129
Intermediates in Nucleophilic Aromatic Substitution. 11.' Temperature-Jump Study of the Interaction of 1,3,5Trinitrobenzene with Aliphatic Amines in 10% Dioxane-90 % Water. Concurrent Nucleophilic Attack on the Aromatic Carbon and on the Nitro Group?z Claude F. Bernasconi Contribution from the University of California, Santa Cruz, California 95060. Received May I , 1969 Abstract: Solutions of 1,3,5-trinitrobenzene and n-butylamine, piperidine, or pyrrolidine in 10% dioxane9Ox water are characterized by three relaxation times. It is shown that 71 arises from Meisenheimer complex formation (XH and X-) between 1,3,5-trinitrobenzeneand the amine, r2most probably from an oxyhydroxylamine (YH and Y -), and 73 from a Meisenheimer complex formation (Z-) between 1,3,5-trinitrobenzene and the hydroxide ion. Various rate coefficients and equilibrium constants involving the different complexes are evaluated. The relative rates of formation of a Meisenheimer complex by the three amines conform to the familiar reactivity pattern in nucleophilic aromatic substitution reactions by these amines. The rates of Meisenheimer complex decomposition are practically the same for the three amines; intramolecular hydrogen bonding to one or both o-nitro groups is believed to be mainly responsible for this result. The pK values of the Meisenheimer complexes formed by piperidine and pyrrolidine are very close to the pK of the respective amine; in the case of n-butylamine the pK of the complex is significantly lower. The general reactivity pattern in the oxyhydroxylamine series is similar to the one in the Meisenheimer complex series and is interpreted along similar lines. The various rate and equilibrium constants vary with amine and amine hydrochloride concentration and depend also on whether the compensating electrolyte is NaCl or (CH&NCl.
C
onventional kinetic studies have provided us with considerable insight into the general mechanism of activated nucleophilic aromatic substitution reactions. The main feature is that the nucleophile attacks the aromatic substrate-generally activated by one or several nitro or other electron-withdrawing groups-to form a high-energy intermediate, often referred to as a Meisenheimer complex, which can either proceed t o products in a second step or revert t o reactants, as illustrated in eq 1,
As far as reactivities are concerned, the amount of information which can be derived from conventional kinetic studies is limited. When the nucleophile is an anion, kl is the only rate coefficient of the three elementary steps which may be determined separately provided that k2 >> k-1, a condition which has t o be inferred in(1) Part I : C. F. Bernasconi, J . Amer. Chem. SOC., 90, 4982 (1968). (2) This investigation has been supported in part by Public Health Service Research Grant GM 14647 from the National Institute of
General Medical Sciences. (3) For reviews on the subject, see (a) J. F. Bunnett and R. E. Zahler, Chem. Rev., 49,273 (1951); (b) J. F. Bunnett, Quart. Reo. (London), 12, 1 (1958); (c) J. Sauer and R. Huisgen, Angew. Chem., 72, 294 (1960); (d) S. D. Ross, Progr. Phus. Org. Chem., 1, 31 (1963); (e) E. Buncel, A. R. Norris, and K. E. Russell, Quarr. Rev. (London), 22, 123 (1968).
d i r e ~ t l y . ~When ,~ the nucleophile is a primary or secondary amine, it is sometimes possible to determine also the ratios k2/k+ and k3B/k-l, provided that k2/ k-l > [H+], so that the intercepts in Figures 9 and 10 are approximately k-, [H+]/KxH and k-2 [H+]/KyH, respectively. For the same change in solvent k-1 [H+]/KxHdecreases by a factor of 5.6 whereas k-2 [H+]/KuHincreases by a factor of 1.3.39 This represents quite a dramatic difference in behavior indeed considering the rather small change in solvent. A question which might arise is why this species has not been found in other solvents, by other methods. The strong solvation requirements of the oxyhydroxylamine, particularly in its anionic form by a polar protic solvent, as borne out by these experiments, which show that even a slight modification of the solvent tips the balance of relative stabilities strongly in favor of the MC, may be the principal answer;40thus, water appears to be a unique solvent favoring oxyhydroxylamines. KYH, k,, and k.-2. By similar reasoning as with the MC's, electronic effects on KYH are expected to be comparable for the three oxyhydroxylamines and any differ(39) The meaning of pH in 20% dioxane-80% water being uncertain, the intercepts rather than k-I/Kxrr or k-l/Kya values are compared directly here. This has no bearing on the argument however. (40) In 50% dioxane-50z water, not even a trace of 7 2 can be detected, implying that the oxyhydroxylamine formation has become negligible relative to MC formation.
