Organometallic Chemistry Volume 34

A Review of the Literature Published between January 2004 and December 2005

By I. Fairlamb, J. Lynam

The Royal Society of Chemistry

Copyright © 2008 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-0-85404-353-8

Contents

Preface Ian Fairlamb and Jason Lynam, 7,
Samarium enolates and their application in organic synthesis Iain M. Rudkin, Laura C. Miller and David J. Procter, 19,
Metal boryl compounds and metal-catalysed borylation processes: synthetic applications and mechanistic considerations Todd B. Marder, 46,
Organometallics in ionic liquids — catalysis and coordination chemistry Tilmann J. Geldbach, 58,
Groups 1 and 11: the alkali and coinage metals J. V. Morey and A. E. H. Wheatley, 74,
Group 2 (Be–Ba) and Group 12 (Zn–Hg) Felipe García and Dominic S. Wright, 92,
Scandium, yttrium and the lanthanides John G. Brennan and Andrea Sella, 111,
Group 14: silicon, germanium, tin and lead Richard A. Layfield, 155,
Organo-transition metal cluster complexes Mark G. Humphrey and Marie P. Cifuentes, 166,

CHAPTER 1

Samarium enolates and their application in organic synthesis

Iain M. Rudkin, Laura C. Miller and David J. Procter

DOI: 10.1039/b606111g

1. Samarium enolates — an introduction

Metal enolates are amongst the most important organometallic species in synthetic chemistry. The generation of lithium enolates, for example, using strong lithium amide bases and reaction with carbon electrophiles represents a cornerstone of synthetic organic chemistry. Whereas the chemistry of many metal enolates is well understood and extensive structural studies have been undertaken, the chemistry of lanthanide enolates is a little studied area. Over the past 25 years, the widespread use of samarium(II) iodide (SmI2) in organic synthesis has brought the chemistry of samarium enolates to the fore as many processes using the popular reducing agent involve the formation and reaction of these organometallic species.

This review will discuss the role that samarium enolates play in organic synthesis drawing on illustrative examples from the recent literature. In the majority of cases, enolates are directly or indirectly formed by the reaction of substrates with SmI2; a brief introduction to the reagent is therefore given in the following section. The remainder of the review is organised according to the method used to generate the samarium enolate with a section dedicated to the asymmetric protonation of samarium enolates. In recent years, samarium enolates have begun to find application in solution and solid-supported polymer synthesis. This area lies beyond the scope of this article and has recently been reviewed.

One of the few structural studies on a samarium enolate was reported by Hou in 1994. The reaction of a samarium-benzophenone dianion species with bulky phenol 1 led to protonation of the dianion species at the para-position of the aromatic ring to give the samarium(III) enolate complex 2 (Scheme 1). An X-ray crystallographic study revealed that this complex possessed a trigonal bipyramid structure with one aryloxy substituent and two benzophenone moieties equatorial and two HMPA ligands at the apical vertices. When heated in toluene overnight, 2 isomerised to samarium complex 3 (Scheme 1).

2. Samarium(II) iodide in organic synthesis

The first applications of the single electron transfer reagent SmI2 in organic synthesis were reported by Kagan in 1977. In his seminal study, Kagan carried out a thorough investigation of the organic transformations that could be performed using the reagent. The reagent now enjoys a privileged status amongst reducing agents for synthetic organic chemistry.

Due to the reagent’s inclination to revert to the more stable samarium(III) oxidation state, it operates as a single electron donor. This property enables SmI2 to mediate both radical and anionic processes or, most commonly, a combination of the two. The reagent has been used for a wide range of synthetic organic transformations such as functional group interconversions, inter- and intramolecular carbon–carbon bond-forming reactions and powerful cascade reactions that can rapidly increase molecular complexity. The popularity of SmI2 in part arises from its ability to carry out transformations in a highly chemoselective and stereoselective manner. Added to this, the reactivity and selectivity of the reagent can be modified through the use of co-solvents or additives, thus increasing the scope of this already versatile reagent.

