
The group of Buchwald at MIT has now published full details of their research on the use of an inexpensive copper catalyst for amidation of aryl halides – the Goldberg reaction (Klapars A et al, J Am Chem Soc, 2002, 124, 7421).
The choice of ligand (a diamine such as NN dimethyl ethylene-diamine, or NN dimethyl cyclohexane diamine) is crucial as is the choice of copper salt – the air stable copper iodide gives best results with a wide range of substrates although cuprous oxide and chloride also give acceptable results for some substrates.
The choice of base is the most important factor, however. For aryliodides potassium phosphate is the best base, whereas for iodides a weaker base such as potassium carbonate works best. An interesting experiment using KHMDS gave clues about the reaction mechanism. If the KHMDS is added all at once, the conversion is below 1%, whereas if it is added slowly a 92% yield is observed. This indicates that the rate of deprotonation of the amide has to match the rate of the amidation (this would be a good way to scale up the process too!). The mechanism is shown below – with excess deprotonated amide an unreactive cuprate complex is formed.















