0 Hits

  • Previous / Next

1.1.16 (Aza)-Morita–Baylis–Hillman Reactions

DOI: 10.1055/sos-SD-204-00330

Hatakeyama, S.Science of Synthesis: Asymmetric Organocatalysis, (20121673.

General Introduction

The MoritaBaylisHillman (MBH) reaction is essentially a three-component procedure involving the coupling of the α-position of an activated alkene, such as an acrylate, with a carbon electrophile, such as an aldehyde, under nucleophilic amine or phosphine catalysis thereby producing synthetically useful multifunctional products. When an imine is used as the carbon electrophile, the procedure is called the aza-MoritaBaylisHillman (aza-MBH) reaction (Scheme 1). The MoritaBaylisHillman reaction was discovered in 1967 when Morita and co-workers disclosed the phosphine-catalyzed reaction of an aldehyde with an activated alkene in a French patent.[‌1‌,‌2‌] In 1972, Baylis and Hillman subsequently reported a basically similar amine-catalyzed reaction in a German patent.[‌3‌] However, since these discoveries were not followed by substantial journal publications, the MoritaBaylisHillman reaction did not attract much attention at the time. It was only in the early 1980s that organic chemists started realizing the enormous synthetic advantages of the MoritaBaylisHillman reaction and the first reports of aza-MoritaBaylisHillman reactions appeared in the literature. Thus, it was found that aromatic N-tosylimines undergo a MoritaBaylisHillman-type reaction with ethyl acrylate under 1,4-diazabicyclo[2.2.2]octane-catalyzed conditions.[‌4‌] Now the MoritaBaylisHillman reaction, including the aza version, is regarded as one of the most promising CC bond-forming reactions in terms of synthetic utility, atom economy, and operational simplicity, an importance evidenced by a quantum jump in the number of publications dealing with these reactions since the late 1990s.[‌5‌‌12‌]

Scheme 1 The MoritaBaylisHillman and Aza-MoritaBaylisHillman Reactions[‌6‌]

The MoritaBaylisHillman reaction involves an exquisite Michaelaldolretro-Michael reaction process as depicted in Scheme 2.[‌13‌‌15‌] Michael addition of the nucleophilic catalyst (Nu) to an activated alkene 1 forms an enolate 2 (step 1), which in turn undergoes an aldol reaction with the carbon electrophile 3 to furnish a zwitterionic intermediate 4 (step 2). A proton transfer from the α-carbon to the alkoxide (step 3) followed by the release of the catalyst from the enolate 5 (retro-Michael) affords the MoritaBaylisHillman adduct 6 (step 4). On the basis of pressure dependence, rate, and kinetic isotope effect data, it is proposed that the rate-determining step should be the aldol addition (step 2) and this mechanistic view has long been accepted.[‌14‌,‌15‌]

Scheme 2 Mechanism for the MoritaBaylisHillman Reaction Proposed by Hill and Isaacs[‌14‌,‌15‌]

However, from detailed kinetic evidence and computational studies, a modified mechanism has been proposed wherein two kinds of rate-limiting proton-transfer processes are present in step 3 (see Scheme 3).[‌16‌,‌17‌] In this proposal, step 3 consists of an alcohol-catalyzed pathway, via the transition state 7, involving a six-membered proton transfer from an alcohol (R3OH) to an alkoxide with concomitant generation of an enolate, alongside a non-alcohol-catalyzed pathway via the transition state 8, which embodies a proton transfer from a hemiacetal alkoxide to an enolate.[‌18‌] The first pathway becomes operative in the presence of a proton donor (R3OH), whereas the second pathway, involving the hemiacetal intermediate, dominates in the absence of a proton donor. The rate acceleration observed in a protic solvent such as methanol supports the involvement of the transition state 7 in the rate-determining step. In addition, the autocatalysis observed in the MoritaBaylisHillman reaction in aprotic media can be understood by assuming the formation of the species 7, but in this case the MoritaBaylisHillman product 6 (R3OH) acts as a shuttle to transfer a proton from the α-position of the carbonyl group to the oxyanion. The intervention of the transitional species 8 is supported by the fact that the MoritaBaylisHillman reaction is second order with respect to aldehyde in aprotic solvent; moreover, the frequent production of a dioxanone 9, along with the MoritaBaylisHillman ester 6, also implies the participation of species 8. Due to the fact that the reaction is second order in aldehyde, the non-alcohol-catalyzed pathway is sensitive to the nature of the aldehyde so that, even in the presence of protic species, this latter pathway can become predominant in the reactions involving reactive aldehydes.

