Hello, I encountered a difficult situation with finding a transition state in Gaussian 09w. I have a molecule Ni(C5H5)2, which reacts with surface OH groups. As a replacement for the silicon surface, I used the cluster Si(OH)4. In the first reaction, we have:
Here, the calculations were performed at the DFT B3LYP/6-31G level. The transition state was found using QST2, and after the calculations, there was one imaginary frequency. Then, I performed the IRC calculation, and it turned out that along the reaction coordinate axis, the graph is built from the products to the reactants (instead of from reactants to products).
In the second part of the reaction, it is assumed that:
In this reaction, I cannot find the transition state using QST2. The reactants and products have been optimized. These are the first and second half-cycles in ALD reactions.
Does anyone know how to help with this?
When you did the IRC, did you do it with both signs? In principle, one of them should lead to the educts and one to the products and as far as I recall, there was no systematic with respect to the sign.
When QST2 is not successful, QST3 with a guessed transition state would be the next logical option.
In the first half of the reaction, I had the interaction Si(OH)3-OH + Ni(C5H5)2 => Si(OH3)-O-NiC5H5 + C5H6. After searching for the transition state QST2 (I had previously optimized them), I showed the possible reaction pathway. There was one imaginary frequency, and then I performed the calculation in IRC. I am attaching the data from the IRC.
If it's just that educts (reverse IRC) and products (forward IRC) are flipped, that's not a problem. To my extent of knowledge, the original input of the QST is not propagated into the IRC, so the direction would be lost. As long as the right substances are on the ends, no matter the order, the calculation has worked.
- Badges
- Science topic