The Neber Rearrangement

What the Neber rearrangement does

Take a ketone, turn it into an oxime, tosylate the oxime oxygen, and hit it with base. Instead of the amide you'd get from a Beckmann, you get an α-amino ketone — an amine and a carbonyl on adjacent carbons. The whole trick is that a leaving group parked on nitrogen gets displaced by a carbanion generated on the carbon next door, and the only way those two atoms can bond is by closing a three-membered ring.

The reaction runs in three conceptual moves:

1. Activate the oxime. The oxime OH is a poor leaving group, so it is converted to an O-tosylate (or mesylate, or the classic picryl/2,4-dinitrophenyl ether). Now the nitrogen carries a superb leaving group: ⁻OTs.
2. Deprotonate and close the ring. Base removes an α-proton — the C–H on the carbon adjacent to the C=N. The resulting carbanion swings onto nitrogen, kicking out tosylate and forming a strained 2H-azirine (a three-membered ring with an internal C=N).
3. Hydrolyze. Aqueous acid adds water across the azirine's C=N, opening the ring to an α-amino ketone. If you stop before this step, you can isolate the azirine itself.

    R-C(=N-OTs)-CH₂-R'   ──base──▶   [2H-azirine]   ──H₂O/H⁺──▶   R-C(=O)-CH(NH₂)-R'
       oxime tosylate                strained ring              α-amino ketone

The α-amino ketone is the workhorse product — it is a precursor to α-amino acids, to substituted amines, and (via cyclization) to nitrogen heterocycles like pyrazines and imidazoles.

The mechanism, arrow by arrow

Every step is a two-electron polar process. There are no radicals and no carbocations — this is a base-driven, carbanion-mediated cyclization, which is exactly why it behaves so differently from the acid-driven Beckmann.

1. α-deprotonation. The base (ethoxide) removes a proton from the α-carbon. That α-C–H is weakly acidic because the resulting carbanion is stabilized by the adjacent C=N (it is essentially a nitrogen-analog of an enolate — an "aza-allyl" or metalated ketimine anion). Aryl and other anion-stabilizing groups on the α-carbon make this step easy; a benzylic α-C–H is ideal.
2. Intramolecular S_N2 on nitrogen. The carbanion lone pair attacks the sp²-nitrogen from the back side of the N–O bond. The tosylate departs as ⁻OTs. Carbon and nitrogen are now bonded; combined with the pre-existing C=N, this forms the three-membered 2H-azirine ring. This is the rate- and selectivity-determining event, and it is the point of highest ring strain (~40 kcal/mol of strain energy in the azirine).
3. Hydration of the azirine. The azirine's C=N is a strained, electrophilic imine. Water (or hydroxide) adds to the ring carbon; proton transfers and ring-opening give a 2-amino ketone. Net, the imine nitrogen becomes a primary amine and the imine carbon becomes a carbonyl.

  step 1 (deprotonate):
        R–C(=N–OTs)–CH₂R'   +   EtO⁻   →   R–C(=N–OTs)–CH⁻R'   +   EtOH

  step 2 (ring-close, lose OTs⁻):          N
                                       ╱ ‖
        R–C(=N–OTs)–CH⁻R'   →       R–C──CH–R'      +   TsO⁻
                                     (2H-azirine: C=N double bond, C–C single bond)

  step 3 (hydrolyze C═N):
        azirine   +   H₂O/H⁺   →   R–C(=O)–CH(NH₂)–R'

Notice what is not happening: no group migrates from carbon to nitrogen (that's Beckmann), and no external nucleophile forms the ring (it's intramolecular). The carbon skeleton is fully preserved — the same carbons that were in the ketone end up in the amino ketone.

Regiochemistry and stereochemistry

The defining feature of the Neber is where the selectivity is decided. For an unsymmetrical ketoxime with α-protons on both sides of the C=N, the base removes the proton that gives the more stabilized carbanion — typically the more acidic, more accessible, benzylic or otherwise anion-stabilized position. That carbon becomes the amino-bearing carbon of the product.

• Not oxime-geometry-controlled. The Beckmann marches to a single product dictated by which group is anti to the leaving oxygen; mix your E/Z isomers and you get a mixture of amides. The Neber ignores oxime geometry: both the E- and Z-oxime tosylate deprotonate to give the same aza-allyl anion, which closes to the same azirine. You get one α-amino ketone regardless of the starting oxime E/Z ratio.
• Stereochemistry at the new stereocenter. The α-carbon becomes a stereocenter in the product (it now bears H, NH₂, R', and C=O). Under standard achiral conditions the amino ketone is racemic, because the planar α-carbanion closes the ring onto nitrogen from either face with equal probability. Asymmetric variants using chiral bases or chiral phase-transfer catalysts can render the ring closure enantioselective, delivering enantioenriched azirines and amino ketones.
• Chemoselectivity. The reaction needs an α-C–H. An oxime with no α-hydrogen (e.g., from a diaryl ketone, or with a fully substituted α-carbon) simply cannot form the azirine and instead undergoes competing Beckmann fragmentation or elimination.

