Relative Reactivity and Kinetic Pattern of Aniline and N-Methylaniline as Nucleophiles in Aromatic Substitution (SNAr) Reactions THOMAS A. EMOKPAE, CHUKWUEMEKA ISANBOR Chemistry Department, Faculty of Science, University of Lagos, Akoka Lagos, Nigeria Received 29 January 2003; accepted 22 October 2003 DOI 10.1002/kin.10188 ABSTRACT: Kinetic results are reported for the reactions of 4-nitrophenyl-2,4,6-trinitrophenyl ether 3 with aniline and N-methylaniline in dimethyl sulphoxide, acetonitrile, methanol, and benzene. The reactions gave the expected 2,4,6-trinitrodiphenylamine and were base catalyzed in all the solvents. Both nucleophiles showed the same kinetic pattern under the same reaction conditions but aniline was found to be considerably more reactive than N-methylaniline. The greater catalytic efficiency of aniline over N-methylaniline is consistent with the proton transfer mechanism of the base-catalyzed step. Dichotomy of amine effects in aromatic substitution (SNAr) reactions is discussed. C© 2004 Wiley Periodicals, Inc. Int J Chem Kinet 36: 188–196, 2004 INTRODUCTION The general mechanism of aromatic nucleophilic sub- stitution reactions when either primary or secondary amines are the nucleophiles is given in Scheme 1. Equa- tion (1) is the steady-state expression for the observed second-order rate constant kA expressed in terms of the component steps in Scheme 1. kA = k1(k2 + kB[B]) k−1 + k2 + kB[B] (1) Correspondence to: Thomas A. Emokpae; e-mail: temokpae@ yahoo.com. Contract grant sponsor: ICSC World Laboratory, Lausanne. c© 2004 Wiley Periodicals, Inc. A salient feature of this mechanism is that the interme- diate 1 can proceed to the product spontaneously (k2) or through general base catalysis (kBB). If no catalysis is observed, the inference can be made that the formation of the intermediate 1 is rate-limiting and the condition k2 + kB [B] � k−1 prevails. In this case, the measured overall constant kA is equal to k1. When this condition does not hold, decomposition of 1 into product is rate limiting and the reaction is base-catalyzed; the kinetic form then depends on the relative magnitude of k−1 and k2 + kB[B]. Provided that k−1 � k2, Eq. (1) de- scribes generally a dependence of kA on [B], which is linear in low nucleophile concentrations but changes to a plateau as the base concentration is increased. At low [B] values, k−1 � k2 + kB [B] and kA responds linearly to base concentration. The second-order rate ANILINE AND N -METHYLANILINE AS NUCLEOPHILES IN AROMATIC SUBSTITUTION REACTIONS 189 Scheme 1 constant then obeys an equation such as (2) kA = k ′ + k ′′ [B] (2) If k−1 ≈ kB [B] within the experimental range of base concentrations, kA depends hyperbolically on base concentration. In aromatic substitution (SNAr) reactions, when a substrate containing an ortho-nitro group reacts with primary and secondary aliphatic amines of the same basicity, quite often, the reactions with secondary amines are base-catalyzed whereas the correspond- ing reactions with primary amines are not. The sit- uation with primary and secondary aromatic amines is not so clear cut since it has been found that some aromatic nucleophilic substitution reactions involving aniline as nucleophile are base-catalyzed while the corresponding reactions of N -methylaniline are not. Kavalek et al. [1] have reported that the reaction of N -methylaniline with 1-fluoro-2,4-dinitrobenzene in acetonitrile is not based-catalyzed by N -methylaniline whereas the same reaction with aniline exhibits base catalyses by the amine. On the basis of the observed order of halogen mobility (Cl > F) for the reactions with N -methylaniline, these