Dear Suman,

I suggest that you run your calculations without symmetry.  Your compound has only 39 atoms which is considered as a medium size molecule and not a large one. The calculations should be fast. In the following link you have examples which show the keywords to include in the input file:

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

Hoping this will be helpful,

Rafik

Dear Suman,

although it would save a lot of time I also suggest to run your calculations without symmetry. This way you have the chance to end up with the lowest energy structure. TD-DFT with a large basis set and 39 atoms can take its time and I'm not too surprised that it takes a few days. However, that you end up with an error suggests that you start your frequency calculation with the wrong structure/symmetry. You can TD-optimize different structures using a smaller basis set and do a frequecy calculation also with a smaller basis set. just to test if the structure is at a minimum. If one of these calculations converges you take this structure and optimize it with your desired basis set. After that the frequencies with the desired set.

If your frequency calculations yield negative frequencies manually adjust the coordinates of these frequencies and redo the optimization.

Good luck

Robert

 an important example can be fined on:

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

in this example  "Emission (Fluorescence) from First Excited State (n→π*) of Acetaldehyde"  can be modified easily for your purpose.

Dear Suman GR,

for rapid optimization which will reduce the time of calculation, I advise you to write also the key word "scf=novaracc". it allows you to get around redondant calculations. you will find out this information in "Exploring chemistry with Electronic structure methods" of James B Foresman, second edition. the book is free of charge, it is also on line for he who needs it.

I hope that will help you.

Dear Suman,

It is not astonishing that your calculations are taking time but, the 39 atom system is not too big to be dealt with above mentioned basis set. However, to reduce computational time go for the excited state optimization without symmetry constraints. Additionally, you can also lower the basis set for frequency calculations and if you are getting desired results then go for higher basis set. 

Dear Mohsin

I tried without modifying the symmetry, i.e I took directly the previous required check file and optimised and continued to frequency calculation to confirm whether the optimised geometry is minimum or not but still it is consuming a lot of time. any how I'll try with lower basis set and let me see. If you have any further suggestion you can drop it in.

Thanking you 

Dear Suman,

Please let me know the number of processors, memory and the code you are using while carrying out your calculations?

Dear Mohsin,

It is intel (R) core (TM)  i3-3220 CPU with 8 GB RAM and I didn't understand what sort of code you are asking for. Hope my answer is clear

Thanking you.

Hi Suman,

By code, I mean to say whether you are using Gaussian, ADF etc for the computation and by memory I mean, the job memory/number of cpu's you are providing and not the system configuration.

Oh sorry, the job memory/number of cpu's, I'm giving default while preparing input using gaussview and code as I have mentioned in my query it is gaussian 09

Thanking you

Dear Suman,

Oops overlooked your query.What you need to do is, increase the number of processors (6 or 8 ) because by default it takes into account a single processor and increase the job memory too (512mb/1gb). I'm pretty sure that you will get the results swiftly.