J. Chem. SOC., Faraday Trans. I, 1982, 78, 2157-2165 Hydrogen Bonding and Proton Transfer in Hydrido-bis-phenolate Complexes in Acetone BY ZENON PAWLAK AND BOGUSLAW NOWAK Institute of Chemistry, University of Gdansk, 80-952 Gdansk, Poland AND MALCOLM F. Fox* School of Chemistry, Leicester Polytechnic, Leicester LE 1 9BH Received 2nd September, 198 1 The homoconjugation, (ArO),H-, and heteroconjugation, Ar’O- . . * HOAr, (where Ar is aromatic) with proton transfer have been determined in acetone at 298 K. Tetra-alkylammonium phenolates were titrated with a variety of phenols to given homocomplexes and heterocomplexes. Potentiometric data give the overall equilibria constants, KO, proton-transfer constants, KpT and formation constants, K f . Two types of heterocomplexes were studied.When ArO- is a weaker base than Ar’O-, the complexation occurs without proton transfer, as confirmed by the low KO values for the reaction. The overall equilibrium constants, KO, are large when ArO- is a stronger base than Ar’O-, as both the equilibrium proton-transfer constant (KpT) and equilibrium formation constant (Kf) of the hydrogen bond are included in the measurement of KO = KfKPT. It has been shown by many authors studying the proton transfer in molecular B - HA BH+A- complexes that symmetrical hydrogen bonding occurs when the difference of ApK, (H,O) (ApKa being the difference between pK, values for the acceptor and donor) falls within the range - 2 to +7.5. For instance, symmetrical hydrogen bonds in the systems phenol-substituted anilines in cyclohexane, benzoic-acid-substituted pyridines in acetonitrile, and carboxylic acid with base as solvent have been observed at the ApK, (H,O) values of - 1.75,’ 3.752 and 2.33 in water, respectively.The conductiometric, spectrophotometric and potentiometric paH measurements (where paH is the hydrogen-ion activity) in mixtures of phenols with their tetra- alkylammonium salts4> shows stable complex formation of hydrido-bis-phenolate. Formation constants for homo- and hetero-complexes in acetonitrile have been determined. In conductivity studies some phenols in acetone6 exhibited a considerable associative ability to form complex anions, (ArO),H-. In this study the phenols were found to exhibit a stronger interaction in acetone [anion (ArO-)-molecule (ArOH)] than carboxylic acids of approximately the same strength in water.The enthalpy changes for the formation of hydrogen-bonded complexes of the form (RCOO),H- and (ArO),H- have been determined in propylene carbonate as solvent by a calorimetric m e t h ~ d . ~ The hydrogen-bond energy of homocomplexes decreases almost linearly with decreasing acidity of the proton donor. The ratio of the slopes of the curves for hydrido-bis-carboxylate and hydrido-bis-phenolate is ca. 4.5. This result may mean that substituent effects in aromatic acids are attenuated to a large degree through charge delocalization on hydrogen bonding to the negative phenoxide or benzoate ions. 21572158 PROTON TRANSFER I N PHENOL COMPLEXES I N ACETONE Anions with a localized charge can be stabilized in polar aprotic solvents either by homoconjugation : ArO-+ HOAr + (ArO),H- or by heteroconjugation without proton transfer or, alternatively, with proton transfer ArO- + Ar’OH ArO- * * * HOAr or ArOH * - * -0Ar’ (2) where Ar’OH is a weaker or stronger acid than ArOH.8!9 Studies on paH in acetone solution showed that the interaction between phenols (Ar’OH) and phenolates resulted in two distinct types of products.One would be a hydrogen-bonded complex, Ar’OH - - .