Partition coefficient ion pairs

For the parhtioning of the latter ion, the ion-pair partition coefficient may be defined as ... 

The use of distribution coefficients for the QSAR treatment of ionizable compounds has been extended to consideration of ion-pair partitioning into biolipid phases. Two experimental methods for determining ion-pair partition coefficients are described. One is a single-phase titration in water-saturated octanol, in which case (for acids) log Pj = log P + pKa - pKa. The other is a two-phase titration (octanol/water) from which the ratio (P + 1)/(Pj + 1) can be calculated. An example outcome is that the uncoupling activity of phenols can be represented by an equation in log instead of log D and pKa. 

If [X] is the concentration of AH in the aqueous phase (Figure 8), the concentration of AH in the octanol is P times this. The aqueous-phase concentration of the ion is [X] times the degree of dissociation. Multiplying this product by the ion-pair partition coefficient Pj (or P -) gives the concentration of the ion pair in the octanol. The actual amount of species in a phase is given by its concentration times the volume of the phase. At the pK, the amount of neutral species equals the amount of ionized species. Setting the sum of the two terms in the left quadrants equal to the sum of the two terms on the right, one can derive Equation 20. The equivalent e ression for bases is Equation 21. 

However, as stated above, the partition coefficients measured by the shake-flask method or by potenhometric titration can be influenced by the potenhal difference between the two phases, and are therefore apparent values which depend on the experimental condihons (phase volume ratio, nature and concentrahons of all ions in the solutions). In particular, it has been shown that the difference between the apparent and the standard log Pi depends on the phase volume raho and that this relationship itself depends on the lipophilicity of the ion [80]. In theory, the most relevant case for in vivo extrapolation is when V /V 1 as it corresponds to the phase ratio encountered by a drug as it distributes within the body. The measurement of apparent log Pi values does not allow to differentiate between ion-pairing effect and partihoning of the ions due to the Galvani potential difference, and it has been shown that the apparent lipophilicity of a number of quaternary ion drugs is not due to ion-pair partitioning as inihally thought [80]. 

When the ion-pair partitioning is indicated in the quadrant diagram (below) it becomes obvious that a circle of equilibria is present. Knowing the octanol pKa, the log P and the aqueous pKa should allow one to calculate the partition coefficient of the ion pair. From these equilibria one can write that the difference in log P between the acid and its salt is the same as the difference between the pKa s (Equation 9). The closer the pKa s, the more lipid soluble the ion pair will be, relative to the acid. Internal hydrogen bonding or chelation that stabilizes an ion pair will affect the octanol stability more than the aqueous stability, where it is less needed, and so will decrease the delta pKa. Chelation should therefore favor biolipid solubility of ion pairs. Ultimate examples are available in some ionophores. This is one of the properties of some of the herbicides I pointed out earlier. 

We decided to try a direct titration in water-saturated octanol. What we hoped to achieve, if not duplicating literature values, was to obtain partition coefficients proportional to "true" values so that regression analyses could be run and would be meaningful. Secondly, this should be a rapid method to assess structural features in a series for their effect on ion-pair partitioning. 

The recent work of Wang and Lien (29) illustrates that ion-pair partitioning occurs to a greater extent than previously realized. Partition coefficients calculated from measurements made on partially ionized compounds depend not only on the pH, but on the buffer used. They may vary by more than one log unit. The authors derived equations to correct log P to octanol/water values, but these can still be off by several tenths of a log unit. A preferable solution would be to know the log Pj and account for ion-pair partitioning. 

Ageneraiized, weighted, noniinear ieast squares procedure is deveioped, based on pH titration data, for the refinement of octanoi-water partition coefficients (iog P) and ionization constants (pKa) of multiprotic substances. Ion-pair partition reactions, self-association reactions forming oligomers, and formations of mixed-substance complexes can be treated with this procedure. The procedure allows for CO2 corrections in instances where the base titrant may have CO2 as an impurity. Optionally, the substance purity and the titrant strength may be treated as adjustable parameters. The partial differentiation in the Gauss-Newton refinement procedure is based on newly derived analytical expressions. The new procedure was experimentally demonstrated with benzoic acid, 1-benzylimidazole, (+)-propranolol, and mellitic acid (benzenehexacarboxylic acid, AH6). 

