Chapter 28: Retrosynthetic Analysis

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement, not replace, the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Imagine you're faced with this huge challenge.

You need to create a completely new molecule.

Maybe it's a potential drug, you know, something to fight a disease or perhaps a material with some amazing new properties or even just a better greener dye.

It's really not about just sort of randomly mixing chemicals and hoping for the best.

No, definitely not.

That's not chemistry.

That's alchemy, right?

This is about a systematic,

a really creative way to design molecules.

And the powerful,

almost, well, artistic problem -solving tool chemists use for this, it's called retro -synthetic analysis.

Retro -synthetic analysis.

Okay.

And today we're taking a deep dive into those core principles.

Right.

You'd find similar stuff in, say, Chapter 28 of Clayton Greaves and Warren's Organic Chemistry.

Okay.

Our mission really is to show you how chemists don't just know reactions, they design them.

It's like solving a really complex puzzle.

Right.

But instead of starting with a jumble of pieces, you actually start with the finished picture.

The target molecule.

Exactly.

The target.

And you work backward, meticulously, piece by piece, figuring out the most logical way it could have been put together.

And this really matters, doesn't it?

It's not just theory.

Oh, absolutely not.

Think about it.

Almost every big improvement in quality of life over the last, say, 50, 100 years,

medicines,

materials,

it traces back to new molecules.

Molecules designed using these very ideas.

Understanding retrosynthesis is, well, it's fundamental to seeing how chemistry builds our world.

Okay, let's unpack this core idea then.

Thinking backward, how does retrosynthesis actually, you know, change the game for a chemist?

It's a complete shift in perspective, really.

Instead of asking, okay, what happens if I mix A and B?

Yeah, the usual forward approach.

Right.

You start with your target molecule, the thing you absolutely want to make.

That's fixed.

Okay.

Then you systematically dissect it.

You work backwards, step by step, aiming for simple, easy to get starting materials.

It's like running the film in reverse.

So it's not A plus B makes C.

It's more like, hmm, C can be made from A and B.

Exactly.

That's the mindset.

And there's a special way to write that down, a specific notation.

Yes.

We use what's called a retrosynthetic arrow.

It looks different from a normal reaction arrow.

It's open -ended.

Right.

And it signifies a disconnection.

That's the mental process, the imaginary slicing of a bond, often shown with a little wiggly line.

A disconnection.

Right.

Like conceptually undoing a bond.

Precisely.

Could you give us a quick example?

Just something simple.

Sure.

Think about a common insect repellent, maybe an ester.

Okay, ester.

Our chemical intuition, built up from seeing reactions,

tells us that ester bond can be formed by joining an alcohol and, say, an acetyl chloride.

Makes sense.

So retrosynthetically, we draw the ester, put in that special arrow,

and show it disconnecting to an alcohol piece and an acyl chloride piece.

It's like identifying the two key ingredients that could have formed that part.

And often, we even write the reason above the arrow, like CO ester,

just to keep track of our thinking.

That's smart.

Like leaving notes on your own plan.

Now, when you do these disconnections, you're not immediately jumping to specific chemicals, are you?

It's more abstract.

That's where synthons come in.

Exactly right.

Synthons are these idealized fragments.

Think of them as conceptual pieces.

Each one has a sort of inclined polarity, a plus or a minus charge, and they stand in for the real reagents we'll use later in the actual synthesis.

So like a placeholder?

Kind of.

For example, if you're making an amamide, like paracetamol, you know it can come from an amamine in some kind of acetyl group.

That acetyl group synthon, it's electrophilic.

That's the key idea.

Got it.

Now, whether you actually use acetyl chloride or acetic anhydride in the lab,

well, that often depends on practical things.

Like how easy is it to handle the byproducts?

But the synthon in the planning stage lets us focus on the fundamental reactivity the roles these pieces play without getting bogged down in those specifics too early.

So synthons help you map out the chemistry of the break, not the exact lab recipe immediately.

That seems really useful.

OK, so just to quickly recap for you listening,

the target molecule, TM, that's your goal.

Retrosynthetic analysis is the whole backward thinking process.

The retrosynthetic arrow is the symbol for that backward step.

A disconnection is that imaginary bond break.

Synthons are the idealized pieces.

A conceptual fragment.

And reagents are the actual chemicals you'd grab off the shelf.

Exactly.

Now here's the tricky part.

Oh, it's always the tricky part.

Well, choosing where to make that disconnection, that's often the hardest bit.

It requires intuition, experience, and creativity.

But thankfully, there are guidelines.

OK, lay it on us.

Guideline one.

How do you decide where to cut?

Guideline one.

Disconnections must correspond to known reliable reactions.

This is like the golden rule.