-OAr, with the proton still attached to the oxygen of the original Ar’O-, whilst for the other, the complex would be formed by proton transfer to the oxygen of ArO-, ArOH. * .-OAr. The hydrogen is transferred to the equilibrium position of proton-transfer anionic bridges, (-OH - . . -0-) + (-0- - - - HO-), and is determined from a study of the hydrogen- ion activity, paH.were confined to the determination of the formation constants, K,, of the homo- and hetero-complexes (hydride-bis-carboxylate). In this paper we progress further to the determination of additional equilibrium constants, overall equilibrium constants, KO, and the proton-transfer constants, KpT, and the relationship between formation constants, Kf. We give special attention to proton transfer between anion-proton donors and the interpretation of these data in terms of the acidity scale of non-aqueous solvents. Previous studiess$ EXPERIMENTAL Acetone was purified and rigorously dried.I0 All phenols used (table 1) were recrystallised 2-3 times from methanol or methanol+water mixtures and dried in vacuum over P,O,.The tetra-n-alkylammonium salts of the substituted phenols were prepared by potentiometric titration of a weighed sample of the phenol with a methanolic solution of the corresponding tetra-alkylammonium hydroxide. After evaporation of the solvent under reduced pressure the salts were recrystallised from ethyl acetate and dried in a vacuum over P,O,. Their purity was checked potentiometric titration with 0.1 mol dm-3 perchloric acid in glacial acetic acid. All results fell in the range 99.5-100.5%. Electromotive-force measurements were made with a PHM-52 digital pH meter (Radiometer, Copenhagen) using an S-60 glass electrode (Gliwice, Poland). The reference half-cell was a saturated calomel electrode with a double junction, and the salt bridge was filled with a 1 .O x mol dmP3 acetone solution of tetra-n-butylammonium perchlorate.The electrode was checked every day in picric-acid-picrate buffers. All measurements were carried out at 298k0.1 K. RESULTS CALIBRATION OF THE GLASS ELECTRODE The reversibility of the glass electrode was checked by e.m.f. measurements in buffer solutions containing CBu,NPi = 4 x 1 OP3 mol dm-3 + picric acid, CHPi = 1 x 1 0-1 mol dmP3. The paH values of these solutions were calculated, assuming complete dis- sociation of Bu,NPi in dilute solution,ll and pKgspne = 6.3,” PaHref = P&piflog& where the subscript f stands for CHPi = Csalt. The activity coefficient was calculated from the expression -lOgf= 3.76 d I .TABLE I .