Reagents, activators, and conditions

The Neber is a two-operation sequence — activate, then rearrange — and each has established options:

• Oxime activation. p-Toluenesulfonyl chloride (TsCl) with pyridine gives the O-tosylate — the classic choice. Methanesulfonyl chloride (MsCl) gives the more reactive O-mesylate. Neber's original substrates used O-(2,4-dinitrophenyl) and O-picryl oximes, which have very good leaving groups. Quaternized versions (O-acyl, O-phosphinyl) are also used.
• Base. Potassium ethoxide (KOEt) or sodium ethoxide (NaOEt) in ethanol is the textbook base and works for benzylic substrates. For milder or more sensitive systems, pyridine, triethylamine, DBU, or aqueous carbonate suffice — the more acidic the α-proton, the weaker the base you can use.
• Solvent and temperature. Ethanol or methanol at reflux is common; ring closure typically proceeds between room temperature and 80 °C. The azirine forms in minutes to hours.
• Hydrolysis. A dilute aqueous acid workup (0.5–2 M HCl, or simply water) hydrolyzes the azirine to the amino ketone. To keep the azirine, quench cold and avoid protic acid.

Neber vs Beckmann vs related oxime chemistry

	Neber rearrangement	Beckmann rearrangement	Beckmann fragmentation
Promoter	Base (alkoxide, amine)	Acid / Lewis acid (H₂SO₄, PCl₅, P₂O₅)	Acid, but with a stabilized α-cation/EWG
Key intermediate	2H-azirine (C=N 3-ring)	Nitrilium ion (linear R–C≡N⁺–R')	Carbocation + nitrile fragment
Bond changes	New C–N ring, C skeleton kept	1,2-alkyl/aryl shift C→N	C–C bond breaks into two fragments
Product	α-amino ketone	Amide (or lactam from cyclic oxime)	Nitrile + carbonyl/alkene
Oxime E/Z dependence	None — both isomers converge	Strict — anti group migrates	Strict — anti bond fragments
Selectivity set by	Most acidic α-proton	Which group is anti to OTs	Which C–C bond is anti to OTs
Requires α-C–H?	Yes — mandatory	No	No (needs a cation-stabilizing α)
Named for	P. W. Neber (1926)	E. Beckmann (1886)	Beckmann (variant)
Industrial flagship	Azirine / amino-acid building blocks	Caprolactam → nylon-6	Ring-opening degradations

The single most testable distinction: point an activated oxime at acid and it migrates to an amide (Beckmann); point the same substrate at base and it cyclizes to an amino ketone (Neber). Same starting functional group, opposite promoter, entirely different product.

Worked example: propiophenone oxime tosylate

Start from propiophenone (PhCOCH₂CH₃), the aryl ethyl ketone. Convert to its oxime and tosylate:

  Ph–C(=O)–CH₂CH₃
      │  NH₂OH·HCl, NaOAc          (make the oxime)
      ▼
  Ph–C(=N–OH)–CH₂CH₃
      │  TsCl, pyridine, 0 °C       (activate as the O-tosylate)
      ▼
  Ph–C(=N–OTs)–CH₂CH₃
      │  KOEt, EtOH, reflux         (α-deprotonate → close azirine)
      ▼
  2-methyl-3-phenyl-2H-azirine      (the strained intermediate)
      │  H₃O⁺ (dilute HCl, H₂O)      (hydrolyze the C=N)
      ▼
  Ph–C(=O)–CH(NH₂)–CH₃             (α-amino ketone: 2-amino-1-phenylpropan-1-one)

• Regiochemistry. Only the methylene of the ethyl group carries an α-C–H (the phenyl is on the imine carbon and has no α-hydrogen), so deprotonation and ring closure happen there and the amino group lands on that carbon. The product is 2-amino-1-phenyl-1-propanone (the amino analog of the corresponding propiophenone), also called α-aminopropiophenone.
• Why base, not acid. Run the same tosylate with H₂SO₄ and you'd get the Beckmann product (an N-phenyl or N-alkyl amide, depending on geometry). The base flips the whole outcome by generating a carbanion instead of a nitrilium ion.
• Yield. Aryl alkyl systems like this are the Neber's sweet spot — the aryl group on the imine carbon stabilizes the 3-aryl-2H-azirine — and isolated yields of 50–75% of the amino ketone are typical, with the azirine isolable in higher yield if hydrolysis is skipped.

