123713AB lecture03

These are the old slides that made up the ‘traditional’ version of these two units (asymmetric synthesis & total synthesis).

I will annotate these slides and see if they work as the reading material for the course ... bear with me, it is a bit of an experiment.

These notes are NOT comprehensive but supplement your own reading. It is impossible to cover these two areas in just 10 lectures (the original length of this module).

Some of my colleagues would go as far as saying “we don’t”. They would, of course, be wrong. There are two quick answers:

1) We need organic compounds so we need to learn how to make organic molecules.

2) Research and Education. The problems encountered in total synthesis push forward the development of new methodology and teach us the application of chemistry.

1

Here we continue our brief overview of some of the factors effecting substrate control.

Again I must emphasise that there are only 6 lectures on stereoselectivity so they are necessarily brief and only give a taste of this intriguing area.

This lecture moves away from addition to the carbonyl group to look at other forms of substrate control but the key concept is still the conformation of the substrate.

Here we see the diastereoselective epoxidation of two closely related alkenes.

Hopefully you remember what m-CPBA is? It is meta-chloroperoxybenzoic acid. You might have forgotten the mechanism of epoxidation … look it up).

The real question is …

2

The last lecture looked at substrate control. This involves an existing stereocentre within our substrate influencing the diastereoselectivity. The obvious limitation of this methodology is that the substrate must contain a stereocentre and it must influence the reaction to give the desired diastereomer.

One solution to this is to add a stereocentre to control the diastereoselectivity … this is auxiliary control.

Macbecin I is a marine natural product that has anti-carcinogenic properties.

Funnily enough the synthesis on the next slides is going to involve the use of a chiral auxiliary …

3

In this reaction the the isopropyl stereocentre on the oxazolidinone (the 5-membered ring containing both and oxygen and a nitrogen) is used to control the configuration of the hydroxyl functionality.

This example may look like an example of substrate control and in many respects it is. The reason it is given a different classification is due to the fact the controlling stereocentre is not an inherent part of the substrate; it is incorporated into the substrate earlier in the synthesis and, perhaps more importantly, it can be removed after it has influenced the diastereoselectivity of the key reaction …

4

… here is a cartoon trying to shown this.

In this idealised example we are converting an achiral substrate into a pure enantiomer.

We do this by:1) adding a new stereocentre.2) using this to control the diastereoselectivity of the reaction.3) removing the original stereocentre to leave the pure enantiomer.

Ultimately, the chiral auxiliary does not have to control the diastereoselectivity. It can just act as a resolving agent.

As the auxiliary allows the creation of a diastereomer we should be able to separate the two molecules before removing the auxiliary and leaving the pure enantiomer.

5

This is an early example (which I appear to have forgotten to reference but I guess that is what SciFinder or Reaxys) is from the synthesis of a beatle pheromone.

The chiral auxiliary is 8-phenylmenthol and this example demonstrates that it can control the facial selectivity of the addition of the Grignard reagent to a ketone.

The original stereocentre is then removed by reducing the ester to an alcohol.

Ozonolysis then gives a ketone that undergoes ketal formation … you should know the mechanism for each of these reactions.

6

The control of diastereoselectivity relies on a number of criteria being met.

• The magnesium coordinates to both carbonyl groups and this prevents rotation around the central C–C bond.• The carbonyl of the ester eclipses the C–H of the ring to minimise interactions.• A 𝛑–𝛑 interaction (𝛑 stacking) helps the phenyl ring to block on face of the substrate.

Probably the most common family of chiral auxiliaries are the oxazolidinones. These have been used in a huge range of reactions and many different syntheses.

Their popularity arises from their reliable diastereoselectivity, versatility in wide range of reactions and …

7

… their ease of synthesis. They are readily prepared by the reaction of an amino alcohol with phosgene (or a synthetic equivalent). The amino alcohols themselves can be prepared from amino acids (as the slide indicates). Obviously, there are a large number of readily (cheap) available amino acids.

Oxazolidinones permit highly diastereoselective electrophilic reactions of enolates. The imide precursors are readily prepared by reaction with the appropriate acyl chloride (the propionate (above) is very common).

Deprotonation with a strong metal base results in the formation of an enolate. The bulk of the oxazolidinone results in the formation of the Z-enolate. Coordination of the metal with the oxazolidinone carbonyl fixes the conformation of the molecule, preventing rotation around the C–N bond. This ensures only one face of the enolate is blocked.

8

Treatment of the enolate with the appropriate electrophile (this is an example of an alkylation) normally results in a highly diastereoselective reaction.

The isopropyl group of the auxiliary blocks the bottom face of the enolate so the benzyl iodide must approach from the top (Si) face.

Here is an example of an allylation reaction using a phenylalanine derivative.

Again, deprotonation results in the Z-enolate. Coordination of the lithium between the two oxygen atoms fixes the conformation of the substrate. The electrophile then approaches from the top face.

9

A huge number of electrophiles can be used in this reaction.

This example uses an electrophilic source of oxygen to perform a diastereoselective hydroxylation reaction. The reagent is called an oxaziridine.

