Cuprates and conjugate addition — Organometallics

A Gilman reagent, or lithium dialkylcuprate, is made by treating two equivalents of an organolithium with one of a copper(I) salt:

2 R–Li + CuI → R₂CuLi + LiI

Copper is far closer to carbon in electronegativity than lithium is, so the C–Cu bond is much less polar and the reagent is much less reactive. That sounds like a loss. It is the opposite: three things become possible that a Grignard cannot do, and all three are consequences of the reagent being weaker.

One: conjugate addition

An α,β-unsaturated carbonyl offers two electrophilic sites — the carbonyl carbon and the β carbon — and which one a nucleophile picks is the standard hard/soft question.

Nucleophile	Adds	Product
RMgX or RLi	1,2 — at the carbonyl carbon	Allylic alcohol
R₂CuLi	1,4 — at the β carbon	Ketone with a new alkyl group at β

The hard, charge-dense Grignard goes for the site with the most positive charge, which is the carbonyl carbon. The soft, polarizable cuprate goes for the site with the largest orbital coefficient, which is the β carbon. Adding 1,4 gives an enolate first, and that enolate picks up a proton on workup to give the ketone back with the new group installed at β.

This is the single most useful thing in the section, because it means the choice of reagent chooses the product. Same substrate, same R group, two different compounds, and the choice is the metal rather than the conditions. The honest qualifier is that a catalytic copper salt added to a Grignard switches it to 1,4 as well — which is the same rule arriving by another route, since what adds is a cuprate either way.

Two: coupling with alkyl halides

One enone and two nucleophiles carrying the same R group. The hard, charge-dense Grignard adds at the carbonyl carbon; the soft, polarizable cuprate adds at the β carbon, and the enolate it forms protonates on workup to give the ketone back with the new group installed.The carbonyl carbon still carries the larger partial positive charge — that is exactly why the Grignard picks it, and why saying the cuprate “prefers the more electrophilic site” gets the reasoning backwards. The metal is the dominant factor rather than the only one — a catalytic copper salt added to a Grignard will switch it to 1,4, which is the same argument arriving by a different route. What does not move it is temperature.

A Grignard and an alkyl halide in the same flask do not reliably couple. A Gilman reagent and an alkyl halide do:

R₂CuLi + R′–X → R–R′

Only one of the two R groups is transferred; the other is spent. The halide should be methyl, primary, vinyl or aryl — and the last two are the point, because vinyl and aryl halides are inert to S_N2 entirely. A cuprate will couple to them with the double-bond geometry preserved, which no substitution reaction can offer.

Three: stopping at the ketone

The ester problem from the Grignard section — add once, get a more reactive ketone, add again — does not arise here. A Gilman reagent converts an acyl chloride to a ketone and stops, because it is not reactive enough to attack the ketone it just made.

All three abilities have the same cause, and it is worth stating plainly: the reagent's low reactivity is not a limitation it works around but the property being used.

Selectivity as a general principle

Because a cuprate is mild, it tolerates functional groups a Grignard would attack — esters, nitriles, and in many cases a ketone elsewhere in the molecule. When a synthesis needs one site touched in a molecule that has several, the answer is normally a less reactive reagent rather than cleverer conditions.

Organozinc reagents (R–ZnX) make the same trade further still. They are mild enough to be prepared and used in the presence of an ester, which is why they show up in the palladium chemistry of the next section, where the organometallic has to survive a catalytic cycle.

What carries forward

A Gilman reagent is the selective carbon nucleophile: 1,4 to an enone where a Grignard goes 1,2, coupling to vinyl and aryl halides where nothing else will, and one addition to an acyl chloride where a Grignard does two. Whenever a question gives you an α,β-unsaturated ketone and asks which product you get, the first thing to check is which metal is on the nucleophile.