[CP2K:621] Re: How do I generate PSF ?

[Jack Shultz]

Axel,
I ordered one of the books you recomended...also came across this article
"Pseudobond ab initio QM/MM approach and its applications to enzyme
reactions" Yingkai Zhang 2005
Looks fairly comprehendable and relevant to my interests. Thanks for your
help and after I've studied up more I'll probably still have questions.

Jack
On Jan 22, 2008 5:39 AM, Axel <akoh... at gmail.com> wrote:

>
>
> On Jan 22, 2:34 am, "Jack Shultz" <jackyg... at gmail.com> wrote:
> > Axel,
> > I am still trying to figure out what to turn to for tutorials and
> > literature. Whether I should turn to Amber, GROMACs or QE tutorials. Got
> an
>
> each of the tutorials address a different audience and each of the
> packages have different features. so in principle you should go over
> _all_
> of them to learn the most and particularly see the common features and
> differences (at least that is what i did). there is a huge difference
> between
> gromacs and quantum espresso with respect of the implemented methods
> typical uses and capabilities.
>
> > academic license for Amber which is on its way. You said Amber would be
> much
> > better to learn on than CP2K, so I will look over that material first. I
> > understand there is no exact correlation between what I learn in one
> > chemistry package to another...but just trying to figure out where to
> start
> > seems hard.
>
> this is completely understandable. it looks hard, because you are not
> familiar
> enough with the underlying concepts and theory and particularly don't
> know
> the 'slang' that is used. so before starting any tutorials, you should
> take a deep
> look into (introductory) books on MD, molecular modeling and quantum
> chemistry,
> and once you're getting the hang of it, everything will be much more
> obvious.
>
> cheers,
>    axel.
>
>
> >
> > Jack
>  >
> > On Jan 21, 2008 10:42 AM, Axel <akoh... at gmail.com> wrote:
> >
> >
> >
> > > On Jan 21, 2:06 pm, "Jack Shultz" <jackyg... at gmail.com> wrote:
> > > > My goal is to learn how to design a covalent bond-breaking reaction
> > > model.
> > > > So I do what usually works for me with most applications, take an
> > > example I
> > > > know works 2gly_CI-NEB.inp but I try to build the inital and final
> > > images
> > > > using H2O2, H2O & O2 molecular files. I modified these files.
> >
> > > > I only want the final geometry to represent two water molecules and
> one
> > > > diatomic oxygen with sufficient distance to make any weak
> interactions
> > > such
> > > > as hydrogen bonds non-existent So I've displaced them 2 angstroms
> along
> > > x
> > > > and now I tried 1 angstrom along y.
> > > > So I guess I can use that triangle formula since it is only two
> axis.
> > > 2sqr +
> > > > 1sqr = csqr = 5 so c = sqrt 5 < 2.8 ... maybe not enough distance.
> > > > As far as the FF goes, can you point me to an example that has FF
> > > configured
> > > > correctly for a bond-breaking reaction? Moreover, are there any
> >
> > > almost all traditional classical (pairwise additive) force fields do
> > > not support bond breaking.
> >
> > > if you want to look at bond breaking, why not use semi-empirical
> > > or DFT? then you don't need to provide topology information.
> >
> > > in the danger of repeating myself. please go an _first_ study a bit
> > > on how the various modeling methods work, what they can do and
> > > don't try to do too many things at the same time (NEB is really very
> > > much at the end of this 'food chain'). it will pay off in time, and
> > > you
> > > don't have to ask about things, that puzzle you now, but are
> > > essentially trivial.
> >
> > > cheers,
> > >   axel.
> >
> > > > bond-breaking examples I can look at?
> >
> > > > Jack
> >
> > >  > On Jan 21, 2008 2:58 AM, Teodoro Laino <teodor... at gmail.com>
> wrote:
> >
> > > > >  On 21 Jan 2008, at 04:35, Jack Shultz wrote:
> >
> > > > > So I did not displace the water molecules appropriately?
> >
> > > > > what do you mean with appropriately? If that is the geometry that
> you
> > > want
> > > > > to get as your final state.. that's fine.. but don't expect
> > > > > the topology generator to work properly with it..
> >
> > > > > I only tried to seperate them by 2 angstroms along the
> x-axis...maybe
> > > I
> > > > > should move them 1 angstrom along the y-axis as well?
> >
> > > > > a typical hydrogen bond (O1-O2) for bulk water is ~ 2.8 angstrom..
> > >  (as
> > > > > you said yours is 2.0 Angstgrom)..
> >
> > > > > Should that help?
> >
> > > > > Sorry.. I don't understand why/where you are expecting an help
> > > changing
> > > > > the geometry of the final state of the NEB.
> > > > > Perhaps, I was not clear in my last message:
> >
> > > > > You've setup a bond-breaking reaction with a classical FF: 1) You
> > > didn't
> > > > > provide any parameter file for the classical FF (!) (and this
> causes
> > > the
> > > > > abnormal termination of cp2k)
> > > > > 2) You cannot do BB* with CP2K, using a classical Hamiltonian.
> >
> > > > > *(Bond-Breaking)
> >
> > > > > cheers,
> > > > > Teo
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20080125/71080f76/attachment.htm>