802
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 ENGI,%rEERIIVG C H E M I S T R Y ON HARDENED CHRYSALIS OIL B y MITSUMAKU TSUJIMOTO Received August 2, 1915
Raw chrysalis oil is unsuitable for t h e purpose of hydrogenation, as its nitrogenous a n d other impurities largely affect t h e activity of t h e catalyzer, even if we neglect t h e dark color a n d bad odor of t h e oil. Hon7ever: t h e refining of chrysalis oil is b y no means an easy one; b u t t h e process proposed by t h e author, which essentially consists in heating t h e oil with j t o I O per cent b y vol. of dilute sulfuric acid (sp. gr. 1.39) and then treating it with Kambara e a r t h , gives an excellent result.' The refined oil hardened by nickel catalyzer is a white fat which may be used as a useful raw material for soap-making. T h e composition of chrysalis oil has been as yet little investigated. T h e results of experiments published b y t h e author some years ago appear t o be t h e only report on this subject.2 According t o this rep o r t , t h e f a t t y acids of chrysalis oil consist of about 2 5 per cent saturated a n d 7 5 per cent unsaturated acids (iodine value 178.73). Among t h e saturated acids, palmitic acid was identified; stearic acid is probably not present. T h e unsaturated acids consist of oleic, linolenic and isolinolenic acids ; besides them, some isomers of linolic acid are present in a somewhat large quantity. If t h e conclusion of t h e above-mentioned investigation be really t h e case, t h e final product of t h e hydrogenation of these unsaturated acids must be stearic acid. X s t u d y of t h e product i s . i m p o r t a n t from t h e point of view of utilizing t h e hardened chrysalis oil for technical purposes. T h e author made a few experiments which are described below. I-HYDROGEK-ATIOZ
OB
THE
UXSATCRATED
(LIQUID)
FATTY ACIDS O F CHRYSALIS O I L
Fifty grams of chrysalis oil3 were saponified in a flask with 38 cc. of j o per cent aqueous solution of K O H a n d 113 cc. of 96 per cent alcohol, b y warming on a water b a t h ; t h e excess of alkali was neutralized with acetic acid and j o o cc. of 7 per cent aqueous lead acetate solution was stirred into. The resulting lead soap was twice washed with joo cc. of hot water a n d treated with 500 cc. of ether a t 10' C. a n d t h e n filtered (Tortelli and Ruggeri's method). T h e filtrate was t h e n treated with dilute HC1, in order t o decompose t h e lead soap, and was well washed with water: 2 5 0 cc. of t h e ethereal solution of t h e free unsaturated acids t h u s obtained. which contains about 2 0 g. of t h e acids of iodine value 176.17, were transferred into a strong glass bottle; 0 . j g. of Loew's platinum black was added. The bottle was t h e n connected t o a hydrogen h01der.~ On expelling t h e air from t h e bottle b y hydrogen, it was strongly shaken by means of a J . Chem Ind., Tokyo, 17, No. 191; Chem. Rev., 1914, 58. 2 J . toll. Eng., Tokro, 4 (1908). KO. 3 ; J . SOC.Chem. Ind., 1908, 455. 3 This sample of the oil was procured from an oil factory in Ibaraki prefecture. I t had the following properties: acid value 104.84, saponification value 187.03, and iodine value (Wijs) 140.33. 4 Hydrogen was prepared by pure zinc and dilute HzSOI, and passed through the wash bottles containing concentrated potassium permanganate solution and concentrated HISOP.