4-
OK'
k2
k-i
NO2 2-
shorter as expected on the basis of eq 10. This equation
!-= k3[OH-] + k+
(10)
73
pertains under pseudo-first-order conditions and in the absence of arnineq41 From Figure I1 the rate coefficients ks and k--8 were determined. With sodium chloride as compensating electrolyte, k3 = 48 1. M-l sec-', k-3 = 9.55 sec-', and thus the equilibrium constant for MC formation K 3 = 5.02 1. M-I; with tetramethylammonium chloride, k3 = 70 1. M-' sec-I, k-3 = 6.5 sec-I, and K3 = 11.8 1. M - l . The different salt effect exerted by the two compensating electrolytes is noteworthy; it is in aggreement with findings by Bunton and Robinson on the reaction of 2,4-dinitrochlorobenzenewith hydroxide ion. 4 2 k3 is seen to be so small that it cannot contribute significantly to 7 3 at pH < 12. This explains why 7 3 is not increased in the presence of amine41in the experiments reported. Medium Effects. The negative slopes in Figures 2 and 3 and the observation that the calculated ~ 2 - l curves in Figures 7 and 8 d o not perfectly fit with the experimental results were suggestive for some medium effect operating. There appear to be three different factors influencing the kinetic parameters: the amine concentration, the amine hydrochloride, and the compensating electrolyte. In general, relaxation kinetics is not a very amenable method for the study of medium effects except for limiting situations. The reason for this is that in equations such as 3 or l l the rate coefficient of the forward and reverse reaction will generally be affected in an opposite way, so that medium effects tend to compensate (41) In the presence of amine, ka has to be multiplied by a complex correction factor including all the equilibria. (42) C. A. Bunton and L. Robinson, J. Amer. Chem. Soc., 90, 5965 (1968).
Journal of the American Chemical Society / 92:l / January 14, 1970
137
each other in 7,unless one term is much larger than the other and solely determines 7. Thus T~ in the piperidine and pyrrolidine systems does not reveal any significant medium effect outside the limit of error, but 7' in the butylamine reaction does, because here 71' = k-'[H+]/(KxH [Hf] ), without contribution of kl. Interestingly, the medium effect is much more pronounced with tetramethylammonium chloride than with sodium chloride as compensating electrolyte (Figures 2 and 3). It is not clear whether the medium effect is inherently large but greatly compensated by sodium chloride in a certain concentration range, or if tetramethylammonium chloride introduces an effect of its own. Intuitively, the first hypothesis seems more reasonable because tetramethylammonium chloride is a better model for the amine hydrochloride and should be a more suitable compensating electrolyte. This is also consistent with data obtained by Bunton and Robi n s ~ n 'on ~ the effect of a series of electrolytes on the reaction of aniline with 2,4-dinitrochlorobenzene. Analysis of the data at different pH values, with (CH&NCl compensating electrolyte, demonstrates that
+
both the amine and the amine hydrochloride contribute about equally to a decrease in TI-', i.e., both tend to stabilize the M C with respect to reactants. Several nucleophilic aromatic substitution reactions by amines, in a variety of solvents, have been found to proceed faster in the presence of high amine concentrat i o n ~ , implying ~ ~ ~ ' ~ a stabilization of the intermediate relative to reactants.44 The problem is very complex and there has been no agreement as to the precise nature of this stabilization. The stabilization of the M C by the amine hydrochloride on the other hand might be due to hydrogen bonding to the rather strongly negatively charged nitrogroups ;20 that intramolecular hydrogen bonding plays a role has been shown previously.
