2D Slab Modeling Without Vacuum and Region-Based QM/MM Setup

[Elham M.Goliaei]

Dear CP2K Developers and Users,

I am writing to ask two questions related to CP2K simulations:

1. 2D Slab Without Vacuum
   I am interested in modeling a 2D slab of liquid without introducing vacuum along the non-periodic direction (e.g., the z-axis). I have already set the Poisson solver to PERIODIC XY and changed the cell periodicity from XYZ to XY. My question is:
   Are these two settings indeed sufficient to correctly model a 2D slab without any vacuum in the non-periodic direction, or is additional setup required?

2. Region-Based QM/MM Partitioning
   Is it possible in CP2K to define the QM, QM/MM and MM regions based on spatial regions (e.g., z-axis areas) rather than using fixed atom indices? This would be especially helpful for systems like liquids, where atoms move across layers during simulation.

Thank you, 

Elham

[Marcella Iannuzzi]

Hi Elham, 

1 Even with PERIODIC XY, the simulation cell has to be defined such that sufficient vacuum space is present to ensure that the charge density goes to zero at the boundary and the  poisson solver correctly works. 

2 No  it is not possible to define the regions as volume partitions. To account for switches of molecule-type between QM and MM one should use adaptive QM/MM :

Mones L, Jones A, Götz AW, Laino T, Walker RC, Leimkuhler B, Csányi G, Bernstein N. The adaptive buffered force QM/MM method in the CP2K and AMBER software packages. J Comput Chem. 2015 Apr 5;36(9):633-48. doi: 10.1002/jcc.23839. Epub 2015 Feb 3. PMID: 25649827; PMCID: PMC4351341.

Regards

Marcella