sosman: problems of the oxides of iron 61 



tures of the ordinary forms of hematite and magnetite are not 

 thus transformable, then by geometry alone, without the aid of 

 chemistry, could we predict that there must be a break in the 

 solid-solution series between these two oxides? 



A third problem of the hematite-magnetite series concerns 

 the escaping tendency or fugacity of the oxygen which dis- 

 sociates from the oxide. What determines the equilibrium pres- 

 sure of this oxygen for a given composition of the solid solution? 

 It is interesting to note the peculiar form of the oxygen pressure 

 curve in this series. The middle portion is fairly flat, and the 

 pressure changes slowly with change of composition. As the 

 composition of magnetite is approached, however, the pressure 

 begins to fall rapidly, and, conversely, it rises rapidly as the 

 composition Fe 2 3 is approached. We have been able to show 9 

 that the dissociation pressure of the first small portion of oxygen 

 from pure Fe 2 3 at 1200° is of the order of magnitude of an 

 atmosphere of oxygen, although the pressure rapidly falls to a 

 few millimeters as dissociation progresses. 



From a theoretical standpoint there are two ways of looking 

 upon this form of dissociation curve. On the one hand, we may 

 consider that the curve is asymptotic to the axis of ordinates; 

 in other words, the initial composition of the crystal corresponds 

 exactly to the formula Fe 2 3 , and the dissociation pressure of 

 the first few atoms of oxygen is very high. These atoms come 

 from all parts of the crystal, and their removal brings into play 

 forces which greatly restrain the fugacity of all the remaining 

 oxygen atoms, causing a rapid fall in the dissociation pressure. 



On the other hand, we may consider, as Langmuir 10 and others 

 have shown, that there are free chemical forces at the surface 

 of the crystal, and that these forces hold atoms of oxygen "ad- 

 sorbed" on the surface when the crystal is exposed to air. The 

 initial composition then is Fe20 3 +* ; in which x is a very small 

 number. The fugacity of these surface atoms is high, but not 

 much higher than that of oxygen atoms just within the surface. 

 Oxygen will begin dissociating from interior parts of the crystal, 



9 Hostetter and Sosman. Jour. Am Chem. Soc, 38: 1188-1198. 1916. 

 10 Langmuir, I. Jour. Am. Chem. Soc, 38: 2221-2295. 1916. 



