Acylation of Amines to Amides
An aniline molecule dropped into a flask of acetic anhydride undergoes a transformation. It looks simple on paper. But it hides a beautifully choreographed sequence of bond-making and bond-breaking events. Within seconds, the free amine is converted into an amide. The amine is nucleophilic and basic, with an electron-rich nitrogen lone pair exposed on a benzene ring. The resulting amide is a functional group so electronically muted that it can survive strong acids. It also survives bromination conditions and Friedel–Crafts electrophiles. These conditions would have destroyed or over-reacted with the parent amine. This single reaction, amine acylation, is one of the most important "volume control knobs" in organic synthesis. It does not build a new carbon skeleton. Instead, it changes how reactive nitrogen is, by orders of magnitude. This change happens in a fully predictable and reversible way. The transformation matters because free amines are frequently too reactive for what a synthetic sequence demands next. An aniline subjected to bromination reacts so fast and so non-selectively that polybromination and even oxidation side reactions compete. This happens via the very same activation that makes it a good nucleophile. Acylate the nitrogen first, and the resulting acetanilide directs electrophiles cleanly to the para position. There is none of the runaway reactivity. The same idea recurs throughout medicinal chemistry, peptide synthesis, and polymer science. Nylon, Kevlar, and every polypeptide bond in every protein are amides. They are formed by variations on this exact nucleophilic acyl substitution. The big picture is that amide bond formation involves an amine and an acylating agent. That agent may be an acyl chloride, anhydride, or activated ester. This is a textbook case of nucleophilic acyl substitution. This is the addition–elimination mechanism that dominates all reactions at the carbonyl carbon of carboxylic acid derivatives. The amine nitrogen begins as a good nucleophile. This is built on the same lone pair that makes Amines as Nucleophiles such a productive lens. It ends as part of a resonance-stabilized, poorly nucleophilic amide. So the reaction is simultaneously a synthesis and a deactivation. Understanding why the amide nitrogen becomes so much less reactive is key. This happens via delocalization of its lone pair into the carbonyl π* system. This is the single mechanistic idea that connects amide synthesis to its downstream role. That role is as a protecting and activating strategy. This article works through the full curved-arrow mechanism for both acyl chloride and anhydride acylations. It explains quantitatively why amides are ~10¹⁰ times less basic than amines. It walks through three fully worked examples. These span a foundational reaction, a synthetic (Schotten–Baumann) application, and an exam-style multistep problem. It closes with the scope, limitations, and the historical figures who established this chemistry.
- Amine acylation is nucleophilic acyl substitution: the amine lone pair attacks the acyl carbonyl carbon, forming a tetrahedral intermediate that collapses by ejecting a leaving group (Cl⁻ or RCO₂⁻).
- Acyl chlorides are far more reactive than anhydrides, which are more reactive than esters, because Cl⁻ and carboxylate are better leaving groups and less able to donate into the carbonyl by resonance.
- One equivalent of HCl (or acid, with anhydrides) is generated per acylation; a base (excess amine, NaOH, or pyridine) is needed to mop it up and prevent unproductive protonation of the amine.
- Amide nitrogen lone pairs are delocalized into the carbonyl π* system, so amides are ~10 pKa units less basic than amines and essentially non-nucleophilic at nitrogen.
- This basicity/reactivity drop is exploited industrially and synthetically: acetanilide formation tames aniline for clean, monosubstituted electrophilic aromatic substitution.
- The reaction is reversible in principle (hydrolysis regenerates the amine) but amide hydrolysis is slow, which is exactly what makes amides useful as protecting groups.
- Key idea: Acylation converts a strongly nucleophilic/basic amine lone pair into a resonance-delocalized amide lone pair, and this loss of nitrogen electron density is the whole story — both for making the bond and for why the product behaves so differently from the starting amine.
