Welcome to the lecture designing organic synthesis and imagine you are working on your master's thesis or your PhD thesis and you are in need of synthesizing this molecule, this structure.

You have already checked SciFinder and you found out no one has ever synthesized this molecule. Well, actually in this case it's not true, but maybe you are not satisfied with the protocol which is offered to you in the literature.

Or another scenario, you are in search of a new job, you have your job interrogation and your, well, hopefully later on

boss is asking you, we have a problem in our company, we would like to synthesize this molecule, do you have an idea? And of course this guy, well, does not really believe that you have an optimal synthesis

present like that, but he wants to test, can you deal with a problem like that? Well, so and what we would like to learn during this lecture is, well,

analyzing such molecules, how to synthesize these, maybe also some even more complex molecules. And moreover, during this lecture we will recapitulate our basic knowledge of organic chemistry.

As you will find out, molecules like that are synthesized by all reactions you normally know from organic chemistry 1 or organic chemistry 2, this is basic knowledge of organic chemistry.

So let us start with some key terms.

First of all, retrosynthetic analysis. So retrosynthetic analysis is a technique for planning organic synthesis.

So a technique for planning an organic synthesis or lots of synthesis

by transforming the target molecule like this into hopefully much simpler precursor structures.

First example, this is now our target molecule.

So we want to, by retrosynthetic analysis, to break bonds that this molecule maybe break up into simpler, much simpler partial structures.

Therefore, we are looking for so-called strategic bonds, so let's write down B, strategic bonds which have to fulfill the following criteria.

If we break those bonds, just theoretically, then our molecule becomes simpler or falls apart into two partial structures which are of course then both simpler than that target molecule.

And secondly, most important, we have to know a method, a good method, how to form that bond. Well, in practice, otherwise it doesn't make sense.

Or we have to, of course, to put in some time in research, developing the method. So in this case, it is rather simple and straightforward.

Let's break already two bonds at the same time. So we draw this type of arrow. This arrow is a retrosynthesis arrow.

Well, and if we break that homolytically, that would mean in this case, in this case we are doing that homolytically, the radical center here, radical center there, that means a double bond is formed here.

So we have now this structure, well, acrylic acid, methyl ester. And on the other hand, well, if we have done here radical center, here radical center, then of course a butadiene moiety is formed here.

And well, just draw that and you will find out that we are dealing with a substituted cyclopentadiene.

Now as I said, well, these are strategic bonds because the partial structures which result then by this bond breakage, double, twice bond breakage,

are much simpler than our target molecule. And on the other hand, yes, we know a method to form that bond because well, simply these two structures as compounds, as reagents,

would react in a Diels-Alder reaction and will give this product predominantly, as you know this is a Diels-Alder product with an ando orientation of the ester functionality.

This is the kinetically favored product of this type of Diels-Alder reaction as we know it. Well, currently we don't need to deal with enantioselectivity.

Well, as you see, this is of course a chiral molecule. Well, if we would need to consider enantioselectivity, we would have to apply a chiral Lewis acid, hopefully inducing some enantiomeric excess.

So, well, we would then have a look, do we have these chemicals present or can we buy it? Are they cheap enough?

If not, and this is the case for that one, we would have to think about of course synthesizing that. And now, again, a retrosymphesis arrow, well, has to be drawn and we would have to disconnect another bond.

Well, take one which is rather central. This is then a strategic bond because we have then two parts approximately the same size.

This is usually a rather good idea. So, and well, this is also rather easy.

Cyclopentadiene has one starting component and the other one could be, for instance, benzylic bromide.

Next step will be, we will have a look for how to analyze that systematically. Next expressions are symphons and synthetic equivalents, but here I think it's clear to you that you should have to deprotonate the cyclopentadiene.

Then we have the cyclopentadienyl and iron aromatic system, relatively high acidity of those hydrogens. And this will then react as the nucleophile, for instance, SN2 type process giving this one.

Or, unfortunately, because we are on a basic reaction, conditions double bond isomers of that.

So, if you would try that reaction we could anticipate at least some problems because for that target the double bonds of the diene has to remain at this position and should not migrate in that.

So, simply we don't want to have finally this one as the product. Okay, this is a problem. We would have to anticipate this problem and have to deal with that problem. So, next key terms. Well, a retrosynthesis tree. Let's call that C.

Here for that target molecule the retrosynthetic analysis was rather straightforward. And, well, let's assume here we have one more complex target molecule and we have two ideas.

One is retrosynthetic pathway 1 and retrosynthetic pathway 2 getting two partial structures A and B.

Or, as an alternative, we have A' and B'. So, then A and B as well as A' and B' are not commercially available.

Then we have, again, to analyse that how should we disconnect A to get two more simple structures or B we have just one idea and here it's the same.

