Nov., 1921
T H E JOURNAL OF INDUSTRIAL A N D ENGINEER~NGCHEMISTRY
way, 825 cc. of distillate were obtained, whichcontainedabout 28 g. of furfural. The cellulosic material was filtered by pressure and washed with 2000 cc of water, and again pressed. The filtrate and wash waters from t,his second kilo of cobs were mixed and concentrated to their original volume of 3175 cc. (the 4000 cc. minus 825 cc. blown off when the pressure was released). This solution, with the 2650 cc. obtained from the first kilo of cobs, contained the carbohydrate material from 2 kg. of cobs. Forty-five hundred cc. of this solution, which represented the soluble extracts from 1540 g. of cobs, were heated in the autoclave a t 180" to 1 8 5 O . After 15 min. the autoclave was blown off through the condenser as before. The amount of condensate was noted and an equal volume of water was added to the contents of the digester and it was again heated for 15 min. a t 180" to 185". This was then blown off, the volume of the condensate noted, an equal volume of water added to the digester, and the heating repeated. This was carried out five times, after which so little furfural was being produced that it was not worth while to carry the process further.l The following table denotes the results obtained by five successive heatings of 4500 cc. of soluble
1025
extractives obtained by the method described above. Furfural Furfural Furfural in in ConG. in 5 Cc. Condensate densate (Av.) Grams Percent 0.1454 21.8 2.91
No. 1
Condensate Cc. 750
Phloroglucidc G. from 5 Cc. (a)0.2770 (b10.2738
2
775
(a)0.2610 (bl0.2625
0.1383
20.9
2.70
3
755
(a)0.2137
0.1146
17.3
2.29
(b)0.2180 4
725
(a)0.1310 ( b )0.1240
0.0691
10.2
1.38
5
725
(a)0.0752 ( b ) 0 0760
0.0420
6.1
0.84
If it is desired merely to prepare furfural and not to save the cellulose, the process could be made continuous from the start, following the general procedure just described for the hydrolysis of the soluble carbohydrate extractives. Under these conditions the cellulosic material would be allowed to remain in the digester during the complete process. It should be emphasized, however, that the cellulose may have as great value as the furfural in technical operations and it is of great importance to preserve it from injury by overheating.
The Color Value of a Tan Liquor as a Function of the Hydrogen-Ion Concentration"'" By John Arthur Wilson and Erwin J. Kern LABORATORIES OF A. I?. GALLUN & SONS Co., MILWAUKBE, WISCONSIN
In judging the value of a vegetable tanning extract, tanners often place much weight upon the color imparted to skin by solirtioiis of the extract. This has given rise to a movement to try to devise some satisfactory method for measuring the color value of an extract. In Europe the tendency has been toward matching a solution of the extract with standard colored glasses, but apparently the preparation of absolutely standardized glasses is a serious problem. Moreover, the color of a tan liquor is not necessarily a measure of the color it will give to a skin. I n this country preference seems to have been given to a method involving the observation of the color of strips of sheep or calf skin tanned in a standard solution of the extract. A strip of tanned skin accompanying the analysis report is apt to find greater favor with the tanner than figures on an arbitrary scale, but this method has sometimes been deceptive, because a given tan liquor may impart different colors to samples of skin from different sources. The present paper does not deal with the question of measuring color values, but with factors powerfully affecting such measurements which in the past have been uncontrolled and apparently often quite unappreciated. It has long been known that the presence of lime in a tanned skin will cause a darkening of the color. If the skin is treated with mineral acid soon after tanning, the color brightens, but the effect of acid becomes less pronounced the longer the skin is previously allowed to stand exposed to air. I n our studies on the effect of change of hydrogen-ion concentration upon tanning, we cbtained leather differently colored for each different pH value of the tm liquor. It was immediately apparent that the color of a tan liquor as well as that of the leather is a function of the hydrogen-ion concentration. 1 I n large-scale practice the process may be made continuous by keeping the temperature constant and maintaining a constant level in the digester by continuous addition of water or steam. 9 Presented before the Section of Leather Chemistry at the 62nd Meeting of the American Chemical Society, New York, N. Y ,September 6 to IO, 1921. a Received August 22, 1921.