Where this lives today
Amide bond formation by acylation of amines is arguably the single most industrially performed organic transformation on Earth, because every peptide bond in every synthetic peptide drug (insulin analogs, GLP-1 agonists such as semaglutide, and countless others made by solid-phase peptide synthesis) is an amide formed by a closely related activation-and-acylation logic, even though modern peptide couplings typically use carbodiimide- or uronium-based in situ activators (EDC, HATU) rather than isolable acyl chlorides, precisely to generate a reactive acyl electrophile under mild, epimerization-minimizing conditions. Nylon-6,6 and Kevlar are built from diacyl chloride plus diamine polycondensations that are large-scale, industrial versions of exactly the Mechanism 1 sequence above, run at interfaces (the classic "nylon rope trick" demonstration is literally interfacial Schotten–Baumann polymerization). In pharmaceutical manufacturing, acetanilide-type acylations remain a standard way to protect and deactivate aniline-derived intermediates before performing electrophilic aromatic substitution steps — the same logic worked through in Example 3 — which is exactly the bridge to controlling reactivity that foreshadows the broader theme of Amines as Protecting/Activating Groups in EAS. Beyond synthesis, the extreme robustness of the amide bond (a direct kinetic consequence of the very same resonance delocalization that makes amide nitrogen non-nucleophilic) is why proteins are stable enough to fold, function, and be sequenced, and why amide-based polymers survive as durable materials; understanding amine acylation mechanistically is therefore simultaneously a lesson in synthetic methodology, in protein chemistry, and in materials science.
Ask the tutor: Why does protonation of an amide occur at oxygen rather than nitrogen, and how would you prove this experimentally?
The physical picture
Look first at the electronics of the two nitrogen environments before worrying about mechanism. In a simple amine such as aniline, the nitrogen lone pair occupies an sp³-like (or, for aniline, partially conjugated) orbital and is available for two jobs: acting as a Brønsted base (protonation) and acting as a nucleophile (attack on an electrophilic carbon). Both jobs draw on the same electron pair. In an amide, that lone pair has a third, competing option: conjugation into the adjacent carbonyl. The nitrogen lone pair, the C=O π bond, and the oxygen lone pairs together form a three-center, four-electron delocalized π system across N–C(=O). This delocalization is so favorable (worth roughly 15–20 kcal/mol of resonance stabilization) that the amide C–N bond acquires substantial double-bond character — around 40% by some estimates from rotational barrier measurements (~15–20 kcal/mol barriers to amide bond rotation, versus a few kcal/mol for a normal C–N single bond). This is precisely why amide bonds are planar and why peptide backbones adopt restricted, predictable cis/*trans* conformations. Mechanistically, both acyl chloride and anhydride acylations proceed by the same two-stage logic that governs all nucleophilic acyl substitution: addition of the nucleophile to the sp² carbonyl carbon to give a tetrahedral, sp³ alkoxide-like intermediate, followed by elimination of the leaving group to regenerate a trigonal planar carbonyl carbon — but now bearing nitrogen instead of the original leaving group. The driving force for the elimination step is reformation of the strong C=O π bond, which is temporarily sacrificed during the addition step. Whether that reformed π bond ends up flanked by a good or poor leaving group governs the rate: chloride is an excellent leaving group (weak conjugate base of a strong acid, low-lying antibonding orbital, poor π-donor back into the carbonyl in the ground state) so acyl chlorides react fastest; carboxylate is a good but slower leaving group, and anhydrides react correspondingly more slowly but still efficiently, since carboxylate is still a resonance-stabilized, reasonably stable anion once expelled.