The mechanisms of many SmI2-mediated organic reactions proceed via samarium enolates. Understanding and harnessing the reactivity of these organometallic intermediates is vital to the success of many known transformations and to the future development of powerful, new synthetic procedures.

3. The formation and use of samarium enolates

3.1 Reduction of α-heteroatom substituted carbonyl compounds

The reduction of α-heteroatom substituted carbonyl compounds is an important transformation in organic synthesis. When the reaction is carried out using SmI2 this transformation provides a simple route to samarium enolates.

Although the reduction of α-bromoesters was reported by Kagan in 1980, the first detailed study of the reduction of α-heteroatom substituted ketones with SmI2 was carried out by Molander in 1986. The reduction of a range of α-oxygenated ketones with SmI2 was found to give the parent ketones in good yield (Scheme 2). Molander proposed the intermediacy of samarium enolates or enols that were protonated by the MeOH co-solvent.

In order to explore the chemoselectivity of the reaction, α-acetoxy ketone 4 possessing a primary iodide group was treated with the reagent. Chemoselective reduction of the α-acetoxy group was observed in the presence of the iodide. Surprisingly, no elimination of iodide from the presumed samarium enolate intermediate was observed and 5 was obtained in good yield. This suggests that either protonation of the enolate intermediate by the MeOH co-solvent is fast, or that an alternative mechanism is in operation for the reduction of aryl ketone 4 (Scheme 3).

Molander also found that α-halo, α-sulfanyl, α-sulfinyl and α-sulfonyl cyclo-hexanones underwent smooth reduction with SmI2 to give cyclohexanone in good yield.

In 1989 Inanaga investigated the reduction of a range of α-oxygenated esters. Deoxygenation of both α-acetoxy and α-methoxy esters proceeded well at room temperature using HMPA to increase the reduction potential of SmI2. Inanaga found that the use of a more acidic proton source was required for the reduction of α-hydroxyesters (Scheme 4).

In contrast, the reduction of α-heteroatom substituted amides with SmI2 has only recently been studied by Simpkins.

The following sections discuss the use of samarium enolates generated from a range of α-heteroatom substituted substrates.

3.1.1 Samarium enolates from the reduction of α-halo carbonyl compounds. In 1980 Kagan reported the coupling of ethyl α-bromopropionate with cyclohexanone using SmI2. This was the first example of a samarium Reformatsky-type reaction. Analogous asymmetric, samarium Reformatsky reactions of chiral 3-bromoacetyl-2-oxazolidinones have been described by Fukuzawa. For example, reduction of 6 with SmI2 generates a samarium enolate that then reacts with pivalaldehyde to give the α-unbranched β-hydroxy carboximide 7 in 87% yield and in high diastereomeric excess (Scheme 5). The reaction is synthetically noteworthy as highly diastereo-selective acetate aldol processes are difficult to achieve. The samarium(III) ion is presumed to play an important role in the transition state of the reaction leading to high diastereoselectivity.

In 1986 Inanaga reported the construction of medium and large ring lactones using SmI2-mediated intramolecular Reformatsky reactions. In 1991 Inanaga then developed a general synthesis of medium and large carbocycles by means of the Reformatsky reaction. The cyclisation of the samarium enolate intermediates 8 to give carbocycles 9 is believed to be aided by the large ionic radius, flexible co-ordination and high oxophilicity of samarium (Scheme 6).

In 1991 Molander investigated the diastereoselectivity of SmI2-mediated Reformatsky-type cyclisations and found that they often proceed with high levels of selectivity. For example, treatment of α-bromo ester 10 with SmI2 gave lactone 11 in 98% yield and as a single diastereoisomer (Scheme 7).

The SmI2-mediated Reformatsky reaction has since been used as a key ring forming step in the synthesis of a number of natural products and their precursors. In 1997 Tachibana reported the use of SmI2 in the formation of the fused oxonene ring F of ciguatoxin. The samarium enolate derived from α-bromoketone 12 underwent efficient cyclisation to give 13 in good yield after acetylation of the Refomatsky product (Scheme 8).

