Time is the variable that decides the outcome
The competing pathways have very different time behavior.
- The Li–Br exchange is fast and reversible. It reaches its equilibrium position within minutes at low temperature, and adding more nBuLi only shifts that position; it does not remove butyl bromide from the flask.
- Addition to the metal carbonyl, alkylation of the carbene carbon, and attack on the ethereal solvent are slow but essentially irreversible. They convert starting material and intermediate into products that cannot revert.
Stirring overnight does not change the first process, because it is already at equilibrium. It gives the second group of processes many more hours to run.
What accumulates over the night
- Butylated carbonyl-derived products from nBuLi addition to coordinated CO.
- Butyl–carbene adduct, the same undesired product class you were trying to eliminate.
- Solvent-derived enolates and ring-opened ether fragments from nBuLi attacking THF, together with butane as the protonated by-product.
- Loss of active nBuLi, so the reagent titer you planned on is no longer there when you add the electrophile.
The practical consequence
By the time the mixture is worked up, the aryllithium and the carbene have both been partly consumed, the butyl bromide is still present because its exchange is reversible, and the side products have had hours to build. The overnight stir therefore trades a small, reversible equilibrium effect for a large, irreversible loss of the desired intermediate.
What this rules out
This is why the substitution cannot be treated as a simple reagent swap. The problem is not that nBuLi is too weak to react with butyl bromide; it is that nBuLi reacts with everything else as well, and the unwanted reactions are the ones that accumulate with time. Any viable solution has to remove butyl bromide without introducing a free butyl nucleophile into the mixture.