The core ideas
The rate law for acylation under the standard conditions (excess or catalytic base, non-nucleophilic solvent) is second order overall, first order in each reactant:
where: is the second-order rate constant, reflecting how readily the amine lone pair attacks the specific carbonyl electrophile (acyl chloride ≫ anhydride ≫ ester ≫ amide in electrophilicity). The relative reactivity of acyl donors tracks directly with leaving-group ability and how much the carbonyl carbon is already deactivated by resonance donation from its own heteroatom substituent:
where: each step right corresponds to a leaving group that is a poorer leaving group and/or a substituent (Cl, RCO₂, OR″, NR₂″) that donates more strongly by resonance into the carbonyl, making that carbonyl less electrophilic to begin with. The thermodynamic reason amides are such poor nucleophiles/bases is captured by comparing the basicity of an amine to the basicity of the analogous amide:
where: is the pKa of the conjugate acid; a smaller (more negative) value means the neutral species (amide) is a much weaker base than the amine. The ten-unit gap corresponds to roughly a 10¹⁰-fold difference in basicity — the exact number chemists mean when they say "amides are non-basic." The free energy of the resonance stabilization that causes this can be estimated from the rotational barrier about the C–N bond, measured by variable-temperature NMR coalescence experiments:
where: the barrier reflects the energy cost of twisting the amide out of planarity, destroying the resonance overlap between the N lone pair and the C=O π* — direct experimental proof that the "double bond character" drawn in the minor resonance structure is real and large.
Goal. Convert R–NH₂ + R′COCl into the amide R′CONHR, tracking every arrow, and show why a base (a second equivalent of amine, aqueous NaOH, or pyridine) is required for a clean, high-yielding reaction.
Step 1 (nucleophilic addition). The amine nitrogen lone pair (HOMO, essentially a filled sp³ orbital) attacks the acyl chloride's carbonyl carbon (LUMO = π* of C=O, further lowered in energy by the inductively electron-withdrawing chlorine). One curved arrow runs from the N lone pair to the new N–C σ bond; simultaneously the C=O π bond curls onto oxygen, because carbon cannot exceed four bonds. This generates a tetrahedral alkoxide intermediate in which nitrogen now bears a formal positive charge (it has donated its lone pair and gained a bond, going from three bonds/one lone pair to four bonds/no lone pair) and oxygen bears the negative charge.
Step 2 (collapse / elimination). The tetrahedral intermediate is not the product — it must collapse to regenerate the thermodynamically favorable C=O π bond. An oxygen lone pair pushes back down to re-form the π bond, and this electron flow pushes the C–Cl σ bond electrons entirely onto chlorine, expelling it as chloride ion. This is the elimination half of addition–elimination: the same tetrahedral-carbon logic used whenever a leaving group is ejected to restore a carbonyl.
Step 3 (deprotonation). The immediate product of Step 2 is an ammonium amide — R′C(=O)–N⁺H₂R — still bearing a proton on nitrogen and a positive charge, because forming the new N–C bond in Step 1 used the lone pair, leaving nitrogen with four bonds and formal +1 charge that is retained through the collapse. A base (chloride ion itself is weakly basic and can do some of this; but usually excess starting amine, added pyridine, or added NaOH is far more effective) removes this N–H proton, giving the neutral amide product and regenerating aromaticity/planarity of the nitrogen lone pair delocalized into the new carbonyl.
Step 4 (why base is essential, not merely helpful). Without a base to intercept it, the HCl generated (net: one H⁺ liberated, one Cl⁻ liberated) will protonate unreacted starting amine — amines are much better proton acceptors (pKaH ≈ 10–11) than the tiny residual amount of product amide (pKaH ≈ 0) or than water. Protonated amine (R–NH₃⁺) has no lone pair available for nucleophilic attack, so every equivalent of amine consumed as "HCl sponge" is an equivalent removed from productive acylation. Practically, chemists therefore use two equivalents of amine (one becomes product, one becomes ammonium chloride salt) or add a non-nucleophilic/weakly nucleophilic auxiliary base such as pyridine, triethylamine, or aqueous NaOH (Schotten–Baumann conditions) so that the precious amine substrate is not wasted as a proton sponge.
Go deeper · Derivation B · 2 minMechanism 2 — Acylation by an anhydride, and why it is slower but still clean
Goal. Show the addition–elimination pathway for (CH₃CO)₂O + R–NH₂ → CH₃CONHR + CH₃CO₂H, and explain quantitatively why anhydrides are less reactive than acyl chlorides yet remain excellent, controllable acylating agents.
