Esters from Acid Chlorides and Alcohols
A flask of octanoyl chloride, a splash of pyridine, and a slow drip of a hindered secondary alcohol. Within minutes, at 0 °C, a clean ester forms in near-quantitative yield. There is none of the equilibrium hand-wringing that plagues Fischer esterification. This is the acid-chloride (or anhydride) route to esters. It is the workhorse method every synthetic chemist reaches for when yield, speed, and mild conditions matter more than atom economy. Fischer esterification asks an alcohol to displace a poor leaving group (hydroxide, effectively) from a carboxylic acid. It requires forcing acid catalysis and removing water to shift an equilibrium. By contrast, the acyl chloride route uses a spectacular leaving group — chloride. It runs downhill, irreversibly, often at or below room temperature. The reason this matters is entirely about the shape of the reactivity ladder among acyl derivatives. Acid chlorides sit at the top of electrophilicity among common carbonyl compounds. The carbonyl carbon is flanked by an electron-withdrawing, weakly-donating chlorine. Chloride is an outstanding leaving group once the tetrahedral intermediate collapses. Alcohols are weak nucleophiles by themselves. But they become potent acyl acceptors when a base is present to deprotonate them (or at least to mop up the HCl produced). The whole transformation becomes a textbook example of nucleophilic acyl substitution. It proceeds by addition–elimination at a trigonal-planar carbonyl carbon. This is mediated by a tetrahedral intermediate whose collapse is essentially irreversible. The big picture connects two ideas already familiar. First, acyl derivatives interconvert down a reactivity ladder (acid chlorides → anhydrides → esters → amides → carboxylates). Each step releases a better leaving group and forms a more stable product. Second, esters themselves are moderately electrophilic but far less so than acid chlorides. So the reaction, once it reaches the ester stage, essentially stops. It does not run past the ester to something even less reactive under these conditions. This article works out the full mechanism of acylation by acid chlorides and anhydrides. It explains why bases like pyridine, triethylamine, and DMAP are needed, and how DMAP works catalytically. It treats the classic Schotten–Baumann conditions. It also shows how sterics and electronics on both partners control rate and yield. By the end you will be able to draw, arrow-by-arrow, the full addition–elimination sequence for acyl chloride and anhydride acylation. You will predict which base to use and why DMAP accelerates hindered esterifications. You will rationalize relative rates across primary, secondary, tertiary, and phenolic alcohols. And you will recognize the practical failure modes. These include competing HCl-catalyzed side reactions, racemization risk at α-stereocenters, and the special problem of esterifying phenols and hindered acids.
- Acid chlorides and anhydrides acylate alcohols by nucleophilic acyl substitution: addition of RO⁻/ROH to the carbonyl gives a tetrahedral intermediate, which collapses by ejecting Cl⁻ (or RCO₂⁻).
- The reaction is irreversible because chloride (or carboxylate, in the anhydride case) is a far better leaving group than alkoxide, so the equilibrium lies overwhelmingly toward ester + HCl.
- A base (pyridine, Et₃N, or aqueous NaOH in Schotten–Baumann conditions) is essential — it removes the HCl byproduct, which would otherwise protonate the alcohol and, worse, catalyze reversal or side reactions.
- DMAP (4-dimethylaminopyridine) is a nucleophilic catalyst: it attacks the acid chloride first to form a highly reactive acylpyridinium ion, which is then intercepted by the alcohol — crucial for hindered or unreactive alcohols.
- Rate order across alcohols follows steric approach control: primary > secondary > tertiary, and phenols are markedly slower and often need stronger bases (pyridine, or NaOH/Schotten–Baumann) because ArO⁻ is a weaker nucleophile than RO⁻.
- Anhydrides are milder, more selective acylating agents than acid chlorides — useful when a less violent reaction or a "spent" carboxylate byproduct (rather than corrosive HCl) is preferred.
- Key idea: acyl chlorides and anhydrides acylate alcohols via addition–elimination through a tetrahedral intermediate, and the reaction is driven irreversibly forward because Cl⁻ (or RCO₂⁻) is a dramatically better leaving group than the alkoxide that would have to leave in the reverse direction.
