Higher wavelength corresponds to lower energies. You are looking for one of first excitations (most likely first excited state). There is no point to request 12 24 or 48 states, i think default 3 is more than enough in this case. Reaching 200 nm is meaningless. If you dont have excited state with wavelength you expect reason might be in functional, however i am wondered that any DFT would fail too bad for such simple system as pentacene.

Dear Muath

Here i attached TDDFT results of pentacene

(B3LYP/6-31G* level of theory)

You can see the absorption peaks in range of 400 - 600 nm

Hope this help

Dear Sunwoo Kang

Thanks a lot for your reply,

Which basis set you are using ? I used lanl2dz with the following command in Gaussian to get my results :

#p td=(nstates=12) rb3lyp/lanl2dz scrf=(solvent=acetone) geom=connectivity

Dear Dmitry Sharapa

Thanks a lot for your answer

lanl2dz has no point for this system. Take something traditional, from Pople (6-31), Dunning (cc-pVXZ) or def2-XXXX families. Some diffusion would be fine.

geom=connectivity and connectivity at the end of input are useless for DFT, as well as R at rb3lyp. These willnot bring error in calculation, just pointless. Decreasing number of states will have significant positive effect on convergence speed without any effect on energy of first excited state.

Anyway, your pictures are almost same - first peak around 630 and second at 380. Still might be complex to compare with experiment - pentacene is purely soluble in anything.