shown a method for separating gasoline vapor from air. PARAFFIN HYDROCARBONS-Any combination of t h e gaseous hydrocarbons, methane, ethane, propane a n d t h e two butanes, can be separated b y first liquefying t h e mixture at t h e temperature of liquid air, removing t h e methane a t t h a t temperature, t h e ethane a t a temperature not higher t h a n -140' C., t h e C., propane a t a temperature not higher t h a n -120' C. t h e butanes a t a temperature not higher t h a n -9;' leaving t h e pentanes or higher paraffins as a final residue. O L E F I N E HYDROCARBONS-Any combination O f t h e gaseous olefine hydrocarbons, ethylene, propylene, and t h e butylenes, can be separated. T h e ethylene can be removed a t a temperature not higher t h a n -140' C., t h e propylene at a temperature not higher t h a n --I zoo C., leaving t h e butylenes as a residue. P A R A F F I N A N D O L E F I N E HYDROCARBONS-Any COmbination of t h e paraffin a n d olefine gaseous hydrocarbons can be separated. The methane can be removed a t t h e temperature of liquid air. T h e ethane a n d ethylene can be removed a t a temperature not higher C. T h e propylene a n d propane can t h a n -140' be removed a t a temperature not higher than-120' C., leaving t h e butanes and butylene as a residue. The pairs of constituents can be analyzed as such b y combustion in oxygen, a n d from this d a t a t h e percentage of each one determined. The authors have separated ethylene a n d ethane, boiling point only 10' C. a p a r t , b u t t h e operation is too tedious t o be of practical value. B E N Z E N E V A P O R i n air or in mixture with other gases, such as coal gas, can be separated a t t h e t e m C.). I n t h e perature of liquid carbon dioxide (-78' case of benzene vapor a n d other gases t h a t represent t h e final or t h e only separation desired in mixtures, it is most convenient t o measure their pressure after t h e removal of t h e other constituents. T h e volume present can be calculated from this pressure. For instance in t h e case of benzene vapor in air or coal gas, t h e mixt u r e is introduced into the liquefaction bulb a t atmospheric pressure. It is t h e n cooled by surrounding the liquefaction bulb with liquid carbon dioxide a n d removing as much gas as is possible with a vacuum pump. Next t h e stopcock on t h e liquefaction bulb is closed, t h e refrigerant removed, and t h e pressure of t h e benzene read on a manometer attached t o t h e liquefaction bulb. If a represents t h e pressure of t h e atmosphere a n d b t h e partial pressure of t h e benzene vaporl t h e percentage of benzene b y volume becomes I O O b/a: water vapor is removed by absorbing same in phosphorus pentoxide. G A S O L I N E V A P O R I N AIR-It was found necessary t o adopt a temperature lower t h a n t h a t of liquid carbon dioxide in order t o separate gasoline vapor from air, hence liquid air was used since it has t o be used a n d time is consumed in obtaining temperatures between those of liquid carbon dioxide a n d liquid air. T h e partial pressure of t h e gasoline is determined in t h e same way as is t h a t of t h e ben-
