146 Mixter — Heat of Formation of the 



sodium as follows : -J.') grams of vanadium trioxide and 15 

 grams of sodium were heated in a heavy hrass cylinder closed 

 with a screw plug. The reaction was very violent and the 

 plug was partly melted. The metallic sodium remaining was 

 removed by alcohol and the residue left was washed with 

 water, alcohol and ether and dried in vacuo. It weighed 15 

 grains. Next, this 15 grams and sodium were heated in an 

 iron cylinder until sodium nearly ceased to burn about the 

 screw cap. On opening the cold cylinder the contents ignited 

 on top and it was at once placed in alcohol. But little hydro- 

 gen came off. The black powder obtained was washed with 

 water, digested with dilute hot hydrochloric acid which dis- 

 solved some of it, and finally washed with water and dried. 

 It was found on analysis to be almost pure vanadium trioxide. 

 Undoubtedly sodium reduces vanadium trioxide, but is not as 

 well adapted for the reduction as potassium, which does not 

 react as violently probably because the latter metal distills at a 

 lower temperature than sodium. 



The following preparations were made in an iron cylinder 

 of little greater capacity than the bulk of the mixture. Potas- 

 sium was placed in the bottom of the cylinder and vana- 

 dium trioxide in a bulky powder on top except in one case 

 mentioned later. The cylinder was kept at a red heat until 

 potassium ceased to burn about the screw cap, and then quickly 

 cooled in water. The black product was digested and washed 

 with cold water and dried in vacuo over sulphuric acid. A 

 weighed portion of the product was converted into the 

 pentoxide for the determination of the vanadium. 



Preparation "H. " — 7 grams of vanadium trioxide and 5 

 grams of potassium. Product, 2 grams, containing 75*9 per 

 cent of vanadium ; V 2 2 has 76 - l per cent. 



Preparation " I. " — Vanadium trioxide, 12 grams pressed 

 down in the cylinder, potassium 7 grams ; yield 7 grams. 

 Vanadium content 72*5 per cent. Equal molecules of "V 2 2 

 and V 2 3 contain 72*1 per cent. A determination of the heat 

 of reaction of " I " with sodium peroxide showed that the 

 oxides were not combined. 



Preparation " J. " — 25 grams vanadium trioxide and 15 

 grams of potassium. The cylinder unfortunately was imper- 

 fect and opened when cooled under water and the product was 

 mixed with iron scale. The black powder which was sepa- 

 rated from it by floating yielded on analysis 76"3 per cent of 

 vanadium including a trace of iron. On dissolving in hydro- 

 chloric acid a little hydrogen came off. 



Preparation " K. " — 34 grams of vanadium trioxide and 23 

 grams of potassium. There was a small hole in the top of the 

 cylinder which was closed with a plug when potassium ceased 