Avdeef, A. pH-metric logP. Part 1. Difference plots for determining ion-pair octanol-water partition coefficients of multiprotic substances. Quant. Struct.-Aaiv. Relat. 1992, 11, 510-517. 

The approach taken here is to use the lattice strain model to derive the partition coefficient of a U-series element (such as Ra) from the partition coefficient of its proxy (such as Ba) under the appropriate conditions. Clearly the proxy needs to be an element that forms ions of the same charge and similar ionic radius to the U-series element of interest, so that the pair are not significantly fractionated from each other by changes in phase composition, pressure or temperature. Also the partitioning behavior of the proxy must be reasonably well constrained under the conditions of interest. Having established a suitable partition coefficient for the proxy, the partition coefficient for the U-series element can then be obtained via rearrangement of Equation (2) (Blundy and Wood 1994) ... 

Scherrer, R. A. Crooks, S. L., Titrations in water-saturated octanol A guide to partition coefficients of ion pairs and receptor-site interactions, Quant. Struct.-Act. Relat. 8,59-62 (1989). 

Note that no ion pair is assumed in the aqueous phase. The overall distribution ratio of the charged species is then a combination of the partition coefficient of the charged species and the distribution ratio of the ion pair. 

The thermodynamics of the extraction mechanism is extremely complex. In the initial equilibration of the ion pairs (Scheme 1.6) account has to be taken not only of the relative stabilities of the ion-pairs but also of the relative hydration of the anionic species. Assuming the complete non-solvation of the ion-pairs, the formation of the ion-pair [Q+Y] will generally be favoured when the relative hydration of X- is greater than that of Y. However, in many cases, the anion of the ion-pair is hydrated [8-11] (Table 1.1) and this has a significant effect both on equilibrium between the ion-pairs in the aqueous phase and the relative values of the partition coefficients of the two ion-pairs [Q+X ] and [Q+Y ] between the two phases. 

J.C. Kraak, H.H. van Rooij, and J.L.G. Thus, Reversed-phase ion-pair systems for the prediction of n-octanol/water partition coefficients of basic compounds by high-performance liquid chromatography , J. Chromatogr., 1986, 352,455. 

Figure 8. A quadrant diagram to aid derivation of equation 20. V is the volume of octanol or water phase [X] is the concentration of AH in the aqueous phase P is the partition coefficient of AH P is the partition coefficient of the ion pair A"Na+ pKa is the aqueous single-phase pKa pK is the two-phase pKa.

Table IV. Partition Coefficients of Ion Pairs Determined by Two-Phase Titrations ...

These new methods for determining the partition coefficients of ionized species are still e3q>erimenta1, but they are presented in a spirit that they may stimulate thinking and further refinement. Single-phase titrations with HCl in octanol have only recently been run. A possible concentration dependency of pKa in the single-phase titrations has been suggested by a referee and will be looked for. Further refinement of the two-phase titrations may incorporate ion-pair association constants. 

Absorption of some highly ionized compounds (e.g., sulfonic acids and quaternary ammonium compounds) from the gastrointestinal tract cannot be explained in terms of the transport mechanisms discussed earUer. These compounds are known to penetrate the Upid membrane despite their low Upid-water partition coefficients. It is postulated that these highly lipophobic drugs combine reversibly with such endogenous compounds as mucin in the gastrointestinal lumen, forming neutral ion pair complexes it is this neutral complex that penetrates the Upid membrane by passive diffusion. 

Liquid-liquid extraction is used extensively and successfully (6). If the analytes are acidic or basic, as is often the case when HPLC is the analytical method selected, appropriate ionization suppression can be employed to affect the desired extraction. Back extraction of the analytes into an appropriately buffered aqueous volume can then serve to isolate and concentrate. Anionic and cationic surfactants, or so-called ion-pairing reagents, can be added prior to extraction to increase the partition coefficients of the trace organic ionic compounds. 

---