-HETERO- AND HOMO-CONJUGATION OF PHENOLATES (ArO-) WITH SUBSTITUTED PHENOLS (Ar'OH), OVERALL EQUILIBRIUM CONSTANTS (KO), FORMATION CONSTANTS (Kf) AND PROTON-TRANSFER CONSTANTS (KPT) IN ACETONE AT 298 K log KPT log KOb 1% Kf calcd calcd calcd phenol, ArOH (pK,AC)" quaternary salt, R,N+ArO- [eqn (8)l [eqn (711 [eqn (91 1 3,5-dichlorophenol 2 2-nitrophenol 3 2,4,6-tribromophenol 4 3,5-dichlorophenol 5 2-nitrophenol 6 2-nitrophenol 7 3,Sdinitrophenol 8 2,4,6-tribromophenol 9 pentachlorophenol 10 2,4,6-tribromophenol I I 2-nitrophenol 12 3,5-dichlorophenol I3 2,4,6-tribromophenol 14 pentachlorophenol I5 2,4,6-tribromophenol 16 pentachlorophenol 17 pentachlorophenol 18 2,4-dinitrophenol 19 pentachlorophenol 20 2,5-dinitrophenol 2 1 2,4,6-tribromophenol 22 2-nitrophenol 23 2,4,6-trichlorophenol 24 3,5-dichlorophenol 25 2,6-dichlorophenol ( 1 5.7) ( 1 8.3) (19.8) (22.3) (22.5) (22.7) (23.9) (21.1) heterocomplexes, (Ar'OH0Ar)- (C4H9),N 2,4-dinitrophenolate - 7.0 (C,H,),N 2,4-dinitrophenolate - 6.6 (C,H,),N 2,4-dinitrophenolate - 5.4 (C2H5),N pentachlorophenolate - 4.4 (C,H,),N 2,5-dinitrophenolate - 3.5 (C,H5),N pentachlorophenolate - 4.0 (C,H,),N 2,5-dinitrophenolate - 2.9 (C2H5),N pentachlorophenolate - 2.8 (C,H,),N 2,4-dinitrophenolate - 2.6 (C,H,),N 2,5-dinitrophenolate - 1.3 (C,H,),N 2,4,6-trichlorophenolate + 0.2 (C,H,),N 2,6-dichlorophenolate + 1.2 (C,H,),N 2,4,6-trichlorophenolate + 1.4 (C,H,),N 2,5-dinitrophenolate + 1.5 (C,H,),N 2,6-dichlorophenolate + 2.8 (C,H,),N 2,4,6-trichlorophenolate + 4.2 (C,H,),N 2,6-dichlorophenolate + 5.6 homocomplexes, ArOHOAr- (C,H,),N 2,4-dinitrophenolate 0 (C2H5),N pentachlorophenolate 0 (C,H,),N 2,5-dinitrophenolate 0 (C,H,),N 2,4,6-tribromophenolate 0 (C,H,),N 2-nitrophenolate 0 (C,H,),N 2,4,6-trichlorophenolate 0 (C,H,),N 3,5-dichlorophenolate 0 (C,H,),N 2,6-dichlorophenolate 0 3.04 k 0.08 2.95 0.07 2.98 & 0.07 3.27 0.08 3.52 & 0.08 3.23 f 0.06 3.78 f 0.09 3.71 kO.08 3.42 f 0.07 3.79 5 0.09 3.58 f 0.06 4.33 & 0.10 4.70 & 0.12 4.30 k 0.10 4.85 & 0.1 1 5.42 & 0.14 6.25 f 0.12 3.41 f 0.07 4.28 f 0.09 4.59 & 0.10 4.32 f 0 .12 4.12 f 0.09 4.20 f 0.09 4.02 f 0.12 3.50 f 0.10 3.04 2.95 2.98 3.27 3.52 3.23 3.78 3.71 3.42 3.79 3.38 3.13 3.30 2.80 2.05 1.22 0.65 3.41 4.28 4.59 4.32 4.12 4.20 4.02 3.50 a Values from ref. (13); if log KPT < 0, log KO = log Kf for systems 1-10 and 18-25.w 0 X2160 PROTON TRANSFER I N PHENOL COMPLEXES I N ACETONE The glass electrode was calibrated every day in a picrate buffer. For our electrode13 paH = (E,’ - E )/ W = (765 - E ) / 42.5 where Wis the Nernst slope, and E,’ and E are the apparent potential of the reference electrode and the measured potential, respectively. DETERMINATION OF THE PROTON-TRANSFER CONSTANTS, Kpp THE FORMATION CONSTANTS, Kf, A N D THE OVERALL EQUILIBRIUM CONSTANTS, KO The reaction between the proton donor Ar’OH and proton acceptor ArO- in an aprotic solvent may lead to the formation of hydrogen-bonded complexes with proton transfer (PT) or without proton transfer. A general scheme for the formation of the heterocomplexes can be written as follows : KPT ArO- + Ar’OH + ArOH + Ar’O- (3) K € Ar’O- + HOAr S Ar’O- - HOAr (4) KO ArO- + Ar’OH f: Ar’O- - * HOAr where KO, KPT and Kf are the equilibrium constants of the overall reaction, the proton-transfer constant and the formation constant, respectively.The overall equilibrium constant KO is related to KPT and Kf by KO = KPTKf (6) and was calculated from the potentiometric data using eqn (7) adapted by us14 from the study by Kolthoff and Chant~oni:~ Ko = CR4NCArO-r2 - r(cAr’OH + CR4NCArO-) + CAr’OH/r(CR,N+ArO- - CAr’OH) (7) where r = aH