On Tuesday, December 20, 2022 at 07:51:03 PM GMT, Grigoriy
Zhurko reg_zhurko_-_chemcraftprog.com <owner-chemistry|*|ccl.net> wrote:
I need to compute the pKa values of some
carbon acids and phenols, and it is a common practice to add 1-3 explicit water
molecules, bonded with polar groups of the studied molecule, to better take into
account the specific solvation effects (in addition to CPCM model). I tried to
add many water molecules (14) for better results – see attached picture.
The results are interesting. At one side, possibly very big error is produced by
the existence of many local minimums with different numbers of h bonds. How
these minimums should be named in paper – maybe
“conformations”, “configurations”,
“geometrical configurations”, “local
minimums”?
Then I found, that if the computation
of different acids is performed with same geometrical configurations of 5 h2o
molecules bound to –COOH, and with same configurations for neutral
molecule and anion, possibly the results are quite good (a common correlation
can be obtained for carbon acids and phenols). So I want to publish these
results; but I need to read more works when this approach (adding many explicit
water molecules) was used. Can you suggest such papers?
By the way, using my program Chemcraft is important for such jobs,
because the “Structures merger” utility in in allows one to
transfer several H2O molecules in same configuration from one acid to
another.
Grigoriy Zhurko