Study of the Hammett equation for the ionisation of substituted benzoic acids and phenols in dioxanndash;water mixtures

摘要

1977 1513 Study of the Hammett Equation for the lonisation of Substituted Benzoic Acids and Phenols in Dioxan-Water Mixtures By Dattatraya V. Jahagirdar," Dryanoba N. Shelke, and Ratnakar G. Deshpande, Department of Chemistry, Marathwada University, Aurangabad-431 002, India The thermodynamic pKvalues of 17 substituted benzoic acids and 21 substituted phenols were determined in 20.40, 60, and 80v/v dioxan-water at 30 f 0.1 'C and p 0.1M (NaCIO,) by the Calvin-Bjerrum potentiometric titration technique. The ' practical ' pK values of substituted benzoic acids were also determined in 85 and 90 v/v dioxan-water mixtures. The Hammett substituent constants were calculated and the data for the meta-and para-derivatives were utilized to give Hammetl reaction constants.It was observed thst dp/dD (D = dielectric con- stant) was not regular. In the case of the benzoic acids, the p values decrease above 0.44 mole fraction dioxan. The p values are, however, insensitive to the medium for ortho-substituted compounds. Attempts have been made to explain these observations by considering the structure of the dioxan-water solvent. THEeffects of substitution on the dissociation constants of organic acids have been extensively studied by many workers since the proposal of the Hammett equati0n.l The Hammett p value is known to be dependent on the nature of the ~olvent.~?~ Explanations of the effect of solvent on p values and on rates and equilibria of re-actions have been attempted. The importance of solvent reorientation in the activation process has also received attention in recent year^.^^^ This paper presents a systematic study of ionisation of substituted benzoic acids and phenols in dioxan-water, The choice of dioxan as a cosolvent was made for two reasons: (i) the relatively higher solubility of various compounds in dioxan and (ii) the possibility of obtaining dioxan- water mixtures covering a wide range of dielectric L.P. Hammett, J. Amer. Chem. SOL, 1937, 59, 96. Y. Kindo, T. Matsui, and N. Tokura, Bull. Chem. SOC. Japan, 1969, 42, 1037. 3 J: E. Leffer and E. Grunwald, ' Rate and Equilibria of Organic Reactions,' Wiley, New York, 1963. p. 148. C. D. Ritchie, G. A. Skinner, and V. G. Badding, J. Amer. Chem. SOL, 1967, 89, 2063.constants between 7 and 78. The experimental data can be used in the study of co-ordination complexes of aromatic carboxy- and hydroxy-compounds in solution. EXPERIMENTAL All chemicals were either of analytical reagent grade or properly purified. Substituted benzoic acids and phenols were obtained from B.D.H. and Schuchardt. 1,4-Dioxan was purified by the standard procedure.6 The details of instrumentation and the procedure adopted for potentio- metric titrations were given in an earlier paper.' The B values (pH-meter readings in dioxan-water) were converted into H+ values by applying the corrections proposed by Van Uitert and his co-w~rkers.~~~ 5 G. Kohnstam, ' The Transition State,' Special Publication No. 16, The Chemical Society, London, 1962, p.179. A. I. Vogel, 'A Text Book of Organic Chemistry,' Longman, London, 1956, p. 177. 7 D. V. Jahagirdar and D. D. Khanolkar, J. Inorg. Nuclear Chem., 1973, 35, 921. 8 L. G. Van Uitert and C. G. Haas, J. Amer. Chem. SOC., 1953, 75, 451. L. G. Van Uitert and W. C. Fernalius, J. Amer. Chem. SOC., 1954, 76, 5888. 1514 KESULTS The proton-ligand formation number, +iA,was calculated by the method of Irving and Rossotti.lO The expression log ii~/(l-A) = pK, -pH was solved for A values between 0.8 and 0.2. On an average 5-6 points were taken for each graph. All the points invariably fell on a straight line, whose correlation coefficient was very close to 1. Two sets of potentiometric titration experiments were performed for each system.The precision of the experimental data was checked by comparing the A values at different pH in the two sets. The standard deviation of these for two representative systems, i.e. benzoic acid in 20 v/v dioxan- water and o-nitrophenol in 40 v/v dioxan-water were 0,029 and 0.019 in pH ranges 4-5 and 7-8, respectively. The pK values are presented in Table 1. TABLE1 (a) pK Values (f0.01-0.04) of substituted beiizoic acids. Temp. 30 f0.1 "C; p 0.1~(NaClO,) yoDioxan (v/v) ,Substituen t 20 40 60 80' None 4.83 5.67 7.12 9.28 0-C1 3.72 4.64 6.12 8.22 m-C1 4.60 5.30 6.63 8.74 p-Cl 4.49 5.18 6.66 8.89 o-Br 3.94 4.55 5.93 7.91 m-Br 4.66 5.11 6.78 8.42 p-Br 4.53 5.24 6.83 9.03 0-1 4.21 4.70 6.14 8.14 m-I 4.19 5.11 6.61 8.28 P-= 4.39 5.37 6.95 8.98 0-N02 3.11 3.97 5.28 7.63 -NO2 4.12 4.64 6.02 8.12 P-NO2 4.04 4.66 5.99 8.15 0-NH2 5.16 6.10 7.49 9.63 m-NH2 4.93 5.82 7.15 9.19 P-NH2 5.51 6.44 7.90 10.21 3,5-(N02)2 3.44 3.68 5.33 7.12 (b) pK Values (f0.01-0.06) of substituted phenols.Temp.30 amp; 0.1 "C; ~0.1~(NaC10,) None 10.96 11.50 12.36 14.70 0-c1 8.97 9.86 10.99 12.78 m-C1 9.67 10.43 11.67 13.56 p-Cl 9.75 10.61 11.51 13.64 u-NO~ 7.33 8.12 9.23 11.38 m-NO2 8.82 9.14 10.60 12.35 P-NO2 7.33 8.09 9.03 10.86 o-NH~ 10.45 10.96 12.46 14.67 m-NH, 10.54 11.43 12.64 14.70 P-NH2 11.12 11.85 13.67 15.27 o-OH 10.05 10.63 11.90 13.72 m-OH 10.06 10.86 12.31 14.34 fi-OH 10.81 11.47 12.78 14.86 2,3-C1, 8.34 9.06 10.26 12.41 2,4-c12 8.62 9.35 10.97 12.77 2,5-c12 7.88 8.96 10.59 12.50 2,6-C12 7.51 8.25 9.45 11.52 3,4-C12 9.13 9.95 11.11 13.27 3,5-C12 8.94 9.51 10.25 12.70 2,4-(NO2)2 4.52 4.93 5.95 7.71 4-Cl-2-NOS 6.89 7.48 8.58 10.45 DISCUSSION The accuracy of the experimental data was checked by comparing the pK values obtained by extrapolating the straight line plot of pK against mole fraction dioxan 10 H.Irving and H. Rossotti, J. Chem. Soc., 1954, 2904. 11 L. G. Sillen and A. E. Martell, ' Stability Constants of Metal Ion Complexes,' Special Publication No. 17, The Chemical Society, London, 1964; Supplement No. 1, Special Publication No. 25, 1971. J.C.S. Yerkin I1 to zero mole fraction with the known literature values.ll The agreement between the two is satisfactory.The ApK (pKparetlt -pKsllbs) values for substituted benzoic acids and phenols at different dioxan-water percentages were obtained from the pK values in Table 1 (ApK in water for benzoic acid is equal to 0). For substituted benzoic acids, the ApK values were also determined in 85 and 90 dioxan-water mixtures I. -0.05~(NaClO,). These ApK values represent the differences between the ' practical ' and not the ' thermo-dynamic ' values as correction factors are not available for such high percentages of dioxan. Despite this limitation, these values are expected to give a correct estimate of ApK as the magnitude of the correction factor would probably be the same for the parent and its substituted derivative at high percentages of dioxan.The pK values for phenols were not determined at these percentages because of' the insensitivity of the glass electrode above pH 12.0. Calcdations of Hammett Reaction Constalzts.-The plots of ApK against CT were drawn to give p values at different dioxan-water percentages. Two straight line plots, one for meta- and para-derivatives and the other for ortho-derivatives, were obtained. The Hammett equation being strictly valid for meta- and para-sub- stituted compounds, p values were calculated by the method of least squares from the linear plot for these compounds. The correlation coefficients were found to be between 0.92 and 0.98. The p values are set out in Table 2. The observed value for phenol in aqueous TABLE2 Hammett reaction constants (p) for substituted benzoic acids and phenols in dioxan-water mixtures.Temp. 30 f0.1 "C; 0.1~(NaC10,) P Dioxan Benzoic acids Phenols' 0 1.00 2.23 20 1.07 2.41 40 1.30 2.74 60 1.42 2.83 80 1.68 3.12 85 1.37 90 1.29 a Lit. l2 1.00. Lit. l2 2.26. medium is in good agreement with the literature value.I2 The p values show a rising trend up to 80 dioxan and then register a fall. Wynne-Jones l3 obtained an inverse relationship between p values and the dielectric constant of the solvent. Kilpatrick and his co-worker l4 suggested that this relationship was limited to solvents with D 25. In our studies, a plot of p against 1/D was nonlinear over the entire range dioxan-water compositions (Figure 1).The plot of p against mole fraction (N) dioxan, however, exhibited a different pattern (Figure 2). For benzoic acid, a linear plot was obtained up to 60 l2 H. H. Jaffe, Chew. Rev.,1953, 55, 191. l3 W. F. K. Wynne-Jones, Proc. Roy. Soc., 1933, A140, 440. l4 M. Kilpatrickand J. H. Elliot, J. Phys. Chem., 1941,45,459. dioxan, while points corresponding to 80, 85, and 90 dioxan were falling roughly on another straight line. All the points from 0 