The difficulty of convergence

In Gibbs mode calculation, by the nature of the algorithm, it is sometimes difficult to achieve convergence particularly in the case that gas phase is about to appear but not yet.

For such critical cases, it is recommended to calculate in Helmholtz mode. Then you can tell whether the gas phase will appear or not by comparing the sum of activities (vapor pressure) with outer pressure. Note that in this calculation you should set the volume small enough for the condensed phase to be left some amount in Last condition. Anyway, in Gibbs mode calculation, gem will face theoretical difficulty when all species are zero in the gas phase. gem tries to evade such calculation but the evasion is not perfect, so it is better to inspect the situation by Helmholtz mode calculation.
If you encounter the case that the gas phase results zero in Gibbs mode while Helmholtz calculation predicts the phase should appear, please inform Kagaku Gijutsu-Sha of the case with its detail. We will keep it for future reference.

The same difficulty will be occurred for mixture phase in either Gibbs or Helmholtz mode.

One way is to increase Number for Too Many Rounds (Default is 100).
The other way is to fluctuate the initial conditions (Temperature, mol, P/V) . You would know what is happening around there.