In addition to a potential excessive level of heavy metals in 1 carried over from the Negishi reaction, the kilo-lab runs were faced with some residual yellow-to-brown discoloration carried over from the Friedel−Crafts acylation step. To address the color issue the workup procedure after the final coupling reaction was modified to include treatment with active carbon.
Thus, the solid material obtained after precipitation of the product with ethanol was redissolved in DCM and treated with Darco KB at 35 °C. After filtration through a Celite bed, the DCM solution was washed with 5% aqueous ammonium chloride to remove residual DMF.The supporting information has more details, including the dissolution of the crude material in 80 liters of dichloromethane and the addition of 2.3 kilograms of Darco KB, stirring for an hour at 30°C (is that all it took?), followed by filtration through a bed of Celite. Man, that must have been a mess... Glad it worked for the authors!
[Incidentally, this article is an interesting example of the non-profit pharma model at work, in that it is a BMS compound, licensed to an organization (the International Partnership for Microbicides) for use in the developing countries.]
1. Pikul, S.*; Cheng, H.; Cheng, A.; Huang, C.D.; Ke, A.; Kuo, L.; Thompson, A.; Wilder, S. "Synthetic Process Development of BMS-599793 Including Azaindole Negishi Coupling on Kilogram Scale." Org. Process Res. Dev. ASAP, dx.doi.org/10.1021/op400012p