Step 1 (nucleophilic addition at one of the two equivalent carbonyls). The amine lone pair attacks either carbonyl carbon of the symmetric anhydride (both are electronically identical, so there is no regiochemical ambiguity). The arrow runs N lone pair → new N–C bond; the attacked C=O π bond curls onto that oxygen, generating a tetrahedral alkoxide intermediate. Crucially, the anhydride carbonyl carbon is less electrophilic than an acyl chloride's, because the adjacent ester-like oxygen (the bridging O of the anhydride) donates electron density into the carbonyl by resonance, partially quenching the π* LUMO before the amine ever arrives. This resonance donation is the electronic reason anhydrides react measurably slower than acid chlorides — it is a ground-state effect on the electrophile, not a transition-state effect on the nucleophile.
Step 2 (collapse, expelling carboxylate). The tetrahedral alkoxide collapses exactly as before: an oxygen lone pair reforms the π bond, and the electrons of the C–O(bridge) bond are pushed entirely onto the bridging oxygen, which departs as a carboxylate ion (CH₃CO₂⁻) still attached to the other acetyl group. Carboxylate is a substantially better-stabilized (resonance-delocalized over two oxygens) leaving group than an alkoxide would be, and only modestly worse than chloride — which is exactly the compromise that makes anhydrides "fast enough to react cleanly, slow enough to control."
Step 3 (proton transfer, no base strictly required). Unlike the acyl chloride case, the byproduct here is acetate/acetic acid rather than HCl — a substantially weaker acid (acetic acid pKa ≈ 4.8, vs. HCl pKa ≈ −7). The immediate N-protonated ammonium amide intermediate is deprotonated by the departed carboxylate ion itself (acetate is a much better base than chloride), so anhydride acylations often proceed cleanly with only mild or no added base, though pyridine is still routinely added to buffer the medium and speed the reaction (pyridine can also act as a nucleophilic catalyst, attacking the anhydride first to generate a more reactive acylpyridinium species — the mechanistic basis of DMAP catalysis in esterifications and acylations).
Step 4 (comparative rate rationalization). Putting Steps 1–3 together: the anhydride's carbonyl is a poorer electrophile (more resonance-donated) but its leaving group (carboxylate) is almost as good as chloride and it generates a much weaker, easier-to-handle acid byproduct. Acyl chlorides win on raw electrophilicity and leaving-group ability (chloride's conjugate acid, HCl, pKa ≈ −7, is an outstanding leaving group by the Hammond-postulate logic that a more stable anion leaves faster), but anhydrides win on selectivity and mildness — which is why anhydrides (particularly acetic anhydride) are the default laboratory reagent for acetylating sensitive or precious amines, including in Schotten–Baumann-type protocols and in classic protecting-group chemistry.
Worked examples
Example 1 — Foundational: acetylation of aniline with acetyl chloride. Aniline (C₆H₅NH₂) is treated with acetyl chloride (CH₃COCl) in the presence of one equivalent of pyridine, in dichloromethane at 0 °C. Step 1. The aniline nitrogen lone pair (delocalized somewhat into the ring, but still nucleophilic — aniline's pKaH is 4.6, much lower than an alkylamine's 10–11, precisely because of this same ring conjugation, yet still nucleophilic enough for acylation) attacks the acetyl chloride carbonyl carbon. Arrow from N lone pair to the new N–C bond; arrow from C=O π bond to the oxygen, generating the tetrahedral intermediate:
Step 2. Collapse: oxygen lone pair reforms π bond, chloride is expelled.
Step 3. Pyridine (the added base, pKaH ≈ 5.2, chosen deliberately because it is a good general base but a poor nucleophile toward the already-formed amide and non-competitive as an acylation substrate under these conditions) removes the N–H proton, delivering neutral acetanilide and pyridinium chloride:
Limiting-case check: if no base were added, unreacted aniline (pKaH 4.6) would compete with pyridine (pKaH 5.2) for the liberated proton; since they are comparably basic here, both mechanisms operate, but pyridine is used in excess/as solvent precisely so it dominates the proton-scavenging role and aniline is preserved for acylation rather than consumed as ammonium salt. The product, acetanilide, has pKaH ≈ 0 (O-protonation) — confirming that the amide product cannot re-protonate or interfere further.