So, and then it goes maybe on and on until we have really simple starting materials.

And, as you can see, some kind of tree is evolving. It might be rather complicated, the result of our considerations. So, next important key terms, a synthon. So, what is a synthon? It's a structural fragment with an assigned reactivity.

So, this is already D. Then E, a synthetic equivalent. This is a reagent or

a substrate within our synthesis which corresponds to the synthon in terms of its reactivity.

I think this will become clearer with an example.

Paramethoxyacetophenone, of course a molecule that you can buy.

But let us assume that we have to design a synthesis for this compound. So, where do we have the strategic bond? Let's say we focus, we will focus on this one.

We disconnect that and our first idea, as it will turn out, not a good idea, is the homolytic bond disconnection.

And homolytic would mean, well, we need some kind of radical reactivity here and, on the other hand, also radical reactivity there.

So, this must not necessarily be radicals. These are the synthons, partial

structures of that which we have assigned a reactivity, radical, radical reactivity here. And then we would have to figure out what kind of reagent or substrate could deliver such a reactivity.

So, we would have to think about what could we introduce which will deliver radicals under certain reaction conditions, photochemically also.

But why is that not a good idea in this case? Simply because if we have a mixture of radicals, how would we get the selectivity that a cross-coupling will occur?

We would have as competing reactions, of course, the formation of homo-coupling products. So, homolytically disconnection is always to be considered if we can divide

our target molecule, in terms of a retrosynthetic analysis, into identical parts. Otherwise, normally not. So, despite the homolytic disconnection, we should better consider heterolytic disconnections.

So, the first one, let's call that retrosynthetic plan 2.

Let's disconnect that bond that we have an acceptor reactivity at this position.

So, A should be an abbreviation for acceptor reactivity and on the other hand, donator reactivity at this position.

So, D means donator reactivity. Alternatively, to A for acceptor reactivity and D for donator reactivity, we could

put in an plus for acceptor reactivity and a minus for donator reactivity.

This is describing the reactivity, not naturally, and an ion or a cation could be, but must not necessarily be. This is just a partial structure of the synthon with an assigned reactivity. So, if we think about how could we get a donator reactivity for that carbon, where we have that carbonyl group,

it's part of a carbonyl group with the electronegative oxygen diminishing electron density at this position.

So, the natural reactivity of the carbonyl group here is not a donator reactivity but an acceptor reactivity. On the other hand, this structure represents an electron rich aromatic system with a plus M effect of that methoxy group.

That means we have relatively high electron density in the para position.

So, the natural reactivity of this structure is donator reactivity at the para position. So, this disconnection is again not a good idea because we would have to assign here, or we have designed in this case,

the opposite reactivity of the natural reactivity of these partial structures. Therefore, we should now just choose retrosynthetic plan 3, donator reactivity, best idea,

the same as the natural reactivity at that position and acceptor reactivity there. The next step again, these are the sünfunds and we have to translate now to the synthetic equivalents.

So, what kind of substrate or reagent would really show that reactivity and have that partial structure?

Well, it could be for instance an organometallic species like that with lithium or magnesium halide.

That would mean we have a very high delta minus here, very high reactivity. On the other hand, well, with an acceptor reactivity here it's always a good idea to put on an X for a leaving group

and then write down what X could be. It could be a chloride, could be a bromide, maybe an acetate, then we would have the anhydride.

Or maybe a lower reactivity, just an astro.

Well, therefore first idea of our analysis is taking such an organometallic species and let that react with either one of that.

Well, we would notice that a lot of energy will evolve in that type of reaction because this has a very high reactivity and actually this is not necessary.

Maybe it's a better idea just taking anisole and performing an electrophilic aromatic substitution in that case.

Well, putting that together this and the acetyl chloride or the acetic anhydride, no reaction will occur normally in that case. We have to add a catalyst and then you know what kind of reaction we are talking about.

By these considerations we have reinvented the Friedel-Crafts oscillation. So that means Friedel-Crafts oscillation.

So, as I said, in these cases where we have assigned a reactivity opposite to the natural reactivity we have a problem.

We need the other reactivity, so we need the so-called unprolong. Indeed, also in English it's unprolong.

So, therefore, we now get to the next key term and that's unprolong.

So, unprolong of reactivity is meant by that and you can call it polarity inversion.

Or maybe one of the eldest examples is of this type. Let's assume an N-propyl bromide. We have a partially negative charge at the bromide and this carbon is positively polarized.

Therefore, we actually have an acceptor reactivity at this methylene group, acceptor reactivity.

If we treat it with magnesium in ether, a Grignard reagent is formed now with

a delta plus at the magnesium and a partially negative charge at the CH2 group. That means we have donator reactivity at this position.