workers concluded that the decomposition of zwitterionic intermediate to prod- uct constitutes the rate-limiting step of the reaction. Hirst et al. [2] reexamined the same reaction and came to the conclusion that there was mild catalysis by N -methylaniline. The reaction was also strongly catalyzed by 1,4-diazabicyclo-[2.2.2]-octane (Dabco). When the nucleophile was changed to aniline, the plot of kA against aniline concentration was curvilinear and passed through the origin; hence the uncatalyzed path- way was negligible. On the contrary, the reaction of 2,4-dinifluorobenzene with N -methylaniline in ethanol was catalyzed by acetate ion, whereas no catalysis was observed in the reaction of the same substrate with ani- line [3,4]. Hirst [5] has indicated that catalysis by ani- lines as nucleophile is difficult to interpret. Akinyele et al. [6] have shown that because of the greater acid- ity of the amino hydrogen atoms of aniline, compared with that of n-butylamine or piperidine, catalysis of the first step of the reaction can take place as in structure 2. Here Y is a base which may be the nucleophile or even chloride ion To circumvent this problem, we decided to use a sys- tem devoid of such complication. In an earlier investi- gation of steric and electronic effect on the mechanism of nucleophilic reactions of some phenyl-2,4,6- trinitrophenyl ethers [7] we made a preliminary com- parison of the reactivity of aniline and N -methylamine in dimethyl sulfoxide (DMSO) and acetonitrile. Here in we report detailed kinetic studies of the reactions of these two nucleophiles with 4-nitrophenyl-2,4,6- trinitrophenyl in various solvents. Our aim was to deter- mine whether the dichotomy of amine effects prevalent in the reactions of aliphatic and alicylic amines exist in SNAr reactions involving aromatic amines. EXPERIMENTAL The substrate 4-nitrophenyl-2,4,6-trinitrophenyl ether was prepared by the reaction of picryl chloride with 1 equiv of base in the presence of an excess of 4- nitrophenol in aqueous ethanol. The reaction product 2,4,6-trinitrodiphenylamine and it N -methyl deriva- tive were prepared by reaction of picryl chloride with fourfold excess of the appropriate amine in ethanol. Recrystallization was from ethanol. 1NMR data, melt- ing points, and C,H,N analysis are given in Tables I and II. The purification of solvents, aniline, and 190 EMOKPAE AND ISANBOR Table I 1H NMR Shifts in CD3CN and Melting Points for Reactant and Products 1H NMR Shiftsb mp (◦C) Compounda H3,5 H2′ H3′ H4′ Other Found Lit7 3 9.11 7.15 8.26 – – 158 157 7a 8.96 7.15 7.33 7.25 9.96NH 179 178 7b 8.85 6.86 7.26 7.00 3.27(Me) 127 108 a Compound 3 is 4′-nitrophenyl-2,4,6-trinitrophenyl ethers, 7a is 2,4,6-trinitrodiphenylamine, and 7b its N-methyl derivative. b Ortho coupling, J 7–8 Hz is observed. N -methylaniline has been described previously [7]. 1H NMR spectra were measured with Varian Mercury 200 MHz or Varian Unity 300 MHz. The details of spectrophotometric determination of the rate constants have already been given [7]. RESULTS AND DISCUSSION Reactions of the substrate with aniline in all the solvents proceeded without the observation of intermediates to give the expected 2,4,6-trinitrodiphenylamine in quan- titative yield. UV and NMR spectra at the completion of the reaction were identical with that of the expected substitution product in the reaction medium. Kinetic measurements in acetonitrile and DMSO were made with aniline and with solutions containing aniline and Dabco. Reactions in methanol were carried out in the presence of the amine and with amine containing amine hydrochloride. With these concentrations