Carboxylic Acids

Formic acid takes its name from the Latin for ant, formica, since it was first isolated from exactly that source; acetic acid takes its own name from acetum, vinegar; and butyric acid, found in rancid butter, from butyrum. The heavier fatty acids, twelve to eighteen carbons long, occur throughout nature as glycerol esters inside natural fats, and hexanedioic acid, adipic acid, is a starting material for nylon-6,6, one of the most widely manufactured synthetic fibres in the world. Every one of these compounds carries the same functional group, the carboxyl group, -COOH, literally a carbonyl group fused onto a hydroxyl group on the very same carbon, and despite looking, on paper, like the simple sum of two functional groups already covered in this thread, it behaves as something genuinely new: more acidic than either an alcohol or a phenol, and reactive in ways neither one alone can manage.

Common names, given to many carboxylic acids long before any systematic naming scheme existed, still carry the suffix -ic acid and trace back to Latin or Greek names for their natural source. IUPAC names replace an alkane's own final -e with -oic acid, numbering always starting from the carboxyl carbon itself, and a molecule carrying more than one -COOH group adds the multiplicative prefix dicarboxylic, tricarboxylic and so on to the parent chain's own name, with the -COOH positions marked by locants before that prefix. Structurally, the bonds around the carboxyl carbon all lie in one plane, separated by roughly 120 degrees, exactly the trigonal geometry already established for a plain carbonyl group; but the carboxyl carbon itself is measurably less electrophilic than an ordinary carbonyl carbon, since the hydroxyl oxygen's own lone pair donates into the carbonyl system by resonance, spreading positive character away from the carbon and softening its own electrophilicity, the first hint that a carboxylic acid's chemistry is never quite the same as either a plain carbonyl's or a plain alcohol's alone.

Primary alcohols and aldehydes both oxidise readily to carboxylic acids, using potassium permanganate in neutral, acidic or alkaline medium, or potassium dichromate and chromium trioxide in acidic medium, the specific combination of CrO3 with dilute sulphuric acid known as Jones reagent. Alkylbenzenes supply an aromatic route: vigorous oxidation with chromic acid or potassium permanganate burns away the entire side chain, however long, down to a single carboxyl group directly on the ring, primary and secondary alkyl groups both oxidising this way while a tertiary group, with no benzylic hydrogen for the oxidant to grab hold of, survives untouched. Nitriles hydrolyse to carboxylic acids by way of an amide intermediate, mild conditions stopping the reaction cleanly at that amide stage if desired, and Grignard reagents react directly with carbon dioxide, dry ice, to form a carboxylate salt that acidification converts to the free acid, a route that, together with the nitrile route, is specifically useful for extending an alkyl halide's own carbon chain by exactly one carbon. Acyl halides and anhydrides hydrolyse straightforwardly to carboxylic acids with water, more readily still with aqueous base followed by acidification, and esters hydrolyse the same way, acidic conditions giving the acid directly, basic conditions giving the carboxylate salt first.

Carboxylic acids up to nine carbons are colourless liquids with genuinely unpleasant odours; heavier ones are practically odourless, waxy solids, their low volatility keeping any smell from reaching the nose at all. Their boiling points sit above even a comparable alcohol's, a claim worth pausing on, since alcohols already boil unusually high: a carboxylic acid molecule forms two hydrogen bonds at once, to a second acid molecule on either side, associating into genuine dimers so stable that they survive not just in the liquid but into the vapour phase and in non-hydrogen-bonding solvents too. Small carboxylic acids, up to four carbons, mix with water in every proportion by hydrogen-bonding directly with it, a solubility that fades steadily as the hydrocarbon chain grows and its own hydrophobic character starts to dominate, benzoic acid, the simplest aromatic carboxylic acid, ending up nearly insoluble in cold water despite carrying the same -COOH group every soluble small acid does.

Carboxylic acids behave like both alcohols and phenols when it comes to electropositive metals, releasing hydrogen gas and forming a salt, and like phenols when it comes to alkalies, forming a salt and water; but unlike either one, they react even with weak bases like carbonates and hydrogencarbonates, fizzing out carbon dioxide, a genuinely useful diagnostic reaction for confirming a carboxyl group's presence in an unknown compound. Dissolved in water, a carboxylic acid dissociates into a resonance-stabilised carboxylate anion and a hydronium ion, and the equilibrium constant for that dissociation, the acid dissociation constant Ka, is conventionally reported instead as pKa, the negative logarithm of Ka, purely for convenience: the smaller the pKa, the stronger the acid, strong acids sitting below 1, moderately strong acids between 1 and 5, weak acids between 5 and 15, and extremely weak acids above 15. On this scale, carboxylic acids land as genuinely strong performers among organic compounds, stronger than water, stronger than phenols, and very much stronger than alcohols, acetic acid's own pKa of 4.76 sitting nowhere near ethanol's 15.9 or even phenol's 10.