From acceptor reactivity to donator reactivity we have obviously a polarity inversion. This is a very old but very nice case of unprolong. Then, such an aldehyde, acetic aldehyde.

We have acceptor reactivity here as the natural reactivity.

If we need donator reactivity there at that carbon we have a couple of methods to achieve that. This one is the classical method by Dieter Zebach who introduced the term unprolong and he was a professor at the ETH in Zurich.

So, a dithiocetal is formed and with n-butyllithium this is acidic enough to be deprotonated.

Therefore, we get a lithiation at this carbon. Now we have a delta minus here. Therefore, donator reactivity at this position and it can react then as a nucleophile.

Of course, after it has react as a nucleophile, for instance with this electrophile and

X as a leaving group, then we have to get rid of this protecting group. There are various methods for that. Not all of them are nice ones.

More examples how to change the reactivity of a polarity at this position of aldehydes.

More examples from organic chemistry too. Do you remember that a silane chemistry you certainly have discussed at that stage of your studies?

So, it's about this type of transformation. Well, an aldehyde reacts becoming to, it dimerizes actually.

So, this C-H bond is added to the carbonyl bond of another aldehyde, an a silane.

So, methods you should know for that type of reaction are always catalyzed by something.

For instance, catalyzed by a cyanide. So, the idea here is the cyanide is added to the aldehyde functionality.

Now, that proton is alpha to the cyano group becoming more acidic. You already have the base and you have then this as a reactive intermediate.

You have there the nucleophilic center. It attacks another aldehyde. And finally, all the retro process will start and potassium cyanide will be eliminated. So, you can apply that indeed in as a catalyst.

And on the other hand, there is the Stetter reaction and the Stetter catalyst. These are thiazolium salts or where you have this proton becoming somewhat acidic and under basic reaction conditions.

You have carbenoid intermediates which then perform a reaction similar to what we have discussed here with the cyanide. Well, this is basic stuff from organic chemistry 2 and if you don't

have that in mind, then please recapitulate that with your organic chemistry 2 material. More examples for Umpolum.

For such an acetyl group, we can assign a donator reactivity to that CH3 group.

Why the donator reactivity? Well, just because from our, I should better, the synthon is this one.

Donator reactivity here. That is the synthon. It's a partial structure and we need donator reactivity here. We can translate that to this synthetic equivalent, just an acetyl group under basic or even acidic reaction conditions.

Under acidic reaction conditions, you will form the enol in equilibrium.

Under basic reaction conditions, it will just be deprotonated and you will form the enolate. And the enolate of course is highly nucleophilic at this position.

Well, this is that carbon transferred to real highly nucleophilic center and well, being in accord with this synthon.

How could we achieve an Umpolum at this position? So, acceptor reactivity here.

As I told you, the most straightforward idea is just add a leaving group at this position. X and again, X could be chloride, bromide and so on.

Unfortunately, as you know, chloride and bromide at that position, alpha to a carbonyl group, these are usually lacrimators and we would like to avoid working with them.

Well, maybe sometimes it's a bit better to have a protecting group here in addition. And there is another intriguing idea, how you get a synthetic equivalent for that type.

It's this one, having a double bound methylene group here in olefin.

So, what would be then the reaction sequence you would have to perform? With the leaving group here, you have there the acceptor reactivity.

So, now add the nucleophile, make the substitution reaction, now the reaction with acceptor reactivity has been performed.

Well, you need the carbonyl group here. The transformation of that olefin into a carbonyl group has to be done and this can be achieved simply by an ozonolysis.

Finally, with this result. So, the next symptom, acceptor reactivity in better position to a carbonyl group.

This is considered as the natural reactivity simply because as synthetic equivalent you can choose an alpha, better and saturated carbonyl compound.

A nucleophile will attack, a soft nucleophile will attack at the better position in a conjugate addition process.

Simple reaction working very well as you know from organic chemistry too.

So, what has to be done if we need the opposite reactivity, umprolong again?

Well, we have to think about the synthetic equivalents again. Well, that could be having a protecting group here, an acetal and there a rhenate reagent

or a bit more complicated an enol ether at a cumulene lithiated here at this position.

This has indeed been done. Or a double lithiated compound, this one.

You can achieve this type of double methylation with strong organometallic bases.

So, F, G, G, the last key term we have to talk about is the so called retron.

So, what is a retron? A nice definition is it's the minimal molecular substructure that enables a certain transformation.

Again, this will become clearer with an example, a butadiene and an olefin if

appropriately substituted will react in a Diels-Alder reaction to give a cyclohexene moiety.

This cyclohexene moiety is the retron of a Diels-Alder reaction. So, retron of a Diels-Alder reaction.

Which now means if we find in our target molecule, which we are analysing in a retrosynthetic analysis, if we find there a cyclohexene moiety as a partial structure.