in large ex- cess of the substrate concentration, first-order kinetics was observed. Reactions in Acetonitrile Plots of second-order rate constants vs aniline con- centration pass through the origin and curve with de- creasing slope as aniline concentration is increased Fig. 1. This implies that the uncatalyzed pathway k2 in Scheme 2 is relatively unimportant. Hence, Eq. (1) reduces to Eq. (3), where kAN represents the pathway Table II CHN Analysis of the Reactant and Products Calculated (%) Found (%) Compound Mw C H N C H N 3 350.20 41.12 1.73 15.99 40.97 1.67 15.88 7a 304.22 47.33 2.65 14.41 47.23 2.65 18.37 7b 318.25 49.02 3.17 17.60 48.95 3.14 17.57 catalyzed by aniline. kA = kobs [Aniline] = k1kAN[Aniline] k−1 + kAN[Aniline] (3) An equivalent form is Eq. (4) kA = K1kAN[Aniline] 1 + kAN k−1 [Aniline] (4) Provided k−1 � kAN [Aniline], values of kA data in Table III allow the calculation of k1 0.26 ± 0.08 dm3 mol−1 s−1, kAN k−1 4 ± 1 dm3 mol−1, and K1kAN 1.03 ± 0.03 dm6 mol−2 s−1. At a constant aniline concentra- tion, values of the second-order rate constant kA in- creased linearly with Dabco concentration as shown in Fig. 2. The slope of this plot allows the calculation of a value for K1kDabco of 3.4 dm6 mol−2 s−1. Values are summarized in Table IV. With N -methylaniline, we found that studies by UV–vis spectroscopy of the reaction of 3 (5 × 10−5 mol dm−3) with excess nucleophile in acetonitrile was miserably slow and did not yield the expected sub- stitution product. This may be due to trace quantities of impurities in the solvent or the amine which was redistilled. However, a 1HNMR study in CD3CN us- ing substrate concentration (0.04 mol dm−3) with N - methylaniline in large excess showed the development of bands over several days attributable to the expected reaction product. Integration of the bands due to the reactant and the product allowed the progress of the reaction to be monitored. Values of the first-order rate coefficient kobs were calculated to be 1.1 × 10−6 s−1 and 1.0 × 10−5 s−1 when the concentrations of the nu- cleophile were 0.25 and 0.85 mol dm−3, respectively. These results indicate that the value of the second–order rate constant kA is linearly dependent on the concen- tration of the amine. The value k ′′ (K1kN-MeAN) calcu- lated from the slope of such a plot is 1.6 ± 0.2 10−5 dm6 mol−2 s−1. The corresponding value for aniline is ca 105 higher than that of N -methylaniline. Reactions in Dimethyl Sulphoxide A plot (not shown) of the second-order rate constant kA vs aniline concentration was linear with positive inter- cept. The intercept of such a plot represents the product of the equilibrium constant K1 for the formation of 4 and the rate constant for its uncatalyzed decomposition to product k2. The positive slope indicates the presence of a base-catalyzed route. At a constant aniline con- centration however value of kA increased linearly with Dabco concentrations. The results are best analyzed in ANILINE AND N -METHYLANILINE AS NUCLEOPHILES IN AROMATIC SUBSTITUTION REACTIONS 191 Figure 1 Plot of kA vs [Aniline] in acetonitrile at 25◦C for the reaction of 4-nitrophenyl-2,4,6-trinitrophenyl ether 3 with aniline. Experimental values are denoted by �, and the curve is generated by kA = 1.03[aniline]/(1 + 4[aniline]). Scheme 2 192 EMOKPAE AND ISANBOR Table III Kinetic Results for the Reactions of 3 with Amines in Various Solvents at 25◦C Acetonitrile [Aniline] (mol dm−3) 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 kA (10−2 dm3 mol−1 s−1) 0.93 1.98 2.78 3.62 4.26 5.02 5.68 6.24 [Dabco] (mol dm−3) 0.01 0.02 0.03 0.04 0.05 0.06 ka A (10−2 dm3 mol−1 s−1) 3.81 6.84 10.00 