Dear Amir Hossein Mohammadzadeh ,

This problem arises when there are arrangements of three atoms that form an angle of exactly 180 degrees and these atoms are used to define some dihedral, which obviously cannot be defined in such situations. Right before the error message you should have in the output the list of linear three-atom arrangements that you have in the structure, just modify the angles to make them slightly smaller than 180 degrees (175 or so) and the calculation should run with no problem.

Hope you find it helpful

Dear Amir Hossein Mohammadzadeh ,

I agree with Enrique Manuel Arpa . I often fell on this type of problem and have succeeded in solving this latter using "opt=cartesian" with coordinates given in xyz format.

For your purpose, I already studied a clathrate and I give you my optimized xyz file just below (M062X/Def2-TZVP, gaz phase).

The problem with your calculation, similarly to what was stated before, is only a failure in the redundant coordinate system. Gaussian automatically builds redundant coordinates at the start of every calculation. During the course of an optimization these can become close to 180 deg and then the algorithm fails. There is in general nothing wrong with your calculation, the algorithm just cannot continue this way. It is not a problem at all.

1. The linear bend problem can be solved using Cartesian coordinates: Opt=Cartesian

2. It might happen occasionally that some of the internal coordinates move out of their comfortable range, like close to linear bond angle, too long bond, etc.This cases may cause the internal coordinate transformations to fail. One possible solution is to continue the optimization, using a copy of the previous checkpoint file and the following options in the command line: Opt=ReadFC Geom=(CheckPoint, NewDefinition) Gueass=Read

This will read the last geometry from the checkpoint file, rebuild the coordinate system and very likely eliminates the problem.