How to make coordinate file

[sumit agrawal]

Hii,

I have recently installed cp2k in windows. I just did some exercise which is available on cp2k website. Now I want to do some bulk phase vibrational analysis. I want to make a coordinate file in which 100 water molecules and one methanol molecule is present. This methanol molecule is present in between water molecules. So how to make a coordinate file for this system.

Can I use gauss view to make the coordinate file?

[Vasyl Stotskyi]

Dear Sumit Agrawal,

You can create some model in Avogadro. There you can pre-optimize it with force fields, and save as .xyz file. In CP2K, then you can read coordinates from .xyz file. Please, read manual for this (https://manual.cp2k.org/cp2k-7_1-branch/CP2K_INPUT/FORCE_EVAL/SUBSYS/TOPOLOGY.html).

Best regards,

Vasyl

--

You received this message because you are subscribed to the Google Groups "cp2k" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CALSDoYad_FGm4Qh5SRvZcMun3fe%3D5q1xpxeW3T5SoMcAoTH52A%40mail.gmail.com.

[sumit agrawal]

Dear Vasyl,

Thanks for the answer. So I have to manually draw 100 molecules of water and methanol in Avogadro? Actually,  I am more comfortable with gauss views. So can i make a coordinate file in gusse view then pre optimize in the avogadro ? 

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/FC05854F-A2E3-45BD-9677-A12F225F42A8%40getmailspring.com.

[Vasyl Stotskyi]

Dear Sumit Agrawal,

Sure, if you can save it in one of avogadro readable formats (.xyz, .cif, ....).

If initial geometries of molecules in gaus view will be fine, then you don't have to optimize it in Avogadro, i.e. if you will not have O-H bonds with more than 1.5 angstrom  length or overlap of molecules, or huge vacuum between molecules. Avogadro optimization is the most useful only for such bad cases.

Best regards,

Vasyl

On Oct 17 2020, at 12:40 am, sumit agrawal <sumitag...@gmail.com> wrote:

Dear Vasyl,

Thanks for the answer. So I have to manually draw 100 molecules of water and methanol in Avogadro? Actually,  I am more comfortable with gauss views. So can i make a coordinate file in gusse view then pre optimize in the avogadro ? 

On Fri, Oct 16, 2020 at 9:04 AM Vasyl Stotskyi <vasyl.s...@gmail.com> wrote:

Dear Sumit Agrawal,

You can create some model in Avogadro. There you can pre-optimize it with force fields, and save as .xyz file. In CP2K, then you can read coordinates from .xyz file. Please, read manual for this (https://manual.cp2k.org/cp2k-7_1-branch/CP2K_INPUT/FORCE_EVAL/SUBSYS/TOPOLOGY.html).

Best regards,

Vasyl

On Oct 16 2020, at 11:27 pm, sumit agrawal <sumitag...@gmail.com> wrote:

Hii,

I have recently installed cp2k in windows. I just did some exercise which is available on cp2k website. Now I want to do some bulk phase vibrational analysis. I want to make a coordinate file in which 100 water molecules and one methanol molecule is present. This methanol molecule is present in between water molecules. So how to make a coordinate file for this system.

Can I use gauss view to make the coordinate file?

--

You received this message because you are subscribed to the Google Groups "cp2k" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CALSDoYad_FGm4Qh5SRvZcMun3fe%3D5q1xpxeW3T5SoMcAoTH52A%40mail.gmail.com.

--

You received this message because you are subscribed to the Google Groups "cp2k" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/FC05854F-A2E3-45BD-9677-A12F225F42A8%40getmailspring.com.

--

You received this message because you are subscribed to the Google Groups "cp2k" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CALSDoYaSwE%3DyxwswTkq6U5OouRZPiuev3yRJDydn%2B37kF-G4%2Bg%40mail.gmail.com.

[sumit agrawal]

Dear Vasyl,

Thanks a lot!!.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/E1FD0EA3-5F69-4894-B38E-7FA74A2088EE%40getmailspring.com.

[Lucas Lodeiro]

Dear Sumit Agrawal,

I suggest the Packmol software to construct initial configurations, It is a simple and tractable, where you can decide where put some particular molecules, the siz of the box, the number of different molecules, and there is a criteria to ensure your molecules are not so near between them. For little boxes is useful, but for big systems is mandatory to use a program like Packmol.

http://m3g.iqm.unicamp.br/packmol/home.shtml

Regards

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/E1FD0EA3-5F69-4894-B38E-7FA74A2088EE%40getmailspring.com.

[sumit agrawal]

Dear Lucas Loderio and Travis

Thanks for the suggestions. I have tried packmol today. In packmol I can easily make my input geometry and it is very easy to use. I have given a tolerance value equal to 2 for my above system. Now I can use this input file for cp2k calculation. 

I have one more question: Should I need to perform geo-opt before computing the vibrational spectra?

Thanks a lot !!!

On Fri, Oct 16, 2020 at 6:54 PM Travis <polla...@gmail.com> wrote:

Hi,

To add, as a general practice for condensed phase simulations: build a box with Packmol and equilibrate it with a classical force field that does reasonably well. Use several well-separated snapshots as initial configurations for CP2K. For a simulation with a simple solute in it, well-separated might mean configurations taken every 2*tau or 3*tau where tau is the solvent-solute residence time. This can be short for weakly interacting species but many nanoseconds for very strongly interacting species or viscous solutions. You may even need to simulate at elevated temperatures to reduce the residence time. If you don't know how to compute that from the output of your preferred molecular dynamics code, I would recommend LAMMPS for which there are many community-developed analysis tools. Ed Maginn's group at Notre Dame has a well-rounded analysis suite hosted on Github... PyLat or something like that. There's no unique answer for what 'well-separated' means, this is just the criteria we use.

For this particular system, since it sounds like this might still be an exercise, you can take one of the CP2K example files for (H2O)128, convert one of the H2O to methanol using Avogadro or GaussView, and run from there. If the concentration really matters for your case, then go the Packmol and classical force field route.

-T

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ee5fea3f-2f5c-42f2-8fb3-c6ab1042e860n%40googlegroups.com.

[sumit agrawal]

Okay....Thanks a lot!!!

On Sat, Oct 17, 2020 at 1:25 AM Travis <polla...@gmail.com> wrote:

Hi,

Packmol simply packs a box of some size with a randomized geometry. There will likely be overlapping atoms and other bad contacts. You'll need to at least optimize it but preferably equilibrate the liquid structure with a reasonable force field before using it with CP2K.

-T

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/24cb5e6e-ba82-45ef-be8f-90b97c9f3253n%40googlegroups.com.

[sumit agrawal]

Hii,

I tried to make a protein, containing a solvation cell via packmol. Here the solvation cell is water. Now I want to change the solvation cell. I want a N2 molecule, containing a solvation cell. How to do it via packmol? Since in packmol only water pdb file is available.

I am attaching a input file of a protein, containing water molecules.

[Lucas Lodeiro]

Hi,

You have to make pdb files for each molecule that you need to put in a box via packmol. If you can make a water-protein cell, the water-N2 cell is the same, just change the protein.pdb file to N2.pdb and rum packmol.

Here are some useful examples to make differents types of boxes and interfaces: http://m3g.iqm.unicamp.br/packmol/examples.shtml

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CALSDoYY1zZUFK0OxSNi-Dqkh9wMfw-_qo1%2BQkM%3DrufvRK09e8w%40mail.gmail.com.

[sumit agrawal]

Thank you... It's really helpful. 

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CAOFT4PLme_F95CUF0BjDyHE%3D%3DPSkB9DKeRR0c-FgeVV-6fsi4w%40mail.gmail.com.