[gmx-users] abnormal MSD values

[Riccardo Concu]

Dear gmx-users,
I’m having some problems trying to understand  anomalous diffusion coefficient values calculated with g_msd.
MSD was calculated for two box with a run of 100ns using these settings 

integrator               = md
tinit                    = 0
dt                       = 0.002
nsteps                   = 50000000;1ns
comm-mode                = Linear
nstcomm                  = 10
comm-grps                = System
nstxout                  = 0
nstvout                  = 0
nstfout                  = 0
nstlog                   = 1000
nstenergy                = 80
nstxtcout                = 750
xtc-precision            = 1500
xtc-grps                 = System
energygrps               = NAP NA WAT MET SI3 SI- I LI+ LI-
nstlist                  = 20
ns_type                  = grid
pbc                      = xyz
periodic_molecules       = no
rlist                    = 0.9
coulombtype              = PME
rcoulomb                 = 0.9
vdw-type                 = Cut-off
rvdw                     = 1.0
fourierspacing           = 0.12
fourier_nx               = 0
fourier_ny               = 0
fourier_nz               = 0
pme_order                = 6
ewald_rtol               = 1e-5
optimize_fft             = yes
Tcoupl                   = v-rescale
tc-grps                  = NAP NA WAT MET SI3 SI- I LI+ LI-
tau_t                    = 0.1   0.1  0.1  0.1  0.1  0.1    0.1 0.1 0.1
ref_t                    = 298 298 298 298 298 298 298 298 298        
Pcoupl                   = Parrinello-Rahman
Pcoupltype               = Isotropic
tau_p                    = 1
compressibility          = 4.5e-5
ref_p                    = 1.01325
gen_vel                  = yes
gen_temp                 = 298
gen_seed                 = 173529
constraints              = all-bonds
constraint-algorithm     = Lincs
lincs-order              = 4
lincs-iter               = 1

The first box was a cube with the side length 8.6 nM and comprises 9 types of molecules with these MW and number:

     NUMBER MW Charge Diff 
      A 10 229.26 NEG 0.2262 
      Meth 7736 46.09  0.4248 
      Na 29 22.98976 POS 0.0116 
      H2O 310 18.01528  0.0339 
      B 9 233.2893  0.007 
      C 9 234.29724 NEG 0.0041 
      D 10 587.82778 POS 0.0037 
      E 10 586.81984 POS-NEG 0.0063 
      I 20 126.90447 NEG 0.0048 





The second box was a cube with the side length of 4.11 nM and comprises the same molecules but with a considerable reduction in concentration of water and methanol:
     NUMBER MW Charge Diff 
      A 10 229.26 NEG 0.7074 
      Meth 1130 46.09  0.9537 
      Na 29 22.98976 POS 0.1017 
      H2O 230 18.01528  0.1909 
      B 9 233.2893  0.014 
      C 9 234.29724 NEG 0.0167 
      D 10 587.82778 POS 0.0104 
      E 10 586.81984 POS-NEG 0.0107 
      I 20 126.90447 NEG 0.0107 


Hence, the diffusion coefficient in the smallest box is higher than in the bigger box.
Is this result normal?
Thnx in advance,
Riccardo