Vol. 8. No. 9
mechanical contrivance. After 31,"2 hrs. Shaking, a loss of about 2900 cc. of hydrogen was observed on t h e holder. Here t h e hydrogenation was stopped for a time. On evaporating off t h e ether, a residue amounting t o 1 7 . j 2 g. was obtained. It was a brown-yellow crystalline mass which when melted formed a brownred liquid; i t melted a t j6.2' C., having t h e neutralization value 188.92 a n d iodine value 45.91. The hydrogenation was apparently incomplete; b u t before continuing the operation, i t was found better t o remove t h e unsaponifiable a n d coloring matters from t h e product. Eleven grams of t h e above product were saponified with j o cc. of 8 per cent alcoholic solution of S a O H ; t h e n j g. of N a H C 0 8 a n d about j o g. of pure sand were thoroughly mixed with i t . The mass was dried, powdered and exhausted in a Soxhlet extractor with petroleum ether. The crude unsaponifiable matter thus extracted was 2 . 2 2 per cent. T h e soap i n t h e extractor was dissolved in hot water and decomposed with dilute HC1 a n d t h e n taken u p with ether. The ethereal solution of t h e f a t t y acids which appeared brownish yellow, was decolorized with animal charcoal, a n d made u p t o 2 j 0 cc. by adding ether; t h e n adding 0.3 g. of t h e platinum black. it was hydrogenized for 2 hrs. in t h e same way as before ( t h e reading of t h e volume of hydrogen was omitted). On evaporating off t h e ether, 8.3 g. of t h e hydrogenated acids were obtained. The white crystalline mass had a melting point of 68 t o 68. j oC., neutralization value 19j.19 a n d iodine value 0 . This product is therefore a saturated compound, which in its m. p. and neutralization value nearly coincides with stearic acid ( m . p. 69.3' C., neutralization value 197.5, molecular w t . 284). A mixture of t h e product with about a n equal quantity of pure stearic acid melted at 68 t o 68.3' C. In order t o perform t h e fractional crystallization of t h e acids, 5 g. of t h e hydrogenated product were dissolved i n IOO cc. of 90 per cent alcohol a n d separated into three portions successively as follows : (1) 4 . 2 i g . ; white laminae with pearly luster; m. p . 69.5 to 70' C . ; neutralization value 197.82; mean mol. wt. 283.59. A mixture with pure stearic acid melted at 69.5 to 69.7' C . (2) 0.21 g . ; m. p. 68' C.; neutralization value 197.20. (3) Residue left on evaporating the mother liquor. 0.41 g.; a little colored solid; m. p. 50' C . ; neutralization value 177.42.
The lorn. m. p . a n d neutralization value are probably due t o t h e accumulation of t h e impurities in this part a n d also t o t h e esterification of the acids on evaporating off alcohol. The result of t h e elementary analysis of ( I ) was as f olloa~s: 0.1245 gave 0.3487 COz and 0.1439 HzO; C = 76.39; H = 12.84 ClsHseOz requires C = 76.06; H = 12.68.
1
Therefore, t h e substance was confirmed t o be stearic acid. From t h e above experiment, i t was concluded t h a t t h e hydrogenated product of t h e unsaturated f a t t y acids of chrysalis oil consists mainly of stearic acid.
Sept., 1916
T H E J O U R N A L OF 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
11-ON T H E S A T U R A T E D (SOLID)
F A T T Y ACIDS OF
C H R Y S A L I S OIL
Although t h e saturated acids have no direct relation t o hydrogenation, a n experiment supplementary t o t h e former investigation1 was performed as follows : T h e impure lead soap of t h e solid f a t t y acids obtained in t h e previous experiment on t h e liquid acids, was carefully detached from t h e filter paper a n d heated i n a beaker with dilute HC1. T h e acids were t h e n dissolved in ether a n d decolorized with animal charcoal. T h e crude solid acid which was left on evaporating off t h e ether, formed a yellow colored crystalline mass. T h e yield amounted t o 11.j g. By dissolving i n IOO cc. of 96 per cent alcohol, it was fractionally crystallized into five parts. (1) 3.78 g . ; granular crystals; on melting. solidifies t o a grayish white mass; m. p. 55-55S0 C.; neutralization value 205.09. (2) 1.40 g.; m. p. 55.5-56' C.; neutralization value 209.30. (3) 0.77 g.; somewhat granular crystals, faintly yellow; m. p. 56.5' C.; neutralization value 210.70. (4) 0.21 g.; pale yellow laminae; m. p. 57' C. ( 5 ) 3.61 g.; residue from the mother liquor; an orange-yellow soft mass.
