In principle, ATK can do such simulation because the molecular dynamics for several types of ensemble have been already implemented in ATK. (refer to
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.moleculardynamics.html)
You have to first find out the key parameters used in the simulation of that paper.
1). The supercell size;
2). The number of water molecules used in their supercell;
3). The ensemble method;
4). The temperature, time step, number of simulation steps in the MD simulations.
In the initial configuration, you can randomly put a couple of water molecules (the number of water molecule should be close to that in the reference you mentioned) close to the calcite surface. And then do a NVT simulation with sufficient steps to achieve an equilibrium state.
"
Can ATK then calculate:
Total energy of calcite with n adsorbed water molecules
Total energy of calcite in vacuum
Number of adsorbed water molecules
Total energy of isolated water molecule
"
It is sure that ATK can do.
"how do I build Mg 2+ and SO4 2- and how do I substitute one Ca2+ ion with an Mg2+ ion? And how do I substitute one CO3- ion with an SO42- ion? And then can ATK calculate the energies?"
Put MgSO4 molecule (i.e., Mg, S, and O atoms) into the supercell.
Replace Ca atom with Mg atom. CO3---> SO4.
Because the whole system is neutral, it is no necessary to specify whether the species are cation or anion ions. The charge distribution around atoms will be automatically determined during the self-consistent calculations.
Yes, it can do.