INDICATOR EFFECT Two tan liquors were prepared, one from gambier and the other from quebracho extract. TOeach was added sufficient phosphoric acid to hring the p H value to 2.5, as determined by the hydrogen electrode. The phosphoric acid was added to act as a buffer in preventing large changes in pH value upon long standing. To equal portions of each, sodium hydroxide was added to give series of tan liquors ranging in pH value from 3.0 to 12.0 and all having a tannin content of 1 per cent, as determined by the authors' new method.1 The gambier series varied in color from light straw a t p H = 3.0 to a very deep red a t 12.0. The quebracho series was similar in color excepting that the liquors of lower p H value had a touch of violet. Either series resembled a standard series of colors used in the indicator method of determining hydrogen-ion concentration, except for the fact that a light precipitate formed in all liquors having a pH value of 4.0 or less. The difference in color was evidently a true indicator effect, for any member of the series could be made to match any other member simply by bringing it to the same pH value. All members of either series appeared practically identical when brought to a pR value of 3.0. This complete reversibility of color change, however, was not found when liquors a t higher pH values were allowed to stand long exposed to air. OXIDATION EFFECTS Two complete series of each extract were poured into test tubes; the tubes of one series of each were tightly stoppered, while the others were left uncovered. Next day the liquors in the stoppered tubes showed practically no change, but the others had become darker in color, the more so the higher the pH value. When the liquors in a series not exposed to air were all brought to a pH value of 3.0, they all assumed practically the same color. But when those of a series that had been exposed to air were all brought to 3.0, they did not assume the same color, but were darker the higher the pH I
THISJOURNAL, 12 (1920). 465, 1149,
1026
THE JOURNAL
OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY
Vol. 13, No. 11
keeping it exposed to air when its pH value is greater than 10 apparently prevents its precipitation when brought to 3; all such liquors remained brilliantly clear. The addition of a great excess of acid, however, caused all liquors to precipitate, while any precipitate could be completely redissolved by the addition of sufficient alkali. Another interesting fact is that the liquors exposed to air when their pH values lay between 8 and 9 gave much trouble with the hydrogen electrode. After bubbling hydrogen through them for only a few minutes, the voltage would fall rapidly towards zero. Even when brought to a pH value of 3.0, the liquors still gave this trouble and we had to check the results by meane of indicators. No such trouble was encountered with liquors exposed to air a t pH values below 7 or above 10. Apparently pH = 9 is a critical point in the oxidation of tan liquors. It may prove valuable to make much more extended studies of the changes going on at this point. EFFECT UPON LEATHER The changes in color of the tan liquors parallel in a rough way the changes in color imparted to hide. A hide tanned in a liquor having a high pH value not only comes from the liquor very dark in color, but this color continues to darken upon exposure to air. SUMMARY The color value of a tan liquor depends upon its hydrogenion concentration when used. A change in pH value produces a change in color of both liquor and leather. Tan liquors change in color like indicators with change in pH value, but over a range from 3 to 12. This change in color is completely reversible, if the liquors are not long exposed to air. Liquors exposed to air continue to darken in color, the more so the higher the pH value, but this change is not reversed by lowering the pH value. Liquors exposed to air a t pH values of about,9 give bulky precipitates when their pH values a,re brought to 3, and such liquors tend to poison the hydrogen electrode.
5 6 7 a g i o i i PI! Value during 3 Days Esrposnre to Air FIO.I-SHOWINQ How TENDENCY OF A TANLIQUOR TO FORM A PRECIPI4
TATE W R E N BROUQET TO A V A L U E DURING A PERXOD OF
pH
V A L U E OF
3
V A R I E S WITH I T S
pH
EXPOSURE TO AIR
value-during the period of exposure to air; furthermore a precipitate settled out from those whose pH values had been in the vicinity of 9. This precipitate formation is very curious. A complete series of each extract was allowed to stand exposed to air in shallow dishes for 3 days; the liqucrs were then made up to original volume and poured into 100-cc. graduate cylinders. Each was brought to a pH value of 3.0 by the addition of hydrochloric acid and allowed to stand over night. Next day the volume of precipitate from 100 cc. of original liquor was read from each cylinder. The resnlts are shown in Fig. 1. Keeping a solution of either extract exposed to air while its pH value is 9 causes it to yield an enormous precipitate when its pH value is subsequently brought to 3.0. But
The Tannin Content of Pacific Coast Conifers' By R. H.Clark and H. I. Andrews DI~PARTMENT O F CHEMISTKY,
UNIVERSITY OF
This investigation was made to determine how the tannin content of western hemlock (Tsuga heterophilla) and spruce (Sitka) bark varies with the month of the year in which the tree is cut. The samples of bark were taken from trees, either standing or just felled. I n no case had the logs been in the water. Samples for nine months of the year were analyzed for total solids, solubles, non-solubles, tannins, nontannins, and moisture. The bark came from Kingcome Inlet, B. C., from the vicinity of Kingcome River. As nearly average a sample as possible was sent from the woods.2 The samples for the months of May and June were held about 2 mo., in stoppered bottles, before the analyses were made; the effect of such seasoning might possibly have raised the tannin content slightly, as the effect of seasoning All douglas fir was found by Benson and Jones3 to be. values recorded are the average of two separate extractions. The extraction and analyses were carried out by the standard methods of the International Association of Leather Trades' Chemists, as described by Lunge and K e s ~ n e . ~ Received June 6, 1921. We wish t o thank the Powell River Co., Ltd., for their kindnessin providing the samples. a THIS JOUKXAL, 9 (1917),1096. 4 "Technical Methods of Analysis," Part I, Vol. 111. 1
2
BRITISHCOLUMBIA, VANCOUVER,
CANADA
The average sample shipped from the woods weighed about 4 Ibs. This was resampled, and about 300 g. were ground up in an ordinary meat chopper until it all passed through a 20-mesh sieve. I n most cases 25 g., in some 20 g., of the ground bark were extracted, as this amount was found t o give the required 3.5 to 4.5 g. of tanning material per liter. The extract from all the bark was quite clear and remained so for some time, with the exception of a few spruce samples, which deposited a very slight precipitate. The total solids were determined by drying 100 cc. of the extract a t 100'. The total solubles were found by evaporating 100 cc. of the extract which had been filtered through a Berkfeld filter until the filtrate was clear by the reflected and transmitted light of an electric lamp. The first 200 to 300 cc. of filtrate to pass through the filter was rejected, to avoid any correction for absorption. The tannin was d e termined by shaking the extract in a mechanical shaker with hide powder, previously tested and chromed. A clear solution of the non-tannins was left, which was evaporated, and the non-tannins weighed. The tannin absorbed by the hide powder is given by the difference between the total solublea and the non-tannins. I n the case of hemlock, the tannin content rises quite
