quantum-espresso/examples/example17
giannozz 7abc7cdebb Minor changes to scripts used to check and save examples
All examples updated again (sorry), this time with ifort 8
Only pwcond refuses to work with this particular compiler


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@2057 c92efa57-630b-4861-b058-cf58834340f0
2005-07-28 17:26:51 +00:00
..
reference Minor changes to scripts used to check and save examples 2005-07-28 17:26:51 +00:00
README Created unified "examples" directory replacing old "cp_examples", 2004-06-10 09:37:46 +00:00
run_example NEB and Fourier String dynamics: the quasi-Newton Broyden's second method has 2005-02-25 14:51:41 +00:00

README

This example shows how to use pw.x to calculate the minimum energy path (MEP) in
a simple activated reaction i.e. the collinear proton transfer reaction :

H2 + H  <==> H + H2

The MEP is obtained by means of the Climbing Image Nudged Elastic Band (CI-NEB)
method and two different climbing image algorithms are used (highest-TS and
manual).

(for the meaning of the cited input variables see the appropriate INPUT_* file)

The symmetric reaction path H2 + H  <==> H + H2 is calculated in three 
different ways:

1) The path connecting the initial and the final configurations is discretized
   with an odd number of images (7) so that the standard CI_scheme (highest-TS)
   will give rise to a symmetric MEP (3 images on the left of the saddle point
   and 3 images on the right). Note that in this system the use of the climbing
   image is not neccesary. Indeed using CI_scheme = "no-CI" the result is the 
   same.

2) The path connecting the initial and the final configurations is discretized
   with an even number of images (8) and no climbing image is used. The
   resulting path is symmetric, but no image is at the saddle point.

3) The path connecting the initial and the final configurations is again 
   discretized with an even number of images (8), but the "manual" CI_scheme is
   used so that the resulting path is asymmetric. The image 5 now is at the
   saddle point.
   
Given the low accuracy of these calculations (plane waves cut-off and 
thresholds) and the small box employed, these results can not be compared with
experiments or with other "ab initio" calculations.