Hasnain Hafiz
Dear all,
I am having an issue with the SMD solvation model and need your suggestion.
I am calculating the reduction potential of Ethylene Carbonate (EC) and its
complex with Li+ (Li+EC). According to the literature (attached paper) the reduction
potential of EC is smaller than that of Li+EC because of the high electron
affinity of Li+ cation. I am doing NWChem DFT calculation with B3LYP functional
and 6-31++G* basis set and SMD solvation model. The reduction potential (in V) with
respect to Li/Li+ scale is is defined as following:
Energy (neutral molecule) - Energy (reduced anion) – 1.4
My NWChem calculation with SMD gets the following reduction potential:
EC = 0.57 V and Li+EC = 0.16 V.
For sanity check, I repeat the calculation with Gaussian 09
code but with PCM solvation model keeping the B3LYP and 6-31++G* basis set. It
gives me reduction potential EC = 0.21 V and Li+EC = 0.56 V, which is almost exactly
match with the result of Hou’ et al.
I am wondering if anyone notice this problem before and is there any available solution
for this. I am attaching my input files and optimized geometries for your
reference.
Best regards,
Hasnain
Edoardo Aprà
Hasnain Hafiz
Thank you very much for your reply. The gas phase energies are consistent between SMD and PCM models and the reduction potentials are practically same between these models. The gas phase reduction potentials of EC and Li+EC from the SMD model in NWChem are -1.72 V and 2.52 V respectively. The gas phase reduction potentials of EC and Li+EC from the PCM model in Gaussian 09 are -1.72 V and 2.53 V respectively. Besides EC, we have done tests with Fluro Ethylene Carbonate (FEC) molecule and found the same issue with the SMD model. One of our collaborators from The Molecular Sciences Software Institute at Viginia Tech independently did these tests as well and confirmed this issue with the SMD model. We kindly request you to check this problem if that is a possibility. We will be grateful to you if you can do that.
Best regards,
Hasnain