The Robinson Annulation

What the Robinson annulation does

Annulation means "ring-building." The Robinson annulation is the classic way to bolt a fresh six-membered carbocycle onto a molecule that already has a ketone — and it does so by welding together two reactions you already know: a Michael addition followed by an intramolecular aldol condensation. Both are base-catalyzed enolate chemistry, so the whole thing can run in a single flask.

The three moving parts of a textbook Robinson annulation are:

• The nucleophile (Michael donor). A ketone with an acidic α-hydrogen — typically a cyclohexanone, a 1,3-diketone, or a β-ketoester. Base pulls off the α-proton to make an enolate.
• The Michael acceptor. An α,β-unsaturated ketone, almost always methyl vinyl ketone (MVK, CH₂=CH-CO-CH₃). Its alkene is the electrophile; its methyl ketone becomes the second carbonyl.
• The base. Hydroxide, alkoxide, an amine, or the chiral organocatalyst L-proline. It deprotonates twice — once for the Michael step, once for the aldol closure.

The net result is a cyclohex-2-enone: a six-membered ring carrying an α,β-unsaturated ketone (an enone). That enone is a gift — it's a handle for further chemistry (conjugate additions, reductions, more annulations), which is why the reaction is so central to building polycyclic frameworks like steroids and terpenoids.

The mechanism, arrow by arrow

Follow the electrons. There are three phases: Michael addition, intramolecular aldol addition, then E1cb dehydration.

Phase 1 — Michael addition (builds a 1,5-diketone).

1. Enolize the donor. Base removes an α-hydrogen from the ketone. The C-H bonding electrons flow to make an enolate; the negative charge is delocalized onto oxygen.
2. 1,4-conjugate addition. The nucleophilic α-carbon of the enolate attacks the β-carbon (the terminal CH₂) of MVK. The alkene π electrons shift onto the MVK carbonyl, generating a new enolate on the acceptor.
3. Proton transfer. That enolate picks up a proton (from solvent or the conjugate acid of the base). You now have a 1,5-diketone — two C=O groups separated by three carbons. This spacing is the whole trick: it is exactly right to close a six-membered ring.

Phase 2 — intramolecular aldol addition (closes the ring).

4. Enolize the other side. Base removes an α-hydrogen on the far side of the 1,5-diketone, generating a new enolate.
5. Ring closure. That enolate carbon reaches across the molecule and attacks the other carbonyl carbon intramolecularly. The C=O π electrons collapse onto oxygen as an alkoxide. A new C-C bond forms — and with it, the six-membered ring. The product at this stage is a β-hydroxy ketone (an aldol).

Phase 3 — dehydration (delivers the conjugated enone).

6. Enolize α to the remaining ketone. Base removes the α-hydrogen between the ketone and the newly formed C-OH.
7. E1cb elimination. The enolate expels hydroxide as a leaving group (this is an E1cb, not E2 — the carbanion/enolate forms first, then loses OH⁻). A C=C double bond appears, in conjugation with the ketone.
8. Thermodynamic sink. The final α,β-unsaturated ketone (cyclohex-2-enone) is conjugated and low-energy. Losing water is what makes the whole cascade irreversible and drives it forward.

  MICHAEL:   R₂C=C(O⁻)–  +  CH₂=CH–CO–CH₃   →   1,5-diketone
             (enolate)        (MVK acceptor)      (O=C···C···C···C=O)

  ALDOL:     enolize one carbonyl, its α-C attacks the OTHER carbonyl
             →  β-hydroxy ketone with a NEW 6-membered ring

  DEHYDRATE: E1cb loss of H₂O  →  cyclohex-2-enone (conjugated enone)

Every bond made is a C-C bond formed by an enolate attacking an electrophilic carbon — first an alkene (conjugate addition), then a carbonyl (aldol). That symmetry of logic is why one base and one flask can do all of it.

Reagents, catalyst, and real conditions

The classic laboratory recipe is deliberately staged, because the two halves want different conditions.

