Read an Excerpt

The Chemistry of the Moritaâ"Baylisâ"Hillman Reaction

The Royal Society of Chemistry

Morita–Baylis–Hillman Reaction

MEI-XIN ZHAO, YIN WEI AND MIN SHI

1.1 Introduction

The formation of carbon–carbon bonds is one of the most fundamental reactions in organic chemistry and, therefore, has been and remains an important challenge and a fascinating area in organic synthesis. Numerous reactions for the formation of carbon-carbon bonds have been discovered and exploited. Recent progress in organic chemistry has clearly established that the development of a reaction is dependent on two main criteria: atom economy and selectivity (chemo-, regio- and stereo-). Among the carbon–carbon bond-forming reactions, the Morita–Baylis–Hillman (MBH) reaction has become one of the most useful and popular routes, with enormous synthetic utility, promise and potential. The origin of Morita–Baylis–Hillman reaction dates back to 1968 to a pioneering report presented by Morita (phosphine catalyzed reaction) and, subsequently, Baylis and Hillman described a similar amine-catalyzed reaction in 1972. Although this reaction is promising and fascinating, unfortunately, it was ignored by organic chemists for almost a decade after its discovery. At the beginning of the 1980s, organic chemists such as Drewes, Hoffmann, Perlmutter, Basavaiah started looking at this reaction and exploring various aspects of it. Especially, since the mid-1990s, particularly in last decade, this reaction and its applications have received remarkable growing interest, and the exponential growth of this reaction and its importance are evidenced by numerous research papers and several major reviews. The reasons for the rapid growth of MBH reaction can be attributed to its several advantages:

1. the starting materials are commercially available and the reaction is suitable for large-scale production;

2. atom-economic nature;

3. MBH adducts are flexible and multifunctional;

4. usually involves a nucleophilic organocatalytic system without the heavy-metal pollution;

5. mild reaction conditions.

The Morita–Baylis–Hillman (MBH) reaction can be broadly defined as a condensation of an electron-deficient alkene and an aldehyde catalyzed by tertiary amine or phosphine. Instead of aldehydes, imines can also participate in the reaction if they are appropriately activated, and in this case the process is commonly referred to as the aza–Morita–Baylis–Hillman (aza-MBH) reaction (Scheme 1.1). These operationally simple and atom-economic reactions afford α-methylene-β-hydroxy-carbonyl or α-methylene-β-amino-carbonyl derivatives, which consist of a contiguous assembly of three different functionalities.

Although several major reviews have discussed MBH/aza-MBH reaction and their applications in synthesis, it was difficult to completely overview the chemistry of MBH reaction due to a boom of research results in recent years. We hope that this book will satisfy the expectations of readers who are interested in the development of the field and looking for complete and up-to-date information on the chemistry of MBH reaction.

1.2 Mechanism of Morita–Baylis–Hillman Reaction

1.2.1 Amine-catalyzed Mechanism

Indisputably, a thorough understanding of reaction mechanism can lead to a better design of ligand or catalytic system. Whilst the elementary steps of the MBH reaction were postulated in the earliest publications, the fine details of the reaction, in particular those controlling asymmetric induction, have been highlighted only recently, and remain at the core of mechanistic discussions. In 1983, Hoffmann first proposed a mechanism for the MBH reaction, which was refined by kinetic studies by Hill and Isaacs and others. Their proposed mechanism is described in Scheme 1.2. The first reaction step I involves 1, 4-addition of the catalytic tertiary amine 1 to the activated alkene 2 (α,β-unsaturated carbonyl compounds, nitriles, etc.) to generate the zwitterionic aza-enolate (3). In step II, 3 forms intermediate 5 by adding to aldehyde 4via an aldolic addition reaction. Step III involves intramolecular proton shift within 5 to form 6, which subsequently generates the final MBH adduct and releases the catalyst 1via E2 or E1cb elimination in step IV. Owing to the low kinetic isotopic effect (KIE = 1.03 [+ or -] 0.1, using acrylonitrile as electron-deficient alkene and acetaldehyde as carbon electrophile for the MBH reaction) measured by Hill and Isaacs and the dipole increase by charge separation, step II was initially considered as the MBH rate-determining step (RDS, Scheme 1.2).

