Active Coordinates in the Jaguar Input File

In transition state searches, it is useful to be able to specify which coordinates are "active": that is, they are involved in the changes in bonding from the reactants to the products of the reaction. To specify that a coordinate is active, add “#active” to the end of the line (after one or more spaces).

As an example, the coord section

&coord
C1  C2  #active
C1  C3  #active
C2  C4  #active
C3  C4  #active
&

specifies that the C1–C2, C1-C3, C2-C4, and C3-C4 bonds are considered active in the transition state search. This might be the case if the C1-C2 and C3-C4 bonds in the reactants are broken to form the C1-C3 and C2-C4 bonds in the products.

The active coordinates can be used for three separate purposes:

• If transition state vetting is requested (through the keyword ts_vet), they are used rather than attempting to determine active bonds from the structures specified in the zmat sections.

• They are used to refine the Hessian at the start of the calculation, if nhesref is set to −1.

• They are used to select an eigenvector to follow in transition state optimizations (igeopt=2) if itrvec=−6 . This is done by choosing the vector with the lowest eigenvalue that satisfies the overlap criterion for a transition vector:

  

  where x_i are the active coordinates, v_j is the eigenvector of interest, and η is the cutoff specified by the keyword ts_vet_olap_cut.