Hello Friends,
hello b.d.,
I am really tired of checking different commands to see" the solidification
on a simple solid layer kept in the undercooled liquid ".
if you cannot handle frustration, then i suggestion you find a new
profession. the nature of scientific work is that you try to understand
a problem that you don't understand yet. that in turn _will_ cause
frustration since you don't understand the problem and will have a
lot of failed attempts until you gather the sufficient clues that will
make you understand what is going on. once you understand a
problem, it will be easy and you spend little time on it. thus in total,
you'll be most of the time frustrated, so you better get used to it.
purpose:
I am going to see the "solid cap" on the solid substrate.but I have not seen
yet!!!!!!!!!!!!!!!!!!
so what?
method:
At first, I found the melting point of the system and then I put the system
how did you determine that? finding the (true) melting point of a small, finite
system is not at all trivial.
at the required undercooling .both substrate and surrounded liquid are made
from the same material (same sigma and epsilon) .I did not move the group
of"center atoms" to perform as a solid substrate in the liquid.
if you use the same epsilon throughout and reduced units,
why set epsilon to such a large value?
I have checked time step size,nvt,nve,npt commands to see the solidification
on the substrate but it is not possible!!!!!!!!!!!!!!!!!!!!!!
you have a _very_ small number of particles and run only a short trajectory.
if you want particles to become solid you have to wait until the kinetic
energy is transferred from the solidified particles to the still moving ones.
as far as i can see that process is going on. what exactly did you expect?
you have to forget the nice sketches usually shown in text book. this is
not how things happen, at least not in the time scale that you are looking at.
theoretically , I should see the solid cap on the solid substrate but
unfortunately I have not seen this phenomena yet.
when I watch final movie I think that there is not enough attraction between
substrate and unedrcooled liquid.
since you are using reduced units this is irrelevant. all that changes
by changing epsilon is your melting point. think about that for a bit
and have a look at some stat mech textbook.
I have attached the file .please look at it and tell me what I have not
considered yet.please help me. I need a help. I am really disappointed.
phase transitions in finite systems is a non-trivial problem. you have
to deal with fluctuations that effectively don't fit into your system.
there are several systematic procedures to approach such properties,
but you have to know what you are doing. in general, MD follows - like
most computational methods - the GI-GO principle.
cheers,
axel.