Amides from Acid Chlorides and Amines

Interfacial Schotten–Baumann acylation is literally how nylon-6,6 fibers are demonstrated in the classic "nylon rope trick": a solution of adipoyl chloride in an organic solvent is layered under an aqueous solution of hexamethylenediamine (with NaOH as base to scavenge HCl), and a polyamide film forms instantly at the interface, which can be drawn upward as a continuous rope. Every amide bond in that polymer forms by exactly the addition–elimination mechanism developed here, molecule by molecule, at the liquid–liquid interface. In pharmaceutical synthesis, acid-chloride amidation remains a default method whenever the corresponding carboxylic acid can be converted cleanly to its chloride (typically with thionyl chloride or oxalyl chloride) — local anesthetics of the "-caine" family (lidocaine, procaine-type analogs), many benzamide and sulfonamide-adjacent drug scaffolds, and numerous peptide-bond-forming steps in small-molecule (non-solid-phase) synthesis all rely on this chemistry when racemization risk and functional-group tolerance permit acid chloride formation. Because acid chlorides are so reactive, this method is typically reserved for substrates without epimerizable stereocenters alpha to the future amide carbonyl, since the highly electrophilic acyl chloride and moderately basic conditions can occasionally promote unwanted enolization — a limitation that motivates milder coupling reagents (carbodiimides, uronium/phosphonium reagents) for stereochemically sensitive peptide couplings, but the underlying tetrahedral-intermediate mechanism taught here is the conceptual ancestor of every one of those "activation" strategies. In materials science beyond nylon, Schotten–Baumann-type interfacial polymerization is also the industrial method for producing aromatic polyamides (aramids such as Kevlar-type fibers, via terephthaloyl chloride and diamines) and polyurea/polycarbonate-adjacent materials, and it underlies many surface-functionalization protocols in which an acid chloride is used to graft amide-linked functional groups onto amine-terminated surfaces, nanoparticles, or dendrimers. Biologically, understanding this addition–elimination mechanism at a reactive acyl carbon is also the conceptual bridge to enzymatic acyl transfer (acyl-CoA chemistry, serine protease acylation of the catalytic serine), where the same tetrahedral-intermediate logic operates with far better-tuned leaving groups and exquisitely positioned general bases — foreshadowing exactly the reverse process of Amide Hydrolysis, where water rather than an amine performs the nucleophilic addition and a proton-shuffling network takes over the role played here by external base.

Ask the tutor: How does the mechanism change if the nucleophile is water instead of an amine — that is, what does this same tetrahedral-intermediate logic look like running as amide hydrolysis?

Picture the carbonyl carbon of an acid chloride, $\mathrm{R\text{-}C(=O)\text{-}Cl}$. It is $sp^2$-hybridized, with a $\pi^*_{C=O}$ orbital lying in the same plane as, and conjugated with, chlorine's lone pairs and stabilized (weakly) by resonance donation from Cl. That donation is feeble compared to the analogous donation in an amide, because chlorine's $2p$-like lone pairs are held in a $3p$ orbital that overlaps poorly with the carbon $2p_z$ of the $\pi$ system — a bad size/energy match. The carbonyl carbon in an acid chloride is therefore strongly electrophilic: little resonance donation cushions the partial positive charge, and the inductive pull of chlorine adds further electrophilic character. An amine nitrogen carries a nucleophilic lone pair in an $sp^3$ orbital. When the amine approaches the carbonyl carbon along the Bürgi–Dunitz trajectory (roughly $107°$ off the C=O axis, the angle at which the nitrogen lone pair achieves best overlap with $\pi^*_{C=O}$ while avoiding the developing filled orbitals), the nitrogen lone pair donates into $\pi^*_{C=O}$. This pushes the carbonyl $\pi$ electrons fully onto oxygen, converting the trigonal, $sp^2$ carbon into a tetrahedral, $sp^3$ carbon bearing four substituents: R, Cl, the new N, and an alkoxide-like O⁻. This tetrahedral intermediate is the pivotal species of the whole mechanism — the same species that appears (formed and destroyed by different partners) throughout ester, anhydride, and amide interconversions. The tetrahedral intermediate has a choice of three groups it could expel to collapse back to a trigonal carbon: R⁻ (never — carbanions are terrible leaving groups and are never ejected), the newly added amine (the microscopic reverse of addition), or chloride. Chloride wins overwhelmingly: as the conjugate base of a strong acid ($\mathrm{p}K_a(\mathrm{HCl}) \approx -7$), Cl⁻ is an exceptionally stable, weakly basic, highly polarizable anion, and it is expelled with the alkoxide oxygen's lone pair re-forming the $\pi$ bond to give the trigonal amide carbonyl. The amide product is stabilized by strong nitrogen lone-pair donation into $\pi^*_{C=O}$ (good size match between N's $2p$ and C's $2p_z$), giving the C–N bond significant double-bond character and rotational barriers of $15$–$20\ \mathrm{kcal\ mol^{-1}}$ — the signature of Amide Resonance and Stability. It is this stabilization of the product carbonyl, not any special stability of an amide-forming transition state, that makes the overall transformation strongly exergonic and effectively irreversible.

