I am trying to perform a transition state optimization on the attached sn2 reaction and everything I've spit back out is very ugly (output had multiple bond triads, large migration of F and Cl to mid chain, and other big n-no's).
I have read several of the G09 manual pages that talk about freezing geometries but it wasn't clicking. I was wondering if I could see an example file that shows various examples of how the frozen geometry actually looks inside the file. I wanted to constrain some bond angles on the SN2 to do a partial optimization of the other portion, then proceed to hopefully find the final state with one unconstrained optimization.
In the end of input (after the atomic coordinates) your should choose the parameter (bond, angle etc.) ant try the parameter value and symbol F
For example
1 2 1.005 F
2 3 4 151.03 F
Dear Skyler,
Even though obtaining the TS associated with a SN2 reaction does not normally need to freeze any coordination but if you would like to do any freezing during the optimization you can do as follows:
you should cut the bond distance (presented as B1, B2, etc.) or angle (presented as A1, A, etc.) in the molecular specification section then paste them at the end of molecular specification after a blank line. For example, consider the structure of benzene with the following jgf file:
C
C 1 B1
C 2 B2 1 A1
C 3 B3 2 A2 1 D1 0
C 4 B4 3 A3 2 D2 0
C 1 B5 2 A4 3 D3 0
H 1 B6 6 A5 5 D4 0
H 2 B7 1 A6 6 D5 0
H 3 B8 2 A7 1 D6 0
H 4 B9 3 A8 2 D7 0
H 5 B10 4 A9 3 D8 0
H 6 B11 1 A10 2 D9 0
B1 1.39516000
B2 1.39471206
B3 1.39542701
B4 1.39482508
B5 1.39482907
B6 1.09961031
B7 1.09965530
B8 1.09968019
B9 1.09968011
B10 1.09976099
B11 1.09960403
A1 120.00863221
A2 119.99416459
A3 119.99399231
A4 119.99845680
A5 120.00431986
A6 119.98077039
A7 120.01279489
A8 119.98114211
A9 120.01134336
A10 120.00799702
D1 -0.05684321
D2 0.03411439
D3 0.03234809
D4 -179.97984142
D5 179.95324796
D6 179.96185208
D7 -179.99643617
D8 -179.99951388
D9 179.98917535
Assume that we are going to optimize this structure in which B1 and A6 should be frozen. B1 and A1 should be removed and included at the end of gjf file after a blank line leading to the new gjf file as follows:
C
C 1 B1
C 2 B2 1 A1
C 3 B3 2 A2 1 D1 0
C 4 B4 3 A3 2 D2 0
C 1 B5 2 A4 3 D3 0
H 1 B6 6 A5 5 D4 0
H 2 B7 1 A6 6 D5 0
H 3 B8 2 A7 1 D6 0
H 4 B9 3 A8 2 D7 0
H 5 B10 4 A9 3 D8 0
H 6 B11 1 A10 2 D9 0
B2 1.39471206
B3 1.39542701
B4 1.39482508
B5 1.39482907
B6 1.09961031
B7 1.09965530
B8 1.09968019
B9 1.09968011
B10 1.09976099
B11 1.09960403
A2 119.99416459
A3 119.99399231
A4 119.99845680
A5 120.00431986
A6 119.98077039
A7 120.01279489
A8 119.98114211
A9 120.01134336
A10 120.00799702
D1 -0.05684321
D2 0.03411439
D3 0.03234809
D4 -179.97984142
D5 179.95324796
D6 179.96185208
D7 -179.99643617
D8 -179.99951388
D9 179.98917535
B1 1.39516000
A1 120.00863221
Now, this file is ready to be used by Gaussian in which benzene will be optimized with B1 and A1 as constants (frozen). It should be noted that in order to consider such freezing during the optimization including "opt=z-matrix" in the command line is mandatory. For example:
#b3lyp/6-31g* opt=z-matrix
Good luck.
Thank you for the clear and easy to follow instructions Saeedreza. To make sure, essentially I can freeze any part of a geometry by just moving that parameter to the end of the file after inclusion of 1 newline as long as I've specified Z-matrix in my optimization? No extra F's or any other special tweaks with ModRedundancy or etc?
Yes, as I mentioned you can freeze any parameters you would like. F`s or any other considerations do not necessary to be included. If your problem remains unresolved contact me via following mail in order to more help:
- Badges
- Science topic