f / a ; h and a4 a n d 4 are values at midpoint (CR,N+ArO- = CAro.H). quaternary ammonium salts are presented in fig. 1 (a) and (b). by then Plots of paH against log CAr’oH/CR4N+ArO- of mixtures of phenols with different If we express the ionization constant, K,, of an acid ArOH in acetone medium (S) ArOH + S f SH+ + ArO- Ka = [SH+l EAro-1 fSH +fArO-/[Sl [ArOH1 f S f A r O H - We assumefAr,, andf, to be equal to 1 at low concentrations and By replacing [SH+]f,,+, [ArO-] and [ArOH] by aH+, CR,N+ArO- and CArOH, respec- tively, and taking logarithms we obtain For a medium point (subscript t, at CR4N+ArO- = CArOH) the equation may be written as follows13* l4 This equation is correct for the homosystems R4N+ArO- + ArOH.pK,Ac = paH4 - log &.Z . PAWLAK, B. NOWAK AND M. F. FOX 1 I I I Lr - 0.6 - 0.3 0 0.3 0.6 log (CA~'OH/~K~N+A~O- 1 2161 -0.6 -0.3 0 0.3 0.6 log (CA~'OH /CR~N+A r o - ) FIG. 1 .-Relationship between log CAr,OH/CR,N+ArO- and paH in acetone at 298 K. Numbers identify the systems listed in table 1.2162 PROTON TRANSFER I N PHENOL COMPLEXES I N ACETONE 14 2 2 - I 1 I I I 1 I n - - m 2 20 9 ; t" 2 16 \ 0 9 0 18 -6 - 4 -2 0 2 4 6 AC 'PKtc = PK;&eptor) - pKa (donor) FIG.2.-Relationship between paHt(Cphenol = Cquaternary salt) in homocomplexes and heterocomplexes in acetone at 298 K plotted against ApK,Ac, where ApK,AC = PK%cceptor)- ~Kaq2oonor). Numbers identify the systems listed in table 1 . Hence, we consider the reaction of an acid, Ar'OH, which involves the following ArO- + Ar'OH $ ArOH + Ar'O- reaction for which the equilibrium constant, KpT is KpT = KAc a ( A r ' 0 H) 0 H) * (8) Hence the values of K,, the equilibrium constants for the formation of hydrogen bonding of heterocomplexes with proton transfer, may be found, since log Kf = log KO -log K p T .(9) In the case when KPT is close to, or less than, unity, the value of the overall equilibrium constant, KO, is equal to that of the formation constant, KO = K,. For instance, this case if found in systems where In mixtures of R,N+ArO- with a non-conjugated phenol Ar'OH in acetone, where the proton transfer is not complete, the paH change is relatively small, but the decrease in paH is sharp where the proton transfer is complete, fig. 1 (a). For systems in which the proton is attached to the proton-donor group Ar'OH (ApK,Ac < 0), the plot of the function is linear. function is non-linear, and the curve has a sigmoidal shape. The plots of paH of a mixture of > pKfGrOH), table 1. psH =f(log CAr/OHICArO-) In heterosystems in which proton transfer occurs (ApKkC > 0) the plot of the PaH = Alog CAr/OHICArO-)Z.PAWLAK, B. NOWAK A N D M. F. FOX 2163 4.5 2 00 - 4.0 3.5 I I I I I I I I I * 16 17 18 19 20 21 22 23 AC pKa(pheno1) FIG. 3.-Formation constants, log K,, of homocomplexes, (ArO),H-, in acetone at 298 K plotted against pK$&,,. Numbers identify the systems listed in table 1. an acid and salt without proton transfer, pK&ceptor) < p K ~ ~ o n o r , are linear in fig. l ( a ) and (b), curves 1-10. Some heterosystems in which pK$&.ceptor, > pK$&,nor) [fig. l(a) and (b), curves 13-17] are indicative of the proton-transfer reaction, e.g. curve 17, where ApK,Ac = 5.6: C,Cl,OH + Cl,C,H,O- + C,Cl,O- * * HOC,H,C1,. Consequently, the paH greatly