Where this lives today
Acid-chloride and anhydride acylation of alcohols is the backbone of large swaths of practical synthetic chemistry, not a museum-piece method. In pharmaceutical process chemistry, acetylation of hydroxyl groups (for example, converting a diol intermediate into a mono- or di-acetate to mask reactivity during a later step, then removing the acetate by hydrolysis) is done almost exclusively with acetic anhydride or acetyl chloride plus a tertiary amine base, precisely because the reaction is fast, clean, scalable, and — crucially, as shown in Worked Example 3 — does not touch stereocenters. Aspirin itself (acetylsalicylic acid) is manufactured industrially by acetylating the phenolic OH of salicylic acid with acetic anhydride, a direct large-scale application of anhydride acylation chemistry, generally catalyzed by a trace of sulfuric or phosphoric acid or run thermally, illustrating that anhydrides can even acylate phenols efficiently when reaction conditions (heat, acid catalysis) are tuned appropriately. In natural product and fragrance chemistry, DMAP-catalyzed acetylation of hindered terpene alcohols (menthol, borneol, and related cyclic secondary/tertiary alcohols) is standard practice for making the corresponding esters, many of which are themselves valuable fragrance or flavor compounds (menthyl acetate, bornyl acetate) — directly building on Mechanism 2 above. In carbohydrate and nucleoside chemistry, selective acylation (mono-acetylation of a primary hydroxyl in the presence of secondary hydroxyls, exploiting the steric rate differences discussed in the limiting-cases table) is routinely used to install temporary protecting groups, later removed by base-mediated hydrolysis or Transesterification with a different alcohol under basic or enzymatic catalysis. In polymer chemistry, diacid chlorides (e.g., terephthaloyl chloride) react with diols under Schotten–Baumann-type interfacial polycondensation conditions to manufacture polyesters industrially, exploiting exactly the irreversibility argument developed here — HCl is continuously removed by the aqueous base phase, driving the polymerization to high molecular weight. Finally, understanding why this acylation is essentially one-way, in contrast to the genuinely reversible transesterification equilibrium between an ester and a different alcohol, sets up the mechanistic logic needed to analyze that later, closely related transformation, where the same tetrahedral-intermediate machinery operates but now both directions are kinetically accessible because both leaving groups (two different alkoxides) are comparably matched.
Ask the tutor: Why doesn't the ester product get attacked back apart by the chloride ion still floating around in solution?
The physical picture — thorough prose
Picture the carbonyl carbon of an acid chloride, R–C(=O)–Cl. It is -hybridized, trigonal planar, with a π system spanning C=O and a lone pair on chlorine feeding weakly into the empty π orbital of the carbonyl — weakly, because chlorine's 3p orbitals overlap poorly with carbon's 2p orbital, and because chlorine is powerfully inductively withdrawing. The net effect: the carbonyl carbon in an acid chloride is more electrophilic than in an ester or amide, where the attached oxygen or nitrogen donates its lone pair far more effectively into the π system, quenching electrophilicity. This is the same reasoning that ranks acid chlorides at the top of the reactivity ladder in Interconverting Acyl Derivatives Down the Ladder. When an alcohol's oxygen lone pair attacks that carbon, it adds into the π* orbital of the C=O bond, in the classic Bürgi–Dunitz trajectory (nucleophile approaching roughly 105° off the C=O axis, anti to the developing negative charge). The carbon rehybridizes from to ; the π electrons move onto oxygen, generating a tetrahedral alkoxide-oxyanion intermediate that also bears a full positive charge on the newly-attacked oxygen (since it started as neutral ROH). This intermediate is a fork in the road: it can either kick back out the alcohol (reverting to starting material) or eject the chloride, using the developing lone pair on the ring oxygen to re-form the C=O π bond and expel Cl⁻ as a full anion. Because Cl⁻ is dramatically more stable (weaker base, better leaving group — its conjugate acid HCl has ) than an alkoxide leaving group would be, collapse strongly favors chloride expulsion. The product is a protonated ester (an oxocarbenium-stabilized species, i.e., the ester with an extra proton on the carbonyl or ester oxygen depending on how you draw the proton transfer), which a base then deprotonates to reveal the neutral ester. The role of the base is therefore twofold and worth separating clearly. First, mechanically, it deprotonates the oxocarbenium/oxonium intermediate (or the alcohol beforehand, in the Schotten–Baumann variant) so the forward reaction is not stalled by a proton sitting on the product oxygen. Second, thermodynamically and kinetically, it mops up the HCl released, preventing HCl from re-protonating unreacted alcohol (which would kill its nucleophilicity) and preventing acid-catalyzed side reactions such as SN1 ionization of the alcohol (a serious problem for tertiary or allylic/benzylic alcohols) or acid-catalyzed racemization at adjacent stereocenters.
