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
1096
creased by the high temperature of the sewage, averaging 84’ F. during August. There is a noticeable correlation between the number of bacteria and the amount of copper, as shown in Table 111. I n order t o determine whether the disinfecting action is greatest a t the time when the copper content is greatest, samples were collected a t the Experiment Station from 6 A.M. t o g P.M., on August IO. Four-ounce samples were incubated a t 20’ C. with
Vol. 9, No.
12
age, and remained colloidal and opalescent. T h e relatively high iron content of the one o’clock sample produced a floc of iron hydroxide which coagulated the suspended matter and clarified the sewage. The results obtained by incubating diluted samples show that in some cases,notably a t P.M. and 6 P.M.,the diluted samples were decolorized before the undiluted ones. This shows how impossible i t is to apply the “dilution method” for the determination of the oxygen demand of this sewage. By incubating t h e composite sample of raw sewage with methylene blue, however, i t is possible t o obtain a rough measure of the amount of germicide present. The Sunday sewage will reduce the methylene blue t o t h e leuco-base within 6 hrs. ; some week-day composites have not produced the colorless compound for more than 20 days. The effect of the copper on the various processes being studied remains t o be seen, since operation has not been continued sufficiently long to allow any conclusions to be drawn. The company producing t h e copper-bearing wastes has investigated the recovery of the copper and acid, and claims that such recovery would not be profitable because of t h e great concentration necessary. If the wastes interfere seriously with the solution of the sewage disposal problem, their removal will probably be considered again. The writer wishes t o express his thanks t o Mr. Sturges for the bacterial data contained herein, and t o Prof. Winslow for suggestions and advice. S E W A G S EXPERIMENT STATION NEWH A V E N , CONNECTICVT
TANNIN CONTENT OF PACIFIC COAST TREES By H. K. BENSONAND FRANK M. AM
Time RG.
EM.
111
methylene blue, t h e time required for decolorization being a rough measure of the strength of the disinfecting action. Diluted samples were also incubated, and bacterial counts made (Table V and Fig. 111). TABLE V-DISINFECTION OF PARTS PERMILLION
AUG. 10, 1917 BacGas Formers METHYLENE BLUE teria per CC. Cop- Undi- 1 sewage per cc. Chlor- Alka370 c. ine linity Iron per luted 2 water 20” C. 6 3,700,000 100,000 120 0.1 0.56 30 94 24 6 1.52 3750 136 0 . 2 2;500:000 1 ,000: 000 72 24 1.36 3100 160 5 . 0 120 48 1 . 8 2.88 2900 101 19,000 100 192 168 2800 43 0.3 4.80 144 60 2750 114 0 . 4 2 . 7 2 559,000 1,000 168 144 4 . 0 0 0.2 2800 98 240 240 5.5 5.60 3075 6 187,000 10,000 72 1.92 24 0.6 3200 86 120 24 ..... ..... 1.60 3325 91 0.7 39,000 100 96 48 1.92 88 0.2 3375 144 ..... 150 3.52 3450 9.3 0 . 1 45,000 1,000 100 216 130 103 0 . 4 2.88 80 72 95 63 1 . 4 1.28 33,000 1,000 96 0.64 24 0.2 52 92 ..... 6 24 0.0 0.44 50 96
.
Time 6 A.M. 7 8 9 10 11 12 1 P.M. 2 3 4 5 6 7 8 9
JONES
Received August 28. 1917
SBWAGB BY C O P P E R ,
HOURS
..... .....
..... .....
.....
.....
.....
.....
.....
The time of decolorization varies directly and the alkalinity and bacterial content vary inversely, with the amount of copper. The physical appearance and tests showed markedly the effect of the copper. The samples collected a t 6 A.M., 8 P.M., and g P.M. became septic within a few days, developed a strong odor and lost their colloidal appearance. The remaining samples, with the exception of the one collected a t I P . M . , retained their original odor, t h a t of fresh sew-
I n a previous article it was shown’ t h a t the tannin content of Douglas fir slabwood when properly selected contained sufficient tannin t o be considered a suitable raw material for the extract industry. I n order t o ascertain what other species common t o the Pacific Northwest might be available for extract manufacture this study was undertaken. M E T H O D S O F E X T R A C T I O N A N D ANALYSIS-The Samples in each case were taken from four or five trees of the same species, were first chipped in a Mitts & Merrill “hog”, carefully quartered and the final sample finely ground in a Koerner drug mill. The extractor was of the same type as used in the previous study, extraction carried on for 16 hours and analyzed according t o t h e method of t h e American Leather Chemists’ Association (1914). All calculations are made on a moisturefree basis. E F B E C T OB S E A S O N I N G O N D O U G L A S FIR
The sample of sawmill slab reported in t h e previous study (1. c.) was still available, having been kept in a loosely stoppered jar. It was thought of interest t o ascertain its tannin content after a year’s storage. The average of seven analyses is given in Table I for comparison with previous results. From these results, i t would seem t h a t t h e effect of seasoning is t o bring about hydrolysis and possibly ITHIS JOURNAL,
I (1915), 915.
