It was in the fall of 1973 that Geoffrey Wilkinson made one of his too-infrequent visits to the Central Research Department at E.I. duPont de Nemours and Company. On this occasion he announced that he had prepared hexamethyltungsten. At the time I was interested in adding the dianion of cyclooctatetraene to early transition metals in order to provide complexes of potential catalytic interest [I], and in the process had rediscovered Juvinall’s preparation of TaMqCl, [2]. Since the C,H, complexes I obtained did not appear to be potentially interesting catalysts, I decided to try to prepare pentamethyltantalum by adding methyllithium to TaMe,Cl,. Pentamethyltantalum, like hexamethyltungsten, turns out to be an unpredictable and potentially vicious material when isolated as a solid, owing to its ready decomposition to give methane and (under some conditions) hydrogen 131. Since Zr(CH,Pli), and Ti(CH~Ph)~ had been known for several years [4], I was not surprised to find that pentabenzyltantalum could be prepared, and that it was a relatively stable red crystalline species [3b,S]. I was surprised when I attempted to extend the list of pentaalkyls to pentaneopentyltantalum. Trineopentyltantalum dichloride was prepared straightforwardly, and upon addition of 2 equivaIents of neopentyllit~um to it in pentane a lovely orange solution was produced, from which a low melting orange crystalline product could be isolated virtually quantitatively [6a]. Its mass spectrum revealed that it was not pentaneopentyltantalum, but a compound whose mass was consistent with the formulation T~(CHB~‘)(CH,BU’)~. Paul Meakin was kind enough to run a 13C NMR spectrum of Ta(CHBut)(CHzBu’), that same night; I asked him to be sure to extend the window to 300 ppm downfield of TMS. He called the next morning to tell me that there was indeed a most peculiar doublet located at 250 ppm with a coupling constant of only 90 Hz. Since this was the first “carbene” complex that did not contain some sort of stabilizing substituent(s) (usually 0 or N) [7], I did not think the low “olefinic” carbon-hydrogen coupling constant was especially significant. Extensive studies did not result in any firm conclusions regarding the mechanism of formation of TatCHBut)(CH*But)~, and it is still uncertain whether Ta{CH,Bu’),Cl spontaneously loses neopentane or is dehydrohalogenat~ by a fifth equivalent of a lithium reagent to give the final product [6b]. The salient feature of the reaction in either case is removal of an a-hydrogen from one neopentyl group by another
Read full abstract