13.50 17.80 20.10 Methanol [Aniline] (mol dm−3) 0.005 0.01 0.02 0.03 0.04 0.05 0.06 kA (10−2 dm3 mol−1 s−1) 2.60 3.00 3.81 4.55 5.30 – – kb A (10−2 dm3 mol−1 s−1) – 0.75 1.30 1.85 2.40 2.95 3.60 [N-Methylaniline] (mol dm−3) 0.04 0.06 0.08 0.10 kA (10−4 dm3 mol−1 s−1) 2.15 2.74 3.35 3.93 kc A (10−4 dm3 mol−1 s−1) 1.15 1.68 2.19 2.70 Benzene [Aniline] (10−3 mol dm−3) 0.5 1.0 1.2 1.5 1.8 2.0 kA (10−3 dm3 mol−1 s−1) 0.5 1.8 2.5 3.9 5.5 6.8 [N -Methylaniline] (10−2 mol dm−3) 2.0 3.0 4.0 5.0 kA (10−4 dm3 mol−1 s−1) 0.53 0.90 1.36 2.08 Dimethyl sulfoxide [Aniline] (mol dm−3) 0.06 0.08 0.1 0.15 0.2 kA (10−1 dm3 mol−1 s−1) 3.6 3.6 3.7 4.1 4.5 a Kinetic results for the reactions with aniline (0.01 mol dm−3) and various concentrations of Dabco. b Contains 0.1 mol dm−3 aniline. c Contains 0.1 mol dm−3N-methylaniline hydrochloride. Figure 2 Plot of kA vs [Dabco] in acetonitrile at 25◦C for the reaction of 4-nitrophenyl-2,4,6-trinitrophenyl ether 3 with increasing concentration of Dabco at constant 0.01 mol dm−3 [Aniline] (r = 0.996). ANILINE AND N -METHYLANILINE AS NUCLEOPHILES IN AROMATIC SUBSTITUTION REACTIONS 193 Table IV Summary of Rate Dataa for the Reaction of 3 with Aniline and N-Methylaniline K1k2 K1kAn K1kDabco kAn/k2 Amine Solvent (dm3 mol−1 s−1) (dm6 mol−2 s−1) (dm6 mol−2 s−1) kAn/kDabco (dm3 mol−1) Aniline DMSO 3.2 × 10 0.65 0.9 0.72 2.0 Acetonitrile – 1.03 3.4 0.30 – Methanol 2.23 × 1 0.77 – – 34.6 Methanolb 1.67 × 1 0.564 338 N -Methylaniline DMSO 3.0 × 10 9.0 × 10−6 – – 3.0 Acetonitrile – 1.6 × 10−5 – – – Methanol 9.60 × 1 2.98 × 10−3 – – 31 Methanolc 1.24 × 1 2.58 × 10−3 – – 208 a Values quoted are ±10%. b Obtained with addition of 0.1 mol dm−3 aniline. c Obtained with addition of 0.1 mol dm−3 N -methylaniline hydrochloride. terms of the processes shown in Scheme 2. Making the assumption that the zwitterion can be treated as a steady state intermediate leads to the rate expression of Eq. (5), where kAN and kDabco represent k3[B] for the respective bases. kobs = k1[AN][k2 + kAN[AN] + kDabco[Dabco]] k−1 + k2 + kAN[AN] + kDabco[Dabco] (5) If k−1 � k2 + kAN[AN] + kDabco[Dabco], then Eq. (6) applies kA = kobs [Aniline] = K1(k2 + kAN[AN] + kDabco[Dabco]) (6) The value obtained for K1k2 is 0.32 dm3 mol−1 s−1 while the values for K1kAn and K1kDabco are 0.65 dm6 mol−2 s−1 and 0.9 dm6 mol−2 s−1, respectively. Based on the acceleration produced by added Dabco, we con- clude that at high aniline concentrations, there was some evidence of weak base catalysis. Similar obser- vation has been made by Crampton and Robotham [8]. The greater susceptibility of an amine nucleophile to base catalysis in acetonitrile than in DMSO may be traceable to the nature of the solvent. The first step in adduct formation (Scheme 1) is the formation of the zwitterion and this involves the production of charges. DMSO is much better than acetonitrile at solvating charged polarizable species, such as the zwitterion. The values of overall equilibrium constants for adduct for- mation are 104 higher in DMSO than acetonitrile [9]. However, DMSO is also a good hydrogen bond accep- tor so that NH2 + protons will be strongly hydrogen- bonded to the solvent. This will reduce values of rate constants for proton transfer from zwitterions to base. There is evidence that values of the rate constants for such proton transfer are ca 104 lower in DMSO than in acetonitrile [9]. Hence the increases observed in k ′ and k ′′ on going from acetonitrile to DMSO are a combina- tion of increases in K1 values and the reduction in k2 and kAN values. The low ratio of k ′′/k ′ in DMSO may reflect solvent-assisted intramolecular proton transfer as depicted