Corrected version:

The calculation of logD should always be a two step procedure, consisting of the calculation of all species of the compound of interest (let us call it X) in water and in the organic phase (usually wet n-octanol, but it may be alkane, or whatever) and the calculation of the partition coefficient of the neutral species.

Usually the compound will be in a neutral form in the organic phase. Let usconsider this form as the reference form of X. Now you can calculate the partition coefficient P or logP with various approaches. For this part often group contribution methods as CLOGP or logKow (from EPA) will do a good job, as long as you stay in their range of validity, i.e. standard organic chemistry without tautomerisation.

But for logD you need in addition the portion of the compound X which is in this reference state in water. In water it can protonated, deprotonated, zwitterionic, tautomeric, ....
Usually protonation and deprotonation is most important (but the other cases can happen as well). For the portion of protonated and deprotonated species you need to define the pH of water, usually set to 7.4.
Then you need the pKa and pKb of the compound and use the Hendersson-Hasselbalch equation in order to get the portion of the different species. Now youcan correct the neutral partition coefficient by the speciation equilibrium. In other words: You have to calculate the free energy of all possible species of X in water (neutral, maybe multiple protonated, multiple deprotonated, zwitterion and tautomeric) at pH=7.4, build the partition function Z relative to the energy of the reference state, and calculate D = P/Z.

Perhaps read the articles on the blind prediction challenge SAMPL5 on the logD(cyclohexane/water), which have just appeared as a special issue of JCAMD. (Un)fortunately the bet method in that contest is not free. It is our COSMOtherm program.

Again, for standard organic chemistry you will find free software doing the job. But if you have some more complicated compounds (as those 53 drugs), you need a method which takes into account all possible complications.

Best regards

Andreas

Am 21.10.2016 um 01:27 schrieb Yingduo Gao yingduo * yahoo.com:

-- -------------------------------------------------- Prof. Dr. Andreas Klamt CEO / Geschäftsführer COSMOlogic GmbH & Co. KG Imbacher Weg 46 D-51379 Leverkusen, Germany phone +49-2171-731681 fax +49-2171-731689 e-mail klamt[#]cosmologic.de web www.cosmologic.de [University address: Inst. of Physical and Theoretical Chemistry, University of Regensburg] HRA 20653 Amtsgericht Koeln, GF: Prof. Dr. Andreas Klamt Komplementaer: COSMOlogic Verwaltungs GmbH HRB 49501 Amtsgericht Koeln, GF: Prof. Dr. Andreas Klamt