This then suggests making use of a Diels-Alder reaction as maybe a key step in our synthesis.

Let's try examples for that.

This is now our target molecule. We notice that here we have a cyclohexene moiety.

Therefore, we should consider breaking these two bonds simultaneously giving rise to these two much simpler structures.

Malenic anhydride and this one, well, benzothiophene with two exocyclic double bonds.

Okay, this will be rather sensitive. This you can buy. This will have some tendency for polymerization, especially under acidic reaction conditions. Maybe we can generate this in situ in the presence of the malenic anhydride.

So, that means generating that is, well, again analysing how to synthesize that and maybe we can synthesize that by an elimination reaction.

Okay, of course, you can synthesize a butadiene moiety by an elimination reaction or appropriate elimination reaction.

Having for instance acceptor reactivity here, donator reactivity there. Well, how could we get to this situation? Well, we can think about lots of possibilities to achieve that.

Having, for instance, a leaving group here and getting methylation at this position will, for instance, eliminate lithium halide.

Several ways for achieving similar reactions have been realized in literature. One rather simple process was that one. Starting from benzothiophene, the bisbromo methylation was achieved in just one step, one reaction step.

And then potassium iodide, for instance acetone, was applied. This is again a name reaction you might know. No, Finkestein.

So, it's a Finkestein reaction which will give this bis-iodo-methyl benzothiophene.

Wow! And just heating it up a little bit will eliminate molecular iodine.

This is generated and it will be trapped immediately by the dienophile which should be already present in the reaction solution.

Next example, as an exercise for you, this is your target molecule.

Please try a retrosynthetic analysis while keeping in mind we are still dealing with the retron of the diels-alder reaction.

So, in this target molecule we notice we have a six-membered ring since it's a cyclohexane derivative. We know six-membered rings are easily synthesized by these alder reactions but not cyclohexane but cyclohexene.

Okay, one other advantage of the diels-alder reaction is that stereoselectivity cis-trans isomers,

or certain cis-trans isomers as dienophiles are then translated in the product in a certain diastereoselectivity. So, but for the diels-alder reaction we are missing the retron of the diels-alder reaction.

So, we have the cyclohexane, we need the retron of the diels-alder reaction, so let's introduce a double bond at the right position. For instance here, then we have the retron of the diels-alder reaction and it's okay to

do so because we know a nice method to actually get the transformation done from here to there. This is simply a hydrogenation reaction with some catalyst, some transition metal catalyst.

So, and now with that retron of the diels-alder reaction in here, we can get that disconnection within our retrosynthetic analysis.

So, this would be the dienophile we need together with butadiene to get to this product.

Well, if we interpret this as an enantiomer, then we should add, for instance,

chiral Lewis acid, chirality marked with a star there to get some enantioselective induction. So, if you don't...

find the return, we can generate that return as one step in our retrosynthetic analysis. Another example, a steroid system. Here we have a cyclohexene moiety.

Making plans for synthesizing that, we could notice retron of its other reactions. Well, could be an idea making that disconnection here and there.

So as retrosynthetic analysis, this would be correct in principle, but not a good idea

because this is more complex, far more complex than that. Therefore, these two bonds are certainly not strategic bonds. In fact, there are publications known where some kind of similar

transformation like that was done from here to there to form such interesting macro cycles.

And this is the last reaction we talk about today. This is from Eckhart Winterfeld's group in Hanover. He had a nice publication, I think in the late 70s. He started

with this diene, this steroid, and let this react

with propagulic aldehyde just in toluene and reflux temperature. So what will happen?

Dienophile diene, a Diels-Alder reaction will take place. Well, just a sketch for a Diels-Alder product. So we have something like that, ring system here

and ring system there and a cyclohexadiene moiety there. And if you disconnect

at these two bonds, then an olefin will be formed here and an aromatic system there.

Under very moderate reaction conditions with a catalyst, we could isolate that. But under these reaction conditions toluene reflux then Diels-Alder reaction and retro Diels-Alder reaction

took place in just that, well, as a domino process. And well, finally giving rise to this

interesting anza compound. Anza compounds are those, for instance, having a benzene ring and, well, something like that, as you will see it now attached. So I abbreviate the structure

there. So this was the product which has been observed and just one proof of,

well, I think they have the x-ray structure actually. But spectroscopically you will find

that this vinylic hydrogen, normally giving in the NMR, the proton NMR, a signal around 4.5 or 5 ppm has a, well, significant upfield shift to about 3.5 ppm. This is the anisotropic shift

introduced by this aromatic system where the hydrogen is sitting above the aromatic plane.

So I think this is enough for today. We will go on tomorrow and as part of tomorrow's lessons will be the analysis of this molecule as a target. And if you wish you can

already try it by yourself. Thanks for listening. See you tomorrow.