The Comparison of KMn with log Poct

Log Poct represents the prime reference parameter as lipophilicity descriptor in QSAR; correspondingly, many investigators compare their TLC data (both RM and RMw) by means of Collander equations with octanol/water partitioning data. Some typical examples for such correlations are summarized in Table 3. We list only correlations applying RMw, since this is the only case, for which a numerical comparability with log Poct can be expected. The quality of correlation increases significantly when chemically

Table 8.2. /?Mw = a + b log . Correlations of RP-TLC data (flMw) with RP-HPLC data (log £w) are listed. Provided the stationary phases used in TLC and HPLC are very similar, the regression data (slope b and intercept a) indicate a rather good 1:1 fit; see lower four rows. In the case of different stationary phases (upper four rows) slopes and intercepts deviate significantly from 1 and 0.

Table 8.2. /?Mw = a + b log . Correlations of RP-TLC data (flMw) with RP-HPLC data (log £w) are listed. Provided the stationary phases used in TLC and HPLC are very similar, the regression data (slope b and intercept a) indicate a rather good 1:1 fit; see lower four rows. In the case of different stationary phases (upper four rows) slopes and intercepts deviate significantly from 1 and 0.

Stationary phase RP-HPLC RP-TLC

Modifier

a

b

r

n

Compounds

Reference

Bondapak

Silicone

Acetone

-1.49

2.11

0.947

44

Cardiac glycosides

[67]

C18

LiChrosorb

Silicone

Acetone

2.11

1.68

0.938

28

Cardiac glycosides

[67]

C18

Corasil C18,

Oleyl alcohol

Methanol

-1.97

0.95

0.938

11

Benzodiazepines

[68]

silylated

Porasil C,

Oleyl alcohol

Methanol

-1.01

1.01

0.966

11

Benzodiazepines

[68]

oleyl alcohol

LiChrosorb

RP 18-HPTLC

Methanol

0.01

1.39

0.978

45

Phenols

[23]

C18

Bondapak

KC 18 F

Acetonitrile

0.18

0.89

0.931

21

Benzo-dia-ze-pines

[60]

C18

Bondapak

KC 18 F

Methanol

0.28

1.09

0.954

17

ß-carbolines

[69]

C18

various C18

RP 18

Methanol

0.11

0.93

0.987

23

Various compounds

[24]

silicagels

Table 8.3. RMv/ = a + b log PM. Correlations of RP-TLC data (RMw) with experimental otanol/water partition coefficients (log Poct) are listed. In general, comparability is better in the case of chemically homogeneous sets of test compounds. Provided ODS material is used, acceptable correlations are even found for chemically diverse sets (see Eqs. (25) and (26)).

Table 8.3. RMv/ = a + b log PM. Correlations of RP-TLC data (RMw) with experimental otanol/water partition coefficients (log Poct) are listed. In general, comparability is better in the case of chemically homogeneous sets of test compounds. Provided ODS material is used, acceptable correlations are even found for chemically diverse sets (see Eqs. (25) and (26)).

Stationary phase

a

b

r

n

Compounds

Reference

Equation

Silanized Kieselguhr

2.19

0.86

0.994

6

Hydrocoumarins

[52]

11

Silica-gel/octanol

0.17

0.51

0.868

31

Cardenolides

[55]

12

Silica-gel/silicone oil

0.97

0.54

0.693

31

Cardenolides

[55]

13

Silica-gel/silicone oil

-1.21

0.93

0.961

28

Phenols

[14]

14

Silica-gel/silicone oil

0.25

0.55

0.975

15

Penicillins

[31]

15

Silica-gel/silicone oil

1.25

0.63

0.914

23

Cardenolides

[67]

16

Silica-gel/silicone oil

0.49

0.37

0.916

40

Amines

[63]

17

Silanized precoated

0.47

0.26

0.861

48

Various drugs

[42]

19

Silanized precoated

0.32

0.80

0.991

11

Arylaliphatic acids

[41]

20

C 8-HPTLC

-1.26

1.21

0.985

17

NSAIDs

[59]

21

C 18-HPTLC

-0.35

1.07

0.983

17

NSAIDs

[59]

22

C 18-HPTLC

-0.24

1.10

0.963

28

Phenols

[23]

23

KC 18F

1.24

0.60

0.929

39

Amines

[63]

24

C18-TLC

-0.06

1.00

0.983

40

Simple compounds

[24]

25

C18-TLC

0.26

0.93

0.961

49

Various drugs

[70]

26

related compounds are used for comparison. The RMvt data used for Eqs. (14-17) in Table 3 have been measured under almost identical experimental conditions; corresponding intercepts (a) and regression coefficients (b) differ profoundly and indicate a putative decrease in significance when merging all these compounds for comparison. Nevertheless, the following equation for 415 chemically diverse heterocycles shows an acceptable correlation between RM and log P, but the slope proves the significant deviation from a 1:1 fit [50],

In contrast to the above conditions intercepts near 0 and regression coefficients approaching 1 are obtained in comparisons of log P«, with /?Mw data estimated on plates with ODS as stationary phase (Eqs. (19-26)). The numerical agreement of log Poa with RMw data is impressively demonstrated for a test set of 121 chemically diverse molecules (identical with the test set in Chapter 23 of this volume) by the following equation:

A linear Collander relation between log Paa and RMw data is shown in Fig. 3. It should however be noted that such a linearity is not generally given. Pliska et al. [20] using cellulose plates and f-butanol as modifier could show nonlinear relations between log Poct and RMw for a set of amino acids.

Figure 3. Plot of RMv/ versus log Poct data for a set of 121 chemically diverse molecules including both simple organic standard compounds and complicated drug molecules. TLC-data have been estimated on ODS material as stationary phase and methanol as modifier. The solid line represents the theoretical 1:1 fit and indicates linearity between RUw and log P„a data; for details, see [70].

log Poet

Figure 3. Plot of RMv/ versus log Poct data for a set of 121 chemically diverse molecules including both simple organic standard compounds and complicated drug molecules. TLC-data have been estimated on ODS material as stationary phase and methanol as modifier. The solid line represents the theoretical 1:1 fit and indicates linearity between RUw and log P„a data; for details, see [70].

0 0

Post a comment