filtered through a filter paper; t h e precipitated lead salt was washed well with alcohol a n d decomposed b y dilute hydrochloric acid. T h e free acid was taken u p with ether, a n d washed free from t h e mineral acid; t h e ether was t h e n evaporated and a white solid acid was left. I t was dissolved in 2 5 cc. of 90 per cent alcohol a n d cooled t o room temperature for 30 min. The mother liquor was filtered off; t h e deposited crystals were dried b y pressing between dry filter papers. It was t h e n twice dissolved in 1 2 . 5 cc. a n d 6 cc., respectively, of 90 per cent alcohol b y repeating t h e same t r e a t m e n t as above. The final deposit weighed 0.35 g. and consisted of nacreous laminae; m. p. 69.5-70' C . ; neutralization value 194.76. This substance was, therefore, identified as stearic acid. No saturated acid higher t h a n stearic was found in t h e hardened chrysalis oil. This result is in a good accordance with t h a t obtained in t h e previous experiment. IV-INNER
A mixture of 3 g. consisting of 2 . 2 g. of ( I ) a n d 0.8 g. of (2) was dissolved in I O O cc. of 96 per cent alcohol a n d fractionally precipitated with magnesium acetate into the following fractions: (1) 0.9895 g.; m. p. 55.5' C.; neutralization value 203.22; mean mol. wt. 276.05. (2) 0.4920 g.; m. p. 56' C.; neutralization value 205.43; mean mol. wt. 273.09. (3) 0.4305 g.; m. p. 59' C.; neutralization value 207.98; mean mol, wt. 269.74. (4) 0.4785 g.; m. p. 60.5-61O C . ; neutralization value 215.25; mean mol. wt. 260.16. (5) No precipitate was obtained by adding an excess of the precipitant. Therefore, i t was abandoned. (The acid contained in this p a r t corresponds t o 0.6095 g.)
The result is not decisive, b u t i t is certain t h a t a n acid or acids higher t h a n palmitic are present. T h e mean molecular weight of about 2 7 0 seems t o point t o t h e presence of daturic acid, C17H34O2. But as b y repeated precipitations, t h e m. p. a n d neutralization values are regularly changed a little, i t is more probable t h a t the 'substance under examination consists of a n eutectic mixture of stearic a n d palmitic acids. 111-DETECTION
OB H I G H E R S A T U R A T E D F A T T Y ACIDS I N H A R D E K E D CHRYSALIS OIL
T o decide whether t h e hardened chrysalis oil contains s a t u r a t e d acids higher t h a n stearic, t h e author applied t h e method Proposed b y H. Kreis a n d E. R o t h ? which h a d been modified a n d applied t o t h e analysis of t h e hardened oil b y W. Normann a n d E. Hugel. 3 Ten grams of t h e mixed f a t t y acid obtained f r o m a hardened chrysalis oil4 (an oil f r o m Nagano prefecture hardened b y means of nickelcatalyzer) weredissolved in IOO cc. of 96 per cent alcohol in a flask. By heating it on a water b a t h , 0 . 7 5 g. of lead acetate dissolved in j o cc. of alcohol was added. On cooling t h e SOlUto the room temperature '*), it was Soon
' LOC. Cil.
