Pronlem in using gmx msd and mol option to find msd for COM

[saiqa]

I am working on MD simulation using gromacs for solid state electrolytes for polymer and run successfully all the simulations. Also did some fine analysis like RDFs calculations for Li and TFSI ions. All results go comparable to literature. But the problem is when calculating MSDs and Diffusion coefficients for Li and TFSI ions. The trends are same but the values which i calculate using gmx msd commands are one point less than present in literature. So, i have two questions.

1. Whether calculating msd for normal atom and for COM of that atom makes any difference?
2. I am unable to use command mol if i want to calculate msd for COM of Li or TFSI. Please can you write a command line for how to find msd for COM of a molecule.
3. May be i am making some error in the index file. Can you share any sample index file used for calculating MSD dor COM of a molecule.

[jalemkul]

@saiqa

Whether calculating msd for normal atom and for COM of that atom makes any difference?

For a single atom, its COM is the same as its normal (x,y,z) position, by definition.

@saiqa

I am unable to use command mol if i want to calculate msd for COM of Li or TFSI. Please can you write a command line for how to find msd for COM of a molecule.

This is what the -mol flag does.

@saiqa

May be i am making some error in the index file. Can you share any sample index file used for calculating MSD dor COM of a molecule.

There is nothing special about such an index file. You should specify the relevant group for analysis in its own group. If using -mol (which also requires a .tpr file passed to -s), the program will split the index group into molecules.

[saiqa]

Thank you for your response. I followed your instructions and run the gromacs command using mole but now encountered with a new error (shown in attachment). I couldn’t understand why this is happening. Please help me if you can understand the reason for this error. Waiting for your kind response.