The pun orgaNOmetallic chemistry in the intro text is courtesy of Nicholas Leadbeater who is discussing the merits of metal-free coupling reactions in an essay accompanying a research article by Sun et al. in Nature Chemistry (DOI). An earlier blog introduced Leadbeater advocating a metal-free Sonogashira reaction back in 2003. But these so-called metal-free reactions raise suspicions because minute quantities of metal may be present in any solvent or reagent , see for example the findings of Buchwald & Bolm in another blog. In his essay Leadbeater explains the preparations for the 2003 experiment were meticulous with an (unfortunate) student performing the reaction quarantined (but for how long!) in demetallized laboratory fitted with new glassware and fresh reagents. But when the lab moved from England to the US the reaction started to behave oddly and eventually traces (50ppb) of palladium were found lurking in the sodium carbonate.
Nevertheless Leadbeater fully supports the new Sun et al. research. They reacted bromoanisole with benzene (the solvent) in the presence of potassium tert-butoxide and a catalytic amounts of phenanthroline. Traces of Pd, Cu and Fe (10 ppb - 10 ppm) were found in the reagents using ICP-MS but deliberately adding these metals in larger quantities to the reaction did not improve conversion. They also had the reaction performed in other laboratories just to be sure.
Also in 2010 Liu et al. (DOI) used potassium tert-butoxide and another diamine N,N-dimethyl-ethylenediamine in a very similar reaction. Earlier in 2008 Yanagisawa et al. had already noted that nitrogen heterocycles could be coupled with just the base (DOI) In all reports the proposed mechanism is based on a aryl radical anion intermediate. In the Sun report the amine and butoxide base can be replaced by trusted radical initiators azobisisobutyronitrile and tributyltin hydride.