Many observations in the MBH reactions could be explained by the above mechanism; however, it failed in some critical cases. First, the mechanism did not provide any clue as to why stereocontrol is so difficult in MBH reactions. Privileged nucleophilic chiral catalysts, which in the past have usually allowed good results in related asymmetric transformations, afforded only modest asymmetric induction. This fact pointed out that the basic factors governing the reactivity and selectivity of catalysts in MBH reaction were not fully understood. Other observations, such as the rate acceleration by the build-up of product (i.e., the autocatalytic effect) and also the formation of a considerable amount of "unusual" dioxanone byproduct, such as 11 (Scheme 1.3) in the MBH reaction of aryl aldehydes with acrylates, warned of the limits of the discussed mechanism.

Recently, McQuade et al. and Aggarwal et al. have re-evaluated the MBH mechanism using both kinetics and theoretical studies, focusing on the proton-transfer step. According to McQuade, the MBH reaction is second order relative to the aldehyde and shows a significant kinetic isotopic effect (KIE: kH/kD = 5.2[+ or -]0.6 in DMSO). Interestingly, regardless of the solvents (DMF, MeCN, THF, CHCl3) the KIE were found to be greater than 2, indicating the relevance of proton abstraction in the rate-determining step. Based on these new data, McQuade et al. proposed a new mechanism for the proton-transfer step (Scheme 1.3), suggesting the proton transfer step as the RDS. Soon after, based on their kinetic studies, Aggarwal also proposed that the proton transfer step is the rate-determining step but only at its beginning (≤20% of conversion), then step II is the RDS when the product concentration builds up and proton transfer becomes increasingly efficient. Apparently, the MBH adducts 10 may act as a proton donor and therefore can assist the proton-transfer step via a six-membered intermediate (Scheme 1.3). This model also explains the autocatalytic effect of the product.

In addition, the Aggarwal proposed model shed light on the asymmetric catalysis of the MBH reaction. It suggested that all four diastereomers of the intermediate alkoxide are formed in the reaction, but only one has the hydrogen-bond donor suitably positioned to allow fast proton transfer, while the other diastereomers revert back to starting materials. These mechanistic studies directed attention to the proton-donor ability of the catalyst. If either the Brønsted acid or the Lewis base could be appropriately positioned on a chiral molecule, the Lewis base would react with the substrate (Michael addition), while the acid in an asymmetric environment would allow the chiral proton transfer. The Brønsted acid remains hydrogen-bonded to the resulting enolate in the enolate-addition step to the aldehyde, and finally ensures efficient proton transfer in the rate-determining proton abstraction step. The action of the Brønsted co-catalysts, which are often employed in MBH reaction, is not limited to a role in proton transfer step. It rather promotes conjugate addition by binding to the zwitterionic enolate, and stabilizing these intermediates.

This new kinetic evidence has stimulated further theoretical studies on the MBH mechanism, conducted initially by Xu and Sunoj. Recently, Aggarwal performed an extensive theoretical study, which supported their own kinetic observations and those of McQuade about the proton transfer step. They proposed that the proton-transfer step can proceed via two pathways: (i) addition of a second molecular of aldehyde to form a hemiacetal alkoxide (hemil) followed by rate-limiting proton transfer as proposed by McQuade (non-alcohol-catalyzed pathway) and (ii) an alcohol that acts as a shuttle to transfer a proton from the α-position to the alkoxide of int2 (Scheme 1.4).