• Michael step — mild and often cold. Catalytic base: KOH or NaOH in EtOH/MeOH, sodium ethoxide, or a secondary amine (pyrrolidine, piperidine) that forms an enamine donor. Run at 0 °C to room temperature. Mildness matters: MVK polymerizes under strong base, and you want clean mono-addition.
• Aldol + dehydration — hotter, stronger. Add more base (KOH, NaOEt, or NaOH) and heat, often to reflux. The dehydration is the step that needs the push; heat drives off water and locks in the conjugated enone.
• MVK generated in situ. Because MVK is volatile, toxic, and polymerization-prone, it is frequently released slowly from a stable precursor: a Mannich base such as 4-(diethylamino)butan-2-one, used as its methiodide salt. Under the reaction base, the quaternary ammonium undergoes a Hofmann-type β-elimination to release MVK exactly as fast as it's consumed — a slow-release trick that suppresses polymerization.
• Enantioselective variant. Swap the achiral base for L-proline (typically 3–35 mol%) in the aldol step. Proline forms an enamine and steers the ring closure through a chair-like transition state with a hydrogen-bonded carboxylic acid, delivering one enantiomer of the enone. This is the Hajos-Parrish / Eder-Sauer-Wiechert asymmetric annulation.

A representative achiral procedure: dissolve the 1,3-diketone or cyclohexanone in ethanol, add catalytic KOH, add MVK (or its Mannich precursor) dropwise at 0 °C, stir; then add more KOH and reflux 2–12 h; acidic workup and recrystallization give the fused enone in 50–85% yield.

Scope, selectivity, and stereochemistry

• Donor scope. Any carbonyl with an acidic α-H: cyclohexanones, cyclopentanones, 1,3-diketones (very common — the flanking carbonyls acidify the central C-H to pKa ~5–11 and give clean regiochemistry), β-ketoesters, and even nitroalkanes as surrogate donors.
• Acceptor scope. MVK is the standard; ethyl vinyl ketone, methyleneketones, and other enones extend the ring substitution pattern. Substituted acceptors let you install substituents on the new ring.
• Regioselectivity of the ring. Because the 1,5-diketone geometry dictates a 6-membered closure, you almost always get a cyclohexenone. Which enolate closes (thermodynamic vs kinetic) can be steered by base and temperature; thermodynamic enolate control usually predominates under the equilibrating basic conditions.
• Stereochemistry. The Michael step can set a stereocenter; the aldol closure sets ring-fusion geometry (cis vs trans decalin-type junctions). With L-proline, the asymmetric aldol delivers high enantiomeric excess (often 90%+ ee in the Hajos-Parrish product), which is why the reaction anchors so many steroid syntheses that must be single-enantiomer.

Robinson annulation vs related ring-forming methods

	Robinson annulation	Diels-Alder	Simple intramolecular aldol
Bond-forming logic	Stepwise ionic: Michael then aldol	Concerted [4+2] cycloaddition	Single intramolecular enolate + carbonyl
New C-C bonds	Two (one per step)	Two (simultaneously)	One
Reagents	Enolizable ketone + enone (MVK) + base	Diene + dienophile	A dicarbonyl within one molecule
Ring made	Cyclohex-2-enone (carbonyl built in)	Cyclohexene	Any ring size the chain allows
Driving force	Dehydration to conjugated enone	Aromaticity / two σ bonds from two π	Dehydration (if it condenses)
Stereocontrol	Enolate geometry; L-proline for ee	Endo rule, stereospecific suprafacial	Enolate + transition-state geometry
Flagship application	Steroid & terpenoid ring systems	Terpene & polycyclic frameworks	Ring closures in synthesis

Worked example: the Wieland-Miescher ketone

The most famous single application is the synthesis of the Wieland-Miescher ketone, a bicyclic enedione that is the standard chiral starting material for steroids and terpenoids.