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
Dec., 1917 TABLE I-EFFECT
OF SEASONING ON
TANNIN CONTENT
FIRSLABWOOD
PERCENTAGES
Analyzed by Thompson in 1914 14.92 13.02 1.90 7.10 5.92 6.91
.......... ........... ................. ......... .............. .............
Total Solids.. Soluble Solids. Reds.. Non-Tannins.. Tannins.. Moisture..
OF
DOUGLAS FIR
Same sample analyzed by Jones one Year later 15.8 13.7 2.1 6.2 7.5 10.0
,oxidation, thereby increasing the tannin content b y 1.6 per cent (27 per cent of the former). The total solids are increased and the non-tannin decreased, each ~ . per g cent. P A R T I-CONIFERS W E S T E R N L A R C H (Larix 0ccidentaZis)-The sample was taken from four trees, averaging 28 in. in thickness, growing in the Whitman National Forest in Oregon. The trees were felled in December, 1910,and transported by skidding and log train t o the station. The cross-sections were cut 18 in. from the ground and showed a very wet and pitchy surface. The wood is dull red in color and the bark averaged z1/4in. in thickness. The sample was dried for one week a t 40' C. before extraction. The analyses were made in March, 1916. The tannin content of this, as well as of the other species tested, is given in Table 11. The wood extract was bright red in color and unimportant from the standpoint of either tannin content o r dye. The extract from the bark was a dull red and gave a pleasing light red color t o the tanned hide. Inasmuch as this species ranks ninth in t h e order of most prevalent trees in the states of Oregon and Washington, often occurring in pure forests in Northeastern Washington, Northern Idaho, and Northwestern Montana, is a prolific seeder and well adapted t o reforest burned-over areas, its importance as a source of tannin must not be overlooked. Investigations carried on in the U. S. Forest Products Laboratory also show' t h a t this species produces a high yield of alcohol by hydrolysis. Considering the high tannin content of its bark and t h e exceptional yield of alcohol, it would seem t h a t this species might be especially adapted t o chemical utilization, rather than for lumber and timber purposes.
TABLE11-TANNIN CONTENT OF PACIFIC COASTSPECIES Total Soluble PERCENTAGES: Solids Solids Western Wood(a) 28.9* 27.6* Bark(b) 20.5 17.0 Larch Western Wood(c) 19.6 16.9 Yellow Pine{Bark(c) 22.3 19.3 Western Wood(c) . . . . . . 4 . 2 4.1 Hemlock {Bsrk(c)....... 20.8 19.5 Wood 8.3 7.1 Dogwood Bark 8.3 7.9 Wood ........ 1 0 8 9.1 Cottonwood{Bark.. 25:3 24.3 Wood 3.5 3.3 Alder Bark 10.2 9.6 ( a ) Results are averages of 2 extractions. ( b ) Results are averages of 8 extractions. (c) Results are averages of 4 extractions.
...... .......
{ {
{
.... .......
......... ......... ....... ......... .........
NonTan- Tan- MoisReds nins nins ture None* 23.9* 6.7* 7.8* 3.5 6.4 10.6 11.0 2.7 8.0 8.9 3.1 8.3 10.9 4.0 0.1 3.1 1.0 9.5 1.3 8.57 10.93 9 . 8 1.2 5.8 1.3 5.0 0.4 6.2 1.7 5.0 1.7 7.3 1.8 6.5 1.0 19.6 4.7 2.0 0.2 2.6 0.7 6.8 0.6 6.2 3.3 9.8 Analysis by W. E. Gailey.
....
*
YELLO w PINE ( P i n u s ponderosa)-This sample was obtained from the Whitman National Forest, the crosssections being cut 18 in. in diameter. The bark averages 2 in. in thickness. The trees were felled 4 months before extraction. 1 THISJOURNAL,
7 (1915), 922.
