Hello for all researchers
I have a system of C2H2/KCl(001) and I did Frequency calculations using Gaussian 09 software. I used the following route section:
#T b3lyp 6-31G(d) Freq SCF=Tight .
My the first question is: Why some values of frequency were negative? what does it mean?
I have please another question about the SCF convergence criterion for my calculation, the following what I got in output file:
Item Value Threshold Converged?
Maximum Force 0.005110 0.000450 NO
RMS Force 0.000680 0.000300 NO
Maximum Displacement 0.165810 0.001800 NO
RMS Displacement 0.027053 0.001200 NO
GradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGrad
I didn't understand why No Converged?
Thanks in advanced
In general, negative frequency values are a result of the process of the Fast Fourier Transform (FFT). They do not have any physical meaning as such (though there is some interpretation as the frequency of a wave moving backward in time).
For all practical purposes, these negative frequencies are treated the same as positive frequencies.
This is a general answer. i am not sure if this is correct with respect to Gaussian software.
Are you talking about Raman Frequencies or something like that?
Hi Safiyah, most likely your problem is that the geometry that you are using for the calculation has not been previously optimized with Gaussian at the level of theory you are using (b3lyp 6-31G(d)). Then, the program does not recognize it as a stationary point in the potential energy surface (and then you get imaginary frequencies). The convergence results that you see are not related to SCF convergence, in a 'very simple' way those values indicate how far your calculation is from being a stationary point.
I would suggest to read this post --> https://www.researchgate.net/post/In_calculated_with_software_of_gaussian_09_if_frequncy_become_negetive_what_do_you_do
or go to the Gaussian webpage regarding the 'opt' keyword --> https://gaussian.com/opt/
Dear Safiah,
Negative vibrational frequencies indicate a transition state. However، the transition state is defined as having one and only one negative (imaginary) vibration frequency. What you are observing probably due to technical error, such as conducting the frequency analysis on a geometry optimized with a different level of theory, or by performing the analysis on a geometry far away from its energy minimum. Frequency calculation should be performed on optimized geometry. A gradual stage of optimisation with small to higher basis set, could really help to avoid the SCF convergence failure.
Regards
Abishek Hariharan thanks for reply.
I have not much experience about calculations using G09, but as I know when you use the keyword "Freq" without any other options, you get Harmonic frequencies, and IR intensities, then you can see modes of vibrations by using GaussView.
So I am talking about IR Frequencies.
Thank you so much for reply Juan V. Alegre-Requena . Yes as you guessed the geometry that I'm using has not been previously optimized with Gaussian at the level (b3lyp 6-31G(d)), because of it is already optimized by using Monte Carlo method. So I did not need to optimize the geometry of the system under studying, just need to calculate IR frequencies.
The links were useful for me, specially the first of them. thanks again for your clear answer.
A frequency, by definition, is how often something happens (i.e. frequency of what, in your case, by the sound of it, the number of vibrations) per unit time. A calculation, therefore, that gets a negative number is perforce wrong, which means there is something wrong with the methodology.
Thank you for reply Melek Hajji .
As you said the geometry I'm using has not been previously optimized with Gaussian at the level used in frequencies calculation, because it is already studied by others who got a stable structure of C2H2/KCl using Monte Carlo simulation. for that I used that stable geometry.
Regards
only complex signals have positive and negative frequencies.
The formal evidence is relation between the cosine function and its complex signals formulation;
cos wt = .5(exp j wt + exp - jwt)
you see a real vibration signal can be represented as a sum of two exponential signal with positive and negative frequency.
real signals have positive frequency and complex signals can have positive and negative frequencies.
Best wishes
- Badges
- Science topic
- Similar topics
- Force