  2-methyl-1,3-cyclohexanedione  +  methyl vinyl ketone
        │  (Michael donor)              (acceptor)
        │  base
        ▼
  triketone (1,5-relationship set up for closure)
        │  L-proline (cat.), then acid
        ▼
  Wieland-Miescher ketone  (bicyclic enedione, single enantiomer)

• Step 1 — Michael. 2-Methyl-1,3-cyclohexanedione (the flanking carbonyls make its central C-H very acidic) adds to MVK under mild base, giving a triketone.
• Step 2 — asymmetric aldol/dehydration. Catalytic L-proline (~3 mol%) forms an enamine and closes the ring enantioselectively; acidic workup dehydrates to the enone. Typical results: ~70–90% yield and 90%+ ee.
• Why it matters. The Wieland-Miescher ketone contains the pre-built A/B ring junction of a steroid with defined stereochemistry, so a few more steps append the C and D rings. It is a launchpad in dozens of published total syntheses.

The proline-catalyzed version — reported by Hajos and Parrish (Hoffmann-La Roche) and independently by Eder, Sauer, and Wiechert (Schering) in 1971 — is one of the founding results of modern asymmetric organocatalysis, decades before the field had that name.

Limitations and side reactions

• MVK polymerization. Methyl vinyl ketone loves to polymerize under base. Excess base or too-fast addition gives dark, tarry oligomers instead of the annulation product. The in-situ Mannich-base slow-release trick is the standard fix.
• Double Michael addition. A reactive donor can add MVK twice, giving a bis-adduct that can't cyclize cleanly. Controlled stoichiometry and mild base curb this.
• Retro-Michael. The Michael step is reversible; under harsh conditions the 1,5-diketone can fragment back. This is usually harmless (the forward aldol/dehydration is the thermodynamic sink) but can erode yield if the aldol is slow.
• Regiochemical ambiguity. An unsymmetrical ketone can enolize on either side, giving competing annulation products. 1,3-Dicarbonyls sidestep this because the acidic site is unambiguous; that's why they are so popular as donors.
• Wrong-ring closures. If the tether length is off (e.g., a wrong-length acceptor), 5- or 7-membered closures compete; matching MVK to a ketone donor is what guarantees the six-membered ring.

History: who and when

The reaction is named for Sir Robert Robinson, the English chemist who reported the sequence in 1935 (with W. S. Rapson) as a route to fused ring systems. Robinson had been building the toolkit of alkaloid and steroid synthesis for years, and this ring-building cascade fit his program of assembling complex polycyclic natural products from simple ketones. He won the 1947 Nobel Prize in Chemistry for his work on plant alkaloids (notably the structure of morphine and the synthesis of tropinone) — the same style of "biomimetic" one-pot ring assembly that the annulation exemplifies.

The reaction's second act came in 1971, when two industrial groups independently found that L-proline could make the aldol step enantioselective — the Hajos-Parrish-Eder-Sauer-Wiechert reaction. That result sat quietly for thirty years until the early-2000s renaissance of organocatalysis (List, Barbas, MacMillan) recognized it as the prototype of enamine catalysis. So a 1935 ring-building trick became a founding experiment of a Nobel-winning field (organocatalysis, Nobel Prize 2021, List and MacMillan).

Industrial and synthetic notes

• Steroids. Nearly every classical steroid total synthesis (cortisone, estrone, testosterone frameworks) uses a Robinson annulation somewhere to install one of the six-membered rings of the steroid nucleus. The Wieland-Miescher ketone is the canonical entry point.
• Terpenoids and natural products. Sesquiterpenes, diterpenes, and other polycyclic terpenoids frequently rely on annulation to build their fused decalin cores.
• A teaching keystone. Because it chains two of the most important enolate reactions, the Robinson annulation is a standard capstone in undergraduate organic chemistry — it forces you to track enolate formation, conjugate addition, aldol closure, and E1cb dehydration all in one problem.
• Safety. MVK is a potent lachrymator and skin/respiratory irritant and is acutely toxic — handle in a fume hood, and prefer the in-situ generation route both for safety and for yield. Strong base and reflux in alcohol solvents are the usual hazards of the ring-closure step.