 |
 |
|
||||||||||||||||||||
|
|
|
||||||||||||||||||||
|
 |
|
||||||||||||||||||||
|
|
 |
|
|
||||||||||||||||||
|
|
Home > Groups > Samir Zard |
|
|||||||||||||||||||
|
|
|
|
|
||||||||||||||||||
|
Research interests
Our research interests are focused on the development of new radical and ionic reactions, and their use in organic synthesis. Some of our work is done in collaboration with the industry and will not be mentioned herein.
Xanthates (dithiocarbonates) are a general and efficient source of various radicals: alkyl, acyl, alkoxycarbonyl, alkoxythiocarbonyl, triphenylstannyl, etc... These radicals can be obtained by homolytic rupture (after chemical or photochemical initiation) of either the C-O bond ("Barton-McCombie type radical") or the C-S bond. Total syntheses of (±)-cinnamolide and (±)-methylenolactocine involve alkoxycarbonyl radicals. Intramolecular trapping of these species leads to saturated lactones, or unsaturated if the xanthate group is eliminated.
Methylenolactocine is much more sensitive than cinnamolide, so that the xanthate elimination could not be run under basic conditions, as this leads to the endo isomer. This issue was solved by heating the adduct with copper powder under reduced pressure, distilling the expected product at the same time. To the best of our knowledge, this synthesis is the shortest currently described.
Unlike most of the other radical methodologies, xanthates allow intermolecular trapping of the intermediate radical species by non-activated olefins . This feature enabled us to achieve a short and convergent total synthesis of (±)-matrine , a tetracyclic alkaloid. As shown below, the key step relies on a radical cascade reaction, creating 4 new bonds (among which one is formed through intermolecular trapping) and two 6-membered rings, in only one step. The major product happened to be (±)-matrine (3:1), while the minor compound corresponds to its isomer (±)-allomatrine, another natural product of the same family.
The main advantage of this method relies on the fact that the adduct is a xanthate too. This allows to move on to new radical reactions with different traps, and to build complex systems from very simple starting materials. In addition, as the half-life of the intermediate carbon radicals is relatively long, cyclisation on aromatic rings, which are usually difficult to obtain with classical methods, are made possible. Examples below illustrate the synthesis of several carbon skeletons : tetralones, indolines, indanes, dihydroisoquinolones (it is noteworthy that use of tetrazoylemethyle radical is a 'premiere').
Xanthates give access to otherwise non feasible radicals, or force stabilised and 'lazy' radicals (propargylic radicals for instance) to react, whereas they are useless with classical approaches because of kinetic reasons. This shows the huge potential of this chemistry for organic synthesis. The two following examples show an intermolecular addition of a propargylic radical, and the synthesis of a highly functionalized trifluoromethylketone. This is the first time such highly electrophilic trifluoroacetonyl radicals have been used for synthesis.
The xanthate group formed in the final adduct allows various later modifications, based on radical or ionic sulfur chemistry. Nevertheless, it is possible to perform a reductive cleavage of secondary xanthates, by simply heating the adduct in isopropanol with stoechiometric quantities of dilauroyl peroxide (added portionwise). This method (see example below), is a very interesting alternative to tributylstannane, more widely used but expensive and toxic.
In some cases, particularly with xanthates coming from carbohydrates, this reduction can be performed through a catalytic hydrogen transfer from cyclohexane. This transformation is probably the most spectacular proof of polar effects in radical chemistry. It also gives access very simply to deoxy-sugars. In the example below, the reduction occurs together with a migration of the acetate from position 2 to the anomeric carbon. We have also demonstrated that iodo-derivatives could be reduced in the same way, if the corresponding radicals is electrophilic but not stabilised by conjugation.
Back to Top Non-radical Xanthates Chemistry In the course of this work, we have discovered a new non-radical reactivity of S-propargyl xanthates. We have indeed shown that these derivatives undergo a [3,3] sigmatropic rearrangment, which leads to an allene, that equilibrates with a new type of betaïne. These species show remarkable properties. The example presented below, chosen among many other transformations studied in our group, illustrates their synthetic potential.
Back to Top Radical vinylations and allylations A new radical allylation reaction, avoiding the use of heavy metal, has been developed. It is based on the fact that an ethylsulfonyl radical ejects a molecule of SO2 to give rise to a very reactive ethyl radical, that can extract an iodine atom from an iodo-substituted aliphatic substrate. The new radical then reacts on the allylsulfone, and transfers the allyl (that can be functionalised) to perpetuate the radical chain. This method is compatible with complex structure, as the efficiency of the process is ensured by the reversibility or degeneration of several steps in the mechanism.
New radical vinylation reactions with ethylvinylsulfones, based on the same mechanism, have also been investigated. In this case, the radical attacks on the α position, and the resulting radical ejects a sulfonyl radical propagating the radical chain.
Back to Top Nitrogen radicals and Nickel Chemistry Several novel methods to generate iminyl-, amidyl-, carbamyl- and other nitrogen centred radicals have been developed. Applications to organic synthesis, in particular to the field of alcaloids, are numerous. Thiosemicarbazone derivatives, that can react with stannyl radicals to give nitrogen centred radicals thanks to the N-N bond breaking, are very promising because of the easy accessibility of their precursors. The example below illustrates the potential of this method.
A short and enantioselective total synthesis of (-)-Dendrobine has been achieved. The key step is based on a radical cascade reaction resulting from the formation of a carbamyl radical (methodology developed in our group), and controls the creation of 3 adjacent asymmetric centres.
Another new and practical method, though less general, uses powdered Nickel with acetic acid. This system can reduce acetates, pivalates and other oxime esters to generate iminyl radicals. As explained previously, these species can be trapped to form N-heterocycles. One of the particularities of this method lies in the fact that, depending on the nature of the final radical, a hydrogen transfer reduction or an electron transfer oxidation can occur. In the case of a tertiary radical, the oxidation (in a reductive medium!) is observed, as shown below.
Carbon-centred radicals can also be generated by reduction of α-haloamides with Nickel. The resulting radicals can be trapped by an internal olefin, and the sequence is followed by an external trapping. This method gives access to a wide range of lactams (even β-lactams and oxindoles). The interesting specific case of 5-endo cyclisation, which is usually a difficult and rare reaction, leads in only one step, via a succession of reactions such as reductive cyclisation, oxydation and elimination, to a bicyclic diene, immediate precursor (after an acid driven Friedel-Crafts reaction) to the tetracyclic skeleton of Erythrina-type alcaloids.
Back to Top Natural Products Total Synthesis
Back to Top A new Approach to α-substituted alkynes Nitrosation of 4-substituted isoxazol-5-one, after fragmentation, leads to α-substituted alkynes. The isoxazol-5-one precursor can be prepared in different ways, notably by addition of an organometallic derivative to the alkylidene-isoxazol-5-one. The latter is obtained by Knoevenhagel condensation of an aldehyde or a ketone on isoxazol-5-one.
Back to Top
|
|
||||||||||||||||||||
 |
|
|
||||||||||||||||||||
|
|
|
|
|
||||||||||||||||||
 |
|
 |
|
 |
|
 |
 |
|
|
|
|
| |
|
|
|
|
|
