Suitably functionalized pyrroles are the basic skeleton of many biologically important substances and numerous synthetic methods of pyrroles have been investigated extensively. However, the synthesis of pyrrole derivatives from Baylis-Hillman adducts was not developed much. Recently, we reported the synthesis of 2,3,4-trisubstituted pyrroles starting from the rearranged aza-Baylis-Hillman adducts (Scheme 1). Meantime we presumed that we could synthesize 1,2,3,4tetrasubstituted pyrrole derivatives by using the synthetic approach in Scheme 2. As shown in Scheme 2, we imagined that the reaction of Baylis-Hillman acetate 1, as the representative example, and secondary amine derivatives 2a-d could give the corresponding SN2' product 3a-d, which could be cyclized to 4a-d under basic conditions. The following acid-catalyzed dehydration and concomitant double bond isomerization of 4a-d would provide desired pyrroles 5a-d. Among the examined conditions the use of K2CO3 in CH3CN gave the best results for the preparation of 4a-d. As expected we could not observe the formation of 3 (except for 3c, entry 3 in Table 1), instead we obtained 4a-d directly in 50-74% yields as inseparable syn/anti mixtures in a one-pot reaction. Based on the H NMR spectra of 4a-d the ratio of syn/anti was 4:1 to 2:1 (footnotes b-d in Table 1), however, we did not confirm which isomer is the major one. For the reaction of 1 and 2c we isolated 3c in 34% yield (entry 3 in Table 1) together with 4c in 50% yield. For the synthesis of compound 4d (entry 4) we used 2d in slightly excess amount (footnote e in Table 1). The following dehydration step of 4a-d was carried out under the influence of p-TsOH (20-40 mol%) in benzene and we obtained the desired compounds 5a-d in 41-64% yields. Isomerization of double bond occurred during the dehydration stage simultaneously to afford pyrroles directly. The results are summarized in Table 1. However, the reaction of 1 and 2e showed somewhat different reactivity as compared with those of 2a-d (Scheme 3). When we carried out the reaction of 1 and 2e in CH3CN at room temperature the reaction did not show the formation of any new compounds in appreciable amounts presumably due to the limited solubility of 2e in CH3CN. Thus we elevated the temperature to refluxing, however, rearranged acetate was the major product in this case. After many trials we could obtain 3e in 74% yield in aqueous CH3CN at room temperature. In aqueous CH3CN the compound 2e was dissolved completely and the rearrangement of acetate group of 1 to the primary position was minimized at room temperature. With this compound 3e in our hand we prepared 4e under the same conditions of Table 1 (CH3CN,
Read full abstract