[CPMD-list] Locating a Transition State

[Jim Kress]

Thanks for your reply.  Per your suggestion, I tried the P-RFO and it works
better than the other method (RFO) I had been using.

Thanks for your help.

Jim

----- Original Message ----- 
From: "Salomon Billeter" <ZRLSRB at ch.ibm.com>
To: "Jim Kress" <cpmd at kressworks.com>; <cpmd-list at cpmd.org>
Sent: Thursday, April 08, 2004 4:50 AM
Subject: Re: [CPMD-list] Locating a Transition State


>
>
>
>
> Hi,
>
> P-RFO (the partitioned RFO algorithm where you treat two subspaces spanned
> by eigenvectors of the Hessian separately) allows you among others to
> select the mode to be followed. The implementation in CPMD has many
options
> which are described in the manual. If your system is large, you could also
> consider using the linear-scaling microiterative scheme (variational
> separation between reaction core and environment, keywords PRFO CORE=...
> and/or PRFO NVAR etc.).
>
> Certainly the most important option for PRFO is the initial number of the
> mode to be followed:
>
> PRFO MODE
>   <number>
>
> You may find this number either by trial-and-error or using a vibrational
> analysis. In this case, you can reuse the Hessian from the vibrational
> analysis using
>
> RESTART HESSIAN
>
> and
>
> PRFO HESSTYPE
>   1
>
> The manual describes the other options, and in the archive of this mailing
> list, you'll find more information on these topics.
>
> Best regards,
> Salomon
>
> _______________________________________________
> CPMD-list mailing list
> CPMD-list at cpmd.org
> http://www.cpmd.org/mailman/listinfo/cpmd-list
>