[gmx-users] Need help creating md.mdp for silica(quartz)?

[Ekatherina O]

I have done simulation water in nanopore based on quartz (I use model 
ClayFF) using Gromacs 5.1.5 and have observed strange behavior. (Here is 
a link to the pictures at the initial time (1.png) and in the modeling 
process.
https://drive.google.com/drive/folders/1GpIyx8___kQ3Zbe36wkSSINxgtKwIWsn?usp=sharing)
Its NVT simulation (see .mdp below) and neutral charge on surface (have 
silanol groups), system is overall neutral.
1) Did I choose the right settings?  What do these atomic displacements 
(lines in the pictures) mean? What restrictions should be imposed on the 
system?
2) In addition, I noticed that all the atoms of the quartz walls that 
are in the vicinity of the box walls move from the upper wall to the 
lower and lower to the upper. How can I avoid this when periodic 
boundary conditions are imposed? Do I understand correctly that this 
involves interaction between the quartz walls?

integrator          =  md-vv
dt                  =  0.001		; = 1 fs
nsteps              =  500000		; = 100  ps
nstcomm             =  1
; OUTPUT CONTROL OPTIONS
nstxout             =  200
nstvout             =  200
nstfout             =  200
nstlog              =  200
nstenergy           =  200
energygrps	    =  System
cutoff-scheme            = Verlet
; NEIGHBORSEARCHING PARAMETERS
;periodic-molecules  =  no
nstlist             =  20
ns_type             =  grid
pbc		    =  xyz
rlist               =  1.4
; OPTIONS FOR ELECTROSTATICS AND VDW
coulombtype 	    =  PME 	; fourierspacing ; pme-order
;		    =  PPPM 	; fourierspacing
;		    =  Ewald	; No parallel Ewald.
rcoulomb            =  1.4
rvdw                =  1.4
; Ewald, PME, PPPM options
;fourierspacing      =  0.12
;fourier_nx 	     =  0 	; 2 3 5 7
;fourier_ny 	=	0
;fourier_nz	=	0
;pme_order	    =  6	; 10 ; 8 ; default = 4
optimize_fft	    =  yes
; OPTIONS FOR WEAK COUPLING ALGORITHMS
Tcoupl              =  nose-hoover
tc-grps		    =  System
tau_t               =  0.5
ref_t               =  298
;Pcoupl              =  berendsen
pcoupltype	    =  isotropic
tau_p               =  0.1
compressibility     = 0.0008076
ref_p               = 800
; GENERATE VELOCITIES FOR STARTUP RUN
gen_vel             =  yes
gen_temp            =  298
gen_seed            =  173529
; OPTIONS FOR BONDS
constraints         =  all-bonds
constraint-algorithm     = lincs


And sorry if I missed similar topic in mailing list, but I couldn't find 
anything similar.


-- 
Yours sincerely,
Ekatherina O.
oeg at isc-ras.ru