f[cHADSKY] DECOMPOSITION OF POTASSIUM CHLORATE 21 



ing to accidents only a few of the bulbs were used long enough to show 

 this change. 



In subsequent experiments, bulbs containing pure sodium chloride 

 and small quantities of potash were heated to 500°. In each case the 

 turquoise blue manganates were formed and the bulb lost considerably in 

 weight. On the other hand bulbs heated to 560°, while a slow current 

 of pure oxygen was passed through them, did not show any increase in 

 weight. The latter experiment was suggested by the statement of Le- 

 'Chatelier in his work on high temperatures, that heated glass aibsorhs 

 small quantities of gases, but the negative result indicates that the ab- 

 sorption of free oxygen is too small to produce a change of weight in the 

 laulbs used. 



The first experiments were conducted at 355° and at 360° for the 

 purpose of determining whether or not it is possible to bring about the 

 formation of a quantity of perchlorate, without at the same time, induc- 

 ing an evolution of free oxygen. Mention may be made of results ob- 

 tained by Fowler and Grant, when silver oxide was heated with potas- 

 sium chlorate. (J. Chem. Sec, 1890). In accordance with these re- 

 sults, perchlorate may be formed at the expense of the oxygen of the 

 silver oxide, while no free oxygen is evolved. It seemed possible that a 

 similar result might be attained without the introduction of a foreign 

 oxide, for since potassium chlorate yields both chloride and perchlorate, 

 together with oxygen during its ordinary decompositions, it might be ex- 

 pected that by gentle heating all the oxygen at first set free from the 

 decomposing molecules would combine with other chlorate molecules. 

 Under such conditions the only products of the reaction would be 

 chloride and perchlorate. In all experiments conducted to test this 

 point, however, no evidence was found of any change without the simul- 

 taneous evolution of free oxygen. At 355°, a temperature just below the 

 melting point of the chlorate, and at 360°, where the chlorate was fused, 

 no loss of weight occurred after the lapse of two or three hours— nor 

 was there any change in the quantity of chlorate. Decomposition, in- 

 deed, was not found to begin below 410°, and at 420° it was so slow that 

 at the end of three hours only 0-48^ of the chlorate had been changed to 

 chloride. Since perchlorate is not decomposed below 500°, the test at 

 420° may be considered one in which the changes taking place are suffi- 

 ciently slow to permit the reaction under discussion if it were a possible 

 one. It was found, however, that only a small percentage of perchlorate 

 was formed, while about 85^ of the oxygen from decomposed chlorate 

 was evolved as free gas. 