in 8. Interestingly the numerical values of K1kAN are close in the two solvents. However, this similarity is likely to be due to the compensation of large increase in the value of K1 and correspondingly large decrease in the value of kAN as the solvent is changed from acetonitrile to DMSO. The reaction of the substrate with N -methylaniline in [2H6] DMSO (Table V) shows a similar pattern to that obtained with aniline with both the uncatalyzed and the base-catalyzed pathways contributing to the reac- tion flux. The result conforms to Eq. (2) with values of K1k2 of 3 × 10−6 dm6 mol−1 s−2 and K1kN-MeAN of 9 × 10−6 dm3 mol−1 s−1. Comparisons with the re- sults for reaction with aniline indicate a factor of ca 105. This is exactly the same aniline/N -methylaniline reactivity ratio found in acetonitrile. The lower value of K1kAN for N -methylaniline is likely to be the result of decrease in both K1 and in kAN. In the zwitterion, 194 EMOKPAE AND ISANBOR Table V Rate Dataa for the Reaction of 3 with N-Methylaniline in [2H6] DMSO at 25◦C [N -Methylaniline] kobs b kA (mol dm−3) (10−6 s−1) (10−6 dm3 ml−1 s−1) 0.4 2.7 6.7 0.6 4.7 7.8 0.8 7.8 9.8 a Measured by integration of 1H NMR bands, with substrate 0.04 mol dm−3. b Values ±10%. there will be considerable steric crowding at the re- action center resulting in a reduction of the value of K1. Release of the steric strain would enhance k−1 for N -methylaniline. Buncel [10] has estimated in the Dabco-induced Meisenheimer complex formation between 1,3,5- trinitrobenzene [TNB] and N -methylaniline or aniline that with N -methylaniline as the nucleophile, k−1 is an order of magnitude greater than in the case of aniline, reflecting the effect of release of steric compression in the zwitteronic intermediate on reversion to reac- tants in the former case. An additional factor may be the role of hydrogen bonding known to occur between the ammonio hydrogen atoms of the intermediate com- plex and the oxygen atoms of the ortho-nitro group. This hydrogen bonding stabilizes the intermediate so that k−1 is reduced, because reversion to reactants in- volves the breaking of the hydrogen bond in addition to the C N bond. This effect will be about the same for aniline and N -methylaniline, but the effect on the expulsion of nitrophenoxide ion will be different, as the hydrogen bond will have to be broken when the nucleophile is N -methylaniline but not when it is ani- line because of the availability of a free transferable proton in aniline. The ratio k2 + k3 [B]/k−1 will def- initely be smaller for N -methylaniline than aniline, and as such aniline may be less prone to base catal- ysis in DMSO. Recently, such intramolecular inter- actions have been shown not to contribute much to the lowering of the K1kAn value for N -methylaniline [7]. Reactions in Methanol The second-order rate constants kA for the reactions of the nucleophiles with 3 increases linearly with increas- ing concentration of the nucleophile, i.e. kA = k ′ + k ′′ [Nucleophile]. Thus, for the reactions of both nucle- ophiles in methanol the decomposition of the interme- diate to product is rate limiting. The reactions with aniline and N -methylaniline have low values of k ′′/k ′ , 35 ± 0.5 and 31 ± 0.5, respectively. The addition of 0.1 mol dm−3 of the hydrochloride of the nucleophile, while decreasing substantially the rate constants of the reactions of the substrate with aniline and N - methylaniline, results in a large increase in the k ′′/k ′ to 338 ± 60 and 208 ± 22, respectively. Inspection of the individual values of k ′ and k ′′ in Table IV shows that the addition of amine hydrochloride has little effects on