From owner-chemistry@ccl.net Fri Feb 25 09:28:00 2011 From: "Patricia Richardson patricia.richardson(~)ed.ac.uk" To: CCL Subject: CCL: UV-VIS spectra values obtained computationally are different Message-Id: <-44020-110225092507-4504-wXyCtGP3xji3lkSM5J/xXA,,server.ccl.net> X-Original-From: "Patricia Richardson" Date: Fri, 25 Feb 2011 09:24:57 -0500 Sent to CCL by: "Patricia Richardson" [patricia.richardson-.-ed.ac.uk] Hi Ashu, CIS tends to systematically overestimate excitation energies, and so people often scale them by a factor of 0.72. See Broo and Holmen for more details. A. Broo and A. Holmen, Calculations and characterization of the electronic spectra of DNA bases based on ab initio MP2 geometries of different tautomeric forms, J. Phys. Chem. 101 (1997), pp. 3589?3600 Scaling your CIS calculated transition energy (280 nm) gives 389 nm, much closer to the TDDFT value. Of course, you should always check that the CIS and TDDFT transitions are giving the same electronic transitions - same orbitals, similar oscillator strengths etc. Certainly high concentrations can lead to the formation of aggregates in solution, which may have a markedly different absorption spectrum than the monomer species. However, could it be that the band at 400 nm simply has a low extinction coefficient, and so you don't see it at lower concentrations? You should be able to measure the extinction coefficients of each band from from your UV-VIS spectra, and use the Beer-Lambert law to get an idea of the concentration/path-length range at which they will be observable. The calculations will also give you the oscillator strength, which is proportional to the extinction coefficient (see Hollas - Modern Spectroscopy). So if there is a big differential between the oscillator strength of your UV peaks and that at 400 nm, that may explain it. I'm assuming you still see the UV peaks in your high concentration spectrum, though they may be completely saturated? In the experimental UV-VIS spectra you'll see transitions to other higher energy electronic states, and you can usually set the number of states you'd like to calculate in your input file, so you can span the full experimental range, which you might need to do if you aren't seeing the UV bands in your TDDFT calculation. Hope this is helpful. Best wishes, Tricia