Once the neutral amide has formed, notice that the nitrogen is still protonated (it picked up a proton implicitly at the ammonium stage, or explicitly if a secondary/primary amine's N–H survives) — actually let's be careful: the tetrahedral intermediate carries a positively charged ammonium nitrogen (three substituents plus the new C bond) immediately after addition, before any proton is lost. That positive charge makes the C–N bond in the tetrahedral intermediate a poor amide-forming bond until a proton is removed from nitrogen — but proton loss from the ammonium nitrogen is fast and can occur before, during, or after chloride expulsion (concerted with expulsion in many mechanistic depictions, or via a separate fast acid–base step). The key point for exam purposes: somewhere in the sequence, one N–H proton is lost to solvent or to added base, and it is exactly this proton, combined with the departing chloride, that constitutes the "HCl" produced stoichiometrically.

The overall stoichiometry, whether the base is a second equivalent of the same amine or an added tertiary amine/hydroxide, is

where:

• $\mathrm{R\text{-}COCl}$ is the acid chloride (acyl electrophile),
• $\mathrm{R'NH_2}$ is the amine nucleophile (primary shown; secondary $\mathrm{R'_2NH}$ behaves identically),
• $B$ is the base (a second equivalent of amine, aqueous hydroxide, pyridine, or a tertiary amine such as Et₃N),
• $BH^+Cl^-$ is the ammonium (or pyridinium) chloride salt byproduct.

Reading this equation: exactly one proton and one chloride are released per amide bond formed, and they must find a home somewhere — either on a sacrificial second amine molecule or on an added base — or the reaction stalls partway through as amine is converted to unreactive ammonium salt. The rate law for the addition step, under conditions where amine is not fully consumed and no significant reverse reaction competes, is second order overall:

where:

• $k_2$ is the second-order rate constant for nucleophilic addition (specific to solvent, R, and the amine's nucleophilicity/steric bulk),
• the tetrahedral intermediate does not appear because its collapse (chloride loss) is fast relative to its formation, so the first step is rate-determining and the steady-state approximation collapses to simple bimolecular kinetics.

The Hammond-postulate-guided energetics along the reaction coordinate can be captured by comparing the barrier for addition, $\Delta G^\ddagger_{\text{add}}$, with the (much smaller, often negative) barrier for chloride expulsion, $\Delta G^\ddagger_{\text{elim}}$:

where: this inequality is precisely the mechanistic justification for why the tetrahedral intermediate does not accumulate and why the overall reaction obeys clean second-order kinetics in $[\mathrm{RCOCl}]$ and $[\mathrm{amine}]$ rather than showing saturation kinetics.

