Enantioselective Nickel-Catalysed Cycloaddition Reactions

1.1 Introduction

Reactions which form multiple bonds, rings, and stereocentres are particularly important tools for the efficient assembly of complex molecular structures. Of the many families of reactions discovered over the past century, cycloaddition reactions hold a prominent place in the arsenal of synthetic methods currently available to organic chemists, and research activity in this field shows no signs of abatement. Among the metals used to catalyse cycloadditions, nickel has been found competent to promote enantioselectively the formation of carbo- and heterocycles of various ring sizes.

1.2 1,3-Dipolar Cycloadditions

Heterocyclic compounds, which represent almost two-thirds of all known organic compounds, include some of the most significant for human beings. It is not surprising, therefore, that this class of compound has received special attention by chemists to provide selective synthetic access to the enormous variety of structural features typical of this class. The 1,3-dipolar cycloaddition, also known as the Huisgen cycloaddition, is a classic reaction in organic chemistry consisting of the reaction of a dipolarophile with a 1,3-dipolar compound that allows the production of various five-membered heterocycles. This reaction represents one of the most productive fields of modern synthetic organic chemistry. Most dipolarophiles are alkenes, alkynes, or molecules possessing related hetero-atom functional groups such as carbonyls and nitriles. The 1, 3-dipoles can be basically divided into two different types: (1) the allyl anion type such as nitrones, azomethine ylides, and nitro compounds, bearing a nitrogen atom in the middle of the dipole, and carbonyl ylides and carbonyl imines, bearing an oxygen atom in the middle of the dipole, and (2) the linear propargyl/allenyl anion type such as nitrile oxides, nitrilimines, nitrile ylides, diazoalkanes, and azides. Two p-electrons of the dipolarophile and four electrons of the dipolar compound participate in a concerted, pericyclic shift. The addition is stereocon-servative (suprafacial), and the reaction is therefore a [2s + 4s] cycloaddition (Scheme 1.1).

However, the dipole might be stabilised by the adjacent central heteroatom X (nitrogen, oxygen, or sulfur) through resonance, and a non-concerted reaction pathway might also occur. Consequently, in some cases, the original stereochemistry of the alkene is not necessarily conserved, as depicted in Scheme 1.2.

The transition state of the concerted 1,3-dipolar cycloaddition reaction is controlled by the frontier molecular orbitals of the substrates. Hence, the reaction of dipoles with dipolarophiles involves either a LUMO-dipole/HOMO-dipolarophile reaction or a HOMO-dipole/LUMO-dipolarophile interaction, depending on the nature of the dipole and the dipolarophile.

In some cases, when the frontier molecular orbital energies of the dipole and the dipolarophile are very similar, a combination of both modes of interactions can occur. These interactions can also be referred to as either exo or endo, where the endo transition state is stabilised by small secondary n-orbital interactions or via an exo transition state lacking such a stabilisation. However, steric effects can also be important factors for the endo/exo selectivity and override the secondary orbital interactions. Depending on the substitution pattern in the reacting partners, the stereochemical outcome of the process gives rise to either endo or exo cycloadducts. Moreover, the presence of a metal, such as a Lewis acid, in 1,3-dipolar cycloaddition reactions can alter both the orbital coefficients of the reacting atoms and the energy of the frontier orbitals of both the 1,3-dipole and the dipolarophile, depending on the electronic properties of these reagents or the Lewis acid. In particular, the coordination of a Lewis acid to one of the two partners of the cycloaddition is of fundamental importance for asymmetric 1,3-dipolar cycloadditions, since the metal can catalyse the reaction. Furthermore, the Lewis acid may also have influence on the selectivity of the cycloaddition reaction, since the regio-, diastereo-, and enantioselectivity can all be controlled by the presence of a metal–ligand complex. Thus, up to four stereocentres can be introduced in a stereoselective manner in only one single step. In recent years, asymmetric 1,3-dipolar cycloadditions have become one of the most powerful tools for the construction of enantiomerically pure five-membered heterocycles. In particular, the asymmetric 1,3-dipolar cyc-loaddition reaction of nitrones with dipolarophiles, such as alkenes, has received considerable attention over the past 20 years.

Regio- and stereoselective nitrone cycloaddition, followed by reduction of the N–O bond to produce both an amino and a hydroxyl function, allows the synthesis of many products of potential interest. One of the reasons for the success of the synthetic applications of nitrones is that, contrary to the majority of the other 1,3-dipoles, most nitrones are stable compounds that do not require in situ formation. Another synthetic utility of this reaction is the variety of attractive nitrogenated compounds which are available from the thus-formed isoxazolidines. In particular, these products can be easily reduced under mild conditions to give the corresponding chiral 1,3-amino alcohols. The absolute majority of the 1,3-dipolar cycloaddition reactions are diastereoselective and involve chiral alkenes or nitrones. However, the catalytic enantioselective 1,3-dipolar cycloaddition reaction of nitrones has gone through rapid developments during the last 15 years. In particular, metal-catalysed asymmetric 1,3-dipolar cycloadditions have only recently become an important research field. The efficiency of chiral catalysts relies not only on the capability of the enantiopure catalyst to help discriminate between the two p-faces of the dipolarophile, but also on its ability to control both the exolendo selectivity and the regiochemistry as well as the yield. When coordinating to the dipole or the dipolarophile, the Lewis acid catalysts lower the energy difference between the LUM0-H0M0 of the reacting species. The result is that...