Subsequently, a systematic ligand screening was carried out. We were pleased to find that bisphosphine ligands exhibited significant importance in adjusting reactivity and molecular weight (
Table 2). It was found that
L2–L5 gave the desired product
2a in good yield (entries 2–5). In contrast to
L2, L3 exhibited higher molecular weight, probably due to the more steric hindrance of ferrocene skeleton (entries 2 and 3). When
L4 and
L5 were used as ligand, they showed the similar activity, while the more rigid skeleton
L4 showed higher molecular weight and better PDI (entries 4 and 5). When
L6 was used, the Mn was sharply decreased to 6700 (entry 6). Considering the molecular weight and yield,
L4 was regarded as the best ligand for this reaction, furnishing the desired product in 97% yield with Mn value of 25, 400 and PDI of 1.77 (entry 4). Considering the effect of temperature on the reaction, we tried to change the temperature in order to improve the Mn and PDI (entries 7 and 8). Disappointedly, the reaction provided the corresponding product with better PDI but lower Mn when the reaction temperature decreased (entry 7). Elevating temperature, the reactivity decreased compared to 80 ℃ (entry 8). Afterwards, catalyst loading was performed carefully. The catalyst loading had obvious influence on the value of Mn (entries 9 and 10). Regardless of whether CH
3CN was decreased or increased, the most suitable solvent loading was 1.5 mL in view of yield, molecular weight and PDI (entries 11 and 12). When reaction time was prolonged to 48 h, polymer with better PDI was given, nevertheless, the yield of
2a dropped (entry 13). Thus, the optimized reaction conditions were established: CuCl (2 mol%), NaO
tBu (2 mol%),
L4 (2 mol%), CH
3CN (1.5 mL), 80 ℃, 24 h.