This looks, at first glance, like it should run the wrong way: a phenoxide ion spreads its negative charge across five resonance structures, more than a carboxylate ion's own two, and yet carboxylic acids are still measurably more acidic than phenols. The resolution lies in quality, not quantity, of resonance: a carboxylate ion's two resonance structures are exactly equivalent, the negative charge sitting on one highly electronegative oxygen atom in one structure and an equally electronegative oxygen atom in the other, both structures equally low in energy and contributing equally to a genuinely stabilised, delocalised hybrid. A phenoxide ion's several resonance structures, by contrast, are not equivalent to each other at all: only one places the negative charge on oxygen, while the rest place it on ring carbons instead, atoms considerably less electronegative and therefore far less comfortable holding a negative charge, so those carbon-centred structures contribute much less stabilisation than the single oxygen-centred one does. A charge shared equally between two electronegative oxygens beats a charge mostly stuck on one oxygen and occasionally smeared onto a less electronegative carbon, which is exactly why the carboxylate ion ends up more stable, and carboxylic acid more acidic, despite phenoxide's own larger resonance structure count.

An electron-withdrawing group anywhere near the carboxyl group stabilises its own conjugate base further, by inductive or resonance delocalisation of the negative charge, and so strengthens the acid; an electron-donating group does the reverse, destabilising the conjugate base and weakening the acid. Ranked by this inductive effect alone, phenyl pulls least and trifluoromethyl pulls most, giving the increasing order phenyl, iodine, bromine, chlorine, fluorine, cyanide, nitro, trifluoromethyl, and trichloroacetic, dichloroacetic and trifluoroacetic acid all rank among the strongest simple carboxylic acids known specifically because of this cumulative inductive pull. Direct attachment of a phenyl or vinyl group to the carboxyl carbon is a genuine exception worth flagging: naive resonance reasoning predicts this should weaken the acid, by donating electron density into the carboxyl group, yet it measurably strengthens it instead, since the sp2 hybridised carbon doing the attaching is itself more electronegative than an ordinary sp3 carbon, and that electronegativity effect wins out over the resonance donation. Substituents on an aromatic ring follow the same electron-withdrawing-strengthens, electron-donating-weakens logic directly: para-nitrobenzoic acid, with an electron-withdrawing nitro group, carries a lower pKa than plain benzoic acid, while para-methoxybenzoic acid, with an electron-donating methoxy group, carries a higher one.

Heated with a mineral acid like sulphuric acid or with phosphorus pentoxide, a carboxylic acid loses water and forms the corresponding anhydride; heated instead with an alcohol or phenol under an acid catalyst, it forms an ester, through a mechanism worth following closely since it recurs throughout carbonyl chemistry: protonation of the carbonyl oxygen activates the carbon toward nucleophilic addition by the alcohol, a proton transfer within the resulting tetrahedral intermediate turns the original hydroxyl into a genuinely good leaving group, water, which departs to leave a protonated ester that finally loses its own proton to give the neutral ester product. Phosphorus pentachloride, phosphorus trichloride and thionyl chloride all swap a carboxylic acid's own -OH for chlorine, exactly the same substitution already established for alcohols, and thionyl chloride remains the preferred reagent for exactly the same reason: its two by-products are both gases that simply escape, leaving a clean acid chloride behind. Ammonia reacts with a carboxylic acid too, forming an ammonium salt first, which only further heating at high temperature converts into an amide, water lost along the way.

Reduction of a carboxylic acid takes real strength: lithium aluminium hydride does the job, though diborane manages it better still, since diborane leaves other reducible groups like esters, nitro groups and halogens completely untouched along the way, and sodium borohydride, notably, cannot reduce a carboxyl group at all, a genuine contrast worth remembering against every other reduction already established in this thread. Decarboxylation removes the carboxyl group entirely: heating a carboxylic acid's sodium salt with sodalime, a three-to-one mixture of sodium hydroxide and calcium oxide, releases carbon dioxide and leaves a hydrocarbon behind, while electrolysing an aqueous solution of the same salt achieves a related but distinct outcome, Kolbe electrolysis, joining two decarboxylated fragments together into a single hydrocarbon carrying twice the original alkyl group's own carbon count. A carboxylic acid's own alpha-carbon, meanwhile, can be halogenated directly: chlorine or bromine, with a trace of red phosphorus as catalyst, substitutes specifically at that alpha position, the Hell-Volhard-Zelinsky reaction. And on an aromatic ring, the carboxyl group behaves as a deactivating, meta-directing substituent in electrophilic substitution, exactly the same category the carbonyl group of an aldehyde or ketone already fell into, with one further restriction unique to carboxylic acids: Friedel-Crafts reactions simply do not proceed on them at all, since the Lewis acid catalyst, aluminium chloride, ends up binding to the carboxyl group itself rather than activating anything else on the ring.