ETHYL ALCOHOL FROM WOOD WASTE. 17 



measured by means of a gauge protected from the acid vapors, and 

 the temperature was shown by a recording thermometer, the bulb 

 of which projected into the sawdust. The digester was filled and 

 emptied through a pair of concentric manholes in the inner and outer 

 shells. The usual procedure was to make the charge of sawdust 

 equivalent to about 100 pounds dry weight. The exact weight and 

 moisture-content were recorded. The dilute acid was then added, 

 the manhole covers were bolted on, steam was admitted, and rotation 

 was begun. Before the temperature reached 100° C, the air in the 

 inner shell and in the space between the two shells was vented to 

 get a more accurate gauge reading. The admission of steam was con- 

 tinued until the desired pressure was reached and then regulated so 

 that the heating period was always 20 minutes, or as near that length 

 of time as possible. The steam was then throttled to maintain the 

 desired pressure for the necessary time. 



At the completion of the reaction (or in cooks of 15 minutes or more, 

 2 or 3 minutes before the time was up) the rotation was stopped 

 and the vapors were blown off and condensed as rapidly as possible. 

 The time of blow-off varied somewhat, depending on the pressure at 

 which the cook was made and the amount of material in the digester. 

 The condensing and cooling capacity of the coil, however, was not 

 equal to the demands made upon it, so that blowing-off the digester 

 took much longer than it should have done — about 1\ hours from 

 7 or 8 atmospheres to atmospheric pressure. 



The condensed blow-off was weighed and analyzed for volatile 

 acid. The condensation from the steam between the two shells was 

 drained out and weighed. It was also tested qualitatively for dex- 

 trose to detect any leakage through stuffing boxes or flanges from 

 the inner shell. Whenever liquor was present in the inner shell, it 

 was drained out through the blow-off valve, after which the digester 

 was rotated so that the manholes were at the bottom, and the saw- 

 dust was raked out. After the preliminary series of experiments, 

 this material was centrifuged. The digester liquor, centrifugal 

 liquor, and treated sawdust were first weighed and then analyzed 

 for acidity, total solids, dextrose, etc. 



METHODS OF ANALYSIS. 



The following methods of analysis were used: 



MOISTURE. 



About 15 grams of air-dry sawdust or 80 grams of digested or ex- 

 tracted sawdust were weighed into tared crystallizing dishes of glass 

 and dried over night in an oven at 105° C. Although the digested 

 sawdust samples at times charred somewhat, a comparative series in 

 54976°— 22— Bull. 983 2 