The core ideas — the governing relationships
The overall transformation is:
where: R–COCl is the acyl chloride, R′OH the alcohol, and the base (pyridine, Et₃N, or NaOH) sequesters the HCl byproduct so it cannot reverse or side-react. The rate law for the base-mediated (or DMAP-catalyzed) pathway is third order overall, reflecting the two-step addition–elimination with a base-assisted proton transfer:
where: is an overall rate constant lumping together the nucleophilic addition and general-base-catalyzed collapse/deprotonation steps; when the base is also a nucleophilic catalyst (DMAP), the effective mechanism changes to two consecutive bimolecular steps, but the alcohol addition to the acylpyridinium remains rate-determining for hindered substrates. For the tetrahedral-intermediate partitioning that governs whether collapse expels chloride (forward) or alkoxide (reverse, i.e., simple return to starting materials), a Curtin–Hammett-style competition of leaving-group departure rates applies:
where: and are the rate constants for chloride versus alkoxide expulsion from the same tetrahedral intermediate, and the large positive exponent (chloride's transition state lies much lower in energy, since Cl⁻ is a far weaker base / better leaving group than RO⁻) makes the forward (product-forming) pathway kinetically dominant, which is why the ester, once formed, is not attacked back to acyl chloride under these conditions. For anhydride acylation, the analogous leaving group is carboxylate rather than chloride, and the driving force is smaller but still decisively forward, since carboxylate () is still a far better leaving group than alkoxide ():
where: this ~11 unit gap in conjugate-acid translates (via a linear free-energy relationship between leaving-group and departure rate) into carboxylate leaving roughly – times faster than alkoxide from the same tetrahedral center, cementing irreversibility even though anhydrides are milder acylating agents than acid chlorides.
Goal. Derive, arrow by arrow, why pyridine or aqueous NaOH converts an acid chloride and an alcohol cleanly and irreversibly into an ester, and why the reaction is first order in each of acyl chloride, alcohol, and base.
Step 1 — Nucleophilic addition. The alcohol oxygen's lone pair attacks the carbonyl carbon of R–COCl along the Bürgi–Dunitz trajectory (~105° from the C=O axis), donating into the π* orbital of C=O. The π electrons of C=O shift fully onto the carbonyl oxygen. This converts the trigonal-planar, carbon into a tetrahedral, carbon bearing: the original R group, the original Cl, a new O–R′ bond (with a formal positive charge on that oxygen, since it still carries its proton), and an alkoxide-like oxygen (O⁻) from the former carbonyl.
Step 2 — Base removes the proton (general-base catalysis). Before or concurrently with collapse, the base (pyridine or hydroxide) deprotonates the oxocarbenium oxygen (the R′–O⁺H–C fragment), converting it to a neutral R′O–C alkoxy substituent on the tetrahedral center. This step matters kinetically: without it, the positively-charged intermediate is a worse platform for collapse and more prone to simply falling back apart to reactants (protonated alcohol leaving is not favorable, but the charge distribution still influences partitioning).
Step 3 — Collapse of the tetrahedral intermediate: chloride expulsion. The lone pair on the (now neutral) alkoxy oxygen — or, in resonance terms, the negative charge sitting on the original carbonyl oxygen — pushes back down to re-form the C=O π bond, and in a concerted curved-arrow sense, the C–Cl σ bond breaks heterolytically, ejecting Cl⁻. This step is strongly downhill: the transition state leading to Cl⁻ departure is much lower in energy than the transition state that would eject R′O⁻ (returning starting material), because Cl⁻ is stabilized as a free ion far better than an alkoxide would be, and because the C–Cl bond is intrinsically weaker/more polarizable, making its σ* a better acceptor and its heterolysis kinetically easier.
Step 4 — Proton bookkeeping and product release. The liberated Cl⁻ combines with a proton from protonated base (pyridinium or, in Schotten–Baumann conditions, is simply solvated by water/NaOH), and the neutral ester R–CO–OR′ is released. Because pyridinium chloride (or NaCl + H₂O) is thermodynamically a dead end — chloride will not spontaneously re-add to the now much-less-electrophilic ester carbonyl — the reaction does not run backward.
