[CPMD-list] Seeking tips on CPMD involving bond breaking

[Jim Kress]

Thanks Axel.

So, All I need to do is specify LSD in the &CPMD namelist?

Thanks.

Jim

----- Original Message ----- 
From: "Axel Kohlmeyer" <axel.kohlmeyer at theochem.ruhr-uni-bochum.de>
To: "Jim Kress" <cpmd at kressworks.com>
Sent: Thursday, February 12, 2004 2:44 PM
Subject: Re: [CPMD-list] Seeking tips on CPMD involving bond breaking


>
> On Thu, 12 Feb 2004, Jim Kress wrote:
>
> JK> In traditional Quantum Chemistry (e.g. the Spin Restricted Roothaan
and KS
> JK> formulations of LCAO approaches), simulation of chemical processes
involving
> JK> breaking of electron pairs in chemical bonds is problematic.  In order
to
> JK> avoid those problems people usually resort to Spin Unrestricted or
> JK> MultiConfiguration approaches.
> JK>
>
> hi jim,
>
> JK> I am attempting to use CPMD study the dynamics of a chemical process
in
> JK> which chemical bonds are being broken.  What is the 'best' way to do
this
> JK> with CPMD?  Some of the variable in the CPMD manual have a tantalizing
sense
>
> weeeelll, i was always under the impression, that as long as you don't
> have intermediate radicals, you should be somewhat ok with the closed
> shell approach. after all, DFT with plane waves and pseudopotentials
> already introduces a significant error. in case of doubt, you're probably
> better off comparing with an open shell calculation (LSD), and pay the
> price of a much higher computational effort. finally, in some cases the
> ROKS (restricted open shell) approach may give the same quality of result
> at nearly no overhead. there was some discussion of that some time ago on
> the mailing list (i remember, because i just 'harvested' the mailing list
> archive for expanding the hints section in the cpmd manual). you may want
> to dig that up.
>
> JK> of possible utility.  But, in the absence of more information than is
> JK> available in the manual, I am unsure how to apply them.
>
> the LSD stuff is pretty straightforward, IIRC, but not all parts of
> CPMD work with it. the LSE (= ROKS) is far more difficult (a postdoc
> and a phd student in our group are using and working on it), and not
> much of it is documented (you have to pick, e.g., irmgard frank's or
> nikos doltsinis' brains to get reliable information beyond reading
> the source code).
>
>
> hope this helps,
> axel.
>
> JK>
> JK> Any hints or suggestions on how to proceed would be greatly
appreciated.  I
> JK> will summarize all responses to the list.
> JK>
> JK> Thanks.
> JK>
> JK> Jim
> JK>
> JK>
> JK> _______________________________________________
> JK> CPMD-list mailing list
> JK> CPMD-list at cpmd.org
> JK> http://www.cpmd.org/mailman/listinfo/cpmd-list
> JK>
> JK>
>
> -- 
>
>
> =======================================================================
> Dr. Axel Kohlmeyer                        e-mail: axel.kohlmeyer at rub.de
> Lehrstuhl fuer Theoretische Chemie          Phone: ++49 (0)234/32-26673
> Ruhr-Universitaet Bochum - NC 03/53         Fax:   ++49 (0)234/32-14045
> D-44780 Bochum  http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/
> =======================================================================
>
>