CompChem.07.04 Excited Electronic States: Configuration Interaction

[@waatchit]

Dear Prof. Cramer, Thanks for your video lecture series on Computational Chemistry. I have been working on excitation calculation with Gaussian. Simultaneously , I have been trying to understand the concepts in electronic structure calculations. I have been listening to your lectures and reading your computational chemsitry book to gain the basic understanding. I have been having confusion on how to know which level is which orbital in acrolien or in any case. I did the ground state optimization of acrolien with hf/6-31g(d) and then did CSI/6-31G(d) calculation. I got the levels 15 and 16 to be HOMO and LUMO respectively but I am not sure how to know which levels are pi or pi* or non-bonding or sigma or sigma*. Can you please help me . It will be very useful for my understanding. Thank you.

[@ChemProfCramer]

+waatchit you would need to look at the MO itself (either by inspection of the coefficients that multiply the AO basis functions or through use of a visualization tool) in order to assign the character of the orbital(s).

[@waatchit]

+Chris Cramer . Thanks for your response Prof. Cramer. Can you please tell me from where can I know the coefficients of AO basis functions. And with the help of coefficients how can I deduce the character of the orbital(s) In case of visualization tool what software can I use sir. Thank you.

[@ChemProfCramer]

+waatchit electronic structure program output will generally contain the MO coefficients (those ARE the wave function, after all), although you may need to use a keyword to have them appear in the output file. That very much depends on the program. If you have a molecule with all atoms in the (x,y) plane and you see a MO that is composed entirely of atomic p(z) AOs, that would be a pi orbital, for example. There are many visualization packages, some proprietary (e.g., Gaussview for Gaussian), some free (for academics, e.g., MOLDEN).

[@waatchit]

+Chris Cramer . Dear Prof. Cramer , thanks for your response. Sorry for making such a long post. I have tried to understand the orbital character that you have mentioned in this video for acrolein. I ran CIS calculation for acrolien with the following keyword. # CIS/6-31g* #p td=(singlets,nstates=10) pop=(nbo,savenbo) pop=full And I found the largest transition with respect to oscillator strength at 182.56 nm. I have given the detail below. wavelength f 290.67 0.0001 182.56 0.5707 145.9 0.0015 137.21 0.0069 133.71 0 128.97 0.0002 120.6 0.0651 119.27 0.0011 114.6 0.0002 110.37 0.0007 And I also found from calculated "Excitation energies and oscillator strengths:" that the major transition occurs from the level 15 to level 16. I have given the details below for your reference. contd. in the next post (Z)

[@waatchit]

+Chris Cramer contd. from the previous post (Z) Excitation energies and oscillator strengths: Excited State 1: Singlet-A" 4.2655 eV 290.67 nm f=0.0001 =0.000 14 -> 16 0.62460 14 -> 17 0.30486 This state for optimization and/or second-order correction. Copying the excited state density for this state as the 1-particle RhoCI density. Excited State 2: Singlet-A' 6.7915 eV 182.56 nm f=0.5707 =0.000 15 -> 16 0.69101 Excited State 3: Singlet-A" 8.4981 eV 145.90 nm f=0.0015 =0.000 9 -> 16 -0.10202 10 -> 16 0.10028 10 -> 17 0.19091 11 -> 16 0.55930 11 -> 17 0.14963 12 -> 16 -0.23587 14 -> 16 0.11069 14 -> 17 -0.15945 Excited State 4: Singlet-A' 9.0358 eV 137.21 nm f=0.0069 =0.000 13 -> 16 0.57003 13 -> 17 0.17532 15 -> 17 -0.34514 Excited State 5: Singlet-A" 9.2729 eV 133.71 nm f=0.0000 =0.000 10 -> 17 0.11162 11 -> 16 0.25897 12 -> 16 0.56426 12 -> 17 -0.18521 14 -> 16 -0.13885 14 -> 17 0.16889 Excited State 6: Singlet-A" 9.6137 eV 128.97 nm f=0.0002 =0.000 10 -> 16 0.43694 11 -> 17 0.30537 12 -> 16 -0.15505 14 -> 16 -0.17450 14 -> 17 0.36557 Excited State 7: Singlet-A' 10.2804 eV 120.60 nm f=0.0651 =0.000 13 -> 16 0.31465 13 -> 17 0.14243 15 -> 17 0.60068 Excited State 8: Singlet-A" 10.3949 eV 119.27 nm f=0.0011 =0.000 15 -> 18 0.53063 15 -> 19 0.42733 15 -> 20 -0.11590 Excited State 9: Singlet-A" 10.8189 eV 114.60 nm f=0.0002 =0.000 15 -> 18 0.40878 15 -> 19 -0.38419 15 -> 20 0.39311 Excited State 10: Singlet-A" 11.2330 eV 110.37 nm f=0.0007 =0.000 8 -> 16 -0.10727 9 -> 16 -0.24989 9 -> 17 -0.19868 10 -> 16 -0.32422 12 -> 16 -0.14994 12 -> 17 -0.23496 14 -> 16 -0.10438 14 -> 17 0.38248 SavETr: write IOETrn= 770 NScale= 10 NData= 16 NLR=1 LETran= 180. Now I have to find the orbital character of these levels to understand the nature of transition and see if it matches with what is shown in the video. I then generated the cube file for levels 14 (HOMO-1), 15(HOMO), 16(LUMO) and 17(LUMO+1). Then I opened the file using Gaussview. I found the level 14 to be lying on oxygen with two lobes , just like a P orbital, but on the plane of the molecule(if I am right). Is it correct to asusme this to be non-bonding orbitals (may be two lone pairs of oxygen) Then level 15 was lying two c-c double bond,again with two lobes only but perpendicular to the plane of the molecule. (I asusmed it to be Pi orbital). Level 16 was lying on c-0 bond with two 4 lobes (with inverison symmetry),perpendicular to the plane of the molecule. Can I now assume them to be Pi* Level 17 was lying on c-c double bond , with 4 lobes (with inverison symmetry), so I assumed it to be another Pi* orbital. And it looks to me if I assume things the way I have mentioned above I match your video description. Can you please kindly let me know if what I have done is correct. Also I am not sure how to label HOMO-1 as n0, HOMO as Pi(2), LUMO as Pi*(3) LUMO+1 as Pi*4 as show in the video.

[@MengyuZhang-s8x]

Dear Prof. Cramer, Thanks for your wonderful lecture and it helpd me a lot. Could you send me your PPT if possible? Thank you so much!!

[@ChemProfCramer]

Unfortunately, the original ppts are all corrupted at this point. Only the pdfs (available at the listed site) display properly.

[@MengyuZhang-s8x]

Ahah thank you for your reminding!@@ChemProfCramer

[@Oneflyingchair]

How can I determine if a vertical transition is charge-transfer? That is, is there a way I can probe the CT character of an excitation using Gaussian?

[@ChemProfCramer]

sure. take a look at: joaquinbarroso.com/2020/04/06/natural-transition-orbitals-ntos-gaussian/ and you might also be interested in: hal.archives-ouvertes.fr/hal-02878834/document