Non -negotiable.

Absolutely.

Every backward step you draw has to represent a forward reaction that actually works well.

For instance, making an ether.

You disconnect the CO bond because we know reliably the ethers can be formed from an alkoxides and an alkyl halide.

Williams and Nether synthesis, right?

Classic reaction.

Exactly.

You wouldn't disconnect it in a way that suggests, say, a nucleophile attacking a plain, unactivated benzene ring to make the ether bond.

Because that just doesn't really work.

Right.

It's about staying grounded in real chemistry.

Which flows nicely into guideline two.

For compounds made of two parts joined by a heteroatom like oxygen, nitrogen, sulfur disconnect next to that heteroatom.

OK, why there specifically?

Well, those spots, esters, amides, ethers, sulfides.

They're often formed through reactions at the heteroatom or right next to it.

Think substitutions, condensations.

Ah, makes sense.

So if you've seen an amide in your target, like in a melpholide, your first thought is usually break the CN bond.

Or for an ether, break the CO bond.

It points you toward common, reliable synthetic methods.

OK, that covers breaking one bond near a helpful atom.

But what if the molecule's more complex?

Multiple reactive spots.

I can see how that might cause chemoselectivity problems.

Is that the term?

That's exactly the term.

You've got it.

Where a reagent might react at the wrong place.

So how do you navigate that?

Which bond do you break first?

That's where the strategy really comes in.

And it brings us to guideline three.

Always consider alternative disconnections.

And try to choose routes that avoid those chemoselectivity issues.

OK.

And often, maybe even usually, this means disconnecting the most reactive groups first.

Get the tricky bits out of the way early.

It's sort of.

So you have to think ahead.

Imagine a molecule with several ethers and maybe an amann.

If you disconnect an ether bond too early, your forward synthesis might involve trying to say, alkylase something less reactive in the presence of that amann, which is probably more reactive.

Leading to side reactions.

A mess.

Exactly.

Yeah.

So the smarter move might be to deal with that, or you mean disconnection first.

Or protect it somehow, simplifying the molecule and preventing those unwanted reactions later on.

It's about understanding the molecule's reactivity profile.

So it's like chemical chest thinking moves ahead.

And sometimes, you don't actually break a bond initially.

You just swap one functional group for another to make things easier.

That's functional group inner conversion, FGI.

Precisely.

FGI isn't a disconnection itself.

It's a transformation.

You convert group A into group B because group B is easier to disconnect.

Or maybe group A would cause problems later.

Like a chemical disguise.

That's a good way to put it.

It's incredibly useful when a group is too reactive, not reactive enough, or just doesn't fit a standard disconnection pattern.

Making amines is a classic case for FGI.

Right.

You mentioned those earlier.

Yeah.

If you try to make, say, a secondary amine by just adding an alkyl halide to a primary amine, well, good luck stopping there.

Keeps going.

Often.

Yes.

The secondary amine you just made can react again, giving you a tertiary amine.

And that can even react again to give a quaternary salt.

Because the product amine is often more nucleophilic, more reactive than what you started with.

Ugh.

Sounds messy.

So, what's the FGI workaround for amines?

One really common strategy is amide reduction.

Retro -synthetically, you think, OK, this targeted amine could have come from an amide.

OK.

Amines are generally less reactive, more stable than amines, and the CN -amide bond disconnection is very reliable.

Ah.

So you disconnect the amide bond in your plant.

Exactly.

Then, in the forward synthesis, you make the amide, and then you reduce it using something strong like LiOH4, lithium aluminum hydride, or maybe borane back to the amine you wanted.

Clever.

Control the reactivity.

It gives you precise control over where that nitrogen goes, avoiding the over -autolation problem.

Another really powerful FTI for amines is reductive amination.

Reductive amination.

OK.

Here, the amine is thought of as coming from an amine.

Retro -synthetically, you convert the amine to an amine.

Which is like a CN double bond.

Exactly.

And that amine can be disconnected back to the components, usually an amine plus an aldehyde or a ketone.

Ah.

OK.

Then, in the lab, you form that amine or related species like an oxime and reduce it, maybe with sodium borohydride or catalytic hydrogenation.

This is huge in pharma, making things like oxfentanyl, fin -fluoramine,

lots of drugs rely on this.

It's controlled and can even be stereoselective sometimes.

That's really elegant.

Using the intermediate forms, like amines or amines, just to manage reactivity, it makes me think, if you can transform groups and break single bonds, can you sometimes break two bonds in one conceptual step to simplify things even faster?

Absolutely.

And that leads us nicely to guideline four.

Use two -group disconnections whenever you can.

Two -group disconnections?

Why are they so good?

They're highly efficient.