Example 1 (foundational): acetyl chloride plus excess aniline — identifying every species and every arrow. Acetyl chloride, $\mathrm{CH_3COCl}$, is added dropwise to a stirred solution containing $2.2$ equivalents of aniline, $\mathrm{C_6H_5NH_2}$ ($\mathrm{p}K_{aH} \approx 4.6$, a fairly weak but perfectly serviceable amine nucleophile), in dichloromethane at $0\ ^\circ\mathrm{C}$. Step 1. Aniline's nitrogen lone pair attacks the acetyl carbonyl carbon along the Bürgi–Dunitz trajectory. Curved arrow 1: N lone pair → new C–N bond. Curved arrow 2: C=O π → O, generating a tetrahedral alkoxide bearing CH₃, Cl, O⁻, and $\mathrm{^+NH_2C_6H_5}$. Step 2. A second aniline molecule (the "1.2 extra equivalents") acts as the base, removing a proton from the ammonium nitrogen of the intermediate: curved arrow from N–H bond to the second aniline's nitrogen lone pair, forming anilinium, $\mathrm{C_6H_5NH_3^+}$. Step 3. The alkoxide oxygen's lone pair reforms the C=O π bond, expelling chloride: curved arrow from O⁻ lone pair → π bond; curved arrow from C–Cl bond → Cl⁻. Step 4. Product is acetanilide, $\mathrm{CH_3CONHC_6H_5}$, plus one equivalent of anilinium chloride salt. Overall:

Limiting-case check: if only $1.0$ equivalent of aniline were used (no excess to serve as base), the reaction would still proceed for the first roughly $50\%$ of conversion, but thereafter the accumulating HCl would protonate remaining aniline (converting nucleophile to unreactive anilinium), and the reaction would stall well short of completion — exactly why textbook procedures specify $\geq 2$ equivalents of amine, or explicitly add a separate base such as pyridine or aqueous NaOH, whenever only one equivalent of the valuable amine is available.

Example 2 (applied): a chemoselective acylation used in local-anesthetic synthesis — lidocaine-type coupling. Consider coupling 2,6-dimethylaniline with 2-(diethylamino)acetyl chloride — a step conceptually related to the industrial synthesis of lidocaine. The substrate 2,6-dimethylaniline is a hindered, weakly nucleophilic primary aromatic amine (two ortho-methyl groups twist the ring and add steric bulk near nitrogen), while the acylating agent additionally contains a tertiary amine (the diethylamino group) elsewhere in the molecule. The mechanistic question: does the tertiary diethylamino nitrogen, also present in the same molecule and also nucleophilic, interfere by acylating itself intermolecularly (attacking a second equivalent of the acid chloride)? Reasoning through the mechanism. A tertiary amine nitrogen can attack a carbonyl carbon exactly as a primary amine can — the addition step is available to any amine lone pair — but the resulting tetrahedral intermediate, after chloride loss, would carry a quaternary, positively charged nitrogen with no N–H proton to lose. There is no possible neutral, low-energy product: the nitrogen cannot become part of a stable neutral amide, because amides require an N–H (or N-alkyl with only three total substituents) to accommodate the trigonal-planar, resonance-stabilized nitrogen. A tertiary amine attacking an acyl chloride forms a reactive acylammonium salt — itself a competent, transient acylating agent (this is precisely the mechanism by which DMAP catalyzes acylations) — which is rapidly attacked in turn by the more abundant, more thermodynamically favorable primary amine nucleophile, transferring the acyl group onward and regenerating free tertiary amine. Step 1. 2,6-Dimethylaniline's lone pair attacks the acyl chloride carbon (slowed by steric hindrance near nitrogen, but still viable since the electrophile is highly reactive). Step 2. Chloride is expelled from the tetrahedral intermediate (curved arrows as before), giving the anilide directly; any tertiary-amine acylammonium intermediate that forms competitively is simply an unproductive detour that self-corrects by acyl transfer to the more stable, N–H-bearing anilide product. Step 3. Excess base (here, triethylamine or aqueous NaOH added externally, since the substrate's own internal tertiary amine is not present in sufficient excess to serve reliably as base) mops up the liberated HCl. Net lesson: chemoselectivity in polyfunctional amine substrates is governed not by which nitrogen attacks fastest, but by which resulting product is thermodynamically most stable — tertiary-amine acylammonium salts are dead ends that funnel product back toward the stable secondary/primary amide, because Amide Resonance and Stability provides a driving force unavailable to any quaternary ammonium species.