k ′′; the variation is entirely on k ′. In terms of Scheme 1, k ′ (=K1k2) and k ′′ (=K1kB); hence, the addition amine hydrochloride has little effects on the base-catalyzed step kB, consistent with the operation of proton trans- fer mechanism, but gives a reduction in the values of k ′ of approximately 13-fold and eight fold for the re- actions with aniline and N -methylaniline, respectively. This was attributed to increased steric/stereoelectronic effects as a result of added anilinium ion [11]. The rel- atively low values of k ′′/k ′ ratios might reflect some solvent assistance by methanol in the intramolecular proton transfer involved in the k2 step.∗ Reaction in Benzene Benzene is an aprotic, apolar, and scarcely polarizable solvent, which represents an ideal medium to promote the need for base catalysis for the decomposition of the intermediate. Aromatic nucleophilic substitution reactions are therefore more prone to base catalysis in aprotic solvents of low relative permittivity than in dipolar solvents [12]. In benzene, the values of the second-order rate constants kA for the two nucleophile increased rapidly with amine concentration; the plots (not shown) exhibit a curvilinear response, which are concave toward the rate constant axis. The curved re- sponse shows that the order with respect to [Amine] is >2. Further the plots of the quotient kA/[Amine] against [Amine] gave straight lines. In these systems, aniline is more efficient than N -methyaniline in cat- alyzing the reaction. In DMSO, we have shown that the reaction is mildly catalyzed by aniline, while the plot of kA against aniline concentration is curvilinear downward with negligible intercept in acetonitrile and curvilinear upward in benzene. With N -methylaniline as nucleophile, the change in the kinetic form is from one in which the plot is linear with definite intercept in acetonitrile and DMSO to one which is again curvi- linear upward in benzene, a kinetic form that is ob- served quite frequently in SNAr reactions in solvents ∗A referee has suggested that another possibility for the observed decrease in the values of k ′ in the presence of amine hydrochloride may results from a reduction in methoxide concentration (present in equilibrium with the amine) acting as a general base in the deproto- nation of the zwitterions. ANILINE AND N -METHYLANILINE AS NUCLEOPHILES IN AROMATIC SUBSTITUTION REACTIONS 195 of low permittivity. At present, there is controversy as to the origin of the curvature. It is however usu- ally attributed to a term third order in the nucleophile concentration. The mechanistic interpretation of this term is still a subject of active discussion. Banjoko and coworkers [13] have explained the third-order term as being due to reaction occurring through a cyclic tran- sition state containing an eight-member ring formed through a network of inter-hydrogen bonding between two aniline molecules and zwitterionic intermediate as shown in 9. Akinyele et al. [14] gave a plausible mech- anism for the formation of the cyclic transition state originally proposed by Capon and Rees [15]. The con- cept has been developed by Emokpae et al. [16] to rationalize reactions proceeding through cyclic tran- sition states containing either two or three molecules of amine and to distinguish these reactions from those taking place by the specific base-general acid (SB-GA) mechanism. Hirst et al. [6] has, however, explained the upward curvature obtained by Bernasconi and Zollinger [17] in the reaction of p-anisidine with 1- fluoro-2,4-dinitrobenzene in benzene as due to elec- trophilic catalysis of the departure of the leaving group by