Show Posts
1
General Questions and Answers / boundary problem of generating intermediate configurations.
« on: October 28, 2014, 07:18 »
Hello everyone!
I am interested in the BC method of generating intermediate configurations at the beginning of NEB calculation.
However, when I use this method, I encounted a boundary problem. For instance, if I have two positions(A,B), one of that close to one boundary of the box and the other one close to the opposite boundary, in consider of periodic boundary condition, the shortest path from A to B is through the boundary, rather than cross almost the whole box. We put one atom at position A as the initial configuration, and one atom at B as the final configuration. If we use the dist.pl of vtst-scripts to check the distance between these two configurations, the value is very small. While we put them into VNL/Builder and try to create intermediate configurations, it shows a big value of distance.
Because of this, there may be some atoms cross the whole box in the NEB PATH that we created with VNL.
The VTST-scripts however, avoided this problem by doing things as follows:
if ($distance <= -0.5) {
$distance += 1.0;
} elsif ($distance > 0.5) {
$distance -= 1.0;
}
Therefore, the dist.pl and nebmake,pl works well.
How can I use BC method to create intermediat configurations and also avoid boundary problems?
I am interested in the BC method of generating intermediate configurations at the beginning of NEB calculation.
However, when I use this method, I encounted a boundary problem. For instance, if I have two positions(A,B), one of that close to one boundary of the box and the other one close to the opposite boundary, in consider of periodic boundary condition, the shortest path from A to B is through the boundary, rather than cross almost the whole box. We put one atom at position A as the initial configuration, and one atom at B as the final configuration. If we use the dist.pl of vtst-scripts to check the distance between these two configurations, the value is very small. While we put them into VNL/Builder and try to create intermediate configurations, it shows a big value of distance.
Because of this, there may be some atoms cross the whole box in the NEB PATH that we created with VNL.
The VTST-scripts however, avoided this problem by doing things as follows:
if ($distance <= -0.5) {
$distance += 1.0;
} elsif ($distance > 0.5) {
$distance -= 1.0;
}
Therefore, the dist.pl and nebmake,pl works well.
How can I use BC method to create intermediat configurations and also avoid boundary problems?