Search

Rate coefficients have been measured for the Smiles rearrangement of a series of 4- and 5-substituted 2-hydroxy-2′-nitrodiphenyl sulphones in 70%(v/v) dioxane–water at 60.0 °C, and for three substrates at several temperatures. The activation parameters for these substrates have been evaluated, as has the dependence on solvent composition [aqueous dioxane and dimethyl sulphoxide (DMSO)] of the rates for these substrates. The effects of substitution have been assessed by means of the Hammett equation using meta- and para-σ values for the 4- and 5-substituents, respectively, to give ρca.–2.0. All the evidence indicates reaction via a spiro Meisenheimer intermediate with the formation of the intermediate being rate determining. Rate coefficients have been also measured for the Smiles rearrangement of a series of 5-substituted 2-hydroxy-2′,4′-dinitrodiphenyl sulphones in 70%(v/v) dioxane–water at 24.8 °C and for three typical substrates at several temperatures. The activation parameters for these substrates have been evaluated, as has the dependence of rate on aqueous dioxane and aqueous DMSO solvent composition. In aqueous DMSO spiro Meisenheimer complexes are observed as the DMSO content increases. Above 50 mol% DMSO these intermediates are very rapidly and completely formed. The effects of substitution in 70%(v/v) dioxane–water have been assessed by means of the Hammett equation using meta-σ values to give ρca. 1.7. The evidence indicates that, in aqueous dioxane, reaction occurs via the spiro intermediate, with rate-determining decomposition of the Meisenheimer intermediate, and, in aqueous DMSO, the spiro intermediate becomes effectively the ‘initial’ state. The rate coefficients for the formation of the symmetrical spiro complex from 2-hydroxy-2′,4′-dinitrodiphenyl ether in DMSO–water and DMSO-tert-butyl alcohol containing base have been measured. The rate coefficients for the rearrangements of 2-amino-2′,4′-dinitrodiphenyl sulphide and 2-[(2-aminophenyl)thio]-3-nitropyridine have been measured in the same systems. Under these conditions, the ‘initial’ state is the spiro complex and the rate-determining step is the decomposition of the intermediate to form the product.

NN′-Dimethyl-N-(2,4,6-trinitrophenyl)glycinamide (5) is cyclized in methanol under specific base catalysis and gives the spiro-adduct (6). The spiro-adduct is opened by the action of methanolic hydrogen chloride to give the Z- and E-isomers of 2-methylamino-N-methyl-N-(2,4,6-trinitrophenyl)-acetamide hydrochloride (7). In aniline-anilinium chloride buffers the E-isomer of (7) is cyclized to the spiro-adduct (6) with a half-life

Kinetic and equilibrium data are reported for the reversible colour-forming reactions of 2,4,6-trinitro-benzyl chloride with some aliphatic amines in dimethyl sulphoxide and are compared with data for reactions of 1,3,5-trinitrobenzene. With n-butylamine and benzylamine three processes are observable : attack at the 3-position, attack at the 1-position, and transfer of a side-chain proton. For piperidine and pyrrolidine only two-of these are observable, attack at the 1-position being disfavoured for steric reasons. The formation of anionic σ-adducts occurs via zwitterionic intermediates and it is shown that the proton-transfer step may be kinetically significant. Reductions of rate constants for proton-transfer below the values expected for diffusion-controlled reaction are attributed to steric effects which are increased when reaction involves secondary amines and when the CH2Cl group is at the reaction site. Specific effects of chloride ions are noted and are attributed to their interaction with substituted ammonium ions.

The reaction of 2,4,6-trinitrophenetole with aliphatic amines in dimethyl sulphoxide results in the formation of anionic σ-adducts via zwitterionic intermediates. Rapid attack at the 3-position is followed by attack at the ethoxy-substituted 1-position. The 1-adducts formed by reaction with n-butylamine and benzylamine undergo acid-catalysed expulsion of ethoxide to yield N-substituted picramides; that formed by reaction with piperidine is relatively stable. Rate and equilibrium data for these reactions have been determined and compared with data for reactions of related compounds. Increased steric crowding at the reaction centre caused by a change from primary amines to piperidine results in reductions in the rate of proton transfer from zwitterionic intermediates to amine catalyst and in the rate of leaving-group expulsion.