While this example is a diastereoselective azidation.

All these examples display very high diastereoselectivity … all of them >90% de, which would suggest I cut-and-paste …

10

For the oxazolidinone to be a chiral auxiliary it must be easily removed from the substrate after the diastereoselective reaction.

There are a number different methods to do this …

It can be removed by transesterifaction to give an ester …

… or it can be hydrolysed to give an acid.

Note that instead the standard alkaline hydrolysis the optimum results are often achieve with the lithium peroxide. This basic reagent leads to preferential hydrolysis of the correct carbonyl functionality of the imide. Lithium hydroxide can be plagued by ring-opening of the oxazolidinone instead of cleavage of the Csubstrate–N bond.

11

Alternatively, reduction will remove (and destroy) the auxiliary giving you an alcohol.

An example of the use of a chiral auxiliary in total synthesis comes from a synthesis of bistramide A …

12

This example uses an auxiliary derived from valine. Hopefully, you can rationalise the selectivity of the allylation. Deprotonation gives the Z(O)-enolate with the isopropyl group blocking approach of the electrophile from the bottom face. As a result the top (Si) face of the enolate attacks the allyl iodide.

The auxiliary is removed by reduction and the resulting alcohol is subjected to Swern oxidation (temperature must be kept below –35°C to avoid epimerisation) … of course, you all known the mechanism for the Swern oxidation?

The next reaction is the ‘king’ of chemical reactions (see what I did there) … the aldol reaction.

This reaction permits the formation of a C–C bond and potentially two stereocentres. The resulting β-hydroxyketone contains functionality for further elaboration and is a common motif in many natural products.

13

The aldol reaction is a classic organic reaction involving the movement of 6 electrons around a 6-membered ring … a motif that reoccurs throughout organic chemistry.

You should see that the reaction is almost identical to the crotylation reaction and so virtually everything we have learnt about crotylation can be applied to the aldol reaction …

The aldol reaction normally proceeds through a Zimmerman-Traxler chair-like transition state.

The relative stereochemistry is controlled by the geometry of the enolate (cis gives the syn aldol).

The position of everything is fixed except the orientation of the aldehyde; it will be in the pseudo-equatorial position to minimise 1,3-diaxial interactions.

14

The trans enolate [E(O)-enolate] gives the anti aldol product.

Changing the geometry of the enolate changes the position of the methyl group (so that it is equatorial) and this effects the relative stereochemistry.

As before, the position of all the atoms is fixed except the aldehyde (which again prefers to be equatorial).

We have seen how we can control the relative stereochemistry now we just need a method to control which face of the enolate reacts.

This can be achieved with a chiral auxiliary with the isopropyl group blocking approach to one face of the enolate.

Invariably these reactions involve a boron enolate …

15

… but be careful.

Initially, you may think the boron is going to coordinate to the two oxygen atoms of the substrate and this fixes the conformation of the nucleophile as in the previous examples …

… but this is chemistry so of course this does not happen.

The initial step is formation of the enolate; coordination of the boron and the carbonyl group activates the 𝛂-proton.

Deprotonation forms the Z(O)-enolate. The bulk of the auxiliary aids control of the enolate geometry.

For the aldol reaction to occur the aldehyde must coordinate to the Lewis acidic boron and this prevents internal coordination …

16

… the conformation of the substrate is no longer fixed and the auxiliary can rotate around the C–N bond.

Fortunately, one conformation is favoured.

The favoured conformation has the imide carbonyl orientated in the opposite direction to the enolate oxygen. This minimises dipole-dipole interactions (like magnets, dipoles like to oppose one-another).

Once we know the conformation of the substrate we have to rationalise the approach of the aldehyde.

The aldehyde is coordinated to the boron atom. If the aldehyde approaches from the Si face (right hand picture) then there is an unfavourable interaction between the isopropyl group and the substituents of the boron.

17

The alternative is that the aldehyde approaches to the Re face of the enolate (left hand side). This has the isopropyl group and the butyl substituents separated so is favoured.

So the reaction proceeds through the right hand side transition state.

And this explains the observed results.

The geometry of the enolate controls the syn selectivity.

The auxiliary controls the facial selectivity.

As you can see the reaction is incredibly diastereoselective.

18

If you want to find an example of an oxazolidinone being employed in total synthesis then you should probably read the work of Dave Evans … this example is from a beautiful synthesis of cytovaricin.

On this slide the two opposing dipoles are marked (crossed arrow notation) on the transition state. Hopefully, this shows the two substituents blocking one face of the enolate so the aldehyde (and the boron ligands) must approach from the bottom (Re at C2) face.

19

A second example uses a different aldehyde and sets up another two stereocentres of the molecule.

Another fantastic reaction is the Diels-Alder reaction.

This reaction allows the formation of two C–C bonds and up to four contiguous stereocentres.

If you see a non-aromatic six membered ring in a target you almost always can consider using the Diels-Alder reaction to prepare it.

20

The Diels-Alder involves the reaction of a diene and a dienophile.