Mukaiyama reported the use of SmI2-mediated Reformatsky cyclisations in a programme that culminated in an impressive total synthesis of Taxol. α-Bromoketone 14 underwent efficient cyclisation on treatment with SmI2 to give the eight-membered B ring of the target in high yield and with good stereo-selectivity (Scheme 9).

Utimoto and Matsubara have generated samarium enolates, such as 15, from α-bromoesters using SmI2 and have found they undergo efficient aldol reactions. Quenching the samarium enolates with DCl in D2O shows the enolates are stable at -50 ºC but isomerise to the more stable enolate on warming (Scheme 10). The use of two different α-haloesters allows access to more complex samarium enolates before quenching with benzaldehyde.

Linhardt has employed samarium enolate-aldol reactions in a solid phase synthesis of C-sialosides. Sialyl donor 16, immobilised on an amino-functionalised, controlled pore glass support, was treated with SmI2 in the presence of ketone and aldehyde electrophiles, e.g. reaction of 16 with cyclopentanone gave adduct 17 (Scheme 11). Cleavage from the support gave C-glycoside 18 in good overall yield.

Concellón has reported a highly diastereoselective transformation of α-halo-β-hydroxy esters and amides, such as 19 and 20, to E-α,β-unsaturated esters and amides 22 and 23 using SmI2. Following two electron transfers, a samarium enolate 21 is formed which then undergoes elimination. The diastereoselectivity of the elimination has been explained by the intermediate samarium enolate 21 eliminating through a six-membered chelate (Scheme 12).

A similar process involving α-dichlorosubstituted carbonyl compounds has been used to construct (Z)-α-chloro-α,β-unsaturated esters. More recently, Concellón has reported a stereoselective method for the formation of (E)-α,β-unsaturated esters via a sequential samarium enolatealdol reaction followed by an elimination. For example, ethyl dibromoacetate 24 reacts with benzaldehyde to form samarium alkoxide 25 which is reduced to give samarium enolate 26. Elimination then affords (E)-α,β-unsaturated ester 27 in good yield (Scheme 13).

Imamoto has reported the one-pot synthesis of cyclopropanols from carboxylic acid derivatives using samarium and diiodomethane. The reaction proceeds via the preparation of an α-iodoketone, samarium enolate formation and cyclopropanation of the samarium enolate with a second equivalent of diiodomethane and samarium (Scheme 14). In the case of ethyl benzoate, cyclopropanol 28 is obtained in 76% yield. The use of other lanthanide metals led to unsatisfactory results.

3.1.2 Samarium enolates from the reduction of α-oxygenated carbonyl compounds. In 1995 Enholm reported the reductive cleavage of tetrahydropyrans bearing an α-ketone group. Tetrahydropyran 29 was treated with SmI2 and HMPA to produce the samarium ketyl-radical anion before a second equivalent of reagent generated the samarium enolate 30. The enolate was then quenched with benzyl bromide to afford the alkylated product 31 in good yield (Scheme 15).

In 2002 Skrydstrup reported the diastereoselective construction of functionalised prolines by a samarium enolate-aldol cyclisation. Treatment of β-lactam-derived α-benzoyloxy esters, such as 32, with SmI2 led to the generation of a samarium enolate 33, aldol cyclisation and addition of the resultant samarium alkoxide to the β-lactam carbonyl. The efficient sequential reaction gave proline derivatives, such as 34, with high diastereoselectivity and in good yield (Scheme 16).

Procter has developed a linker system for use in phase tag-assisted synthesis based on the reduction of α-heteroatom substituted carbonyl compounds using SmI2.28 The linker has been refered to as a HASC linker (α-heteroatom substituted carbonyl linker). In 2002 an ether HASC linker was used to attach substrates to a polymer support and a solid-phase synthesis of ketones and amides, including 35 and 36, was undertaken to assess the feasibility of the approach. α-Bromo-γ-butyrolactone was immobilised using the linker system and modified to give a range of polymer-supported amides and ketones. At the end of the sequence, traceless cleavage of the HASC linker using SmI2 released amides and ketones from the solid support in good yields and purity (Scheme 17).