the homo-conjugate of the conjugate acid of the nu- cleophile. Recently, Hirst [18] however advanced con- vincing reasons to show that there is only a thin dividing line between the cyclic transition state and the homo/ hetero-conjugate mechanism. For the present reaction, we prefer to interpret the results along the lines sug- gested by Banjoko. Mechanism of Substitution The greater catalytic efficiency of aniline over N - methylaniline may not be unconnected with the greater bulk of N -methylaniline as it affects the mechanism of the base-catalyzed step. Catalysis by Dabco in the sys- tem under investigation is an indication of general base catalysis so that the removal of the ammonium proton from the zwitterionic intermediate is rate-limiting. In dipolar aprotic solvents this can occur substantially ei- ther by a slow, rate-limiting proton abstraction from the zwitterionic intermediate 4 by the base to form the deprotonated intermediate 5, from which the leav- ing group breaks off rapidly or by rapid deprotonation equilibrium followed by a slow detachment of the leav- ing group from 5, which is general acid-catalyzed by the conjugate acid of the amine. The latter pathway, the SB-GA mechanism, has been widely accepted for reactions occurring in DMSO and has been shown to apply in substitutions of several other ring activated alkyl aryl ethers [19]. There is now strong evidence that with phenyl ether and phenyl sulfides rate-limiting proton transfer is from the zwitterion. One argument against the SB-GA mechanism is the failure to ob- serve anionic intermediates such as 5 on the reac- tion pathway. For the reaction of 1-ethoxy-2,4-dinitro naphthalene with aliphatic amines in DMSO, which is widely recognized as a model for the SB-GA mecha- nism, Bunnett and Orvik [20] were able to observe in separate steps the formation of intermediate of struc- ture of type 5 and their acid-catalyzed conversion into substitution products. Related intermediates have been observed during the reactions of several other ring- activated alkyl aryl ethers with amines [21] and there is no doubt that the SB-GA mechanism applies in this system. As Bernasconi et al. [22] have noted that in pro- tic solvents there is evidence that catalysis of alkoxide ion expulsion from Meisenheimer complex is weak or occurs only with acids considerably stronger than R2 + N H2. In water, the pKa of phenol, anilinium, and N -methylanilinium ions are 9.95, 4.62, and 4.84, re- spectively. These values are unlikely to be reduced on transfer to DMSO or acetonitrile. Since the pKa of 4-nitrophenol in water is 7.14, the equilibrium NO2PhO− + R2 + N H2 ↼⇁ NO2PhOH + R2NH is only favored in the thermodynamic sense by 2.3 pK units for the leaving group to be lost in a slow general acid- catalyzed step. This may not constitute enough driv- ing force to compensate for the expense in entropy in incorporating an addition molecule into the transition state [22]. There is already severe steric congestion around the zwitterionic intermediate 5 which will make it harder for another molecule to approach the reaction center. The case against the SB-GA mechanism in our sys- tem is further supported [7] by the effects of sub- stituents on the base-catalyzed pathway in the reac- tions of aniline with X-phenyl, 2,4,6-trinitrophenyl ethers [X = 4-(H,CH3,NO2,Br,Cl),3-NO2)]. If the base-catalyzed pathway involves rate-limiting proton transfer from the zwitterion to base (the kB step), then 196 EMOKPAE AND ISANBOR there should be little dependence on the nature of X. Values of K1kAn obtained in the series vary only by a factor of <3 between the more activating NO2 and