It is a [4+2]-cycloaddition, so is an example of a concerted pericyclic reaction (this means that all bonds are made and broken at the same time).

The electronics of the two reactants are very important. Normally the diene is electron rich and the dienophile is electron deficient (has an EWG attached).

This slide shows the two bonds that are formed.

The regiochemistry of the addition is controlled by the electronics of the substituents (or more correctly the orbital coefficients/frontier molecular orbitals).

For more information look at some of my other lectures:http://www.massey.ac.nz/~gjrowlan/stereo.html - lecture 8http://www.massey.ac.nz/~gjrowlan/adv.html - lecture 5

21

This shows the 4 contiguous stereocentres that are formed during the reaction.

The reaction occurs with stereospecificity and normally very good diastereoselectivity.

The stereospecificity means that the geometry of the diene and dienophile is conserved in the product (so a trans dienophile gives the anti product while the cis gives the syn product).

The diastereoselectivity is quite surprising. Normally the sterically more demanding product (the thermodynamically disfavoured product) is formed.

This suggests that most Diels-Alder reactions are under kinetic control. The reason for this selectivity (often called endo selectivity) is often rationalised by an effect called secondary orbital interactions (although some people argue against this).

22

In an achiral reaction a racemic mixture of the two endo products is formed.

The endo product has the electron withdrawing ester group ‘under’ the diene.

If you do not know the basics of the Diels-Alder reaction you really need to read up on it … it is a really important reaction.

Addition of a chiral auxiliary allows the reaction to occur to give a single enantiomer of the endo product.

The Lewis acid is very important to the success of this reaction. It serves a number of functions:• It coordinates both carbonyl groups locking the conformation of the substrate• It activates the dienophile by making it more electrophilic and this allows the reaction to occur at a much lower temperature.

23

As with all our reactions, we can rationalise the stereoselectivity by inspecting the favoured conformation of the substrate.

The Lewis acid coordinates both carbonyl groups. This requires dissociation of a chloride. We then have 2 x formal +ve charge on O and –ve formal charge on Al. This means the dienophile is formally cationic so highly electrophilic.

The alkene is s-trans to the nitrogen to minimise the interactions between the alkene and the auxiliary.

The isopropyl group of the auxiliary blocks approach from the bottom face so the diene (cyclopentadiene) approaches the Si (C2) face of the dienophile.

24

This example is from the synthesis of a Chinese herbal remedy.

It includes an example of an auxiliary controlled intramolecular Diels-Alder reaction, which sets up the core six-membered ring.

An intramolecular Diels-Alder reaction permits the formation of a bicyclic system with control of 4 new stereocentres.

The reaction occurs identically to the worked example; the Lewis acid ties the two carbonyls together and prevents the auxiliary from rotating around the C–N bond. The auxiliary then shields one face and the diene approaches through the endo transition state.

25

Please excuse me as I plagiarise myself …

This rationalisation comes from the first ever lecture course I delivered way back in 1999.

It shows (badly) the endo selectivity and how the auxiliary controls the approach of the diene.

And as a little light relief the wonderful Sean Connery …

Of course, this is to introduce an example of radical chemistry.

For a long time people though asymmetric synthesis with radicals would be impossible as they are too reactive but this is not the case as we shall see …

26

This example employs an unnatural amino acid derivative (diphenylalanine) to form the auxiliary.

The radical chain reaction is initiated by Et3B/O2, which allows the use of low temperature. The Bu3SnH is the radical chain carrier. (if you don’t know about radical chain reactions look ahead to lecture 9).

R–X is a secondary or tertiary halide.

The ytterbium is a Lewis acid, which functions as normal … it coordinates to both the carbonyl groups fixing the conformation of the substrate. It also activates the enoate making it more electrophilic.

The diphenyl moiety blocks approach of the radical from the bottom fact so it must approach from the top (Si) face.

27

There are many other chiral auxiliaries other than the oxazolidinones. Here is the 8-phenylmenthol auxiliary being used in the Diels-Alder reaction.

Unusually, the enoate is reacting through the s-cis conformation, presumably due to pi-pi interactions between the dienophile and the phenyl group.

The reaction proceeds through endo transition state as normal.

This auxiliary is the Oppolzer camphorsultam. It is an incredibly robust auxiliary.

This is another example of a Lewis acid promoted Diels-Alder reaction.

It is very hard to visualise the diastereoselectivity in this reaction. It is caused by a mixture of electronics (anti to the nitrogen lone pair) and sterics.

28

NSO

OO

M

H

attack Re face

HH

S ON

O

OM

H

LL

approach Re face

HH

S ON

O

O H≡

aux O

If you are interested … this is my explanation …

Like the oxazolidinones the camphorsultam has been used in many different reactions. Here is an example of it being used in a conjugate (Michael) 1,4-addition.

The rationalisation of stereocontrol is the same as with the Diels-Alder reaction.

29

This is my favourite auxiliary, the sulfoxide.

Sulfoxides not only control the facial selectivity but they also give an excellent handle for further functionalisation (or can be simply reductively removed).

There are many auxiliaries … there are plenty of reviews in the literature.

30