﻿ACTION OF SODIUM ON ACETONE. 73 



derivative. Over four hours were consumed in dissolving 0.5 gram 

 of sodium in 5.5 grams of the alcohol. The precipitate was filtered, 

 dried in a current of hydrogen and dissolved in dilute acid, a considerable 

 quantity of condensation products of acetone separating. 0.74 gram of 

 the sodium derivative gave 1.067 grams sodium sulphate or 46.8 per cent 

 sodium. It seems evident, therefore, that sodium acting on diacetone 

 alcohol really does give sodium hydroxide and the condensation products 

 of acetone; the reaction is, however, so radically different from the 

 action of sodium on acetone, and the condensation products so prominent 

 (there are scarcely any when pure sodium acetone is dissolved in dilute 

 acids) that the theory that diacetone alcohol is first formed, subsequently 

 to he acted on by sodium, must be abandoned. There remains only the 

 question then of whether sodium acetone could not react with acetone 

 to give the sodium derivative of diacetone alcohol, although in this event 

 the percentage of sodium found should be very much less than it really 

 is. However, apart from this, in the absence of condensation products 

 of acetone on acidifying, it must be presumed, to account for the acetone 

 which is always found in the acidified solutions of sodium acetone, that 

 the diacetone alcohol would promptly be reconverted into acetone during 

 this process. Rigidly to exclude this supposition we performed the 

 following experiment. 



Experiment 8. — Four grams of diacetone alcohol were dissolved in 

 20 cubic centimeters of ethyl alcohol and 20 cubic centimeters of water. 

 The solution was saturated with sodium acetate and 4 grams of semicar- 

 bazid hydrochloride were added. The solution was then rendered very 

 faintly alkaline and placed in ice-water for one hour, being vigorously 

 scratched with a glass rod. Xo precipitate appeared. To be sure that 

 the conditions were correct for the formation of a semicarbazone, the 

 solution was divided into two equal portions. To portion number two, 

 2 grams of acetone were added, whereupon a precipitate of crystalline 

 needles immediately appeared. Four grams of semicarbazid hydrochloride 

 were now added to this portion of the original solution, so that a suf- 

 ficiency of the reagent should be present to precipitate all acetone and all 

 diacetone alcohol, should it form a semicarbazone. Both the original, 

 unchanged portion and the second one were kept in the ice chest over 

 night. In the morning, portion number one, to which no acetone had 

 been added, contained a very faint precipitate (0.01 gram melting at 

 185° to 186°, it being acetone semicarbazone) so that a very small amount 

 of acetone had been separated by hydrolysis. From portion number two, 

 2.5 grams of acetone semicarbazone were isolated, melting at 1S6°. 



From the above residt it is evident that the acetone semicarbazone 

 obtained in the preceding experiments could not have originated in the 

 sodium derivative of diacetone alcohol as the hydrolysis of the latter 

 body is a time reaction which proceeds slowl} r . 