decreases, and the plot assumes a sigmoidal shape.Calculated values of KpT, KO and Kf obtained from the plots of fig. 1 (a) and (b) are given in table 1 for each system. Plots of paH+ at the point (CR,N+ArO- = CArfOH) in the systems studied as a function ApK,Ac, wherg of mixtures of these phenols with different tetra-alkylammonium salts are presented in fig. 2, and exhibit a pronounced maximum around ApK,Ac = 0. Let us consider the two systems: (1) without proton transfer [3,5-C12ArOH + 2,5(N02),Ar0-, ApK,Ac = - 2.901 and (2) with proton transfer [2,4,6-Br3ArOH + 2,6-Cl2Ar0-, ApK,Ac = + 2.801. The observed peH values are comparable, 22 0.2. From the paHi values (at the point CR,N+ArO- = CAr,OH) as a function of ApK,Ac, fig. 2, the proton concentration has a minimum at ApK,Ac = 0, as a result of the formation of homoconjugate (ArO),H- ions.In fig. 3 the stability constants of the homocomplexes, log K(Aro)?F-, are plotted against pK4irOH) on the acetone scale. The largest increase in stability is observed in the region of 20pK,Ac units. In acetone, the Kf values of homocomplexes of substituted phenols are of the order of lo4 or less. In the series considered in fig. 3, the stability of the homocomplexes increases with pK,Ac up to pK, = 20, and then drops with the pK,Ac of phenols. As can be seen in fig. 4, the plot of the overall equilibrium constant, log KO against ApK,Ac, is linear over the ApK,Ac range from -7 to 0. Above the latter value, log KO markedly increases. Linear plots of log KirOHOAr- against ApK,Ac were obtained for heterocomplexes.An increase in the formation constant, K,, in acetone was found with decreasing ApK,Ac, whereas for ApK,Ac > 0 the stability decreased more markedly. Systems characterised by pK$gonor) (Ar'OH) > pK$!&.eptor) (ArOH) are represented2164 PROTON TRANSFER IN PHENOL COMPLEXES IN ACETONE t1 Y L a g Kf -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 APIEBAC = PK:&ceptor) - PK3ionor) FIG. 4.-Plots of overall equilibrium constants, log KO, and formation constants, log K,, for homocomplexes (0) and heterocomplexes (0) in acetone, at 298 K, as a function ApK,Ac on the'acetone scale. Numbers identify the systems listed in table 1 . by entries 1-10 in table 1 and in fig. 4. For these ApK,Ac is negative. As shown in fig. 4 the overall constants, KO, of the reaction are low under these conditions, as proton transfer does not take place.The intermediate region, where ApK,Ac = 0, corresponds to the formation of homocomplexes, (ArO),H-, with log KO ranging from 3.41 to 4.59. The largest change of overall equilibria occurs in the region where ApK,Ac > 0. In this case, the proton is transferred from the less basic donor Ar'OH to the more basic acceptor ArO-. The overall constant KO does not represent equilibria for the formation of the hydrogen bond alone (Kf), but also includes that of the proton transfer (KpT). The overall equilibrium constant, KO, is related to K p T and Kf by log KO = log K ~ T + log Kf. This difference is illustrated by comparison of systems 8 and 15. The values for system 15 are 2,6-C12C,H,O- 4- 2,4,6-Br,C6H20H -+ (2,6-C1,C6H30H.a * OC6H,Br3-2,4)- where log KO = 4.85 and ApK,Ac = 2.80, while for system 8 one has C6C15O- + 2,4,6-Br3C6H,0H -+ (C6C150 * ' . HOC6H2Br,-2,4,6)- where log KO = 3.71 and ApK,Ac = -2.80.Z . PAWLAK, B. NOWAK AND M. F. FOX 2165 Similar results were obtained for other systems with fApK,AC, namely systems 3, 17 and 5, 16: viz. (3) 2,4,6-tribromophenol+ 2,4-dinitrophenolate, log KO = 2.98 (1 7) pentachlorophenol+ 2,6-dinitrophenolate, log KO = 6.25 ( 5 ) 2-nitrophenol+ pentachlorophenolate, log KO = 3.52 (1 6) pentachlorophenol + 2,4,6-trinitrophenolate, log KO = 5.42 CONCLUSIONS The main interaction in our study of phenolate-phenol by hydrogen bonding shows stable homocomplexes with Kf 2 lo4 and heterocomplexes with Kf = 102-103. The most important implication of this work is that contained in fig.4, showing KO, Kf and ApKi0lvent to have a complex relationship. When ApK is negative or zero, then log KO = log Kf. Formation values Kf at whole range ( -ApK,) are not changed as much. However, when ApK is positive for the systems, then log KO diverges rapidly from Kf. Complexes formed after proton transfer at ApK,AN in the positive range undergo a change information constant more markedly. In future, simple statements concerning ApKi0lvent and log KO for proton transfer should either not be made or should carry the qualification that ApK, is negative or zero. When ApK;Olvent is positive, then both KO and Kf must be given. Further, it is clear that the ApK, scale used must be that for the relevant solvent.Whereas comparisons made in the past have used the water scale for ApK,, maxima in the measured quantities, e.g. in the proton chemical shift, have been taken as showing symmetrical hydrogen-bond formation between acid and base at ApK,Hzo values ranging between - 2 and + 7.5. It is self-evident that symmetry ofhydrogen-bond formation will occur for equal basicity/acidity of the two components, at ApK:OIVent = 0. This has been demonstrated in another paper5 for phenolate complexes in acetonitrile when ApK,Ac was used. Therefore, we urge that KO, Kf and ApKPlvent for proton transfer systems should be interpreted in a more meaningful manner and that ApKi0lvent values used should be those for the relevant solvent. It should no longer be acceptable to use ApK,Wzo values when discussing proton-transfer equilibria in non-aqueous systems. l G. Dobecker and P. Huyskens, J. Chim. Phys., 1971, 68, 295. * S. L. Johnson and K. A. Rumon, J. Phys. Chem., 1965, 69, 74. R. Lidemann and G. Zundel, J . Chem. Soc., Faraday Trans. 2, 1978,73, 788. I. M. Kolthoff and M. K. Chantooni Jr, J. Am. Chem. SOC., 1965, 87, 4428; 1966, 88, 8430. J. Magonski and Z. Pawlak, J . Mol. Struct., in press. Z . Pawlak, T. Jasinski and B. Nowak, Zesz. Nauk. Wydz. Mat., Fiz. Chem., Uniw. Gdariski, Chem., 1972, 2, 5 . Z. Pawlak and R. G. Bates, J. Chem. Thermodyn., in press. Z . Pawlak, Roczn. Chem., 1973,47, 641; 1972, 46, 2069. 2. Pawlak and J. Magonski, J. Mol. Struct., 1980, 60, 179; 1978, 47, 329. lo J. F. Coetzee and D. K. McGuire, J. Phys. Chem., 1963, 67, 1810. l1 M. B. Reynolds and C. A. Kraus, J . Am. Chem. SOC., 1948, 70, 1709. l2 C. M. French and I. G. Roe, Trans. Faraday SOC., 1953, 49, 314. l3 B. Nowak and Z. Pawlak, J . Pol. Chem., 1981, 55, in press. l4 Z. Pawlak, Z. Szponar and G. Dobrogowska, Roczn. Chem., 1974, 48, 501. (PAPER 1 / 1378)