Disordered crystal structure after packing with aqueous phase

Dear all

I want to simulate the interaction between aqueous phase and the surface of a crystal with layered structure. If I run MD on the crystal alone, the crystal still has layered structure after MD. But If I build a aqueous-solid system by using "build layers", I can only run geometry optimization with "conjugate gradient", and still get a disordered soild phase. If I build the system with "amorphous cell->packing", the system becomes just a little bit disordered after geometry optimization, but becomes more disordered after MD. The charge of the system is zero. I ran simulation on similar aqueous-solid system before, there was no such instability. In this system, there are water molecules in the interlayers of crystal, is this the cause of the disordered crystal after packing with aqueous phase?

Regards,

Jingjing