Scheme 3 Proposed Dual Mechanism for the MoritaBaylisHillman Reaction[‌16‌]

In contrast to the MoritaBaylisHillman reaction, only two mechanistic studies on the aza-MoritaBaylisHillman reaction have been reported.[‌19‌,‌20‌] The commonly accepted mechanism involving a MichaelMannichretro-Michael reaction process is illustrated in Scheme 4 (utilizing the reactants, intermediates, and products formulated as 1013). Although the mechanism of the aza-MoritaBaylisHillman reaction is basically the same as that of the MoritaBaylisHillman reaction, there are some relevant dissimilarities. It has been demonstrated that the use of a Brønsted acid (R3OH) as a cocatalyst leads to a marked rate enhancement due to the acceleration of the elimination step (step 3) via a transition state 14.[‌20‌] In this case, the rate-determining step will shift from the elimination step (step 3) to the Mannich type CC bond-forming step (step 2).

Scheme 4 Mechanism for the Aza-MoritaBaylisHillman Reaction[‌19‌,‌20‌]

The MoritaBaylisHillman and aza-MoritaBaylisHillman reactions have intrinsic drawbacks, including a low reaction rate, poor conversion, and limited substrate scope. These issues have now been partially resolved by applying physical or chemical methods.[‌10‌,‌12‌] Whereas attempts to develop highly enantioselective MoritaBaylisHillman and aza-MoritaBaylisHillman reactions have met with difficulty due to the complicated reaction mechanisms involved, several strategically attractive methods have been investigated using a chiral 1,4-diazabicyclo[2.2.2]octane derivative, 2,2-bis(diphenylphosphino)-1,1-binaphthyl (BINAP), and 1,1-bi-2-naphthol (BINOL).[‌21‌‌23‌] In 1997, the first reliable method for the preparation of MoritaBaylisHillman adducts with high enantiomeric purity by the use of the acrylamide 15, which has an Oppolzerʼs camphor sultam as a chiral auxiliary was developed (Scheme 5).[‌24‌] A similar diastereoselective MoritaBaylisHillman reaction of various aldehydes and the chiral acrylamide 16 derived from camphor has also been reported (Scheme 5).[‌25‌] Here it was found that either enantiomer of the MoritaBaylisHillman adduct can be selectively obtained by simply switching the reaction medium from aqueous tetrahydrofuran to dimethyl sulfoxide.

Scheme 5 MoritaBaylisHillman Reactions Using Chiral Acrylic Acid Derivatives[‌24‌,‌25‌]

In order to develop an efficient catalytic asymmetric MoritaBaylisHillman reaction, both the creation of the appropriately restricted chiral environment in the transition state and rate acceleration are required. The first successful catalytic asymmetric MoritaBaylisHillman reaction, employing quinidine (17) as a chiral bifunctional catalyst under high pressure conditions, was reported in 1997 (Scheme 6).[‌26‌] It is proposed that the transition state 18 is governed by hydrogen bonding to rationalize the observed moderate S-selectivity (up to 45% ee) in the reactions of aliphatic aldehydes with methyl vinyl ketone. A bicyclic pyrrolidine catalyst 19 has also been developed; this catalyst also gives moderate enantioselectivity (up to 70% ee) via a chelated transition state 20 in the reactions between aromatic aldehydes and ethyl vinyl ketone (Scheme 6).[‌27‌]

Scheme 6 Pioneering Asymmetric MoritaBaylisHillman Reactions Using Bifunctional Catalysts[‌26‌,‌27‌]

The first breakthrough in the catalytic asymmetric MoritaBaylisHillman reaction was made in 1999 by employing β-isocupreidine (22; β-ICD) as a chiral bifunctional catalyst and 1,1,1,3,3,3-hexafluoropropan-2-yl acrylate (hexafluoroisopropyl acrylate; HFIPA; 21) as an activated alkene (Scheme 7).[‌28‌] It has been proved that both the cage-like tricyclic structure and the phenolic OH of β-isocupreidine, as well as the branched structure of 1,1,1,3,3,3-hexafluoropropan-2-yl acrylate, are all necessary for obtaining a high level of asymmetric induction and rate acceleration in the formation of products 23 and 24.[‌28‌,‌29‌] On the basis of these findings and a computational study, as well as a knowledge of the proton-transfer mechanism in Scheme 3,[‌16‌] a reaction mechanism was proposed that is governed by stereocontrol in the aldol step and where the proton-transfer step from two energetically favored zwitterionic intermediates requires their stabilization by hydrogen bonding.[‌30‌] This mechanism as well as the rationale proposed in Scheme 3[‌16‌] suggests that in order to develop an effective asymmetric MoritaBaylisHillman reaction a catalytic system should be designed that has the capability of controlling both aldol and proton-transfer steps. Indeed, after these discoveries a number of sophisticated catalytic systems were developed,[‌5‌,‌10‌,‌12‌,‌31‌] some representatives of which, including the β-isocupreidine1,1,1,3,3,3-hexafluoropropan-2-yl acrylate method, are described in detail in this chapter (vide infra).

Scheme 7 The β-Isocupreidine1,1,1,3,3,3-Hexafluoropropan-2-yl Acrylate Method[‌28‌‌30‌]