I tried to create a QSAR (Quantitative Structure Activity Relationship) model for my thesis that tries to predict the melting points of certain ZIFs (Zeolitic Imidazolate Frameworks) based on relative Zn-N bond energies obtained from DFT calculations (B3LYP/aug-cc-pVDZ) as well as Labute's Solvent Accessible Surface Area (LASA) and the stoichiometry of a mixed-linker "Xim-" for a ZIF of sum formula Zn(im)_2-x(Xim)_x.
I found out that the calculated bond energies correlate more or less with Hammett-Parameters (meta-position, sigma_m)
and so I got to thinking: What if I had a nice json or csv table i could easily import in python or KNIME (Konstanz Data Miner), that contains a long list of Hammett-Parameters for certain substituents? That way I may not have to perform a DFT calculation for each new substituent X and instead use the Hammett-Parameters, which are already available.
What I found is this publication, which probably has the longest list of Hammett-Parameters available. The problem: It's a PDF and I don't think that gels well with... programmatically importable data?
Now I am willing to manually turn this into a csv/json of my very own, but I'd rather ask first if something like this already exists, if there's a source for such data that I do not know about, or if there is a smart way to convert this PDF into a nicer format that can easily be imported into python.