To investigate the MBH mechanism, Coelho and Eberlin et al. have also used electrospray ionization mass spectrometry [ESI-MS/MS)] to characterize key MBH reaction intermediates. Using ESI-MS-(/MS), they also shed light on the co-catalytic role of ionic liquids in MBH reactions. Stimulated by the mechanistically important new propositions about the proton-transfer step of the MBH reaction just discussed, Coelho et al. have performed complementary investigations on the MBH reaction mechanism via ESI-MS(/MS). New key intermediates for the rate-determining step of the MBH reaction have been successfully interpreted and structurally characterized (Figure 1.1), providing the first structural evidence supporting the mechanistic propositions made by McQuade et al. and Aggarwal et al., based on kinetic experiments and theoretical calculations, for the dualistic nature of the proton-transfer step of the MBH mechanism.

1.2.2 Phosphine-catalyzed Mechanism

The most likely mechanism of the MBH reaction catalyzed by tertiary phosphines is identical to that of the amine-catalyzed reaction except that the initially formed zwitterion 13 can isomerize to phosphorus ylide 14, which can then undergo a Wittig reaction to give olefins 15 (Scheme 1.5). The latter process may require elevated temperatures, since it is not observed in reactions such as the (aza)-MBH reaction involving the more reactive α,β-unsaturated ketones under mild conditions.

This proposed mechanism has been supported by theoretical studies on the aza-MBH reaction between acrolein and mesyl imine catalyzed by tri-methylamine and trimethylphosphine. The relative energies of the crucial transition states for the PMe3-catalyzed reaction have been found to be lower than those of the corresponding NMe3-catalyzed reaction. The kinetic advantage of the PMe3-catalyzed reaction is also evident in the proton transfer step, where the energies of the transition states are much lower than those of the corresponding NMe3-catalyzed reaction. These predictions are consistent with the available experimental reports in which faster reaction rates are in general noticed for the phosphine-catalyzed aza-MBH reaction.

In addition, various phosphonium salts, the key intermediate in MBH reaction, have been synthesized and characterized to verify the phosphine-catalyzed MBH mechanism. Krafft et al. first isolated phosphonium salts 20 from a MBH alkylation and determined its structure by X-ray crystallography (Scheme 1.6).The phosphonium salts 20 exhibit unprecedented trans geometry of the phosphonium salt and acyl group under kinetically controlled conditions, and lack the previously accepted electrostatic stabilization of the zwitterionic intermediate, suggesting that this electrostatic interaction is not the overriding electronic influence defining the stereochemical outcome of the cyclization. Moreover, these results also suggested that the oxygen–phosphorus electrostatic interaction in the transition state, long considered to be a key component in the traditional MBH reaction, is not a requirement for successful MBH alkylation.

Subseqently, Kwon et al. described the synthesis of stable phosphonium enolate zwitterions 22, which have been proposed as intermediates in MBH reactions, through novel three-component coupling reactions of tertiary phosphines, alkynoates and aldehydes (Scheme 1.7). Notably, the reaction of PMePh2 with methyl phenylpropiolate did produce a zwitterion that was observable in solution (NMR spectroscopy) but not isolable; however, no detectable zwitterion was observed in the case of PPh3. These results are consistent with the hypothesis that electron-releasing alkyl substituents on the phosphonium center play a critical role in stabilizing phosphonium enolate zwitterions. Moreover, according to X-ray crystallography, such phosphonium enolate zwitterions (22) established the tetravalent nature of their phosphorus atoms unequivocally, which stands in contrast to those of the well-established pentavalent 1,2-λ-oxaphospholenes, and might explain the instability and high reactivity of phosphonium enolate zwitterions in MBH-type reactions.

Most recently, Tong et al. have isolated a stable phosphonium–enamine zwitterion (23), which has long been postulated as one of the key intermediates in the aza-MBH reaction, from the PPh3-catalyzed reaction between propiolate and N-tosylimine (Scheme 1.8). Slightly different form Kwon's elegant work, PPh3 worked well in this reaction probably because of the introduction of N-tosylimine as the electrophile. Thus, it was believed that either electron-releasing alkyl substituents on the phosphonium centre or electron-withdrawing groups on the anion centre play an important role in stabilizing the phosphonium zwitterions.