the least activating CH3 group consistent with a rate limiting proton transfer mechanism. The results from previous investigation suggest that steric rather than electronic factors determine the rate constant for such proton transfer, which may be slower than the diffu- sion limit [19,20]. The kAn/kDabco ratio of 0.3 shows that despite the large difference by 7 pK units in the basicities of aniline and Dabco their ability to effect the proton transfer from 4 is similar. That the ratio is <1 indicates that Dabco is less sterically demanding than aniline so that it is easier for it to approach the reaction center in the zwitterion. The alternative SB- GA mechanism will require that proton transfer from an ammonium ion to the anionic adduct 5 to be rate determining. The kB step is therefore a product of the equilibrium constant for the conversion of the zwitte- rionic intermediate to the anionic adduct and the rate constant for the general acid-catalyzed expulsion of the nucleofuge (K1kfast). Since the latter term involves loss of the nucleofuge, then a strong dependence on the na- ture of X would be expected. The failure to observe such dependence argues against the SB-GA mechanism in our system. Our conclusion is that in the present system as in re- lated phenyl ethers base catalysis reflects rate-limiting proton transfer from zwitterionic intermediates. The weaker ability of N -methylaniline than aniline to cat- alyze the reactions of 3 in acetonitrile may there- fore be reconciled on the basis that N -methylaniline is less effective in abstracting the proton from 4 be- cause of its greater steric requirement relative to ani- line. In all the solvents, both nucleophiles show the same kinetic pattern under the same experimental con- dition. The dichotomy of amine effects often found in the reactions of aliphatic amine does not exist in our system. Our gratitude goes to Dr. M.R. Crampton for the use of his laboratory and for helpful discussions. BIBLIOGRAPHY 1. Kavelek, J.; Kubias, J.; Sterba, V. Collect Czech Chem Commun 1972, 37, 4041. 2. Bankole, T. O.; Hirst, J.; Hussain, G. J Chem Soc, Perkin Trans 2 1984, 681. 3. Bunnett, J. F.; Randall, J. J. J Am Chem Soc 1958, 80, 6020. 4. Bunnett, J. F. Q Rev 1958, 12. 5. Bamkole, T. O.; Hirst, J.; Onyido, I. J Chem Soc, Perkin Trans 2 1981, 1201. 6. Akinyele, E. T.; Hirst, J.; Onyido, I. J Chem Soc, Perkin Trans 2 1988, 1859. 7. Isanbor, C.; Emokpae, T. A.; Crampton, M. R. J Chem Soc, Perkin Trans 2 2002, 2019. 8. Crampton, M. R.; Robotham, I. A. Can J Chem 1998, 76, 627. 9. Crampton, M. R.; Lord, S. D. J Chem Soc, Perkin Trans 2 1997, 360. 10. Buncel, E.; Eggimann, E. J Am Chem Soc 1977, 99, 5958. 11. Emokpae, T. A.; Uwakwe, P. U.; Hirst, J. J Chem Soc, Perkin Trans 2 1990, 2191. 12. Bernasconi, C. F. In Organic Chemistry Series 1; Zollinger, H. (Ed.); Butterworths: London, 1973; Vol. 3, p. 33. 13. Banjoko, O.; Ezeani, C. J Chem Soc, Perkin Trans 2 1982, 1357. 14. Akinyele, E. T.; Hirst, J.; Onyido, I. J Phys Org Chem 1990, 3, 41. 15. Capon, B.; Rees, C. W. Annu Rep Prog Chem 1963, 60, 279. 16. Emokpae, T. A.; Uwakwe, P. U.; Hirst, J. J Chem Soc, Perkin Trans 2 1990, 2191. 17. Bernasconi, C. F.; Zollinger, H. Hev Chim Acta 1966, 49, 4570. 18. Hirst, J. J Phys Org Chem 1997, 7, 68. 19. (a) Chamberlin, R. A.; Crampton, M. R. J Chem Soc, Perkin Trans 2 1995, 1831; (b) Crampton, M. R.; Chamberlin, R. A. J Chem Soc, Perkin Trans 2 1994, 425. 20. Orvik, J. A.; Bunnett, J. F. J Am Chem Soc 1978, 100, 5530. 21. Crampton, M. R.; Chamberlin, R. A.; Robotham, I. A. J Chem Res 1994, 408. 22. Bernasconi, C. F.; de Rossi, R. H.; Schmid, P. J Am Chem Soc 1977, 99, 4090.