ChemInform Abstract: Invention of New Reactions Useful in the Chemistry of Natural Products

Abstract

The design of new Radical chain reactions useful in the synthesis or modification of Natural Products is discussed. Such reactions can give good yields, and show a selectivity which complements perfectly the ionic reactions in more general use. A flexible system for the production of many kinds of carbon radicals has been invented. It should be applicable for the generation of other kinds of radicals based on elements other than carbon. In the design of these reactions a vital role is played by a disciplinary group (usually the thiocarbonyl group), an idea which is of general application in the invention of radical chain reactions. In a final section a new synthesis of pyrroles is described which is specially designed (by S.Z.Z) for intermediates in the preparation of porphyrins. New reactions useful in synthesis are usually discovered by accident. It is, however, possible to invent reactions by conception (ref. 1). Most of this article is devoted to the invention of new radical chain reactions to give good yields of products. This can be done with the aid of the disciplinary group concept (vide infra). Radical reactions have, of course, an enormous importance in the synthesis of polymers. However, we are concerned here with their use in the chemical synthesis of homogeneous, low molecular weight molecules. In this area of scientific research, radical reactions are not often used, because they are considered to be unselective and to give poor yields of products. It is the purpose of this article to show that well designed radical reactions can give high yields of single products and play an important role in organic synthesis. Radicals react with various functional groups at very different rates which can vary over many powers of ten. By a judicious choice of reagents, solvents etc.. • one can devise a system capable of effecting a highly selective transformation. Radical—radical interactions however (coupling, disproportionation etc...) are extremely fast processes and therefore much more difficult to control. If the sequence of radical reactions is conceived so as to constitute a linear chain process where the propagating steps are so fast that the concentration of radical species remains very small, radical—radical interactions can be largely avoided. Under such controlled conditions, clean high yielding reactions become possible. Moreover radical and radical chain reactions have several advantages over conventional ionic processes : neutral conditions; lower steric effects; lower polar effects; lower tendancy to unwanted elimination reactions; tolerance of many functional groups that have to be protected in ionic chemistry. In ionic reactions s—elimination is a serious problem in the carbohydrate and amino— glycoside antibiotic fields. These important compounds are heavily functionalised and selective deoxygenation or deamimation presented a challenging problem. We shall first consider the problem of deoxygenation. Many years ago, Van der Kerk (ref. 2) discovered accidentally the facile reduction of alkyl halides by stannanes. The mechanism involved is that of a typical radical chain reaction. The stannyl radicals, generated photochemically, thermally or by a chemical initiator (e.g. azoisobutyronitrile — A.I.B.N.) abstract the halogen to give a carbon radical. This latter abstracts a hydrogen atom from the hydride to give the alkane and a stannyl radical, thus propagating the chain (Scheme 1). This reaction has developed 675 676 D. H. R. BARTON AND S. Z. ZARD into a useful synthetic tool, not only for reducing halides but also for making carbon—carbon bonds by interception of the intermediate carbon radical.