Timeline for Gaussian: Relaxed scan with modredundant optimization and dummy atoms
Current License: CC BY-SA 3.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 10, 2018 at 15:01 | history | edited | Martin - マーチン♦ | CC BY-SA 3.0 |
updated with disclaimer
|
| Apr 15, 2016 at 10:08 | comment | added | shrx | @Martin-マーチン no problem, I have already upvoted your answer as I think it is quite informative, even if it doesn't lead to the desired result. If I have further questions I will contact you in chat. | |
| Apr 15, 2016 at 10:01 | history | edited | Martin - マーチン♦ | CC BY-SA 3.0 |
corrected wrong assumption
|
| Apr 15, 2016 at 9:58 | comment | added | Martin - マーチン♦ | @shrx I really don't want to go any deeper into this discussion here. If you want to talk about it, feel free to find me in chat. I answered your question to the best of my knowledge and I am very sorry that I could not help you with your problem. | |
| Apr 15, 2016 at 8:23 | comment | added | shrx |
This is the result of my simultaneous scan over the C-N and O-H distances (animation made with molden): i.imgur.com/NjhSJyN.gif
|
|
| Apr 15, 2016 at 8:14 | comment | added | Martin - マーチン♦ | @shrx Well technically you are scanning only one parameter in your example input. You can do a multidimensional scan by providing more than one scan command in the modredundant section. I am still not sure what you are actually trying to achieve, because I believe that there is no true concerted transition state. | |
| Apr 15, 2016 at 7:51 | comment | added | shrx |
Indeed I'm trying to do the C-N bond formation and simultaneous proton transfer. I tried doing it the simple way like you did, to just move the C and N atoms closer together. However, this does not lead to the desired proton transfer, as evident in your animation. I provided my input file as an example, but I am interested in a general way to do simultaneous scan over more than one variable with the modredundant optimization in Gaussian. For now, I have resorted to a not yet fully-functional python script that simulates a multi-variable scan. I can provide the code if there is interest.
|
|
| Apr 15, 2016 at 5:26 | comment | added | Martin - マーチン♦ | I just realised, that this bug may have resulted from preprocessing of the input file, inserting spaces within a keyword, which the produces a syntax error. It might not be gaussians fault after all... | |
| Apr 15, 2016 at 5:23 | history | edited | Martin - マーチン♦ | CC BY-SA 3.0 |
spell checking and extending the statement about the line length
|
| Apr 15, 2016 at 5:07 | comment | added | Martin - マーチン♦ | @LordStryker Thank you and happy to see you. I could not reproduce this bug anymore, so I am fairly certain they fixed it. Gaussian used a fixed width of 80 char in the input (and still uses that as output for readability, but not everywhere) and break longer lines, apparently that could have destroyed some keywords. My former supervisor told and showed us, but that might have well been the 98 or 94 version. I'll fix that statement. | |
| Apr 14, 2016 at 14:42 | comment | added | LordStryker |
Gaussian sometimes has (still) trouble reading in anything beyond the 80th char. Wow, I never knew this... Can you offer up an explicit example? I'd love to see this behavior. Great answer by the way.
|
|
| Apr 14, 2016 at 14:02 | history | answered | Martin - マーチン♦ | CC BY-SA 3.0 |
