Hi,

I am using gaussian 09 to simulate proton NMR and I am comparing my results with the ones published here:

Atieh, Z., Allouche, A. R., Graveron-Demilly, D., & Aubert-Frécon, M. (2011). Density functional theory (DFT) calculations of the proton nuclear magnetic resonance (NMR) spin-Hamiltonian parameters for serine. Measurement Science and Technology. doi:10.1088/0957-0233/22/11/114015

Unfortunately, I am obtaining a deviation for the solvent simulation which can be explain by the recent changes in PCM models of Gaussian 09 (see here). I have came across with the option g03defaults which allows to reproduce the same results obtained with g03. However, if I set this option to compute the force constants and the resulting vibrational frequencies (option Freq), I obtain the following error:

Warning! Using old PCM defaults because it has been requested by the route generator. These settings can be overridden by following keywords. Analytic 2nd derivatives not available using g03defaults. Error termination via Lnk1e in /pathtogaussian/g09/l301.exe

Does anyone have an idea how to force the calculation of the analytic 2nd Derivatives?

If I skip the option G03Defaults for the vibrational frequency calculations, I get comparable results to the reference above, but the simulations tend to break during a second step which is just a tighter optimization step (before Freq calculations).

Thanks in advance for any help.