Combinatorial synthesis has developed within a few years from a laboratory curiosity to a method that is taken seriously in drug research. Rapid progress in molecular biology and the resulting ability to determine the activity of new substances extremely efficiently have led to a change in paradigm for the synthesis of test compounds: in addition to the conventional procedure of synthesizing one substance after another, new methods allowing simultaneous creation of many structurally defined substances are becoming increasingly important. A characteristic of combinatorial synthesis is that a reaction is performed with many synthetic building blocks at once—in parallel or in a mixture— rather than with just one building block. All possible combinations are formed in each step, so that a large number of products, a so‐called library, is obtained from only a few reactants. Several methods have been developed for combinatorial synthesis of small organic molecules, based on research into peptide library synthesis: single substances are produced by highly automated parallel syntheses, and special techniques enable targeted synthesis of mixtures with defined components. Many structures can be obtained by combinatorial synthesis, and the size of the libraries created ranges from a few individual compounds to many thousand substances in mixtures. This article gives an overview of the combinatorial syntheses of small organic molecules reported to date, performed both in solution and on a solid support. In addition, different techniques for identification of active compounds in mixtures are presented, together with ways to automate syntheses and process the large amounts of data produced. An overview of pionering companies active in this area is also given. The final outlook attempts to predict the future development of this exponentially growing area and the influence of this new thinking in other areas of chemistry.

No abstract

The synthesis and characterization of four new hemicarcerands (1-4) are reported, whose interiors in principle are large enough to embrace such guests as tetraphenylporphyrin or [60lfullerene.Prior publications report syntheses of hemicarcerands with large enough portals and interiors to allow guests like [3.3]paracyclophane to be incarcerated at temperatures of 120-220 "C to give hemicarceplexes stable towards decomplexation in solution at ambient temperatures.1 Mechanical inhibition of hemicarceplex decomplexation has been termed constrictive binding,2 which coupled with host-guest attractive forces accounts for the stability to chemical3 and physical2 manipulation of hemicarceplexes. This paper addresses the question of whether hemicarcerands can be prepared with large enough enforced interiors to incarcerate guests such as tetraphenylporphyrin or [60]fullerene (c60). A complex of the former might act as an oxidation catalyst stabilized by the shell, whereas the host of a capsular complex of c 6 0 should impart interesting new properties to that molecule. We have success- +

The use of OL,G(,OL',E' -tetraaryl-1,3-dioxolane-4,5-dimethanols ( = TADDOLs; 1) as chiral NMR shift reagents ('H, "C, I9F) is described. In many cases, the ratio of enantiomeric aIcohols and amines can be determined under standard conditions of measurement (CDCI, as solvent, room temperature). The preparation and use of a new type of TADDOL, the tetrakis(dimethylamin0) derivative Id, is described. Menthol, octan-2-01, and oct-1-yn-3-01 are partially resolved by crystallization of clathrates with l c and Id.Diols of type 1, readily available from alkyl tartrates in two steps [l], have been found to be useful chiral auxiliaries for the preparation of enantiomerically pure compounds (EPC) by stoichiometric or catalytic enantioselective reactions [ 11 [2] as well as by enantiomer separation of ketones by crystallization [3]. A recent report [4] by Todu et ul. prompts us to describe our experiments aimed at the resolution of alcohols using TADDOLs. l a l b I dSince it is known from crystal structure analyses of TADDOL clathrates that these chiral diols form H-bonds with 0-atoms of C=O groups, and with N-atoms of amines [lc] [3], we thought that, if these interactions were detectable by NMR spectroscopy, then this analytical method might lead the way to efficient enantiomer separation on a preparative scale.A more or less arbitrarily chosen collection of compounds, which are part of current research projects in our laboratories, was used to prepare CDCI, solutions containing 2 equiv. of the 'parent' TADDOL l a and 1 equiv. of the alcohol of interest. The 'H-, I3C-, and I9F-NMR spectra of these mixtures were measured at ambient temperature. Compounds 2-6 (Fig. 1 ) showed nonequivalence of certain 'H-, I3C-, and I9F-NMR signals (arrows) from enantiomers. The resulting chemical-shift differences Ad were sufficiently I )

A facile three-step route has been elaborated leading from furnish novel enantiomerically pure trifluoromethyl-substi-4,4,4-trifluoro-3-oxobutanoate to the trifluoro glycidic ester 1 tuted carboxylic esters, ketones, diols, and epoxy alcohols. The mentioned in the title (0.1-mole scale). Reactions with azide latter ones undergo selective isomerizations by Payne rear-(+ 4,5) and with organometallic compounds such as cuprates rangement (11 -+ 12) in aqueous NaOH/acetone or tert-butyl (+ 3,6), lithium (-+ 7,8), and magnesium derivatives (+ 9 -11) alcohol.