Basically, one functional group actively helps in the disconnection related to another nearby functional group.

It lets you break down bigger chunks of the molecule in a single logical step that corresponds to a powerful known reaction.

Okay, so what are some common examples?

We often see them in well -vary -two disconnections, where you have two functional groups right next door to each other.

Like one and two on the carbon chain.

Exactly.

A really common strategy here involves using epoxides as A2 synthons.

Remember synthons.

Right, the idealized fragments.

A2 means acceptor at position two.

Precisely.

So if your target has, say, an alcohol next to an ether or an amine.

A 12 -vary -two relationship.

Yes.

You can often disconnect that whole unit back to maybe an amine nucleophile and an epoxide electrophile.

Ah, the epoxide ring opening.

Exactly.

And a neat thing about epoxides is that after they react, they form a hydroxyl group.

This usually makes a product less nucleophilic, so the reaction tends to stop cleanly after one addition.

No over -alkylation mess.

Ah, it's Andy.

Very.

It's used all the time, like you're making beta blockers.

Propranolol is a famous example.

Then you also have things like alpha -halo -carbonyl compounds used as A2 synthons.

So a halogen next to a carbonyl.

Disconnecting those leads back to using reagents like alpha -chloroketones in the forward synthesis.

And then there are 1 -carbo -3 disconnections.

These correspond to the reverse of conjugate additions or Michael reactions.

Okay, so groups separated by one carbon now.

Yes, in a 1 -phyl -3 relationship.

Think of a target molecule with, say, a sulfide group and maybe an ester three carbons away.

You can disconnect that using a 1 -phyl -3 to AX disconnection, which suggests forming it via a theyl, adding to an alpha -beta unsaturated ester in the forward direction, Michael addition.

It's a fantastic way to build C -C bonds and set up functionality simultaneously.

So whether it's breaking C -X bonds or building C -C bonds, it always comes back to recognizing these patterns, right?

Based on reactions, we know work.

That's the core idea.

And speaking of C -C bonds, building the carbon skeleton, that's really the heart of synthesis, isn't it?

Oh, absolutely.

Forming carbon bonds is, you could argue, the most fundamental challenge and goal in organic synthesis.

We apply the same retrosynthetic logic.

How does that work for C -C bonds?

Well, one very useful method is alkyne alkylation.

If you see a triple bond in your target, you can often disconnect right next to it.

That retrosynthetic step implies using in a seed -lead anion reacting with an alkyl halide in the forward synthesis.

And alkanes are useful building blocks.

Incredibly versatile, because you can later reduce that triple bond selectively to get either a cis -z double bond or a trans -z double bond, or even reduce it all the way to a single bond, gives you lots of options.

And sometimes if your target just has an isolated double bond, you might even use FGI to an alkyne.

So mentally turn the double bond into a triple bond?

Yeah, it's just in the retrosynthesis plan, because that alkene might reveal a much easier C -C disconnection point.

It's about being flexible in how you look at the molecule.

Then we have 12 -criste C -C disconnections, typically involving alkyl groups next to carbonyls.

Think of an ethyl group right beside an ester carbonyl.

Retrosynthetically you can disconnect that ethyl group.

This suggests adding it via alkylation of the ester enolate in the forward synthesis.

Using those enolate equivalents you mentioned, like malonates.

Exactly.

Things like melonic esters or acetoacetic esters are great starting points, because they form enolates easily and reliably, making those C -C bond formations efficient for building up the carbon framework.

And you mentioned earlier that oxygen groups often help with C -C bond making.

Yes, that's our guideline five.

Convert to oxygen -based functional groups to facilitate C -C disconnections.

Why oxygen specifically?

Because alcohols, aldehydes, ketones, esters, acids,

they are incredibly interconvertible through oxidation and reduction.

And crucially, many of the most powerful C -C bond forming reactions directly involve them.

Think Grignard reactions, Alibol reactions, Wittig reactions.

They nearly all involve carbonyls or alcohols.

So working backward to an oxygen functional group often unlocks the best C -C forming strategies.

And finally, related to that, we have 11 -1 C -C disconnections.

This is classic Grignard chemistry.

Adding to carbonyls.

Exactly.

If you see an alcohol in your target molecule, especially a secondary or tertiary one, you disconnect one of the C -C bonds right next to the carbon -bearing the OH group.

That disconnection points directly to adding a Grignard regent or organolithium to an aldehyde, ketone, or even an ester in the forward synthesis.

It's probably the most common way to build carbon chains and install hydroxyl groups.

Super powerful.

So we've got all these strategies, FGI, different disconnection types,

but how does a chemist actually know what starting materials are, you know, realistic to start from?

You can't just disconnect back to anything, right?

That's a very practical and important question.

