Hammett constants bromide/pyridine

Electronic effects. Nucleophilic attack is favoured by electron-withdrawing groups on the amide and the acyloxyl side chains. Interpolated bimolecular rate constants at 308 K for the series of para-substituted /V-acetoxy-/V-butoxybenzamides 25c, 26b-g and 26i (Table 5) gave a weak but positive Hammett correlation with a constants ip = 0.13, r = 0.86).42,43 These Sn2 reactions are analogous to those of aniline and substituted pyridines with phenacyl bromides, which have similar Arrhenius activation energies and entropies of activation in methanol (EA= 14-16 kcal mol-1, AS = — 27 to —31 calK-1 mol-1) and 4-substituted phenacyl halides afforded a similar Hammett correlation with pyridine in methanol (cr, p — 0.25).175... 

Few other reactions of series of substituted pyridines have been investigated extensively. Dondoni, Modena, and Todesco have measured the rate of N-oxidation of a limited series of pyridines and found a good correlation with normal u-values with a p-value of — 2.23. The A-alkylation of pyridines with alkyl iodides in nitrobenzene has been studied by Brown and Cahn and by Clarke and Rothwell. Unfortunately, the only data available are for the parent compound and for alkyl derivatives, and, since the a-values for the various alkyl groups in a given position are substantially constant, this leaves a correlation of only three independent points. However, the rates of A-alkylation of the j8- and y-alkyl derivatives are so nearly equal that it appears as if no correlation existed. Clarke and Rothwell have also studied the alkylation with allyl bromide in nitromethane at various temperatures, and in this case a more extensive series is available. The authors state that no overall Hammett correlation is obtained however, the j8-substituted derivatives fall on one straight line and the y-derivatives on another one with a different slope. The data are shown in Fig. 2. The line for the j8-compounds, p = — 2.53 0.31, r = 0.95, is seen not to be very good the line for the y-derivatives, p = — 1.42 0.06, r = 0.99, is much more satisfactory. It does not seem likely that the discrepancy is due to the intervention of resonance effects, since in this case one would expect the correlation for the y-derivatives to be poorer than that for the j8-analogs. More extensive studies with a wider variety of substituents would seem very desirable. 

---