to 90 dioxan taken together may, however, be considered to be lying on a parabola. The 4 1 in 1FIGURE Plots of p against 1/D for substituted benzoic acids and phenols (a). Add 1.0 to the y-axis scale for p values(0)of phenols 0.1 0-3 0*5 0*7 Mole fraction 2FIGURE Plots of p against mole fraction dioxan for substituted benzoic acids (0)and phenols (0) relation between p and mole fraction dioxan is given in Table 3.TABLE3 Mathematical equations for Hammett reaction constants. Temp. 30 f0.1 "C; p 0.1~(NaC10,) yoDioxan P-N p-1 ID Benzoic acid (a) 0-80 p = 1.0 + 1.75~p = 0.72 + 23/D (b) 80-90 p = 3.0 + 2.8~ p = 1.30 + 6/D Phenol 0-80 p = 2.26 + 2.8~p = 1.86 + 33/D A decrease in dielectric constant of the solvent should normally lead to an increase in p values. The contrary observation above 80 of dioxan (0.44mole fraction) in the case of benzoic acid is, therefore, interesting. Tokura and his co-workers2 have attributed the variation of p values with the properties of the solvent to the cross interaction of (r with k (where 6 is defined as log KOother so*rent/Kowater).They arrived at this conclusion on the basis of their observation that : l6 M. Pabbo, R. G. Bates, and R. A. Robinson, J. Phys. Chem., 1966, 70, 247. l6 K. Bowden, M. Hardy, and D. C. Parkin, Canad. J. Chem., 1968, 46,2929. (a~/a)~(a/ao),. Our results, however, lead to the = conclusion that the variation of p with the properties of the solvent does not arise by the cross-interaction between p and E. The increase in the p with the decrease in the dielectric constant of the medium may either be due to change in the overall dielectric constant of the medium or due to specific solvation effect. If the ' ring solvation ' of the parent acid is unaltered in the substituted com-pounds, then the plot of 1ogK~~~~f againstsolvent/K~~~.r logKt,h~~nolvent/Kr$zamp; should give a straight line with unit slope.Such linear plots were observed for substituted benzoic acids and phenols with slopes ranging from 0.96 to 1.00 for benzoic acids and around 1.08 for phenols. The effect of solvent on the parent and its substituted derivative is, therefore, almost the same for benzoic acids and phenols. In dioxan-water, there is a greater probability of water molecules occupying the solvation sites in preference to dioxan. However above 80 dioxan, when the molari- ties of the two solvents become almost equal, the dioxan molecules may probably occupy the charged site in the molecule or its anion. In the plots of pK~,ater/PKothersolveiit against 1/D,the points corresponding to D 25 fall outside the straight line.The magnitude of the micro- dielectric constant for the mixture with 80 dioxan as calculated by the procedure of Robinson et aZ.15 was 28. This suggests the orientation of dioxan molecules in the boat form at high percentages of dioxan. ortho-E@ct.-The effect of ortho-substitution on the ionisation of benzoic acids in different mixed solvents has been examined by Bowderi 16917 and Charton.l*~lg The analysis of our results on similar lines showed that the p values (designated as po for ortho-substitution) remain almost constant at 2.3 up to 60 dioxan, fall to 2.1 at SOYo, and then increase to2.4 at higherdioxan percentages. Our results at lower percentages of dioxan are in agree- ment with Bowden's observations.16 The marked sensitivity of p to solvent for the meta- and para-sub- stitution and its near insensitivity to ortho-substitution up to 70 dioxan confirm the findings of Bowden l6that no significant part of transmission of the substituent polar effect occurs through the medium for ortho-substitution.The ortho-effect was not examined for substituted phenols because of limited number of compounds of this type. Principle of Additivity .-The principle of additivity 2o was examined for disubstituted phenols and was found to be invalid. It seems that the principle is not valid for 2,3- and 2,6-disubstituted compounds in dioxan- water mixtures as in water. We record our gratitude to Professor D. D. Khanolkar, Marathwada University, for his keen interest and valuable suggestions. 6/956 Received 20th May, 19761 l7 K. Bowden and G. E. Manser, Canad. J. Chem., 1968, 46, 2941. M. Charton, Progr. Phys. Org. Chem., 1971, 8, 325. l8 M. Charton, J, Amer. Chem. Soc., 1969, 91,6649. 2o J. Shorter and F. J. Stubbs, J. Chem. Soc., 1949, 1181.