Hi,
In this thread I like to put some details concerning the input file, especially what you never should do! This I hope will help to prevent errors which can take long time to figure out. Of course feel free to put your comments or own experiences!
Bernd
---
original message from Bernd
What you never should do!
Re: What you never should do!
Hi,
although it seems reasonable, if you use a thermodynamic .GES file with more components than you need, of if you just for getting started want to disregard a certain component, just to set it's composition to 0:
It is not possible (or at least it is dangerous) to include a component which has the concentration 0. A composition of 0 is only
allowed for stoichiometric phases, and then the corresponding composition needs to be specified for another phase which has solubility for this phase.
On the other hand
it is also not realistic to have a composition of 0 because this would produce an extremely low entropy and therefore an astronomically high driving force for dissolution of such a
component.
A realistic way is to specify the composition according to the level of accuracy of the concentration measurement in an experiment or just to choose it such small that it will
not hurt. But be careful because the value nevertheless can be important for numerical reasons and can affect performance!
---
original message from Bernd
although it seems reasonable, if you use a thermodynamic .GES file with more components than you need, of if you just for getting started want to disregard a certain component, just to set it's composition to 0:
It is not possible (or at least it is dangerous) to include a component which has the concentration 0. A composition of 0 is only
allowed for stoichiometric phases, and then the corresponding composition needs to be specified for another phase which has solubility for this phase.
On the other hand
it is also not realistic to have a composition of 0 because this would produce an extremely low entropy and therefore an astronomically high driving force for dissolution of such a
component.
A realistic way is to specify the composition according to the level of accuracy of the concentration measurement in an experiment or just to choose it such small that it will
not hurt. But be careful because the value nevertheless can be important for numerical reasons and can affect performance!
---
original message from Bernd
Re: What you never should do!
Hi,
another thing which cannot work is defining phase interactions between two phases, which are both stoichiometric for the same composition. This is not possible because no partitioning within this element can be done. It is also like that if you set the phases stoichiometric for purely numerical reasons (e.g. if there is demixing).
Thus you must define 'no_phase_interaction' for the corresponding phase pair in the phase interaction input data!
Bernd
---
original message from Bernd
another thing which cannot work is defining phase interactions between two phases, which are both stoichiometric for the same composition. This is not possible because no partitioning within this element can be done. It is also like that if you set the phases stoichiometric for purely numerical reasons (e.g. if there is demixing).
Thus you must define 'no_phase_interaction' for the corresponding phase pair in the phase interaction input data!
Bernd
---
original message from Bernd