1
The Historical Development of Asymmetric Hydrogenation

John M. Brown

Chemistry Research Laboratory, University of Oxford, 12 Mansfield Road, Oxford OX1 3TA, UK

1.1 Introduction

How did chemists gain the current levels of knowledge and expertise for controlling molecular chirality through hydrogenation or otherwise? The desirability of asymmetric synthesis was recognized in the 1880s by Emil Fischer and others, but practical solutions only arose more than 80?years later. The key reasons are explored here. This brief review has five main Sections 1.2-1.6, covering first the development of ideas underpinning our understanding of asymmetry, then the initial applications to asymmetric synthesis, and also the development of asymmetric heterogeneous hydrogenation of alkenes. The final sections on asymmetric homogeneous hydrogenation of alkenes are limited to work published in or before the early 1980s, in advance of extensive developments, and thus excluding the important inputs of iridium catalysts and more recently early transition metals.

1.2 Early Work on the Recognition of Molecular Asymmetry

Chemistry was an emerging science by the beginning of the nineteenth century with many opportunities for fundamental discovery. At that time scientists crossed disciplines easily; optics and mineralogy played important roles because of the ready accessibility and verifiable purity of solid substances. Malus had invented the first polarimeter in 1808, enabling measurement of both the sense and magnitude of rotation of plane-polarized light [4]. Following this, work by Arago and others on the interaction of polarized light with minerals intensified in the following decade [5]. Haüy had earlier concluded that each type of crystal has a fundamental primitive, nucleus or "integrant molecule" of a particular shape, that could not be broken further without destroying both the physical and chemical nature of the crystal. He had accidentally dropped and shattered a crystal of calcite that enabled him to make the deduction [6]. Biot observed the striking phenomenon that samples of plane sections of rock crystal (a-quartz) rotated the plane of plane-polarized light. Furthermore, some quartz crystals rotated polarized light to the right, and others to the left. The observations obtained by Biot for liquids in his designed polarimeter showed that diverse natural substances, either as liquids or in solution, showed the same phenomenon of optical activity with consistent rotations to the left or to the right for a given substance, which he quantified through Biot's Law. The vapor of oil of turpentine also demonstrated optical activity. By contrast, water, alcohol, and sulfuric acid were inactive [7, 8]. Biot deduced that the response to polarized light was a property exhibited by the individual molecules of the analyte, making a link to Haüy's proposal. The English scientist Herschel was aware of this work and was able to correlate the direction of rotation for a-quartz crystals with the structure of the crystal. He made it clear that the mirror-image pair differed by virtue of the hemihedral faces that were themselves object and mirror image (Figure 1.1) [1, 2]. Well over a 100?years then elapsed before the absolute configuration of an a-quartz crystal was determined by De Vries DeVries, using Bijvoet's recently developed anomalous dispersion method. The laevorotatory form is on the left of Figure 1.1a [3].

Figure 1.1 a-quartz crystals with object and mirror image hemihedral faces; inner part shows packing of sub-units. Created by K-H Ernst and reproduced with permission of Wiley-VCH.

This laid the groundwork for explaining a puzzling observation. Cream of Tartar is a crystalline product isolated as a by-product from winemaking and was widely used in baking and otherwise. It was known to be the dipotassium salt of tartaric acid and showed an optical rotation as expected. A winemaker in the Vosges had isolated a second crystalline product at the same stage of production, and the ensuing isolated acid (racemic acid) had similar properties to tartaric acid but lacked optical activity. This became interesting to the prominent Swedish chemist Berzelius in the late 1820s [9]. He characterized his "paratartaric acid" thoroughly. It was identical to tartaric acid in analytical composition, had the same chemical composition and the same physical properties, and same distinct melting points. His student Mitscherlich [10], by then working in Berlin, discovered that an aqueous solution of paratartaric acid was "indifferent" to polarized light in contrast to the known optical activity of tartaric acid and its salts in solution, although his isolated crystal was active. From the crystal structures of the acids and their sodium ammonium double salts, he concluded that: "the nature and number of atoms, their arrangement and their distance from one another are the same in both bodies." This sparked Pasteur's interest, as he himself acknowledged; he repeated the crystallization of both sodium ammonium salts. Both sets of crystals were hemihedral with a key proviso: the hemihedral faces of the tartrate crystals all had the same sense, while those of paratartaric acid lay either to the left or to the right. Pasteur was able to separate the right-handed from the left-handed crystals on a 5-g scale and examine them separately in solution by polarimetry. Both were optically active but in near equal and opposite directions! This Eureka moment needed rechecking under the critical gaze of his mentor and senior colleague Biot. With that test successfully achieved, his reputation was then secured in 1848 when 26?years old. Later on, he extended this seminal sequence of experiments by resolving paratartaric acid into its enantiomers with the alkaloids quinidine or cinchonidine (CD) [11-13].

All this was carried out without a proper understanding of the molecular structure at the atomic level. The first insights into that came many years later, with the publications of Van't-Hoff and Le Bel in late 1874 [14-16]. Prior to that aliphatic organic compounds were typically drawn (and presumably visualized) as a linear formula string or in-plane with dotted connections (e.g. typical Berichte papers, 1874). Van't Hoff's publication solved the existing problem of isomerism in saturated organic compounds at a stroke - "The theory is brought into accord with the facts if we consider the affinities of the carbon atom directed toward the corners of a tetrahedron of which the carbon atom itself occupies the center." He also explained the existence of enantiomers in the case when a carbon atom had affinities to four different substituent groups. Van't Hoff had claimed to be inspired by the work of Wislicenus on lactic acid, CH3CHOHCO2H, stereoisomerism [11]. Le Bel made similar conclusions and was specifically concerned with explaining the link between tetravalent carbon and optical activity. These analyses proved to be the foundation stone of modern organic chemistry.

The problem of visualizing representations with multiple stereocenters was first solved by Emil Fischer's in-plane notation for sugars, compared with more modern representations in Figure 1.2, [17]. He had determined the relative configurations by chemical means. In order to specify the then unknown absolute configurations, the penultimate carbon in the chain of the dextrorotatory isomer was written with OH to the right, with the aldehyde (or equivalent group) placed at the top of the chain. His guess at the absolute configuration was proved correct, much later, through stereochemical correlation with Bijvoet's X-ray analysis of the sodium rubidium salt of tartaric acid, using zirconium Ka radiation that specifically excited Rb [18].

Figure 1.2 Fischer's representation of the open-chain form of glucose and its modern variants.

Source: Based on Lichtenthaler [17].

1.3 Origins and Early Development of Asymmetric Synthesis [19]

The contrast between optical activity in Nature and its absence in synthesized organic compounds led to the suggestion of a "vital force" that intervened to create natural chirality. This provided a challenge to several late nineteenth century organic chemists, with Emil Fischer prominent. He first suggested that chlorophyll was responsible for the conversion of CO2 into single enantiomers of sugars by plants. Although this was not correct, it made a link between chirality and asymmetric synthesis. He discovered how to convert hexoses into their higher homologues by successive Kiliani hydrocyanation, nitrile hydrolysis to give a lactone, and Na amalgam reduction. He observed significant excesses of one diastereomer of product, formally an asymmetric induction, but he was aware of the possibility of the interconversion of isomers in the course of a multistage reaction, however. To avert this he studied the hydrocyanation of the tetraacetyl salicylaldehyde glucoside, helicin, the glucose moiety acting as a chiral auxiliary. The ensuing cyanohydrin was hydrolyzed and oxidized to the corresponding carboxylic acid with concurrent glycoside cleavage,...