Accurate, wide-range, automated, high-performance liquid chromatographic method for the estimation of octanol/water partition coefficients II: Equilibrium in partition coefficient measurements, additivity of substituent constants, and correlation of biological data.

Occasionally, results from the highly reproducible automated log P(o/w) measurement (ALPM) differ from those determined by shake-flask methods. Several specific examples affording different values are presented. One source of these differences may be curvilinearity in plots of log (t - t0) versus percent methanol, which complicate accurate intercept determinations and, thus, estimates of log P(o/w). Other sources of these differences are presented and discussed, although their cause remains unclear. Equilibrium ALPM log P(o/w) measurements of various phenyl-, methyl-, fluoro-, chloro-, and bromobenzenes, suggest substituent constants are not strictly additive. Moreover, the higher values indicate that calculated values may not be accurate for those compounds having multiple substituents or high log P(o/w) values. ALPM gives better predictability of the in vivo concentration process of 8 or 12 toxicants in fish than the shake-flask method, another HPLC method, or even calculated log P(o/w) values. However, it equally correlates the binding to bovine serum albumin by 34 chemicals as predicted by a combination of shake-flask and calculated log P(o/w) values reported elsewhere.