Step 5 — Why three-component rate law. Because addition (Step 1) requires both R–COCl and R′OH to collide, and productive collapse requires the base to intercept a proton either before or during collapse (Step 2), the observed rate law is third order overall, first order in each component, consistent with kinetic studies of pyridine-catalyzed acylations in aprotic solvents.
Go deeper · Derivation B · 2 minMechanism 2 — DMAP-catalyzed acylation: why the acylpyridinium relay accelerates hindered esterifications
Goal. Show mechanistically why adding a catalytic amount of 4-dimethylaminopyridine (DMAP) dramatically speeds up acylation of hindered secondary and tertiary alcohols that react sluggishly with acid chloride/pyridine alone.
Step 1 — DMAP is the first nucleophile to attack the acid chloride. DMAP's ring nitrogen lone pair (not the exocyclic dimethylamino nitrogen, which is tied up donating into the ring by resonance and is sterically/electronically a poor direct nucleophile at carbonyl carbon) attacks the acid chloride carbonyl in exactly the same addition step as Mechanism 1, Step 1. DMAP is a much better nucleophile toward carbonyl carbon than a simple alcohol because nitrogen is more polarizable and less electronegative than oxygen, and because the pyridine-type lone pair is not engaged in aromatic delocalization (it sits in an orbital in the ring plane, orthogonal to the aromatic π system).
Step 2 — Collapse expels chloride, generating an acylpyridinium ion. The tetrahedral intermediate collapses exactly as before, but now the group that must be pushed off is chloride (again, an excellent leaving group), giving an N-acylpyridinium cation: a pyridinium ring bearing an acyl group on nitrogen, with the para-NMe₂ group strongly stabilizing the positive charge on nitrogen by resonance donation through the ring (push-pull captodative-like stabilization is why DMAP is a far better catalyst than plain pyridine).
Step 3 — The acylpyridinium is a superior electrophile. Counterintuitively, this intermediate is more electrophilic at the acyl carbon than the original acid chloride, because pyridinium nitrogen — now positively charged — is an even worse π-donor into the carbonyl than chlorine was, and it is also an excellent leaving group (pyridine, , is comparable to or better than chloride as a leaving group once protonated pyridinium's stability is considered in this cationic, resonance-stabilized form). This sets up a second, faster addition–elimination cycle.
Step 4 — The alcohol attacks the acylpyridinium. R′OH's oxygen lone pair attacks this activated carbonyl carbon; the tetrahedral intermediate collapses by ejecting neutral pyridine (an excellent leaving group here because it departs as the aromatic, resonance-stabilized free amine, regenerating DMAP for another catalytic cycle) rather than alkoxide.
Step 5 — Catalytic turnover and why this helps sterically hindered alcohols. Because DMAP intercepts the acid chloride before the sluggish alcohol needs to compete for it, and because the actual alcohol-addition transition state (Step 4) is now against a smaller, more reactive, and differently shaped electrophile (a cationic acylpyridinium rather than a neutral acid chloride sitting behind a bulky R group), steric clash in the rate-determining alcohol-addition step is reduced. This is precisely why tert-butanol, neopentyl alcohols, and hindered secondary alcohols that react poorly with RCOCl/pyridine alone acylate efficiently with catalytic DMAP present.
Worked examples
Example 1 — Foundational: acetylation of ethanol by acetyl chloride with pyridine. Acetyl chloride (CH₃COCl) is added dropwise to a solution of ethanol (CH₃CH₂OH, 1.05 equiv) and pyridine (1.2 equiv) in dry dichloromethane at 0 °C. Mechanism walkthrough. Ethanol's oxygen lone pair attacks the acetyl chloride carbonyl carbon (Bürgi–Dunitz trajectory), generating a tetrahedral intermediate bearing –CH₃, –Cl, –O⁻ (former carbonyl O), and –O⁺(H)CH₂CH₃. Pyridine deprotonates the oxocarbenium oxygen (general base). Collapse ejects Cl⁻, re-forming C=O and releasing ethyl acetate, CH₃CO–OCH₂CH₃. Pyridinium chloride precipitates or stays in solution as a salt. Stoichiometric accounting. Using slightly more than 1 equivalent of ethanol and pyridine ensures the limiting reagent is acetyl chloride, driving conversion close to completion; typical isolated yields for this textbook reaction exceed 90%. Limiting case check. If pyridine is omitted entirely, HCl accumulates in solution, protonating unreacted ethanol (converting it to the much poorer nucleophile CH₃CH₂OH₂⁺) and slowing the reaction dramatically — this is the experimental signature that confirms the mechanistic role assigned to base in Step 2 of Mechanism 1.
