Energy Drift - Cations

[isaaraujo]

Dear Dr. Dubbeldam,

I am simulating adsorption of water and salt on a clay slit pore using MD simulations. First I minimized the clay structure + cations in the NPT ensemble using the Baker minimization method. Subsequently I equilibrated the structure in the NPT ensemble. However I notice that during the equilibration, the energy drifts increase gradually at each step. For structures that do not have cations as extra framework, I don't observe this issue. Is there a way to address this problem in RASPA?
I have another question regarding energy drifts in the simulation. For another system that I am simulating and that does not contain any cations I can reach an energy drift around 1x10^-5 during equilibration. However, when production steps start, the energy drift started gradually increasing. What could be a reason for that? 

SimulationType                   MolecularDynamics
NumberOfCycles                         20000
NumberOfInitializationCycles     20000
NumberOfEquilibrationCycles      500000
RestartFile                      yes
PrintEvery      10000
ContinueAfterCrash            no
WriteBinaryRestartFileEvery   2000
Ensemble          NPT

CutOff                           9.0
Movies yes
WriteMoviesEvery 20000
PrintEvery       25000
PrintPropertiesEvery 25000

TimeStep          0.0005
Forcefield                       CLAYFF
RemoveAtomNumberCodeFromLabel yes

Framework 0
FrameworkName superillite_complete20
UnitCells       1 1 1
ExternalTemperature 330.0
ComputeMolecularPressure yes

FlexibleFramework yes
FrameworkDefinitions CLAYFFLEXIBLE
ExternalPressure 5000000.0

Component 0 MoleculeName                 kalium
       MoleculeDefinition           Cations
       TranslationProbability       1.0
       RandomTranslationProbability 1.0
       ExtraFrameworkMolecule       yes
       CreateNumberOfMolecules      0

Thank you
Isa