There is not enough information to give a good answer here. If you did a geometry optimization the threshold for symmetry (since it is never numerically perfect) might be to low. You can specify a lot of options with the key word symmetry. See here:

http://www.gaussian.com/g_tech/g_ur/k_symmetry.htm

It might be good to define a symmetry and fix the point group before running the optimization. How this can be done is described here:

http://www.gaussian.com/g_tech/afc.htm

Which point group does Gaussian09 recognize?

A hint: G09's manual seems to be offline by a lot of time (at least, it cannot be reached by my current position), however it's fully cached into Google.

Thank you  very much Denis and Francesco . I will try to optimize by fixing the symmetry group. I did opt+freq calculation and it shows C1 point group. 

Also if the Gaussian does not show correct point group, is the calculated frequency  wrong ?

It will have some small error, but in real measurements in medium it will also have incorrect point group, so may be it is not so critical. The main problem will be splitting of collective vibrational modes, so you'll need to join vibrational modes with little differences in frequency manually into single active vibration. And there also will be frequency overestimation, that should be corrected: http://cccbdb.nist.gov/vibscalejust.asp

Thank you Daniil for your help.

If G09 sees the molecule as C1 then it means it doesn't recognize any simmetry.

This could have some minor effect on the computed spectrum, as Daniil correctly said, particularly on low frequency modes.

On high frequency modes the effect of lowering the simmetry is really small.

Hi Soma,

Gaussian should recognize Zinc phthalocyanine as having D4h symmetry. If you run your calculation with lower symmetry, the symmetry labels of the vibrational modes will be incorrect (and doubly degenerate modes may come out as being singly degenerate).

You should set the symmetry to D4h symmetry and then rerun the optimization and frequency analysis. If you cannot set the point group to D4h, use the 'very loose' tolerance setting. If this does not come up as an option, you may need to edit the starting geometry you are using to give it D4h symmetry.

Thanks a lot for your help. How can I edit the starting geometry  to give it to d4h symmetry ?

If you are using the molecule builder on Gaussview you can set the point group under the edit tab.

You can also put the command 'PG=D4h' into the route section of your input file.

Hi Chris, If I  set the point group I am not able to run it. Its showing me error. I set the point group to D4h under the edit tab in the point group. I used DFT, B3LYP ,6-31 G . What else I can do ?

What is the nature of the error you are seeing?

Hi Chris ,It shows " Inaccurate quadrature in cal D su ". Also I am attaching the log  file .

Hi Soma,

You will need to run an optimization job on this structure before attempting a frequency analysis. That should fix the problem you are having.

Thanks a lot. I will do that.

Hi Chris,

I can optimized to D4h symmetry group but i can not do frequency calculation It shows "Inaccurate quadrature in cal D su " error.

Did you run a frequency job without the Raman intensities?

With Raman intensity.  What are the other option I can try ?

You can run a frequency job without calculating Raman intensities (freq=noRaman). You will get vibrational frequencies and the infrared intensities. This should tell you if your molecule has been optimized to its ground state structure (you should get no imaginary frequencies).

Try computing the Raman intensities when you are sure you have the correct optimized geometry.

Inaccurate quadrature in calDsu problem could be resolved by the following keyword; int(ultrafine). If the molecule has lot of low frequency modes, its better to keep the integration grid ultrafine.

I recently solved the problem of "Inaccurate quadrature in CalDSu" in Gausian 09 by adding the statement (integral=NoXCTest) to the input command line. This statement stops any checking to the accuracy of quadrature calculations.