Example 2 — Applied: esterifying a hindered secondary alcohol, menthol, with acetic anhydride and DMAP (synthesis of menthyl acetate, a fragrance/flavor compound). (–)-Menthol (a secondary, cyclohexane-ring alcohol with significant steric shielding from the flanking isopropyl and methyl groups) is treated with acetic anhydride (1.5 equiv), triethylamine (2 equiv), and a catalytic amount of DMAP (0.1 equiv) in dichloromethane. Why DMAP is needed here. Menthol's hydroxyl sits on a secondary carbon flanked by an adjacent isopropyl-bearing carbon; with acetic anhydride and Et₃N alone (no DMAP), acylation is slow (hours to days) because the direct addition of menthol's oxygen to the anhydride's already only moderately electrophilic carbonyl is sterically disfavored. Adding catalytic DMAP diverts the mechanism through Mechanism 2: DMAP attacks acetic anhydride first (a bimolecular addition–elimination exactly analogous to the acid chloride case, but now ejecting acetate, CH₃CO₂⁻, rather than chloride), generating an acetylpyridinium ion plus free acetate. Menthol then attacks this small, highly electrophilic acetylpyridinium cation — a much less sterically congested transition state than direct attack on the bulkier anhydride — and collapse ejects pyridine (regenerating catalyst), giving menthyl acetate. Quantitative rate comparison (illustrative). Under comparable conditions, second-order rate constants for DMAP-catalyzed transfer to secondary alcohols are typically reported as – times larger than the uncatalyzed (Et₃N-only) pathway — the exact factor depends on solvent and substrate, but the qualitative asymmetric-input lesson holds regardless of the precise number: a catalytic (substoichiometric, ~10 mol %) loading of DMAP produces a rate enhancement far larger than its own concentration, confirming a true catalytic cycle rather than a stoichiometric activation. Limiting case check. As the catalyst loading of DMAP → 0, the reaction reverts to the slow, purely Et₃N/anhydride pathway of Mechanism 1 (with acetate rather than chloride as the leaving group) — DMAP is not required for the reaction to occur at all, only to accelerate it, which is the hallmark of a true catalyst rather than a stoichiometric reagent.
Example 3 — Exam/competition level: chemoselective mono-acylation and stereochemical fidelity at an α-stereocenter. A chiral secondary alcohol, (R)-1-phenylethanol (PhCH(OH)CH₃, an alcohol with a stereocenter directly at the carbinol carbon), is acylated with benzoyl chloride (PhCOCl) using pyridine as base at –20 °C, in the presence of a second, unprotected primary alcohol elsewhere in a more complex substrate (imagine a diol where only the primary OH should react selectively — but here we ask specifically about stereochemical integrity at the secondary, chiral center). Question posed in the exam context: does the acylation proceed with any racemization at the stereocenter, and why (or why not)? Mechanistic reasoning. The stereocenter is the carbinol carbon itself, bearing –OH, –H, –CH₃, and –Ph. In the acylation mechanism, this carbon is never touched: the bond broken is O–H (loss of the alcohol's own proton, taken by pyridine) and the bond formed is O–C(=O)Ph (new ester linkage at oxygen). All bond-making and bond-breaking events occur at oxygen and at the acyl carbonyl carbon of benzoyl chloride — never at the stereocenter's own carbon skeleton. Since no bond to the stereocenter carbon is broken, the configuration at that carbon is completely retained; the product, (R)-1-phenylethyl benzoate, retains the same descriptor (assuming CIP priorities are not reshuffled by the acylation — check separately, since changing OH to OC(=O)Ph could in principle alter substituent priority ranking, though here O remains the highest-priority atom directly attached, so the descriptor is retained). Why this contrasts with Fischer esterification of the free acid: under strongly acidic Fischer conditions with a tertiary or highly stabilized-cation-forming alcohol, competing SN1 ionization at the carbinol carbon is a real racemization risk. Here, under the mild, non-acidic (pyridine-buffered) acid-chloride conditions, no such ionization pathway is available — HCl is captured by pyridine as fast as it forms, so the acidic medium needed to ionize the alcohol never develops. This is the practical reason acid-chloride/base acylation is the method of choice whenever a stereocenter at or near the carbinol carbon must be preserved. Limiting case check. If pyridine were omitted and HCl allowed to build up, and if the substrate were instead a benzylic or allylic tertiary alcohol capable of forming a stabilized carbocation, one would expect partial racemization via reversible ionization — this is exactly why bases are used even when, formally, "only" a proton needs to be removed from the fast-forming product.