Chem.-Zlg.. 1913, 58. Ibid., 1913, 815. Chem. Rea., 1914, 58. This sample is t h e No. B hardened oil. Its properties are as follows: m . p. 56' C.; acid value 57.16; saponification value 190.71; iodine value 35.81. 2
3
4
803
IODINE
VALUE
OF
HARDENED
CHRYSALIS
OIL
It has been shown by J. Marcusson a n d G. Meyerheim,' t h a t t h e inner iodine values of hardened fish oils are higher t h a n 100, whereas lower values are found in t h e cases of hardened terrestrial animal oils. With a sample of "Talgol" which is of fish oil origin, t h e y found a value of 107. While belonging t o a class of terrestrial animal oils, as chrysalis oil has a high iodine value, it appears not unlikely t h a t t h e inner value of t h e hardened product still exceeds 100, a n d so in this respect it may resemble fish oils. T o determine this, t h e author used t h e following two oils for examination: (A) T h e same sample used in t h e previous experiment 111. (B) A hardened oil having t h e following properties: m. p. 56-56.5" C.; acid value j7.46; saponification value 191.08; iodine value 45.j3. [The origi.naZ chrysalis oil was t h e same as ( A ) , t h e time of hydrogenation only being different.] Three grams of t h e sample were treated i n a manner equivalent t o Tortelli and Ruggeri's method. After keeping a t 6 t o 10' C. for 21/2 hrs., t h e ether solution of t h e lead soap was filtered off and decomposed with hydrochloric acid. T h e iodine values were determined b y Wijs' method. The inner iodine value of sample A? was 87.60; of sample B, I o 3 . 2 2 , nearly approaching t h a t of Talgol. " I n spite of t h e low iodine value (45.53) of Sample B, t h e inner value was comparatively very high, for a sample of t h e hardened chrysalis which has an iodine value near to t h a t of " ~ ~ 1 ~(iodine ~ 1 "value 65 t o 70), a still higher inner value m a y be expected. Of course, t h e process of hydrogenation, i. e., t h e catalyzer] temperature, time, a n d method of stirring, etc., will essentially influence t h e inner iodine value "
so,
angew. Chem., 1914, 201. T h e liquid f a t t y acids of this sample obtained as above, solidified a t the room temperature. It seems, therefore, t o have contained more or less solid f a t t y acids. 12.
2
804
T H E JOURNAL OF INDUSTRIAL A N D ENGIXEERING CHEMISTRY
of the product. T h a t hydrogenation takes place by degre’e according t o t h e unsaturation of t h e f a t t y acids, will not always be anticipated. It will be, therefore, unsafe t o attach too much importance t o the inner iodine value oi a hardened oil without taking into account the iodine value of the oil itself. For hardened chrysalis oil of a n iodine - \ d u e above jo, an inner iodine value exceeding I O O may probably be expected. V-
S U &I MAR Y
The results of t h e present investigations may be summarized as follows: I-The hydrogenated product of the unsaturated f a t t y acids of chrysalis oil consists mainly of stearic acid. 11-Besides palmitic acid. some higher saturated acid or acids are present in chrysalis oil. This substance is probably an eutectic mixture of stearic and palmitic acids. 111-By t h e Kreis and R o t h method, no saturated acid higher t h a n stearic was found in t h e hardened chrysalis oil. IV-An inner iodine value exceeding IOO may probably be expected in t h e case of a hardened chrysalis oil having t h e iodine value above 50. INDUSTRIAL EXPERIYEXT STATION
TOKYO, JAPAX
THE USE OF DIPHENYL GLYOXIME A S AN INDICATOR IN THE VOLUMETRIC DETERMINATION OF NICKEL BY FREVERT’S METHOD R p G. I,. KELLEYA N D J. B. CONANT Received December 2, 1915
A volumetric method for determining nickel in iron and steel as devised b y H. L. Frevert was published in Blair’s “Chemical Analysis of Iron,” 7th edition, 1912. Since t h a t time the method has been constantly in use in this laboratory, b u t inasmuch as some small changes have been made from time t o time i t seems best t o republish t h e method with these modifications. Accordingly me give below Frevert’s method as originally proposed. except for t h e modifications mentioned above, and follow it with a discussion of t h e use of diphenyl glyoxime as an indicator. FREVERT’S METHOD F O R THE DETERMINATIOK