Example 3 (exam/competition level): quantitative kinetic comparison and predicting a rate ratio. At $25\ ^\circ\mathrm{C}$ in acetonitrile, the second-order rate constant for reaction of benzoyl chloride with n-butylamine is measured as

while the corresponding rate constant for benzoyl chloride reacting with n-butanol (an alcohol nucleophile of comparable steric bulk and comparable $\mathrm{p}K_{aH}$ region for the conjugate acid is not directly comparable, but nucleophilicity toward carbonyl carbon is what matters here) is measured as

Question: Estimate the rate ratio and explain it mechanistically; then predict, using the two-step mechanism, what would happen to the product distribution if both n-butylamine and n-butanol (1 equivalent each, no added base) were present simultaneously with 1 equivalent of benzoyl chloride. Step 1 — compute the ratio.

Step 2 — mechanistic rationale for the ratio. Nitrogen is less electronegative than oxygen, so nitrogen's lone pair sits at higher energy (better HOMO–LUMO overlap with $\pi^*_{C=O}$, i.e., a smaller HOMO(Nu)–LUMO(C=O) energy gap) and is more polarizable and more readily desolvated in the addition transition state. Both effects raise $k_1$ (the addition rate constant, rate-determining per the earlier derivation) for amines relative to alcohols of comparable steric profile. This is the same electronic argument used throughout Interconverting Acyl Derivatives Down the Ladder to rank nucleophilicity: amines beat alcohols by roughly 2–4 orders of magnitude toward acyl electrophiles, matching the measured ratio of $\sim1.6\times10^3$ here. Step 3 — predict product distribution in direct competition. With both nucleophiles present at equal concentration and no base added, the amine outcompetes the alcohol for the first equivalent of benzoyl chloride by the ratio computed above — essentially all of the first-formed product is the amide, not the ester. But recall the stoichiometric trap from Example 1: forming the amide releases one equivalent of HCl, and with no added base, this HCl will protonate the remaining free amine (aniline-type or alkylamine, whichever nucleophile is left in excess) far more readily than it protonates the weakly basic alcohol. Once amine is exhausted or fully protonated, any still-unreacted benzoyl chloride is now kinetically funneled toward the alcohol, since protonated ammonium is no longer nucleophilic. The experimentally observed outcome, then, is a mixture dominated by amide but with a measurable yield of ester as a "second-place" product — precisely because HCl selectively deactivates the amine nucleophile partway through the reaction. This is a clean illustration of why competitive-nucleophile experiments without added base can give misleading impressions of intrinsic nucleophilicity ratios unless the HCl-scavenging effect is explicitly accounted for. Limiting-case check: if an external base (e.g., pyridine, 1.0 equiv) were included so that the amine is never protonated, the amide:ester product ratio would climb further, approaching the intrinsic kinetic ratio of $\sim1.6\times10^3$:1, since the amine nucleophile remains fully "alive" throughout the reaction rather than being partially consumed by protonation.

Carl Schotten and Eugen Baumann developed the biphasic acylation procedure independently in the 1880s, seeking a practical, scalable way to acylate amines and alcohols with acid chlorides without the harsh, uncontrolled exotherms and side reactions typical of neat reactions. Baumann's independent work on acylating amino acids and phenolic substrates, aimed at understanding physiological detoxification pathways (his studies on how the body conjugates aromatic compounds before excretion), gave the method an early biochemical pedigree quite distinct from its later dominance in industrial and pharmaceutical synthesis. The genius of the protocol is procedural rather than conceptual: by confining the electrophile and organic nucleophile to one phase and the base to another (aqueous), fast, clean acylation could be achieved at low temperature with a cheap, easily removed base, entirely sidestepping the need for anhydrous glassware or exotic solvents that many other acylation methods of the era demanded.