Example 2 — Applied: Schotten–Baumann synthesis of an amide from a hindered amine, with pH control. A chemist wants to acylate cyclohexylamine (pKaH ≈ 10.6, a simple secondary-carbon-substituted primary amine, non-aromatic and therefore a stronger, more nucleophilic base than aniline) using benzoyl chloride (C₆H₅COCl), employing the classic Schotten–Baumann two-phase protocol: aqueous NaOH plus an organic solvent (or neat with vigorous stirring), keeping the mixture cold and basic throughout. Why two phases and why NaOH, mechanistically. Benzoyl chloride is also reactive toward water (hydrolysis is a competing nucleophilic acyl substitution with hydroxide/water as the nucleophile). The Schotten–Baumann trick exploits a kinetic partition: cyclohexylamine, un-protonated at the interface, is a far better nucleophile toward the acyl chloride carbonyl than water is (nitrogen lone pairs are more polarizable/nucleophilic than oxygen lone pairs of the same approximate basicity, by the general soft-nucleophile argument), so amide formation dramatically outcompetes hydrolysis provided the amine is kept in its free-base (unprotonated) form. That is exactly what the NaOH accomplishes: it (a) neutralizes HCl as fast as it forms, preventing protonation of cyclohexylamine into unreactive cyclohexylammonium, and (b) keeps the aqueous phase basic so any amine that partitions into water remains nucleophilic. Mechanism, step by step.
The product, N-cyclohexylbenzamide, forms in high yield (typically >85% under well-optimized Schotten–Baumann conditions) precisely because the rate constant for amine attack on the acyl chloride vastly exceeds the rate constant for hydroxide/water attack on the same electrophile at the concentrations used, and because NaOH continuously regenerates free-base amine at the interface.
Limiting-case check: if NaOH were entirely omitted and the reaction run in neat water with no base, the HCl generated would rapidly protonate cyclohexylamine (pKaH 10.6, an excellent proton acceptor), converting most of it to unreactive cyclohexylammonium chloride within a few turnovers — the reaction would stall after consuming perhaps ~50% of the amine (one part becomes ammonium salt, one part becomes amide, in the absence of any other base, matching the "2 equivalents of amine" stoichiometric workaround described in Mechanism 1).
Example 3 — Exam/competition level: chemoselective mono-acylation and downstream EAS design. Problem. A synthetic route requires converting p-toluidine (4-methylaniline) into 4-methylacetanilide, then performing a clean mono-bromination, and finally hydrolyzing back to the free amine. Predict the site of bromination and explain, mechanistically, why acylation is a necessary step rather than an optional convenience — and predict what would happen (and why it would fail) if bromination were attempted directly on free p-toluidine. Why acylation is necessary — quantitative reasoning. Free p-toluidine has an activated, electron-rich ring (amino group is a powerful π-donor by resonance, comparable to or stronger than a hydroxyl group in activating power) and the nitrogen itself is a good Brønsted base (pKaH ≈ 5.1, slightly higher than aniline's due to the para-methyl's weak electron donation). Direct treatment with Br₂ suffers from two compounding problems. First, aniline-type rings are so activated that bromination is essentially uncontrollable — polybromination (di- and tribromination) occurs because each successive bromination, although it removes some ring activation, still leaves the ring more reactive than benzene itself; industrially and in the lab, free anilines brominate to the 2,4,6-tribromo product under most conditions without special dilution/temperature control. Second, Br₂ (or the trace HBr generated, or Lewis-acidic FeBr₃ if used) can protonate or complex the amine nitrogen directly; anilinium (protonated aniline) has its nitrogen lone pair tied up in an N–H bond and can no longer donate into the ring by resonance, so the ring reverts toward benzene-like reactivity and the −NH₃⁺ substituent becomes weakly meta-directing/deactivating by induction — an unpredictable, substrate/condition-dependent mixture of directing effects results. The acylation fix, mechanistically. Forming 