1.3 Activated Olefins

During the past 40 years, the Morita–Baylis–Hillman reaction has seen exponential growth in terms of three components, that is, the activated olefins, electrophiles and catalysts. Both substrate compatibility problems and selectivity issues have improved considerably, though they are not yet solved completely, and the range of olefin reagents has been extended. With the exception of the unusually high reactivity of phenyl vinylsulfonate, the reactivity of activated olefins increases with the electronegativity of the activating group, i.e. phenyl vinyl sulfoxide ≈ acrylamides < phenyl vinyl sulfone < acrylic esters ≈ ethyl vinylphosphonate < acrylonitrile < α,β-unsaturated ketones < α-crolein ≈ phenyl vinylsulfonate, as would be expected based on the mechanism of the MBH reaction. The nature of the catalyst does not appear to influence the reactivity order.

1.3.1 Acrylates

Guided by initial studies on the addition of acetaldehyde with ethyl acrylate and acrylonitrile developed by Baylis and Hillman, acrylates were first studied as activated olefins to react with aldehydes for the MBH reaction. To date, acrylates have constituted by far the largest group of activated olefins employed in the MBH reaction, probably due to the versatility of the ester group in further reactions.

In a series of additions of benzaldehyde to alkyl acrylates, it clearly appeared that the reaction rate decreased with steric bulk and with chain length of the alcohol, probably due it impeding the approach of the reagents (Scheme 1.9). The latter effect may also be steric in nature if the chain folded back on itself, or it could be a consequence of a less polar reaction medium since the reactions were carried out with a 30% excess of acrylate and without a solvent.

Esters with electro-withdrawing group at the β-position relative to the oxygen favored the MBH reaction. However, electronic effects are not the only factors since, for example, 2-fluoroethyl and 2chloroethyl acrylates react rapidly, whereas 2-bromoethyl acrylate fail to react. Steric hindrance, as described above, also plays an important role, which could be responsible for the extremely low reactivity of acrylates containing a long-chain alkyl halide due to cluster aggregation around the reation sites. However, why such aggregation did not take place with 6-thiocyanohexyl acrylate is unclear (Scheme 1.10).

Aromatic esters of acrylic acid react more rapidly than aliphatic ones, and there is no simple correlation of substituent σ values with rate (Scheme 1.11). Although it is not easy to interpret the effect of an electron-withdrawing group in the benzene ring, generally speaking it disfavors the MBH reaction, probably due to a decrease of the nucleophilicity of intermediate zwitterion. With strong electron-withdrawing groups, such as 4trifluoromethylphenol (σ = 0.53), 3-cyanophenol (σ = 0.62), 4-cyanophenol (σ = 0.70) and 4-nitrophe-nol (σ = 0.81), only traces of expected adducts are observed after a much longer time.

Aliphatic aldehydes also react with aryl acrylates more rapidly than with alkyl acrylates, but yield instead the cyclic acetals 25, arising from reaction of the initial MBH adduct 24 with a second molecule of aldehyde, exclusively or in a mixture with the normal adducts 24 (Scheme 1.12). More recently, α-naphthyl acrylates have also been shown to have a significant rate accelaration for the DABCO-catalyzed MBH reactions. Either the normal MBH adduct 26 or 1,3-dioxan-4-ones 27 could be obtained by controlling the substrate ratios and reaction time, respectively (Scheme 1.13).

---

Excerpted from The Chemistry of the Moritaâ"Baylisâ"Hillman Reaction by Min Shi, Fei-Jun Wang, Mei-Xin Zhao, Yin Wei. Copyright © 2011 Min Shi, Fei-Jun Wang, Mei-Xin Zhao and Yin Wei. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---