The reactions of the Jackson–Meisenheimer adducts derived from: (a) methoxide ion and 1,3,5-trinitrobenzene; (b) acetone carbanion and 1,3,5-trinitrobenzene, in liquid ammonia solution have been studied by 13C n.m.r. With the adduct (a) the methoxy group is labile as expected, whereas with (b) the C–C bond on the ring sp3 carbon atom remains unbroken, but cyclisation through the γ-carbon of the acetonyl group occurs to form a bicyclo-derivative.

The reactions of several amide and imide anions with 1,3,5-trinitrobenzene (TNB) have been studied in acetonitrile and dimethyl sulphoxide. U.v.–visible and 1H n.m.r. spectra have allowed complete characterization of the N-adducts but the formation of the isomeric O-adducts was not observed. Some N-adducts are found to exist as a mixture of Z and E isomers. In these instances the preferred isomer adopts the E configuration in which the bulky TNB substituent is trans to the carbonyl oxygen. Probable conformations around the N–Csp3 bond of the adducts are discussed in the light of the values of the 3JHNCH coupling constant. Most N-adducts have a thermodynamic stability which is of the same order as that of the TNB–methoxide adducts. This results as well as the failure to observe the O-adducts is shown to be consistent with literature data.

Trihalogenomethyl anions, generated by decarboxylation of trichloro- and tribromo-acetic acid in dimethyl sulphoxide solution, react with added aldehydes. In the presence of 1,3,5-trinitrobenzene, the reaction with aldehydes competes with the formation of the coloured Meisenheimer adduct. The reduction in absorbance from the value in the absence of aldehyde has been used to measure the reactivity of trihalogenomethyl anions towards a series of aldehydes relative to their reactivity towards trinitrobenzene. For 4-substituted benzaldehydes, the reactivities obey a linear ρσ– relationship. The most reactive aldehyde used is only two times less reactive towards CCl3– or CBr3– than hydrogen ions, and it is concluded that, in dimethyl sulphoxide solutions, the reaction between trichloromethyl anions and hydrogen ions is not encounter-controlled. The reactions with aldehydes have been used to prepare several new compounds of the formula RCH(OY)CX3 where R = aryl or pyridyl, X = Br or Cl, and Y = H or COCH3.

Kinetic and equilibrium data are reported for the reactions of 1-chloro-2,4-dinitrobenzene with sulphite ions in dimethyl sulphoxide–water mixtures. Attack at the unsubstituted 5-position yields an observable σ-adduct and is shown to be more rapid by a factor of ca. 12 than attack at the 1-position. The latter process which yields the substitution product is likely to involve intermolecular attack of sulphite on the substrate rather than intramolecular re-arrangement of the 5-adduct.

Kinetic and equilibrium data are reported for the reversible reactoins of 2,2′4,4′,6,6′-hexanitrobibenzyl (HNBB) with aliphatic amines in dimethyl sulphoxide, leading either to σ-adduct formation or to transfer of methylene protons to the base. The most rapid reaction observed with the primary amines n-butylamine and benzylamine and with the secondary amines piperidine and pyrrolidine is attack at the 3(3′)-positions of the substrate. With the primary, but not the secondary amines, this is followed by isomerisation to the thermodynamically preferred 1(1′)-adducts. Slow formation of the blue dianion (7) is observed with the primary and secondary amines and with DABCO. Analysis of the kinetic data indicates that two pathways for the production of (7) involve rate-determining attack of amine on the parent, or on the σ-adducts (2) or (4).

Rate and equilibrium data are reported for reactions of ethoxide ions in ethanol with four alkyl 2,4,6-trinitrophenyl ethers to give isomeric 1,3 and 1,1 σ-adducts. The results indicate the importance of steric factors in this series. Increasing the size of the alkyl substituent causes decreases in values of k3, the rate coefficient, and K3, the equilibrium constant, for reaction at the unsubstituted 3-position, and also causes decreases in values of k1 for reaction at the substituted position.

1 : 1 and 2 : 1 products formed by nucleophilic attack of ammonia in liquid ammonia solutions of 1,3,5-trinitrobenzene and of 2,4,6-trinitrotoluene have been observed by 13C n.m.r. spectroscopy. In the latter reaction three 2 : 1 products are observed: two geometrical isomers formed by attack at the 1- and 3-position of 2,4,6-trinitrotoluene and a structural isomer formed by attack at the 3- and 5-position.