Effect of stannous octoate on the composition, molecular weight, and molecular weight distribution of ethylene glycol-initiated poly(ε-caprolaccone)

Bulk polymerizations of epsilon-caprolactone (CAP) were conducted at 120 degrees C, with and without ethylene glycol (EG) as initiator, using varying concentrations of stannous octoate (SO) as catalyst. In the presence of EG, GPC showed that molecular weight (MW) was determined by the [CAP]/[EG] ratio and not by [SO] or the concentration of adventitious water brought into the reactor via the catalyst. Without EG, MWs were higher but decreased as [SO] was increased; polymerization rates were also lower, and the relationship between MW and conversion suggested the participation of both ring-opening and condensation polymerization. Late EG addition at high CAP conversion reduced MW to a level that was consistent with the [CAP]/[EG] ratio, regardless of the MW of the initial water-initiated polymerization. To aid in structural analysis of the polymers by NMR spectroscopy, two model systems were synthesized using [CAP]:[EG]:[SO] ratios of 0:2:1 and 4:2:1. Additional polymerizations were conducted at 80 degrees C using anhydrous EG as initiator; three were charged with varying amounts of water (0.5, 1, and 2 moles/mole EG), three with EHA (0.025, 0.05, and 0.1 mole/mole EG), and one received no additional reagent. Monomer conversion (by H-1 NMR) vs, time curves showed that all three concentrations of water depressed the rate of polymerization relative to the control; however, the rate had begun to rise for the highest concentration of water. GPC indicated lower MW with increasing water, suggesting the rise in rate was caused by increasing total initiator (EG + H2O). All three concentrations of EHA depressed the rate relative to the control, and again the rate had begun to rise for the highest concentration. GPC indicated essentially no effect of EHA concentration on MW. Matrix-assisted laser desorption/ionization time-of-light mass spectroscopy confirmed that polymerizations conducted in the presence of water consisted of two separate distributions: difunctional chains of higher average MW containing an EG residue, and monofunctional chains of lower average MW containing a water residue (carboxylic acid end group). The control displayed a small fraction of water-initiated chains, which increased as the amount of added water increased. The presence of EHA increased the fraction of water-initiated chains relative to the control, but the concentration of EHA did not affect the fraction of water-initiated chains.