What does it REALLY mean when Perple_X gives you feo, or na2o, or k2o, or 
any other of these species as "stable" in a phase diagram?


Entities named as simple oxides in perplex data bases are (usually) fictive phases 
that serve as place holders for some, now unusual, types of calculations. 
If you get such a fictive phase in a calculation the result is invalid. 
there are two possible causes:

1) The arbitrary properties of the fictive entity make it stable, in which case if you 
increase the free energy of the entity by editing the thermodynamic data file it should 
disappear; alternatively you can simply exclude the entity or delete it from the file.

2) For vertex to give any result it is essential that the bulk composition of interest 
can be defined as a positive linear combination of the phases in the data base. However, 
most data bases do not include phases necessary to describe, e.g., extremely sodic bulk 
compositions. This limitation means that if you specify an extremely sodic composition 
vertex will use the fictive phase na2o regardless of its thermodynamic properties (i.e., na2o is 
required for mass balance). for such a composition if you have eliminated na2o, vertex 
will fail and give you some message that it can't define the bulk composition of interest.