Onium salts, elimination

The onium carbanion formed under phase-transfer conditions is unstable depending on the anion source, and in the absence of an electrophilic reaction partner, degradation of the accumulated onium carbanion in the organic phase may be observed. This is known to proceed via Hoffman elimination, nucleophilic substitution and/or Stevens rearrangement (Scheme 1.4) [4f,6,7]. The direct decomposition of onium salt, as influenced by the strong inorganic base at the interface, may be also operative. 

Diisopropylethylamine is a useful base for this purpose, especially when further transformations of the onium salts are possible. a-Elimination reactions of this type have recently been employed for the synthesis of alkenyl(diaryl)-and alkenyl(triaryl)onium salts of Group 15 and Group 16 elements (Scheme 46)... 

A simple kinetic criterion is applicable in those a-eliminations which lead to the formation of dimeric olefins (see Fig. 2). In basic media, a number of organic halides and onium salts, characterized by the absence of a /J-hydrogen, undergo this reaction. It can be formulated as... 

In Eqs. (5-23) and (5-24), there is no net change of charge after reaction, but for Eq. (5-22) charge is created and for Eq. (5-25) it is destroyed. Further examples of observed effects for solvent changes on rates of mono- and bimolecular eliminations are given by Hughes and Ingold [16, 44]. In most cases studied, haloalkanes and onium salts in ethanol/water mixtures, the observed solvent effects are in the expected direction. 

The solvent-influenced synjanti dichotomy for bimolecular eliminations of acyclic and medium-ring bromides, tosylates, and onium salts has been reviewed [395, 693] and will be mentioned only briefly. As a rule, the s jn-elimination pathway gains importance in non-dissociating solvents, while dissociating solvents facilitate the more common anti-elimination reaction. The more unusual s jn-elimination is favoured in non-dissociating solvents because of ion-pair association, which favours a cycHc six-membered activated eomplex as shown in Eq. (5-151a) see reference [395]. 

The crystal standard of (7r-cp)2Ti(NCO)2 has been determined and confirms that, in the solid state, the cyanate group is here N-bonded it was only in the last stages of the analysis that the 0-bonded alternative could be finally eliminated (i). Hexa-iV-cyanato complexes of ytterbium, erbium, and neodymium have been reported as quaternary onium salts (16) and reference made to the series of tetraethylammonium salts of [Ln(NCO)0] (Ln = Eu-Yb) (22) these results extend Table XLII. The ESR spectra of series of complexes CuL2(NCO)2 (L = an, or substituted an) have been interpreted to show Cu— N(CO)—Cu bridges with no indication of any Cu-0 interactions (19). The ESCA spectra of [M(NCO)4] (M = Mn, Co, Zn) have been recorded (12). 

Apart from the reactions of diazonium salts, a number of other reactions are known in which the C-N bond is broken. The best known of these is the Hofmann elimination of quaternary ammonium hydroxides (Scheme 2.37). An amine is converted by methylation with methyl iodide to the quaternary ammonium salt ( exhaustive methylation ). The iodide, on treatment with moist silver oxide, forms the quaternary ammonium hydroxide which undergoes a bimolecular elimination to form an alkene. The bimolecular elimination of onium salts yields the least alkylated alkene. This substitution pattern is determined by the ease with which a hydrogen atom can be attacked by the base. 

The application of the principle of hyperconjugation to the elimination reactions of saturated molecules, therefore, may be stated as follows Except for the second-order reaction of onium salts, the structure of the principal product obtained from an elimination reaction will be the one having the greater number of a-hydrogen atoms. 

Before leaving our discussion of elimination reactions, the distinction which has been drawn between E2 reactions of alkyl halides and onium salts should be emphasized. When onium salts arc decomposed it has been assumed that the most important step in the process is the formation of a reactive complex between the cation and a base. Once this transition state is attained, olefin formation occurs directly so that the process is effectively irreversible. Consequently, if there is more than one point within a molecule where such complexes can be formed, reaction will proceed predominantly through the one which can be attained most easily. 

Unlike the three fundamental mechanisms of olefin formation already outlined, the a -ji (or ylid) mechanism is only applicable to elimination from onium salts . The base abstracts an alpha hydrogen from the leaving group to form an ylid which subsequently acts as an internal base and abstracts the beta hydrogen atom, viz. 

A number of benzylic halides and onium salts under the influence of base undergo an apparent a-elimination to yield dimeric olefins, viz. 

Scheme 7.29. A representation of two potential pathways for proton loss from a carbocation intermediate on the Sn1-E1 reaction surface. Both pathways are followed. It is supposed that proton loss via the pathway labeled (a), which results in the most highly substituted alkene (the Saytzeff product), occurs preferentially because that is the product formed in highest yield. The alkene resulting from pathway (b), the Hofmann product, is also formed. A. W. Hofmann (1818-1895) was a German chemist who was professor of chemistry at the Royal College of Chemistry in London (1845-1864) and then accepted a post as professor at the University of Berlin. Most of Hofmann s work dealt with amines (Chapter 10). Hofmann found, in contrast to Saytzeff, that the least highly substituted alkene is formed when the elimination is carried out on amine quaternary salts (so-called onium salts). This is, in part, presumably due to the close association between the base and the positively charged onium salt as well as to the removal of the proton in the rate-determining step (cf. the E2 reaction). (Note the 82 18 ratio of products shown here should be considered identical, within experimental error, to the 79 21 ratio of Table 7.9.)...

---