I097
This species ranks second in quantity of the standing timber of Oregon and Washington. The bark on old trees is often to in. thick. The yellow pine is chiefly used in making boxes, but both the quantity of the bark and of the tannin in the bark and wood make the use of the sawmill waste of this species feasible for tannin extract manufacture. W E S T E R N HEMLOCK (Tsuga Heterophil1a)-Various published analyses of the western hemlock bark show t h a t the tannin content of the western species is equal t o or exceeds t h a t of the eastern species. The bark of t h e western hemlock has been regularly used b y tanneries on the Pacific Coast. Inasmuch as considerable second growth hemlock occurs on the logged off lands of the Northwest, a composite sample of four trees taken from the University campus was prepared. One tree was 18 in. in diameter and the other three averaged g in. The bark was in. thick. The trees were felled g months before extraction. The tannin content is less than t h a t found by other investigators, due probably t o the difference in t h e ages of the trees. PART 11-DECIDUOUS
SPECIES
(Cornus nuttal1i)-The western dogwood is essentially t h e same as t h a t of t h e East. The wood is a dull pink and the bark a reddish black. Its dye possibilities have been utilized by t h e Pacific Coast Indians for coloring clothing and moccasins deep red. It ranges from the southern coast of British Columbia through Washington, Oregon and California, usually thriving in low altitudes and damp soils and mixed singly or in groups under all the trees of t h e Pacific slope except the yellow and sugar pines. For the sample, cross-sections of five trees on the University campus were cut 3 ft. from the ground. They ranged from 8 t o 24 in. in diameter, and were air-dried for 6 months. The bark averaged less than '/z in. in thickness and was not easily peeled. The extract is reddish brown in color and imparts a light reddish brown color t o t h e tanned hide in the case of the wood extract and a deeper red from the bark. It is believed t h a t dogwood extract might be used with other tanning extracts t o furnish the dye property. C O T T O N W O O D (Populis trichocarpa)-The cottonwood tree is found growing in damp, sandy land from Alaska t o Mexico. It is a very rapid grower, a tree growing t o 1 5 in. in as many years. The sample was obtained from a tree z ft. in diameter, felled 3 months before extraction. The bark ranged from t o 3/4 in. in thickness. A L D E R (Ahzus 0regona)-This species occurs to a considerable extent as a second growth on the logged off lands of Oregon and Washington. Four trees were cut on the University campus g months before extraction. Cross-sections were taken about 2 f t . from t h e ground. The average diameter of the tree was I O in., the bark averaging '/a in. in thickness. The quality of the tannin is of doubtful value. DOGWOOD
CONCLUSIONS
I-Douglas fir slabwood, selected for tanning extraction, can advantageously be seasoned for one year.
1098
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
11-In addition t o t h e western hemlock bark, t h a t of the western larch and western pine seems t o yield extract satisfactory both in quantity and quality. 111-Of the deciduous species examined, the dogwood alone appears suitable for use in extract manufacture but only as a dye. LABORATORY OF INDUSTRIAL CHEMISTRY UNIVERSITY OF WASHINGTON SEATTLE, WASHINGTON
SATURATED HYDROCARBONS IN BASKING-SHARK LIVER OIL BY MITSUMARU TSUJIMOTO Received July 25, 1917
Basking-shark (Japanese: "Uba-zam6"), Cetorhinzis maximus (Gunner), ( C e f o r h i n i d a e ), is t h e largest of all living sharks and indeed of all fishes. It reaches a length of 14 meters and has an enormous weight. This shark is a dull and sluggish animal of the arctic seas, being occasionally found off the coast of Japan. The liver of basking-shark is very heavy, often weighing as much as I ton and yielding some j cwt. of oil.* Three authentic samples of basking-shark liver oil examined by the writer were pale yellow t o orangeyellow liquids of low specific gravity and contained 41.92-55.51 per cent of unsaponifiable matter. They contained squalene, C ~ O H S O a , highly unsaturated hydrocarbon found by t h e writer in certain Japanese shark liver oils.2 This is important as an instance showing t h a t the occurrence of squalene is not limited t o the liver oils from t h e sharks belonging to the family Squalidae. The experimental details concerning squalene will, as the writer hopes, be published shortly in another paper, together with the results of the experiments performed on the liver oils obtained from some twenty-three species of Japanese sharks. I n the present paper, the writer describes the remarkable occurrence of saturated hydrocarbons in a sample of basking-shark liver oil. Of the three samples of t h e oil above referred to, t h e one from t h e Province Hidachi had the following properties:
. .. ....... ................. ....... .......... .. ........... .. ....... ... .. ....... .......
