Dear Gromacs users,
I am performing a simulation setup consisting of water and zeolite MFI. The simulation setup is shown here in this post. When I run nvt and npt I encounter with an error which is attached hereto. I guess it would be pbc issue but I don’t know how to handle it. I defined particles of 1 and 1954 in the simulation box in blue VDW form.
I would appreciate it if you help me with that.
I would guess that you had a zeolite system with a bond across the periodic boundary. Then you increased the system size and filled that with water. Is that remotely correct? When you increase the system size, the bond across the periodic boundary will be across half the system.
Before increasing the system size you can use gmx trjconv with the -pbc whole option. Hopefully that fixes your problem.
Dear Magnus,
That is correct. I made two boxes of water and zeolite and put it in one box.
Thank you very much
I will test it.
Dear Magnus,
Even with pbc=whole, the same error is shown.
What do you think about this problem? It’s a bit strange!
Thanks
If all molecules are whole before changing the box size it sounds strange, indeed. Can you post VMD snapshots (showing the periodic box, as above) of the systems like this:
-pbc whole-pbc wholeOr is this a molecule that is periodic across the periodic boundary? In that case you will not be able to increase the box size to add water at all. There is no information about the interface of the molecule.
Dear Magnus,
I made the zeolite using the MFI unit cell and expanded it into 3 directions. It should be noted that a predefined box fitting the expanded zeolite exists in the PDB file of the final zeolite after expansion (Figure 1).
1- I performed a short nvt and npt simulation of pure zeolite first with the predefined box size.
2- in the next step I made a water box exactly equal to the Zeolite box. (Figure 2)
3- Then I enlarged the water and zeolite box size and merged the .gro files to make the final box (Figure 3,4).
4- Figure 5 is the zeolite using trjconve before -pbc whole
5- Figure 6 is the zeolite using trjconve after-pbc whole
6- Figure 7 is the mix after -pbc whole. I dont know why the zeolite is a bit out of box after ``pbc whole`.
7- the mixture without -pbc whole is shown previously which is completely fit in the box.
Thanks alot for your help and consideration.
See my comment above (Non bonded interaction is larger than table limit? - #6 by MagnusL). The zeolite structure is periodic, isn’t it? In that case you cannot change the box size.
Yes, it is periodic.
Then, what can I do about that?
To be honest, I have been working on polymers for many years and this is the first time I am working on peridic structures like zeolite. After many efforts and finally your hint, I understood the main reason for the range of errors in this matter (the periodic structure of zeolite). And I can not increase the box size!
Previously I just increased the zeolite box and used gmx solvate to add water to it but I faced other instability problems, too (shown here).
How can I fix this issue?
What is the procedure for adding water to zeolite?
Honestly, I don’t know. The thing with periodic molecules is that they repeat infinitely. The water exposed surface might even have a slightly different structure.
Would there be a way for you to make the zeolite structure non-periodic in one dimension? I.e., represent the whole structure of it in that dimension.
Did your turn on the periodic-molecules mdp option?
Dear Hess,
Yes I did. Here is my nvt.mdp.
;title = OPLS Zeolite NVT equilibration
define = -DFLEXIBLE; -DPOSRES ; position restrain the protein
; Run parameters
integrator = md ; leap-frog integrator
nsteps = 5000000 ; 2 * 50000 = 100 ps
dt = 0.001 ; 2 fs
; Output control
nstxout = 10000 ; save coordinates every 1.0 ps
nstvout = 10000 ; save velocities every 1.0 ps
nstenergy = 10000 ; save energies every 1.0 ps
nstlog = 10000 ; update log file every 1.0 ps
; Bond parameters
continuation = no ; first dynamics run
constraint_algorithm = lincs ; holonomic constraints
constraints = h-bonds ; bonds involving H are constrained
lincs_iter = 1 ; accuracy of LINCS
lincs_order = 4 ; also related to accuracy
; Nonbonded settings
cutoff-scheme = Verlet ; Buffered neighbor searching
ns_type = grid ; search neighboring grid cells
nstlist = 10 ; 20 fs, largely irrelevant with Verlet
rcoulomb = 1.0 ; short-range electrostatic cutoff (in nm)
rvdw = 1.0 ; short-range van der Waals cutoff (in nm)
DispCorr = EnerPres ; account for cut-off vdW scheme
; Electrostatics
coulombtype = PME ; Particle Mesh Ewald for long-range electrostatics
pme_order = 4 ; cubic interpolation
fourierspacing = 0.16 ; grid spacing for FFT
; Temperature coupling is on
tcoupl = V-rescale ; modified Berendsen thermostat
tc-grps = system ; two coupling groups - more accurate
tau_t = 0.1 ; time constant, in ps
ref_t = 300 ; reference temperature, one for each group, in K
; Pressure coupling is off
pcoupl = no ; no pressure coupling in NVT
; Periodic boundary conditions
pbc = xyz ; 3-D PBC
; Velocity generation
gen_vel = yes ; assign velocities from Maxwell distribution
gen_temp = 300 ; temperature for Maxwell distribution
gen_seed = -1 ; generate a random seed
periodic-molecules = yes
I must check it. Honestly, I don’t know.
Thank you dear Magnus
My question was only relevant for analyzing the error behavior. You should never get a segfault.
But your fundamental problem is what Magnus wrote. In your system with water your molecule should no longer be periodic in one direction. So you need to break the periodic bonds there and put in a proper termination at the interface. I’m not an expert on zerolites, so I don’t know how that you look like. For e.g. silica and metals there are often O and OH groups at the surface.
Yes, I think so.
The main problem is that there is a periodic box. Zeolites are made of Si and O in (Si :O = 1:2) form in the structure and should be in the form of SiO4. There is no specific termination for that in all-silica form. What I am looking for.
However, Thanks again for your help.
I would appreciate it if to tell me what to do if you find any solution in the future.
Mohammad