Acknowledgment. I wish to thank Professor J. F. Bunnett for criticism and discussion. (43) See, e.g. (a) J. F. Bunnett and R. H. Garst, J. Amer. Chem. Soc., 87,3875 (1965); (b) H. Suhr, Be?. Bunsenges., 67, 893 (1963); (c) C. F. Bernasconi and H. Zollinger, Helu. Chim. Acta, 49, 2570 (1966). (44) This is to be differentiated from the occasional finding that the
amine acts as a general base catalyst.Bb-h
The Mechanism of Reduction of Alkyl Halides by Chromium ( 11) Complexes. Alkylchromium Species as Intermediates Jay K. Kochi and John W. Powers Contribution from the Departments of Chemistry, Case Western Reserve University, Cleveland, Ohio, and Indiana University, Bloomington, hdiana 47401. Received June 25, 1969 Abstract: Alkyl halides are reduced quantitatively to alkanes by an ethylenediaminechromium(I1) reagent prepared in situ from chromous salts and ethylenediamine in aqueous dimethylformamide solutions. The reduction proceeds via an alkylethylenediaminechromium(II1) intermediate, which is hydrolytically unstable. The kinetics of the formation of the alkylchromium species is first order each in the alkyl halide and the chromium(I1) reagent. The mechanism is postulated to proceed in two steps: a rate-limiting transfer of a halogen atom from the alkyl halide to ethylenediaminechromium(I1) followed by a rapid association of the resultant alkyl radical with a second chromium(I1) species. The second-order rate constant for the latter reaction is estimated as 4 x 107 M-1 sec-1 based on competition studies of the cyclization of the w-hexenyl radical to the cyclopentylmethyl radical. The absorption spectra of various alkylchromium complexes are also examined, and the rates of acetolysis to afford alkane are measured in DMF solutions.
A
reagent useful for the facile reduction of alkyl halides to alkanes was presented in a preliminary report. Chromium(I1) perchlorate and ethylenediamine react rapidly in aqueous dimethylformamide (DMF) solutions to form ethylenediamine-chromium(I1) complexes, which reduced even primary alkyl chlorides to alkanes and aryl bromides and iodides to arenes at room temperature. Indirect evidence suggested the formation of a metastable alkylchromium intermediate.' In this paper, we wish to delineate the scope of the reduction of alkyl halides by the ethylenediaminechromium(I1) reagent, to establish the kinetics, to demonstrate the role of alkylchromium complexes2 (1) J. K. Kochi and P. E. Mocadlo, J . Am. Chem. SOC., 88, 4094 (1966). (2) Other aralkylchromium complexes have been isolated : (a) R. G. Coombs, M. D. Johnson, and N. Winterton, J. Chem. SOC.,7029 (1965); 177 (1966); Chem. Commun., 251 (1965); (b) R. P. A. Snee-
as intermediates, and to elaborate on the mechanism of the reduction.
Results Reduction of Alkyl Halides to Alkanes by Cr"(en). The chromous reagent was prepared in situ by simply treating a solution of chromous perchlorate with stoichiometric amounts of ethylenediamine(en) in aqueous D M F solutions in the absence of air. The organic halide was then added and the reduction allowed to proceed at room temperature. den and H. P. Throndsen, ibid., 509 (1965). (c) Alkylchromium(II1) complexes have also been obtained by metathesis: H. H. Zeiss and R.P.A. Sneeden, Angew. Chem. Intern. Ed., Engl., 6,435 (1967). (3) CrI1 is used to denote chromous ion in aqueous solutions of DMF and other solvents. Hexacoordination with solvent is indicated but no attempt will be made to specify coordination unless pertinent to the discussion,
Kochi, Powers J Reduction of Alkyl Halides