4-methylacetanilide first (via the standard addition–elimination mechanism with acetic anhydride or acetyl chloride, as in Mechanism 1/2) converts the nitrogen lone pair from a "fully available π-donor plus Brønsted base" into a lone pair partially delocalized into the new carbonyl (Mechanism 2's resonance argument). The amide nitrogen is now a substantially weaker π-donor into the ring than a free amine (because some lone-pair density is diverted to the exocyclic carbonyl) — activating the ring far less aggressively, enough to permit clean mono-substitution — while remaining an ortho/para-director (the residual lone-pair donation still dominates over the carbonyl's inductive withdrawal). Simultaneously, the amide nitrogen is essentially non-basic (pKaH ≈ 0), so it cannot be protonated by HBr or complexed destructively by Lewis acids under the reaction conditions; the directing/activating character of the substituent is therefore locked in and condition-independent. Predicting the product. In 4-methylacetanilide, the two ring-activating substituents are the acetamido group (−NHCOCH₃, at C1, ortho/para-director) and methyl (at C4, ortho/para-director, weaker). Both direct incoming electrophiles to the position ortho to the acetamido group that is not blocked by the methyl — i.e., position 3 (equivalently written as ortho to NHCOCH₃ and meta to CH₃). Bromination with Br₂ (mild Lewis acid catalyst optional, since the ring is still activated enough) gives 4-methyl-2-bromoacetanilide (2-bromo-4-methylacetanilide) predominantly, in high yield and with clean mono-substitution. Final hydrolysis. Aqueous acid or base hydrolysis of the amide (reverse nucleophilic acyl substitution — water or hydroxide attacks the amide carbonyl, forms a tetrahedral intermediate, and now nitrogen, a much poorer leaving group than chloride or carboxylate, is ejected only because the reaction is driven forward under forcing conditions — reflux with 6 M HCl or NaOH) regenerates the free amine, now 2-bromo-4-methylaniline, cleanly and without disturbing the newly installed bromine.
Limiting-case check: if a much weaker acylating agent were used (say, an amide itself, RCONR₂′, attempting to transfer its own acyl group to the aniline nitrogen), the reaction would not proceed at any practical rate — amide nitrogen is such a poor leaving group (conjugate acid pKaH ≈ 36 for the amide anion, versus pKaH ≈ 10–11 for an amine or ≈ −7 for chloride) that amides do not acylate other amines under ordinary conditions; this is the mechanistic reason the reactivity order RCOCl > (RCO)₂O > RCOOR″ > RCONR₂″ terminates with amides as essentially unreactive electrophiles, and it is exactly what allows amides to survive as stable, "terminal" functional groups in synthesis.
Limiting cases, scope & common traps
| Scenario | What happens | Mechanistic reason |
|---|---|---|
| Primary amine + acyl chloride, no base | Partial conversion, HCl builds up, yield capped near 50% unless excess amine used | Liberated HCl protonates unreacted amine faster than it is consumed; protonated amine has no nucleophilic lone pair |
| Aromatic amine (aniline) vs. aliphatic amine (cyclohexylamine) | Aniline reacts slower, gives cleaner mono-acylation product | Aniline's lone pair is partly delocalized into the ring already (lower ground-state nucleophilicity, lower pKaH ≈ 4.6 vs ≈10.6), so it is a "softer" nucleophile and less prone to over-acylation/side reactions |
| Secondary amine (R₂NH) + acyl chloride | Forms tertiary amide (R₂NCOR′) normally, still fast | Still has an available lone pair and N–H to lose; sterics can slow but rarely stop the reaction |
| Tertiary amine (R₃N) + acyl chloride | No stable amide forms directly (no N–H to lose); may act as nucleophilic catalyst instead | After addition, the tetrahedral collapse would give an acylammonium salt R₃N⁺–COR′ with no proton to remove — this species is itself a reactive acylating agent (as in DMAP catalysis), not the final product |
| Very hindered amine (e.g., 2,6-di-tert-butylaniline) + acyl chloride | Sluggish or no reaction | Steric bulk blocks nucleophilic approach to the tetrahedral transition state; a smaller/more reactive acylating agent (ketene, or forcing conditions) may be needed |