Cleavage of the HASC linker with SmI2 releases a samarium enolate into solution which is then protonated. As part of their preliminary study, Procter and co-workers carried out model studies evaluating the possibility of trapping the samarium enolate formed on cleavage with carbon electrophiles. Using ketone 37 as a model for an immobilised ketone, reduction in the presence of cyclohexanone and tetrahydropyr-an-4-one resulted in efficient samarium enolate-aldol reactions to give 38 and 39, resepectively. An attempted samarium enolate-Michael process was less successful and gave the expected adduct 40 in low yield (Scheme 18).

Unfortunately, attempts to trap the samarium enolate formed by the cleavage of a linkage to a polymer support was unsuccessful. It was proposed that residual proton sources contaminating the polymer support led to protonation of the samarium enolate prior to reaction with the carbon electrophile.

3.1.3 Samarium enolates from the reduction of α-sulfanyl and selenanyl carbonyl compounds. In 1999 Matsuda utilised an intermolecular samarium enolate-aldol reaction in the first synthesis of herbicidin B. The enolate 42 was generated by the reduction of glycosylsulfide 41 with SmI2. When TLC showed the reduction to be complete, oxygen was passed through the reaction mixture to destroy excess SmI2 before the addition of aldehyde 43. Aldol adduct 44 was obtained in high yield and as a mixture of diastereoisomers (Scheme 19).

In 2000 Skrydstrup utilised samarium enolates in a selective method for the introduction of carbinol side chains into glycine residues in peptides and showed the potential of this approach for peptide library synthesis. The chemoselectivity of SmI2 and the low basicity of the resultant samarium enolate species makes the lanthanide reagent ideal for this application. Treatment of α-pridylsulfide tripeptide 45 with SmI2 at room temperature gave samarium enolate 46 that underwent aldol reaction with cyclohexanone to give modified peptide 47 in good yield (Scheme 20).

In 2002 Shuto and Matsuda utilised a samarium enolate-aldol reaction to construct l’α-branched uridine derivatives. Reaction of the samarium enolate formed by the reduction of selenide 48 with benzaldehyde proceeded with high stereoselectivity to give 49 (Scheme 21).

Samarium enolates can also react with electrophiles on oxygen. In Overman’s 2001 total synthesis of Shahamin K, a samarium enolate was generated from the reduction of α-phenylsulfonyl ketone 50 and the enolate trapped to give enol acetate 51 by the addition of Ac2O and DMAP (Scheme 22).

Procter and co-workers have utilised a sulfur version of their HASC linker system for the solid phase synthesis of oxindoles and tetrahydroquinolones using SmI2 to cleave the linker. The samarium enolates formed by cleavage of the linker have been utilised, for example, cleavage of the sulfone linkage in 52 results in release of an enolate from the support and cyclisation to give tetrahydroquinolone 53 (Scheme 23).

An analogous sulfur HASC linker system has been utilised by Procter and co-workers for the fluorous synthesis of a range of N-heterocycles. Again, the samarium enolate formed on cleavage of the linker can be exploited, for example, removal of the fluorous tag from oxindole 54 generates an enolate that undergoes alkylation in a cleavage-cyclisation sequence to give spirocyclic oxindole 55 (Scheme 24).

3.1.4 Samarium enolates from the reduction of α-amino carbonyl compounds. In 1999 Honda reported that α-aminocarbonyl compounds can be reduced using SmI2 in the presence of HMPA and a proton source. Honda has applied this deamination process to proline derivatives and to the synthesis of a number of naturally occurring alkaloids including a concise enantioselective synthesis of (–)-adalinine 59, a coccinellied alkaloid. Treatment of 56 with SmI2 in the presence of pivalic acid leads to generation of samarium enolate intermediate 57 (Scheme 25). Protonation and lactam formation gives 58, an intermediate en route to (–)-adalinine 59.

(Continues…)Excerpted from Organometallic Chemistry Volume 34 by I. Fairlamb, J. Lynam. Copyright © 2008 The Royal Society of Chemistry. 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.