Specific gravity at 1S0/4O C . . Acid value.. Saoonification-value ....... Toaine value (WijsJ Refractive index at 20° C . . ....... Butyto-refractometer at 20' C ... Unsapcmitiable matter.
0.8839 1.09 102.45 178.30 1,4773 78.2 41.92 per cent
The unsaponifiable matter was, for the most part, a liquid. One hundred grams of the oil were distilled under 5 mm. pressure. At the temperature of 170190' C. (the temperature of the oil bath, 250--270°C.), I O g. (or I O per cent) of a pale yellow liquid distilled over. It had the refractive index 1.4775 a t 20' C. This substance was, therefore, different from squalene. On changing the receiver and raising the temperature L. E. Andes, "Animal Fats and Oils." 1898, p. 194. *THISJOURNAL. 8 (19161, 889. Spinacene, a hydrocarbon found by I
C. Chapman
17.. Chcm. SOC.,66 (1917), 111], in a Portuguese shark liver oil appears to be identical with squalene. The writer has already prepared squalene hexahalogen hydrates of the general formula CaoHao6HX. which crystallize well and serve for the convenient identification of squalene.
Vol. 9, No.
12
of t h e oil bath t o 29c-315~C., a further distillate was obtained a t the boiling temperature of 244-260' C., amounting t o 2 5 g., or 2 5 per cent of the original oil. This latter distillate was found t o consist mainly of squalene. The first distillate was washed with aqueous NaOH solution t o remove free fatty acids, and then 8 g. of this purified substance were distilled under 13 mm. pressure. It distilled over a t 160-166' C., mainly a t 164' C. (the temperature of the oil bath, 2 0 0 - 2 0 5 ' C , ) , leaving a small residue. The distillate was a colorless, mobile liquid which did not solidify even when cooled t o below o o C. I t had the specific gravity 0.7868 a t 15' C., 0.781j a t 20' C.,0.7789 a t 2 8 ' C. (water a t 4' C. = I), and the refractive index 1.4398 a t 20' C. Its iodine value was determined t o be only 4.40, so that i t was nearly saturated. It dissolved in alcohol; when mixed with concentrated sulfuric acid, the acid layer turned brown, and on raising the temperature to about 100' C., the coloration became darker, b u t the distillate on the upper layer was not readily attacked. N o insoluble hydrochloride was formed b y passing dry HC1 into the well-cooled ethereal solution of the substance. When heated under 766 mm. pressure, the substance began to boil a t about 294' C., and mainly distilled over a t 296' C. The distillate thus obtained had the refractive index 1.439: a t z o o C.,which was nearly identical with t h a t of the original substance. 0.2011 g. original substance (distilled under 13 mm. pressure) gave 0.6223 COS and 0.2700 HzO: C
=
84.39 per cent; H
=
15.02 per cent.
Molecular weight determined by the depression of t h e freezing point of benzene solution, (I) 26j, ( 2 ) 252, mean 259. C 8 4 , 8 9 per cent CtrHsa requires.. ClsHai requires.. 84.96 CioH4o requires. . . . . . . . . . . . 84.98
........... ...........
H Mol. wt. 15.11 per cent 240.3 15.04 254.3 15.02 268.3
The substance was, therefore, a hydrocarbon or a mixture of hydrocarbons, the percentage of carbon and hydrogen amounting t o 99.4 per cent. Its elementary composition corresponded t o any of the above-mentioned three paraffins. It is not safe t o say t h a t t h e substance was a single chemical compound, b u t judging from.its boiling point and molecular weight, it appears to be very probable t h a t it consisted mainly of C18H38 or an octadecane. But as the normal octadecane is a solid a t the ordinary temperatures, the substance must be a t least an iso-octadecane. Its comparatively low boiling point may be attributed t o this cause. A quite similar relation has already been observed between solid triacontane and t h e hydrogenation product of squalene, C S ~ H ~a Zliquid , not solidifying even a t --zoo C.' Melting Boiling Point Point OC. OC. (mm.) 2 2 . 5 303 (760) 28 181.5 (15) 317 (760) Normal Nonadecane, ClpH40.. 32 193 (15) 330 (760)
COMPOUXD Normal Heptadecane. C1,Hse.. Normal Octadecane, ClaHa s...
.
1
LOG. cit.
Spec/Bc Gravity -/4'C. 0.7768 (22.5' C.) 0.7768 (28' c.) 0.7774 (32' C.)