| Attempted acylation using a simple amide as acyl donor | Does not proceed | Amide nitrogen is a very poor leaving group (conjugate acid pKaH ≈ 36); no thermodynamic/kinetic driving force |
| Water present during acyl chloride acylation | Competing hydrolysis to carboxylic acid | Water is also nucleophilic toward the same electrophilic carbonyl; Schotten–Baumann conditions rely on amine outcompeting water kinetically, aided by kept-basic pH |
The chemist behind it
The addition–elimination mechanism for acyl substitution, and the specific two-phase acylation protocol still named for its inventors, trace to late-19th-century German organic chemistry, in the era when systematic mechanistic thinking about carbonyl reactivity was just being established through careful kinetic and product studies (well before curved-arrow notation existed in its modern form — that notation itself was popularized decades later by Robert Robinson and Christopher Ingold). Carl Schotten (1853–1910) and Eugen Baumann (1846–1896) independently developed, in the 1880s, the practice of acylating amines and alcohols using acyl chlorides in the presence of aqueous alkali — Baumann's original work was largely on acylating amino acids and related biologically relevant amines, motivated by physiological chemistry questions, while Schotten's contributions refined the general synthetic protocol. Their names became permanently attached to the biphasic base-mediated acylation technique that remains standard in undergraduate and industrial laboratories alike, precisely because it solves the practical problem worked through in Example 2: keeping the amine nucleophilic and un-protonated while HCl is continuously neutralized.
Origins · 1 minEugen Baumann and how this was discovered
Baumann trained in the great German chemical tradition of the 1870s–1880s, a period when organic chemistry was rapidly professionalizing around structural theory (following Kekulé and Butlerov) but had not yet developed a mechanistic electron-based vocabulary — that would wait for the physical-organic chemistry of the 1920s–1930s. Working largely on sulfur- and nitrogen-containing physiological compounds (his other major contribution was to sulfur metabolism and the chemistry of sulfate esters in urine), Baumann needed a reliable, mild way to acylate amino and hydroxyl groups on delicate, often water-soluble biological substrates without destroying them via side hydrolysis or decomposition — precisely the practical motivation behind what became the Schotten–Baumann protocol.
Why does an acyl chloride react so much faster with an amine than an anhydride does?
Two combined effects favor the acyl chloride. First, chlorine is strongly electron-withdrawing by induction and a poor π-donor into the carbonyl (its lone pairs are higher in energy and larger, worse-matched orbitals for overlap with carbon's 2p orbital), so the acyl chloride carbonyl carbon is a more electrophilic (lower-LUMO-energy) target for the incoming amine lone pair. Second, once the tetrahedral intermediate forms, chloride is ejected far more easily than carboxylate because chloride's conjugate acid, HCl, is a far stronger acid (pKa ≈ −7) than a carboxylic acid (pKa ≈ 4–5), meaning chloride is intrinsically the more stable, better-departing anion. Both the addition and elimination steps are therefore faster for acyl chlorides.
Does the acylation of amines ever go in reverse — can amides be converted back to amines?
Yes, by amide hydrolysis, which is mechanistically the reverse addition–elimination sequence run with water or hydroxide as the nucleophile instead of amine. However, amide hydrolysis is intrinsically much slower than ester or acyl chloride hydrolysis because nitrogen is a poor leaving group (its conjugate acid, the amine, has pKaH ≈ 10–11, versus water's pKa ≈ 15.7 as the corresponding comparison for esters — nitrogen actually leaves less readily than alkoxide in neutral conditions, which is why forcing conditions like prolonged reflux in 6 M HCl or NaOH are required). This kinetic sluggishness is exactly what makes amide bonds durable enough to be useful as protecting groups and as the backbone of stable proteins.
Why is pyridine such a common base choice for these acylations rather than, say, triethylamine or NaOH?
