This is not easy to answer without knowledge of the system. However comparing computed spectra with the experiment is always tricky. It sounds like youre only sampling on a single conformation. Is your system rigid enough to justify that?

A few other questions on the computational side

Did you include solvation effects in the geometery optimization?

If yes: How did you include solvent effects?

Are solvent effects considered in the TDDFT calculation? If yes, how?

After all TDDFT reproduces spectral shifts between solvents or the vacuum and a solvent pretty well, but may give absolute excitation energies that may be pretty far off. If  absolute excitation energies are important you should switch to more sophisticated methods like CC2 if your system size allows that.

Mr.  Biswas,

The accuracy of TDDFT for organics is 1-4 nm using M062x functional. Could you provide an example of a system similar to your one?

Before selection of DFT functionals, you should characterize transtion property of target molecule Generally it is diveided  two different types of transition behavior.In charge transfer transition, you need to use range separated functional or functional including high exchange function such as cam-b3lyp, M06-2X, etc ..In normal p-p transition, normal hybrid functional is enough. Sometimes, GGA functional gives a great prediction. After selection of proper functional, you can employ solvent effect.to predict exact value of absorption wavenlength.

If PCM method is not sufficient to obtain exact absorption wavenelngth, you can try to calculate real nature of system with QM-MM

Hope this help

I

Thanks. for your answer @ Sunwoo Kang. I already tried cam-b3lyp and solvent with PCM. I am sorry I could not get "you can try to calculate real nature of system with QM-MM". It would be great f you could explain it little bit more

Dear all,

Thank you for your help. I am sorry for my late reply. Here I am attaching a generic version of the molecule.t  The molecule is  rigid (The crystal structure and optimized geometry are in good agreement).  The molecule that I have attached is little bit weird (Patent issues restricts me). I truly appreciate all of yours help.

Thanks

Bibaswan

@Bojidarka B. Ivanova 

Do I need to use CAM-M062x or only M062x.

@Heiner Schröder,

Molecule is rigid.

A few other questions on the computational side

Did you include solvation effects in the geometery optimization?

Yes

If yes: How did you include solvent effects?

PCM both in ground state and tddft

Are solvent effects considered in the TDDFT calculation? If yes, how?

I am not sure what answer you are looking for but I am using pcm

 Thanks a lot

Try using some Multi-Reference approximation (MR-CI, CASPT, etc.)

Best regards.

Your problem is most likely more fundamental then the choice of the correct method or functional (your functional is fine!). Already for small, rigid molecules a lot can go wrong

10.1021/acs.jpclett.5b00990

Your geometry might not be a good representation of the solute and/or your solvent model is not a good representation of your solvent. If there are important specific interactions between the solvent and your solute (e.g. H-bonds, Pi-interactions, ...) PCM is not well suited. Without switching to an explicit representation of the solvent you might try COSMO-RS. Going to QM/MM got its own set of problems and I wont recommend it to inexperienced users. And its definitely not guaranteed to improve your results in any way.

Please remember this one. The functional including high exchange function generally lead to blue shift behavior of absorption and emission spectra.

I think you need to find best functional to describe your system. This is most important thing. After that you should correct transition property with PCM or direct interaction between solute and solvent

The functional affects the band position within 40-1 nm, but here we are talking about 90 nm shifting, which rather indicates the problem with the initial electronic configuration of the molecule.

In general, the effect of functional on transition property is not over 50 nm. Sometimes, the great wavelength shift over 50 can be achieved with respect to different functional 

Also, my opinion for this question is that proper funtional, which can well describe mole ular transition property, should be adopted at first then need to consider solvent effect or inter or molecular interaction or excited proton transfer or somthing else. 

In case the -X was Fluoro, I used DFT/6-31* for optimization and TDDFT.

here the result

Did you calculate TD DFT with solvent in CPCM model?

@Saha Yes.

Dear Dr. Sunwoo Kang

As you have mentioned in your comments that "The functional including high exchange function generally lead to blue shift behavior of absorption and emission spectra." or it makes ~50nm blue shifts....

Can you please send me any reference about your comment? Recently I am trying some large polymer monomer which also showing the blue shift ~ 60 to 80 nm.

It will be very helpful if you can provide some reference about your comments, Thank you so much Sunwoo Kang

Mamun

I agree with Heiner Schröder

Please take care of every parameter. Further, if yu want to exploit only DFT/TD-DFT, then you can go for functional and basis set assessment..

Regards