This exact motif — an aryl α-amino ketone — is the carbon skeleton of the cathinone family of stimulants and of many pharmaceutical amines, which is one reason the Neber and its azirine intermediates keep showing up in synthesis.

Applications and the azirine payoff

• 2H-azirines as building blocks. The Neber is one of the classic ways to make 2H-azirines, the smallest unsaturated nitrogen heterocycle. These strained rings are springboards: they undergo ring-expansion, cycloaddition, and photochemical nitrile-ylide chemistry to build pyrroles, oxazoles, imidazoles, isoxazoles, and pyrazines.
• α-Amino ketones and α-amino acids. Hydrolysis gives α-amino ketones; further reduction or oxidation ladders these to 1,2-amino alcohols and α-amino acids. This is a non-Strecker entry into α-amino carbonyl chemistry.
• Nitrogen heterocycle synthesis. α-Amino ketones self-condense or condense with 1,2-dicarbonyls to give pyrazines and imidazoles — the Neber feeds these condensations with a clean, single-regioisomer amino ketone.
• Asymmetric catalysis showcase. The intramolecular Neber has become a testbed for enantioselective C–N bond formation: chiral phase-transfer catalysts and chiral Brønsted bases convert prochiral oxime derivatives into enantioenriched 2H-azirines, a reaction highlighted in modern asymmetric-synthesis reviews.

Limitations and side reactions

• No α-C–H, no reaction. The azirine can only close if there is a proton on the α-carbon to remove. Fully substituted α-carbons and oximes of diaryl ketones are dead ends for the Neber.
• Competing Beckmann/elimination. Under the wrong balance of base strength and temperature, the activated oxime can fragment (Beckmann-type C–C cleavage to a nitrile) or eliminate to an α,β-unsaturated nitrile/oxime instead of cyclizing. Electron-poor substrates that stabilize a cation lean toward fragmentation.
• Base-sensitive products. α-Amino ketones are prone to self-aldol condensation and to air oxidation; strongly basic conditions plus a reactive amino ketone can erode yield. Buffered or milder bases mitigate this.
• Azirine fragility. 2H-azirines are strained and can polymerize or ring-open under forcing acid or heat. If the azirine is the target, keep conditions mild and workup cold.
• Racemization. The classic (achiral) reaction gives racemic amino ketone; if a single enantiomer is needed, an asymmetric catalytic variant is required rather than the base-only protocol.

Discovery and mechanistic history

Peter Wilhelm Neber reported the transformation in 1926 in Germany. He was studying what happens when ketoxime p-toluenesulfonates (and the analogous picryl and 2,4-dinitrophenyl oxime ethers) meet base — conditions deliberately unlike the acidic media that Ernst Beckmann had used four decades earlier (1886) to make amides. Neber found that base gave a completely different outcome: an α-amino ketone.

To explain it, Neber proposed a strained three-membered nitrogen ring — a 2H-azirine — as the key intermediate, an audacious suggestion for the 1920s given how little strained-ring chemistry was understood. The proposal held up. Later workers isolated stable 2H-azirines when hydrolysis was avoided, characterized them spectroscopically, and confirmed the base-promoted, carbanion-mediated ring closure. The transformation has carried Neber's name ever since, and the "Neber-type" ring closure is now a standard disconnection in azirine and α-amino-carbonyl synthesis.

Practical and safety notes

• Sulfonyl chlorides. TsCl and MsCl are corrosive lachrymators that liberate HCl on hydrolysis; handle in a hood, keep dry, and quench excess carefully.
• Hydroxylamine. Oxime formation uses NH₂OH (as the hydrochloride) — a reducing agent that is thermally unstable in concentrated/anhydrous form; use the buffered salt and avoid heating neat hydroxylamine.
• Alkoxide bases. KOEt/NaOEt in ethanol are strongly basic and moisture-sensitive; generate or store under inert atmosphere and add to the substrate at controlled temperature to avoid runaway aldol side reactions.
• Azirine handling. Some 2H-azirines are volatile, strained, and potentially shock/heat-sensitive on scale; make and use them promptly rather than storing large quantities.
• Scale. The Neber is primarily a laboratory and fine-chemical method for azirines and α-amino ketones; there is no bulk-tonnage industrial Neber process comparable to the Beckmann's caprolactam route to nylon-6.