Nitriles — Carboxylic Acids & Derivatives

A nitrile has no carbonyl, no leaving group and nothing that looks like the RCO–X skeleton the rest of this chapter is built on. It belongs here anyway, for a bookkeeping reason that turns out to be the practical one.

Same rung as a carboxylic acid

Count the bonds from the nitrile carbon to a heteroatom: three, all of them to the same nitrogen. That is the same count a carboxylic acid carbon carries to oxygen, so the two sit on the same rung of the oxidation ladder — the +3 level shared by the acid, the ester, the amide and the acid chloride. Converting a nitrile into any of them is substitution, not oxidation, and needs no oxidant. Getting off the rung, down to an aldehyde or an amine, needs a reducing agent, exactly as it does for the others.

Where nitriles come from

S_N2 on a primary alkyl halide. RCH₂X + NaCN → RCH₂CN. Cyanide is a strong nucleophile and only a weak base — HCN has a pK_a near 9 — which is the exact combination S_N2 wants. The limits are the usual ones: secondary substrates react sluggishly and give elimination alongside, and tertiary substrates give nothing but alkene.
Dehydration of a primary amide with SOCl₂ or P₂O₅. RCONH₂ → RCN. A sideways move along the rung, as promised.
From a cyanohydrin, made by adding cyanide to an aldehyde or ketone. That route puts an OH on the carbon next door, and the carbonyl chapter covers it.
From an aryl diazonium salt with CuCN, the Sandmeyer reaction. Electrophilic substitution cannot install a CN group at all — there is no electrophilic cyanide reagent — so an aryl nitrile is always made some other way. Sandmeyer is the usual one; dehydrating an aryl amide is another, and an activated aryl halide will take cyanide by S_NAr.

What a nitrile does

Reagent	Product	Through
H₃O⁺ or HO⁻, heat	RCOOH	the amide
LiAlH₄, then H₂O	RCH₂NH₂	full reduction
DIBAL-H, 1 equiv, −78 °C, then H₃O⁺	RCHO	an imine anion
R′MgX or R′Li, then H₃O⁺	RCOR′	an imine

Hydrolysis runs under aqueous acid or base with heat, and it goes through the amide on the way. That intermediate is real enough to catch: milder conditions stop the reaction at RCONH₂.

LiAlH₄ delivers two hydrides and gives a primary amine. Watch the carbon count: the nitrile carbon is the one that becomes the CH₂, so RCN gives RCH₂NH₂ with no carbon gained or lost. This is the row in the reduction table that sits beside the amide row, and for the same reason — nitrogen stays attached.

DIBAL-H, one equivalent at low temperature, stops one hydride earlier. The product of that single addition is a metalated imine anion, and an imine anion is not an electrophile, so it simply sits there until water is added. Hydrolysis on workup gives the aldehyde. The contrast with LiAlH₄ is the same one the reduction section draws for esters: the reagent is crippled, not rationed.

A Grignard or organolithium adds once and cannot add twice, because the intermediate here is an imine salt rather than a carbonyl. Aqueous workup hydrolyzes it to a ketone. That is the whole contrast with an ester, which expels alkoxide to a ketone that is hungrier than the ester was and gets attacked again; a nitrile never generates a carbonyl until the Grignard is gone.

One carbon, four destinations

One starting material, four destinations — only three of them a step off the rung, since hydrolysis to the acid is a sideways move. The carbon skeleton is the thing to watch. Only the Grignard row changes the carbon count, because only there does a new group arrive; the other three rearrange what the nitrile carbon already had.The two reductions differ for a structural reason rather than a stoichiometric one. DIBAL-H adds a single hydride and the product of that addition is a metalated imine anion, which is not an electrophile, so nothing further can attack it however long you wait — the aldehyde appears only when water hydrolyzes it on workup. LiAlH₄ is not stopped by anything, so rationing it to one equivalent does not give you an aldehyde; it gives you a mixture. The reagent is crippled, not rationed.

Put the four together and the nitrile is a one-carbon extension that can be cashed out as an acid, an amine, an aldehyde or a ketone. Combined with the S_N2 route, RX → RCN → RCOOH is the standard way to lengthen a chain by one carbon and finish with an acid.

The Grignard route — RX → RMgX, then CO₂, then H₃O⁺ — reaches the same product with the same one added carbon, and the two are complements rather than duplicates. The cyanide route demands a primary, unhindered substrate but tolerates acidic protons — an O–H or an N–H does not touch it. It is not a general tolerance: cyanide is a nucleophile, so a ketone elsewhere in the molecule becomes a cyanohydrin, which is one of the ways to make a nitrile listed three bullets above. The Grignard route tolerates a hindered or tertiary substrate but is destroyed by any O–H, N–H or carbonyl in the molecule. Look at the substrate, not at the target, to choose.

Spotting one

The C≡N stretch appears near 2250 cm⁻¹, sharp and medium. The 2500–2000 window is otherwise close to empty, which is what makes the peak diagnostic.