Improving MD Results Accuracy

GROMACS version: 2023.1
GROMACS modification: No

Hello.

I run 100 ns MD of Ionic Liquids with GAFF.
The value of density I got deviated by about 3 % from Experiment.

Could one take a look at my production .mdp file and suggest some changes for improving my result as I try hard and hard but no improvement.

See the file below:

title                   = MD equilibration 
; Run parameters
integrator              = md        ; leap-frog integrator
nsteps                  = 50000000     ; 2 * 50000000 = 100000 ps (100 ns)
dt                      = 0.002     ; 2 fs
; Output control
nstxout                 = 500      ; save coordinates every 1.0 ps
nstvout                 = 500      ; save velocities every 1.0 ps
nstenergy               = 500       ; save energies every 1.0 ps
nstlog                  = 500       ; update log file every 1.0 ps
nstxout-compressed      = 500      ; save compressed coordinates every 1.0 ps
compressed-x-grps       = System    ; save the whole system

; remove drifts of the center of mass
comm-mode                = linear	   ; remove COM translation
nstcomm                  = 10              ; number of steps for COM removal
comm-grps                = System          ; COM translation removed from system

; Bond parameters
continuation            = yes       ; Restarting after NPT
constraint_algorithm    = lincs     ; holonomic constraints 
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                 = 20        ; 20 fs, largely irrelevant with Verlet scheme
rcoulomb                = 1.4       ; short-range electrostatic cutoff (in nm)
rvdw                    = 1.4       ; 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
coulomb-modifier         = Potential-shift-Verlet ; shifts potential by constant so potential is 0 at cut-off 
fourierspacing          = 0.168      ; grid spacing for FFT
; Temperature coupling is on
tcoupl		= V-rescale	            ; modified Berendsen thermostat
tc-grps		= System	; two coupling groups - more accurate
tau_t		= 1.0           ; time constant, in ps
ref_t		= 333.15           ; reference temperature, one for each group, in K
nsttcouple               = 1               ; frequency to couple temperature
; Pressure coupling is on
pcoupl                  =  C-rescale     ; Pressure coupling on in NPT
pcoupltype              = isotropic             ; uniform scaling of box vectors
tau_p                   = 4.0                   ; time constant, in ps
ref_p                   = 1.01                   ; reference pressure, in bar
compressibility         = 4.5e-5                ; isothermal compressibility of water, bar^-1
nstpcouple               = 1               ; frequency to couple pressure

;refcoord_scaling        = com
; Periodic boundary conditions
pbc                     = xyz       ; 3-D PBC
; Velocity generation
gen_vel                 = no        ; Velocity generation is off 

Thank You.

The issue is more likely tied to the topology of the molecules in the system rather than .mdp settings. A 3% deviation isn’t terrible, usually a 2% deviation is an acceptable tolerance when validating parameters.

Thank You, Sir.