Limiting cases, scope & common traps
| Scenario | What happens | Why |
|---|---|---|
| No base present | Reaction slows sharply; HCl builds up, protonates alcohol | Protonated ROH₂⁺ is a far weaker nucleophile; also risks acid-catalyzed side reactions |
| Tertiary or hindered secondary alcohol, no DMAP | Slow, sometimes incomplete acylation | Steric clash in the alcohol-addition transition state to the bulky acyl chloride/anhydride |
| Tertiary alcohol, acidic conditions (no base) | Risk of E1/SN1 side products (alkene, rearranged ether/ester) | Carbocation forms from acid-catalyzed ionization of the alcohol before acylation can occur |
| Phenol as nucleophile | Slower than aliphatic alcohols; often needs NaOH (Schotten–Baumann) | ArO–H is more acidic (better stabilized conjugate base) but ArOH itself is a weaker neutral nucleophile (lone pair delocalized into the ring); using NaOH generates ArO⁻, restoring good nucleophilicity |
| Anhydride instead of acid chloride | Milder, slightly slower, cleaner byproduct (carboxylate salt, not HCl gas) | Carboxylate leaving group less activating than Cl⁻, but still far better than alkoxide |
| α-stereocenter at carbinol carbon | Configuration retained | No bond to that carbon is broken during O-acylation |
| Excess water present | Competing hydrolysis of acid chloride to carboxylic acid | Water is itself a nucleophile and often present in higher effective concentration than the alcohol unless anhydrous conditions are used |
The chemist behind it
The Schotten–Baumann reaction — the biphasic acylation of amines or alcohols with acid chlorides in the presence of aqueous base — is named for two 19th-century German chemists, Carl Schotten and Eugen Baumann, who independently and jointly worked out mild, high-yielding acylation conditions in the 1880s. Baumann's original interest was biochemical: he was studying the acylation of amino acids and related biological amines, and needed a method gentle enough not to destroy sensitive functionality, yet efficient enough to acylate quantitatively in aqueous or mixed aqueous-organic media. Schotten extended and systematized the method for a broader range of substrates. Their combined insight — that adding aqueous base alongside the acid chloride lets acylation and neutralization of HCl happen simultaneously, in a biphasic system where the acid chloride reacts faster with the amine/alcohol than it is hydrolyzed by the surrounding water — remains, essentially unmodified, standard practice in every organic laboratory that runs a Schotten–Baumann acylation today.
Origins · 1 minEugen Baumann and how this was discovered
Why is acylation by acid chlorides considered essentially irreversible, while Fischer esterification is a true equilibrium?
The key difference is the identity of the leaving group that would have to depart in the reverse direction. In Fischer esterification, the forward and reverse reactions both involve alkoxide-type leaving groups (water leaving from the tetrahedral intermediate going forward corresponds structurally to alkoxide-type species leaving in reverse under acid catalysis), so both directions are kinetically similar and the reaction sits at a genuine equilibrium, requiring excess alcohol or water removal to shift it. In acid chloride acylation, the forward-collapse pathway ejects chloride — a vastly better leaving group (its conjugate acid HCl has ) than the alkoxide that would have to leave to regenerate the acid chloride. This asymmetry in leaving-group quality between forward and reverse directions makes the acid chloride pathway effectively one-way under normal laboratory conditions.
Do I need to worry about the alcohol's own O–H bond reacting with the base before it attacks the acid chloride?