OF NICKEL
I N STEEL
( A ) SOLUTION O F T H E sAI\IPLE-For ordinary nickel steels, a I-g. sample is taken, b u t mith less t h a n 0.10or more t h a n j per cent of nickel,larger orsmaller samples may be taken. I n t h e absence of more t h a n small amounts of chromium, solution is most rapid in j o cc. of hot dilute nitric acid (sp. gr. I . I ~ ) b, u t with chromium present in amounts greater t h a n 0 . 5 per cent, or under circumstances where chromium carbides are present, more satisfactory results are obtained b y dissolving t h e sample in 60 cc. of dilute hydrochloric acid ( I : I ) with t h e aid of heat. When solution is complete, t h e iron and t h e carbides are oxidized b y adding nitric acid, drop b y drop, until effervescence ceases. Boiling removes the products resulting from t h e decomposition of the nitric acid after which t h e solution is cooled. T h e quantities
VO~. 8, NO. 9
of acid given here are those which are convenient €or use with samples of I g. or less. ( B ) P R E C I P I T A T I O N O F K I C K E L D I X E T H Y L GLYOXIXE--
T h e solution obtained, as described above, is rapidly cooled b y the addition of a lump of ice, after which are added in succession: 1 2 g. of citric acid or equivalent solution, 20 cc. ammonia water (sp. gr. o . 9 0 ) , sufficient solution of dimethyl glyoxime t o precipitate all nickel present and enough more ammonia t o make the solution distinctly ammoniacal. The mixture is thoroughly stirred after each of these additions. The solution of dimethyl glyoxime is prepared b y dissolving 2 0 g. of the reagent in 1300 cc. of ammonia water (sp. gr. 0 . 9 0 ) , after which enough water is added t o bring the volume u p t o 2000 cc. Ten cc. of this solution allow sufficient excess t o completely precipitate 1 . 5 per cent of nickel in a I-g. sample, i. e., about 0.01 j o g. of nickel. (C) T R E A T X E N T O F K I C K E L D I M E T H Y L G L Y O X I b I E PRECIPITATE--When the amount of nickel is small (0.10 per cent or less), time must be allowed for complete precipitation- an hour is usually ample. With amounts larger t h a n this, no danger of lorn results attends immediate filtration. For this purpose an asbestos mat on a z-in. perforated porcelain plate or a Buchner funnel may be used. The solution containing the suspended precipitate will usually have a volume of zoo t o 2jo cc. It should be stirred thoroughly and poured on t o the asbestos mat in such a way t h a t the funnel always remains partly filled with liquid. Strong suction should be avoided. Quantities of precipitate corresponding t o less t h a n j per cent of nickel in a 1-g. sample rarely cause trouble in filtering. b u t the difficulty rapidly increases with larger amounts. When all of t h e precipitate has been transferred t o t h e asbestos, i t is thoroughly washed with water. Both wash water and filtrate are discarded, although t h e latter may be tested with dimethyl glyoxime if i t is believed that all of t h e nickel may not have been precipitated. (D)
SOLUTIOK
AND DECOhIPOSITIOK
O F THE PRECIPI-
TATE-The receiving flask and tip of t h e funnel are next well rinsed with water. With the mat still in place, b u t with suction off, enough nitric acid is added t o in. After a minute, cover the asbestos t o a depth of suction is applied, t h e acid drawn through t h e filter and about as much more added, taking care t o c o \ w the entire surface. At this po:nt there should remain no visible trace of the red precipitate. The mat is now t o be thoroughly washed with water, t h e washings being collected in t h e flask with the acid. The solution so obtained is then transferred t o a 400 cc. beaker in which it is heated t o boiling. Here the solution is ailotved t o cool slightly t o facilitate t h e addition of I g. of either potassium chlorate or ammonium persulfate. The solution is boiled until clear; this usually involves a considerable reduction in bulk, often as much as jo per cent, Insufficient boiling may cause trouble ( I ) through failure t o decompose t h e dimethyl glyoxime, which would reprecipitate when t h e s olution is subsequently made ammoniacal, or ( 2 ) because if t h e solution is not freed from oxidizing products
