[gmx-users] temperature coupling groups for homogeneous mixtures/ ionic liquids?

[Boning Wu]

Dear Gromacs Users,

I am simulating ionic liquids and the mixtures of ionic liquids and neutral
solvents. I used Berendsen or Nose-Hoover temperature coupling for NPT
systems. No constrains or restrains are in the system.

I just find these discussions and documentations about the temperature
coupling groups:

http://www.mail-archive.com/gmx-users@gromacs.org/msg28634.html

http://www.gromacs.org/Documentation/Terminology/Thermostats

I am worried because without realizing the limitations of temperature
coupling, I used to set my tc groups very arbitrarily in my previous
simulations, sometimes I use

tc-grps             =  CATION ANION NEUTRAL
and sometimes I use:

tc-grps             =  system

Here are my questions:

1. Of these two grouping methods, which is better for systems
(a) which is completely homogeneous?
and (b) which has some nano-domain structures like the aggregations of
neutral molecules in ionic liquids with size up to 2 nm ( but no phase
segregation)?

2. Since now it's almost not possible to re-run all my simulations, does it
matter if I use two methods arbitrarily? I have attached an mdp file. Most
work is done by 5.0.4 and 2016.1. How could I check if there is artifact in
my system?

3. Since most of the discussions I found are several years ago, did things
get improved these years? For example, is there anything improved after
cut-off scheme is replaced by Verlet?

Any help would be appreciated!

Here is my .mdp file

title               =  productrun
cpp                 =  /lib/cpp
define              =
constraints         =  none
integrator          =  md
dt                  =  0.001    ; ps !
nsteps              =  2000000  ; 2 ns
nstcomm             =  1
nstxout             =  500000   ; write x to trr file
nstvout             =  500000   ; write v to trr file
nstxtcout           =  1000     ; write x to xtc file
nstfout             =  0
nstlog              =  1000     ; write energy to log
nstenergy           =  1000     ; write energy to edr
nstlist             =  1        ; frequency to update neighbour list
; GPU parameters
;verlet-buffer-tolerance  = -1
;nstcalclr                =  1
;nstcalcenergy            =  1
ns_type             =  grid
pbc                 =  xyz      ; periodic boundry condition
rlist               =  1.5
rcoulomb            =  1.5
rvdw                =  1.5
disre               =  simple
disre-fc            =  1000
coulombtype         =  PME
; Apply long range dispersion corrections for Energy and Pressure =
DispCorr            =  EnerPres
; Spacing for the PME/PPPM FFT grid =
fourierspacing      = 0.08  ; depends on system
; FFT grid size, when a value is 0 fourierspacing will be used =
fourier_nx          = 0
fourier_ny          = 0
fourier_nz          = 0
; EWALD/PME/PPPM parameters =
pme_order                = 6
ewald_rtol               = 1e-05
ewald_geometry           = 3d
epsilon_surface          = 0
optimize_fft             = yes
; Nose-Hoover temperature coupling is on
Tcoupl              =  Nose-Hoover
tc-grps             =  system
tau_t               =  0.2  ; depending on system
ref_t               =  298.0
energygrps          =  system
; Pressure coupling is on
Pcoupl              = Parrinello-Rahman
Pcoupltype          = isotropic
tau_p               =  1.0    ; depending on system
compressibility     =  4.5e-5
ref_p               =  1.0
gen_vel             =  no
gen_temp            =  298.0
gen_seed  =  43697


Thank you,
Boning Wu