Making carbon–carbon bonds — Synthesis & Retrosynthesis

Here is the fact that makes synthesis learnable: in a first-year course there are only about eight ways to make a carbon–carbon bond. Functional group interconversions are numerous and mostly interchangeable, but the carbon skeleton can only be built in a few places — so the C–C disconnections are where a synthesis problem is actually decided.

Learn this list as a list. When you are stuck on a target, the question "which of these eight could have made that bond?" is usually enough to find the route.

The list

Reaction	Joins	Gives
Grignard / organolithium + carbonyl	R⁻ + C=O	Alcohol
Grignard + CO₂	R⁻ + CO₂	Carboxylic acid (one carbon longer)
Acetylide + alkyl halide (S_N2)	RC≡C⁻ + R'X	Internal alkyne
Cyanide + alkyl halide (S_N2)	⁻CN + RX	Nitrile (one carbon longer)
Aldol	enolate + C=O	β-hydroxy carbonyl
Claisen	enolate + ester	β-keto ester
Michael (conjugate) addition	enolate + enone	1,5-dicarbonyl
Diels–Alder	diene + dienophile	Cyclohexene (two bonds at once)
Friedel–Crafts alkylation / acylation	arene + R⁺ or RCO⁺	Substituted arene
Wittig	ylide + C=O	Alkene

Ten rather than eight, but several are variations on one idea — an enolate attacking something electrophilic covers the aldol, the Claisen and the Michael.

Counting carbons is the fastest diagnostic

Before you plan anything, count the carbons in the target and in the permitted starting materials. The difference tells you how many C–C bonds you must form and roughly what size pieces you need.

If the target has the same number of carbons as your starting material, you need no C–C bond formation at all — the whole problem is functional group interconversion, which is the next section.
If it has one more carbon, the candidates are a short list: cyanide (then hydrolysis or reduction), a Grignard on formaldehyde, a Grignard on CO₂, or a Wittig with a one-carbon ylide.
If it has several more, look for a disconnection that splits the target into two pieces of roughly the sizes you were given.

The one-carbon extensions are worth memorizing as a set, because "make this molecule one carbon longer" is an extremely common exam instruction and each route ends somewhere different: RX + NaCN gives a nitrile, which hydrolyzes to an acid or reduces to a primary amine; RMgBr + CO₂ gives the acid directly; RMgBr + HCHO gives a primary alcohol.

Two reactions that build rings

Most of the list joins two pieces into a chain. Two are worth separating out because they make rings, which is otherwise hard:

Diels–Alder forms two C–C bonds in one step and delivers a six-membered ring with defined stereochemistry. If a target contains a cyclohexene, try this disconnection first — it is the highest-value single move in the whole list.
Intramolecular aldol and Dieckmann — the same aldol and Claisen chemistry with both partners in one molecule — close five- and six-membered rings. The retron is a 1,4- or 1,5-dicarbonyl relationship in the open-chain precursor.

Where the new bond can go relative to a carbonyl

The same four atoms three times, with the new C–C bond arriving in a different place each time. A nucleophile lands on the carbonyl carbon; an enolate makes the molecule itself the nucleophile and the bond forms next to it; conjugate addition to an enone lands two carbons out.The middle panel is the one that reverses direction, and that is why it is easy to lose: in the other two the arrow points into the carbonyl compound, and in the aldol, Claisen and alkylation it points out of it. Note also that only the third panel is drawn with a C=C — conjugate addition is not an option a plain ketone offers, it is something the conjugation creates.

Enolate chemistry and carbonyl addition between them cover both sides of a carbonyl, and keeping them straight is worth doing explicitly:

A nucleophile attacking the carbonyl carbon builds a bond at that carbon — Grignard, cyanide, hydride.
An enolate is nucleophilic at the α carbon, so it builds a bond next to the carbonyl.
Conjugate addition to an enone builds a bond at the β carbon, two positions away.

So a carbonyl group gives you three different places to attach a new carbon, chosen by reagent and conditions. That is a large part of why carbonyls dominate synthesis.

Worked example — counting first

Target: 1-phenylpropan-1-ol, PhCH(OH)CH₂CH₃, from benzene and any three-carbon or smaller pieces.

Count. Nine carbons: six in the ring, three in the chain. Benzene supplies the six, so exactly one C–C bond joins the ring to the chain.

Disconnect at the carbinol carbon — the Grignard retron. Two options present themselves: PhMgBr + propanal, or CH₃CH₂MgBr + benzaldehyde.

Choose on availability. Take the first. Benzene brominates to bromobenzene (Br₂, FeBr₃), which gives PhMgBr with magnesium; propanal is a permitted three-carbon piece. The second route would need benzaldehyde, which is not something benzene gives you in one straightforward step at this level — a Friedel–Crafts acylation cannot deliver a plain CHO, because formyl chloride is too unstable to use.

The counting step did the real work: it said immediately that only one C–C bond had to be formed, which pointed straight at the carbinol disconnection.

What carries forward

This list is the skeleton of every synthesis question you will meet. Functional group interconversions decorate it, protecting groups keep it from interfering with itself, and the planning section puts the steps in the right order — but the C–C bonds are the decisions, and everything else is bookkeeping around them.