Not really, in the sense of it being a problem — in fact, in some mechanistic depictions the alcohol is deprotonated first by the base to generate a more nucleophilic alkoxide before it ever approaches the acid chloride, especially when strong bases like NaOH are used (Schotten–Baumann conditions). With weaker bases like pyridine, the alcohol usually attacks while still neutral, and deprotonation of the resulting oxocarbenium happens after or during collapse. Both pathways converge on the same product and the same overall arrow-pushing logic; the precise timing of the proton transfer relative to C–O bond formation is a mechanistic detail that depends on base strength and solvent, not a source of different products.
Why do we usually run these acylations cold (0 °C or below) rather than at room temperature or with heating?
Low temperature suppresses several unwanted pathways that compete with clean acylation: hydrolysis of the acid chloride by trace water (a bimolecular process whose rate is reduced at lower temperature, though the main protection there is really rigorous exclusion of water rather than temperature alone), acid-catalyzed side reactions if HCl transiently builds up faster than the base can quench it, and — for substrates bearing sensitive stereocenters or acid-labile protecting groups — any adventitious acid-catalyzed epimerization or elimination. Acid chlorides are highly reactive electrophiles, so the intrinsic acylation step itself is fast even at 0 °C; cooling mainly buys selectivity and control rather than being required for the reaction to proceed at all.
How does using an anhydride instead of an acid chloride change the practical picture?
Anhydrides are milder acylating agents (the leaving group, carboxylate, is a weaker leaving group than chloride, though still far better than alkoxide), so reactions are somewhat slower and often benefit from catalysts like DMAP, sulfuric acid traces, or simply heating. The major practical advantages are that anhydrides do not release corrosive HCl gas (instead generating a carboxylic acid or carboxylate salt byproduct, easier to handle on scale), and that many anhydrides (acetic anhydride especially) are cheaper, more stable to storage, and less moisture-sensitive than the corresponding acid chlorides. Aspirin manufacture, using acetic anhydride to acetylate salicylic acid's phenolic OH, is the classic industrial illustration of anhydride acylation chosen over acid chloride acylation for exactly these practical reasons.
Can this method acylate amines as well as alcohols, and does the mechanism change?
Yes — acid chlorides and anhydrides acylate amines by the identical addition–elimination mechanism, and in fact amines are considerably better nucleophiles than alcohols (nitrogen is less electronegative and more polarizable than oxygen, and amine lone pairs are typically less hindered by hydrogen bonding in solution), so amine acylation to form amides is usually faster and often does not even require added base if excess amine is used (the amine itself can serve as the base, absorbing the HCl as ammonium chloride). This is exactly the amide-forming step referenced in the reactivity-ladder framework of Interconverting Acyl Derivatives Down the Ladder, where amides sit below esters because amide nitrogen donates its lone pair into the carbonyl far more effectively than ester oxygen, making the amide product much less electrophilic and the overall conversion strongly downhill.
If I use a huge excess of alcohol, could the ester product be re-acylated or otherwise react further?
Under these conditions, no — the ester product, once formed, is far less electrophilic than the starting acid chloride or anhydride (ester oxygen donates its lone pair into the carbonyl much better than chlorine or a second acyl-oxygen does), so it is essentially inert to further attack by excess alcohol at the temperatures and timescales used for acid chloride/anhydride acylations. This is a different regime from Transesterification, where a different alcohol displaces the original alkoxy group from an ester under acid or base catalysis over longer reaction times or higher temperatures — a genuinely reversible process that requires either a large excess of the incoming alcohol or removal of the outgoing alcohol to drive conversion, unlike the essentially one-way acylation discussed here.
What is the single most common experimental mistake when running this reaction, and how does the mechanism explain the fix?
The most common failure is moisture contamination: trace water in the solvent, glassware, or alcohol reacts competitively with the acid chloride (water is itself an excellent nucleophile toward the same electrophilic carbonyl carbon), converting some of the acid chloride to the free carboxylic acid before the intended alcohol ever gets a chance to react, and lowering yield. The mechanistic fix follows directly from the addition–elimination picture: since the first, often rate-determining step is nucleophilic attack at the same electrophilic carbon by whichever nucleophile is present in highest effective concentration or reactivity, rigorously anhydrous glassware, solvent, and reagents (and, in Schotten–Baumann variants, careful phase separation and rapid stirring) are what make the difference between an excellent and a mediocre yield.