search

hg08 / hb_ihb

Instantaneous hydrogen bond dynamics
GNU General Public License v3.0
2 stars 0 forks source link

216w, 343w, 512w, MB-pol results #104

Closed huangchieh closed 3 months ago

huangchieh commented 3 months ago

216 water

case 1: Total time used: 13.20 hours case 2: Total time used: 1.41 hours case orientation: Total time used: .40 hours

{
    "name": "216h2o-240ps-mbx-pol.xyz",
    "system": "216h2o-240ps-mbx-pol",
    "simType": "MB-pol",
    "dt": 0.04,
    "sizeX": 19.71096,
    "sizeY": 19.71096,
    "sizeZ": 98.554799,
    "numAtoms": 648,
    "simTime": 240.0,
    "numFrames": 6001,
    "numSubTrajCase1": 6,
    "subTrajTimeCase1": 40,
    "numSubTrajCase2": 6,
    "subTrajTimeCase2": 40,
    "numSubTrajOri": 12,
    "subTrajTimeOri": 30
}

huangchieh commented 3 months ago

343 water

Last sub-trajectory error

{
    "name": "343h2o-240ps-mbx-pol.xyz",
    "system": "343h2o-240ps-mbx-pol",
    "simType": "MB-pol",
    "dt": 0.04,
    "sizeX": 21.729972,
    "sizeY": 21.729972,
    "sizeZ": 108.64986,
    "numAtoms": 1029,
    "simTime": 240.0,
    "numFrames": 6001,
    "numSubTrajCase1": 6,
    "subTrajTimeCase1": 40,
    "numSubTrajCase2": 6,
    "subTrajTimeCase2": 40,
    "numSubTrajOri": 12,
    "subTrajTimeOri": 30
}

Due to the runtime error for the last sub-trajectory, only first 5 sub-trajectories were calculated for case1 and case2. See details here.

Case1:
- At line 66 of file src/molecules_in_interface.f95
Fortran runtime error: Index '1871' of dimension 2 of array 'surf_info' above upper bound of 1849

Case2: 
- At line 175 of file ghbacf_interface_n_pbc_format2.f95
Fortran runtime error: Index '1871' of dimension 1 of array 'surf_info' above upper bound of 1849
- At line 167 of file ghbacf_interface_c_pbc_format2.f95
Fortran runtime error: Index '1871' of dimension 1 of array 'surf_info' above upper bound of 1849
- At line 177 of file ghbacf_interface_k_pbc_format2.f95
Fortran runtime error: Index '1871' of dimension 1 of array 'surf_info' above upper bound of 1849

Orientation: 
At line 63 of file src/molecules_in_interface.f95
Fortran runtime error: Index '1871' of dimension 3 of array 'surf_info_fortran' above upper bound of 1849

Case 1: Total time used: 7.29 hours (1.4h per sub-trajectory) Case 2: Total time used: 4.98 hours (0.9h per sub-trajectory) Orientation: Total time used: .87 hours

I am debugging.

Anyway, I am running the system of 512w. If problem still exist, then I will solve this problem. Or, I will do nothing.

huangchieh commented 3 months ago

I only calculate the calculated trajectories by changing the above json file to this

{
    "name": "343h2o-240ps-mbx-pol.xyz",
    "system": "343h2o-240ps-mbx-pol",
    "simType": "MB-pol",
    "dt": 0.04,
    "sizeX": 21.729972,
    "sizeY": 21.729972,
    "sizeZ": 108.64986,
    "numAtoms": 1029,
    "simTime": 240.0,
    "numFrames": 6001,
    "numSubTrajCase1": 5,
    "subTrajTimeCase1": 40,
    "numSubTrajCase2": 5,
    "subTrajTimeCase2": 40,
    "numSubTrajOri": 9,
    "subTrajTimeOri": 30
}

Here is the results:

huangchieh commented 3 months ago
  1. The results of 343w are almost no difference to 216w. I guess the results of 512w are very similar.
  2. As d increasing, we selected water molecules on both interface an bulk. This is also true when d is 6 or larger, which means the system we observing are a system including both interface and bulk water. Hence, the react constant rate k, should never like the value of bulk. It makes sense to me that the convergence of the k is above k_bulk because our system is more dynamic than bulk. @hg08
hg08 commented 3 months ago

Also 216water

{
    "name": "216h2o-240ps-mbx-pol.xyz",
    "system": "216h2o-240ps-mbx-pol",
    "simType": "MB-pol",
    "dt": 0.04,
    "sizeX": 19.71096,
    "sizeY": 19.71096,
    "sizeZ": 98.554799,
    "numAtoms": 648,
    "simTime": 240.0,
    "numFrames": 6001,
    "numSubTrajCase1": 6,
    "subTrajTimeCase1": 40,
    "numSubTrajCase2": 6,
    "subTrajTimeCase2": 40,
    "numSubTrajOri": 12,
    "subTrajTimeOri": 30
}
huangchieh commented 3 months ago

Figures C2 shows that the orientation is not sensitive to the thickness. i.e., not all properties are suitable to measure the thickness of air-water interface. It's okey.

hg08 commented 3 months ago

Figures C2 shows that the orientation is not sensitive to the thickness. i.e., not all properties are suitable to measure the thickness of air-water interface. It's okey.

I fit it with double exponential decay function and it shows that the fitting is very good! (See the last figure). Therefore, we can try to use double exopoential functions's parameter to replace Fig 5.

huangchieh commented 3 months ago

What are the results of double exponential decay fitting look like?

hg08 commented 3 months ago

What are the results of double exponential decay fitting look like?

Now, it is done. (y-labe of the right figure is wrong)

huangchieh commented 3 months ago

From the result, I conclude that the orientation dynamics are almost the same for both bulk and surface. Only orientation is not enough to distinguish the surface and bulk.

hg08 commented 3 months ago

If plot the amplitude and the relaxation time, the result is : image

hg08 commented 3 months ago

From the result, I conclude that the orientation dynamics are almost the same for both bulk and surface. Only orientation is not enough to distinguish the surface and bulk.

yes. And the results are consistent to each other.

huangchieh commented 3 months ago

512w

Total time used: 15.34 hours

└─▪ cat 512h2o-240ps-mbx-pol.json
{
    "name": "512h2o-240ps-mbx-pol.xyz",
    "system": "512h2o-240ps-mbx-pol",
    "simType": "MB-pol",
    "dt": 0.04,
    "sizeX": 24.6387,
    "sizeY": 24.6387,
    "sizeZ": 123.1935,
    "numAtoms": 1536,
    "simTime": 240.0,
    "numFrames": 6001,
    "numSubTrajCase1": 6,
    "subTrajTimeCase1": 40,
    "numSubTrajCase2": 6,
    "subTrajTimeCase2": 40,
    "numSubTrajOri": 12,
    "subTrajTimeOri": 30
}