You need to aim for things you can actually buy or make very easily.

Is there a rule of thumb?

Yeah, a rough rule of thumb is that simpler compounds are usually available.

Think up to about six carbons or so with maybe one common functional group, alcohol,

halide, simple aldehyde, ketone.

Simple straight chains might be available up to eight carbons or more.

Basic rings like cyclopentane, cyclohexene derivatives are common too.

And what about things with two functional groups?

Some common ones are definitely available, diavolmalinate, ethylacetoacetate, simple acrylates.

These are standard building blocks.

But beyond that...

You have to check.

You absolutely have to check.

Consulting chemical supplier catalogs, Sigma -Aldrich, Fisher, et cetera, is a routine part of planning any real synthesis.

You need to know what your actual starting points can be.

Makes total sense.

Knowing your starting materials shapes the whole plan.

Now earlier you mentioned classifying synthons using donor and acceptor labels.

Can you expand on that?

Yes.

It's another layer of thinking that helps predict reactivity.

We label negatively polarized synthons, the electron -rich ones, as donor D...

Like nucleophiles.

Essentially, yes.

And positively polarized electron -poor ones as acceptor A.

Like electrophiles.

Right.

And we add a number to show where that reactivity is relative to a functional group.

So an aldehyde carbonyl carbon is electron -poor right at the functional group.

So it's an A1 synthon.

And enolate, though, is electron -rich, a donor, but the reactive carbon is two atoms away from the carbonyl oxygen.

So it's a D2 synthon.

Donor at position two.

Got it.

Why is this labeling useful?

Because it highlights natural reactivity.

Carbonyl compounds, for instance, naturally have this alternating pattern.

A1 at the carbonyl carbon, then D2 at the alpha carbon via enolate,

then potentially A3 at the beta carbon if it's unsaturated.

Ah.

Alternating acceptor, donor, acceptor.

Exactly.

This natural alternation makes synthesizing compounds with functional groups in a 1 ,3 or 1 ,3y relationship often much easier.

They fit the pattern.

Like the aldol reaction.

The aldol reaction is the perfect example.

It connects an enolate D2 with a carbonyl A1 to make a beta hydroxy carbonyl, which has its key oxygens in a 1 ,3 ,3 relationship.

Fits the natural pattern beautifully.

It's amazing how these patterns emerge, but what if you need to make something that doesn't fit, like a 1 ,2 or 1 ,3 ,4 relationship?

They go against that natural alternating pattern.

Now you're talking about umpolung.

Umpolung.

German term.

Yes.

It means polarity inversion or reversed polarity.

It's about finding clever ways to make synthons behave in an unnatural way.

Flipping the polarity.

Essentially.

We need ways to generate synthons, like D1, donor at position 1, where it should be an acceptor like a carbonyl, or A2, acceptor at position 2, where it should be a donor like an enolate.

Those sound tricky.

How do you do that?

Well, cyanide ion, CN, is a classic example of a reagent that acts as a D1 synthon.

It's a nucleophile that attacks carbonals.

We already mentioned epoxides acting as A2 synthons.

These impaling strategies are key for accessing those 1 ,2 and 1 ,3 ,4 functional compounds that don't fit the simple alternating pattern.

It really expands the synthetic toolkit.

So pulling it all together, you can see that retrosynthesis isn't about finding the one true path right away.

No, it seems more exploratory.

Exactly.

It's a powerful method for brainstorming multiple potential strategies for making a target molecule.

You lay out several possible routes.

And then?

Then the refinement begins.

You hit the literature, see what similar things other chemists have done, what worked, what didn't.

You think about practicalities, cost of reagents, safety, yield.

And ultimately you head into the line and start experimenting.

It's this blend of logical planning and hands -on work.

And this whole way of thinking, this systematic backward planning, it's behind everything from, you know, fairly simple fragrance molecules to incredibly complex drugs or new agrochemicals.

Absolutely.

It really hammers home that fundamental idea.

Organic reactions are about electrophiles meeting nucleophiles.

Synthesis and understanding reactions are just two sides of the very same coin.

Hopefully this deep dive has given you a feel for that powerful framework.

You've really seen now how chemists take on this huge task of building molecules.

It's systematic, it's logical, working backward from the goal to find reliable paths.

That mechanistic thinking, understanding how electrons move, the reaction pathways, that's really the essence of organic chemistry.

And what's truly fascinating here is how this approach, this retrosynthetic mindset of working backward from where you want to end up, it isn't just for chemists.

Think about it.

Where else in your life could you apply that?

A big project, a complex problem, maybe even a personal goal.

You start with the end in mind, define that target, and then work backward, breaking it down into manageable steps.

Well, you might just discover your own pathway to synthesis.