Pyridine (pKaH ≈ 5.2) is basic enough to deprotonate the ammonium-amide intermediate and scavenge HCl effectively, but it is a comparatively poor nucleophile toward carbon electrophiles once its aromatic lone pair is considered (its nitrogen lone pair sits in an sp² orbital in the ring plane, somewhat less available than an amine's, though pyridine can still act as a nucleophilic catalyst by forming a transient, highly reactive acylpyridinium species that is subsequently attacked by the amine — a mechanism exploited deliberately in DMAP-catalyzed acylations). Triethylamine is also common and simply a stronger, purely proton-scavenging base with minimal nucleophilic competition. NaOH is reserved mainly for biphasic Schotten–Baumann conditions where its nucleophilicity toward the acyl chloride is suppressed by keeping it in a separate aqueous phase.
What happens if you try to acylate a tertiary amine (R₃N) with an acyl chloride?
The addition step still occurs — the tertiary amine's lone pair attacks the carbonyl carbon, and the tetrahedral intermediate collapses to expel chloride exactly as before. But the resulting species is an acylammonium salt, R₃N⁺–C(=O)R′, because there is no N–H proton left to lose; it cannot become a neutral amide. This acylammonium species is itself a very reactive acylating agent (more electrophilic than the original acyl chloride in many cases, since the ammonium group is strongly electron-withdrawing), which is precisely the mechanistic basis for nucleophilic catalysis by tertiary amines like pyridine or DMAP: the tertiary amine reacts first and transiently, then hands the acyl group off to the actual substrate amine or alcohol.
Why does resonance in the amide make the C–N bond shorter and rotation-restricted, and how do we know this experimentally?
The minor zwitterionic resonance contributor (R–N⁺(H)=C(–O⁻)–R′) places double-bond character directly on the C–N bond, which both shortens it relative to a typical single C–N bond (amide C–N ≈ 1.32–1.35 Å, versus ≈1.47 Å for a simple amine C–N bond) and restricts rotation about it, because rotating out of planarity would break the p-orbital overlap responsible for the resonance stabilization. Experimentally, this is confirmed by variable-temperature NMR: below the coalescence temperature, the two substituents on nitrogen in an N,N-disubstituted amide (e.g., N,N-dimethylformamide) show separate, distinct signals because rotation is slow on the NMR timescale, and the measured rotational barrier (typically 15–20 kcal/mol) matches independent estimates of amide resonance energy from thermochemical and computational studies.
Is acylation of an amine ever used to protect it in a synthesis, the way protecting groups are used for alcohols?
Yes, and this is one of its most important applied roles. Converting a free amine to an amide (commonly using acetic anhydride to install an acetamido group) dramatically lowers both its basicity and its nucleophilicity, allowing the rest of a molecule to be subjected to acidic conditions, oxidations, or electrophilic reagents that a free amine would otherwise intercept or that would simply protonate the amine and shut down whatever reaction depends on it staying neutral. The acetamido group can later be removed by acidic or basic hydrolysis (forcing conditions, since amide hydrolysis is slow) to regenerate the free amine once the sensitive steps are complete — the same underlying idea extended more generally in strategies for Amines as Protecting/Activating Groups in EAS.
Can amines be acylated with carboxylic acids directly, without converting to a chloride or anhydride first?
Not efficiently under ordinary conditions, because a free carboxylic acid, when it meets an amine, undergoes a fast acid–base proton transfer (the amine, a much stronger base, simply deprotonates the carboxylic acid) to give an ammonium carboxylate salt rather than any nucleophilic attack at carbon; the resulting carboxylate anion is both a poor electrophile (extra negative charge disfavors nucleophilic attack) and would need to expel an even worse leaving group (hydroxide/oxide) to form the amide. Direct amide formation from a carboxylic acid and an amine does happen, but only under forcing thermal conditions (heating well above 150–200 °C to drive off water) or, far more commonly in modern practice, via in situ activation of the acid using a coupling reagent (DCC, EDC, HATU) that first converts the carboxylic acid into a much better acylating agent (an active ester or O-acylisourea), effectively recreating the same addition–elimination logic used with acyl chlorides and anhydrides.