[gmx-users] potential energy

[Justin Lemkul]

On 1/16/15 8:56 AM, mah maz wrote:
> Hi Justin,
> I haven't come to a conclusion yet and need your help! I attached a copy of
> a typical g_energy -f ener.edr -o copied from terminal and an .mdp file.
>
>   Bond                        18506.5         50    1625.68   -301.714
> (kJ/mol)
> Angle                       9839.32         13    580.506   -30.2625
> (kJ/mol)
> Proper Dih.                 60199.3         23    845.936    106.894
> (kJ/mol)
> LJ-14                       40461.9         41    1748.27   -227.144
> (kJ/mol)
> Coulomb-14                        0          0          0          0
> (kJ/mol)
> LJ (SR)                    -10117.7        2.6    118.855   -3.85515
> (kJ/mol)
> Coulomb (SR)               -156.169          9     40.823    35.6671
> (kJ/mol)
> Potential                    118733         74    1142.65   -420.414
> (kJ/mol)
> Kinetic En.                 14932.1        7.3    423.825   -50.1192
> (kJ/mol)
> Total Energy                 133665         81    1221.66   -470.533
> (kJ/mol)
> Conserved En.                180812         57    121.277    397.186
> (kJ/mol)
> Temperature                 295.064       0.14    8.37495  -0.990376  (K)
> Pressure                  -0.072961       0.17     17.163  -0.120159  (bar)
> Vir-XX                      4860.45        210    21707.3    26.9299
> (kJ/mol)
> Vir-XY                      4.73233        3.3     1176.9   -2.74323
> (kJ/mol)
> Vir-XZ                      3.18097        7.5    1907.36     4.2031
> (kJ/mol)
> Vir-YX                      4.73147        3.3    1176.89   -2.74818
> (kJ/mol)
> Vir-YY                      5388.86         39     5500.8    202.824
> (kJ/mol)
> Vir-YZ                    0.0509392        3.2    1101.97    3.89409
> (kJ/mol)
> Vir-ZX                      3.18089        7.5    1907.36    4.19967
> (kJ/mol)
> Vir-ZY                    0.0514236        3.2    1101.97    3.89442
> (kJ/mol)
> Vir-ZZ                      4871.76        200    21686.4    31.3387
> (kJ/mol)
> Pres-XX                    0.156061       0.25    25.1574  -0.125195  (bar)
> Pres-XY                  -0.00268517     0.0033    1.36879 0.00150158  (bar)
> Pres-XZ                  -0.00196111     0.0098    2.21263 -0.0154319  (bar)
> Pres-YX                  -0.00268417     0.0033    1.36879 0.00150732  (bar)
> Pres-YY                   -0.511872      0.019    6.38934 -0.0849178  (bar)
> Pres-YZ                  0.000542369     0.0017    1.27895 0.00355122  (bar)
> Pres-ZX                  -0.00196102     0.0098    2.21263 -0.0154279  (bar)
> Pres-ZY                  0.000541808     0.0017    1.27895 0.00355084  (bar)
> Pres-ZZ                    0.136928       0.24    25.1334  -0.150365  (bar)
> #Surf*SurfTen               8.81535        3.2    359.945   -1.26863  (bar
> nm)
> T-CNT                       295.063       0.14    8.44198  -0.997647  (K)
> T-grp1                  295.202       0.21    38.5349   0.485478  (K)
> T-grp2                    294.688       0.32    53.7236   -2.39488  (K)
>
>
>
> dt                  =  0.0001
> tinit               =  0
> nsteps              =  2000000
> nstxout             =  1000
> nstvout             =  1000
> nstfout             =  0
> nstlog              =  1000
> nstenergy           =  1000
> nstlist             =  5
> ns_type             =  grid
> rlist               =  0.9
> coulombtype         =  cutoff
> rcoulomb            = 0.9
> rvdw                =  0.9
> pbc                 = xyz
> integrator          = md-vv
> Tcoupl              =  nose-hoover
> tau_t               =  0.1 0.1 0.1
> ref_t               =  295 295 295
> tc_grps             = CNT grp1 grp2
> gen_temp            = 295
>
> Your answer to any of these questions would be a ton of help!
> These potential energies are the average ones, how can I get the final
> potential of the system? Are average potentials good terms to decide if the
> simulation was wrong?

The .edr file contains energies at every nstenergy steps.  An instantaneous 
value is not terribly useful, though.

> Can the positive potentials be the effect of forcefield inappropriate files?
> The system is a CNT containing some solvent. Can positive potentials be the
> reflection of the system's angle pressure not willing to reform from plane
> graphene system?
>

The outcome indicates that the potential energy is dominated by bonded (i.e. 
intramolecular) interactions, while there are few/weak nonbonded interactions 
among the atoms in the system.  If I'm interpreting what you're saying, there is 
some solvent inside the CNT, and nothing outside?  Generally, most condensed 
phase systems (at least those in water) are dominated by electrostatic 
interactions, whereas your system has a negligible amount of electrostatics. 
The overall effect is a function of what your species are.  But for that, you're 
going to have to tell us what "grp1" and "grp2" are, what their topologies are, etc.

But in reality, there seems to be nothing "wrong" with the simulation; it's just 
atypical, and you can't really compare it to a "normal" simulation with a lot of 
water.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================