Cyclopalladation in pyrroles-some initiating studies

Abstract


Introduction
Cyclopalladation 1 of arenes implies participation by an ortho co-ordinating substituent such as aldehyde, ketone, amide, hydrazino, aminoalkyl, azo, imine, and iminoether, thiazole, or thioalkyl [1][2][3][4] in metallation of the arene.This process then, is a regioselective method for introduction of the metal, with replacement of hydrogen, generating intermediates which can be formulated generally as 1, and which are available for further manipulations involving use of the properties of the organometallic species.
There seem to be few examples of the use of such processes in heteroaromatic chemistry: Grigg demonstrated the formation and use of complexes derived from 3-acetyl-and 4acetylpyridines by reaction with RuH2CO(PPh3)3, for 4-and 3-alkylations, respectively, with alkenes, and extrapolated this to the 4-alkylation of 3-benzoylpyridine and the 2-alkylation of 3acetylindole. 5 To our knowledge, cyclopalladation has been used only once in pyrrole chemistry: 2-dimethylaminomethyl-1-phenylsulfonylpyrrole 2 reacted with lithium tetrachloropalladate(II) to give a solid material assigned structure 3 on the basis of spectroscopic and combustion analytical data, and of its trapping with carbon monoxide/methanol giving ester 4 (Scheme 1). 6heme 1

Result and Discussion
Arising from our interest in the synthesis of fused polycyclic molecules containing pyrrole and/or pyrrolidine rings 7 it occurred to us that the nitrogen in cyclic imines 8 such as 5 might also be capable of promoting cyclopalladation; there are no examples of the use of cyclic imines in such processes.Unfortunately, for our aims, reaction of 5 with lithium tetrachloropalladate(II) produced 6 (Scheme 2), as evidenced by spectroscopic data and confirmed by an X-ray crystal structure determination.Chem3D representations of the solid state structure are shown below, from perspectives chosen to show the geometry around the metal (Figure 1) and the extensively pyramidalised pyrrole nitrogen (sum of bond angles only 349.9 o ).We have previously commented in detail on the pyramidalisation of nitrogen in some N-arylsulfonylpyrroles andindoles. 9However, on reaction of 5 with palladium(II) acetate, the desired type of complex, 7 was obtained in good yield; treatment of 7 with sodium chloride gave the corresponding chlorobridged complex 8 quantitatively (Scheme 3).The regiochemistry of metallation was easily confirmed by nOe experiments involving the remaining, adjacent pyrrole protons.

Scheme 3
The palladium complex 8 behaved just as had been hoped in reaction with bromine and iodine, giving the 3-halogenated derivatives 9a and 9b smoothly and in high yields (Scheme 4).

Scheme 4
In a limited exploration of the further applicability of the cyclopalladated complex 7, it was shown that 7 could be brominated giving 9a 88% yield, acetylated in moderate yield giving 10 and that reaction with tetrabutylammonium cyanide 3 led to the formation of a cyano derivative 11 (Scheme 5) though all attempts to remove residual tetrabutylammonium residues failed (in another context we have encountered difficulties in removing residual tetrabutylammonium fluoride from reactions involving silyl deprotection of heterocycles).No evidence for the formation of 11 was obtained on treatment of 7 with copper(I) cyanide.No useful products were obtained from 7 with cyanogen bromide, methyl vinyl ketone, styrene, benzoyl chloride, allyl bromide, ethyl chloroformate, or cyclohexyl isocyanide.Even with the limited range of successful substitutions described above, we believe that this route is capable of further development into a useful method for the preparation of 2,3-disubstituted pyrroles.

Scheme 5
Having in hand bromide 9a we examined the possibility of utilising the halogen in the formation of a third fused five-membered ring.Imine 9a was reacted with ethyl chloroformate and then, without isolation, the presumed iminium salt 12 treated with sodium cyanoborohydride.A mixture of 13 and 14 was obtained, from reduction and deprotonation, respectively (Scheme 6).Exposure of 13 to t-butyllithium gave the target tricyclic material 15 though only in moderate yield, accompanied by a compound molecular weight 356, into which a t-butyl group had been incorporated, and from which the oxygen had been lost-spectroscopic data are consistent with structure 16 (Scheme 7).In agreement with this interpretation is the finding that amide 15 was quantitatively converted into 16 on exposure to t-butyllithium at 0 o C.