To whom it may concern,
I am going to analysis the “photoelectric behavior” of a deposited structure by LAMMPS QM MM method and I am not sure whether LAMMPS QMMD is suitbale for my case.
Actually, the deposited structure is first produced by LAMMPS MD for more than 1 million atoms on HPC. I mean the system size is not small. Then I would like to analysis their photoelectric behavior with QMMM method.
any comment is highly appreciated.
Yours faithfully,
Bahman
To whom it may concern,
I am going to analysis the "photoelectric behavior" of a deposited structure by LAMMPS QM MM method and I am not sure whether LAMMPS QMMD is suitbale for my case.
most likely not. for several reasons. but there are a bunch of
questions that need to be answered before giving a more specific
recommendation.
before even considering QM/MM, have you done "plain" quantum
calculations on these properties? and what are you aiming to compute
exactly, anyway. you hinting at static calculations, why do you need
QM/MM in the first place? can't this be done with clusters embedded in
pseudopotentials and point charge fields? does this need
multi-determinant calculations? is this an analysis that can be done
with plane wave pseudo-potential calculations? what is your level of
experience with quantum calculations in general?
Actually, the deposited structure is first produced by LAMMPS MD for more than 1 million atoms on HPC. I mean the system size is not small. Then I would like to analysis their photoelectric behavior with QMMM method.
do you have any precedence, i.e. publications from people doing
something similar (who? and what code did they use?)?
i have only peripheral knowledge of the QM side of things and i have
seen people try a lot of crazy things with QM/MM (and fail), but this
sounds a lot more extreme than those.
axel.
Dear Alex,
Thank you very much for your good questions. Actually, I have not any experience with QM simulations right now. I am still searching to find the best solution. My aim is to analyse the photoelectric properties of my resulted atomic structure from MD.
In my research, my notice has been attracted to DFTB and LAMMPS LATTE. It seems that this method is a good choice for large systems but I am not still sure about that to how large system it means ! I am still searching for a good solution as well thinking to response to your questions.
best regards,
Dear Alex,
Thank you very much for your good questions. Actually, I have not any experience with QM simulations right now. I am still searching to find the best solution. My aim is to analyse the photoelectric properties of my resulted atomic structure from MD.
you still haven't explained what you mean by "photoelectric
properties" and what kind of quantum calculations (or properties
derived from them) you want to look at. this should be the *very*
first thing to think about, because that limits the QM methodology
available to you and my hunch is, that nothing that you are currently
looking at is capable of providing what you need.
In my research, my notice has been attracted to DFTB and LAMMPS LATTE. It seems that this method is a good choice for large systems but I am not still sure about that to how large system it means ! I am still searching for a good solution as well thinking to response to your questions.
you are doing things backwards. *first* you need to get the experience
in the quantum calculations part (and that can take a while). this is
not something to pick up quickly, especially if you don't have the
sufficient education in the corresponding theory.
axel.
Thank you for your reply.