Alternating copolymerization of epoxides and cyclic anhydrides: an improved route to aliphatic polyesters.

Aliphatic polyesters constitute an important class of polymers because of their biodegradability1,2 and biocompatability,2,3 which enable their use in drug delivery systems,2 artificial tissues,3 and commodity materials.4 Polyesters such as poly(butylene succinate) are commonly produced through condensation polymerization; however, this method is energy intensive, requiring high temperature and removal of the alcohol or water byproduct to achieve high molecular weight (Mn) polymers.5 Conversely, poly(hydroxyalkanoate)s can be synthesized through bacterial fermentation, a process which is also energy intensive.6 Alternatively, polyesters such as poly(lactic acid) (PLA) and poly( -caprolactone) may be prepared by the ring-opening polymerization of cyclic esters.1c,2,4,7 Although this technique uses mild reaction conditions and avoids the formation of small molecule byproducts, the scope of the polymer architecture is generally limited by the availability of structurally diverse monomers. A different approach, the ringopening copolymerization of epoxides and cyclic anhydrides, has the potential to produce a wider variety of polymer backbone structures.8 However, catalysts reported for this reaction exhibit relatively low activities8a,b,e and produce polyesters with low Mn values.8a-d We have previously reported highly active (BDI)ZnOAc (BDI ) â-diiminate) catalysts for the copolymerization of epoxides and CO2. Additionally, we have shown that (BDI)Zn-alkoxide complexes polymerize lactones and lactides by acyl bond cleavage.10 On the basis of these prior results, we anticipated that (BDI)Zn complexes might serve as active catalysts for the ring-opening copolymerization of epoxides and cyclic anhydrides (Scheme 1). Herein, we report (BDI)ZnOAc catalysts for the synthesis of new aliphatic polyesters with high Mn values and narrow molecular weight distributions (MWD ) Mw/Mn) via the highly alternating copolymerization of epoxides and cyclic anhydrides under mild reaction conditions. Initially, we focused on the copolymerization of diglycolic anhydride (DGA) and cyclohexene oxide (CHO) using 1, a catalyst previously shown to be active for CHO/CO2 copolymerization9b (Scheme 1). Much to our dismay, we were unable to obtain poly(cyclohexene diglycolate) regardless of the reaction conditions (Table 1, entry 1). Investigation of the stoichiometric interaction of 1 with DGA using 1H NMR spectroscopy revealed nearly complete degradation of 1 after 1 h at 25 °C. We hypothesize that DGA reacts with the BDI ligand at the carbon bearing R3, destroying the complex.11 Therefore, we screened complex 2, which has a nitrile group at R3. This complex is stable to DGA for 24 h at 50 °C and is active for the copolymerization of CHO/DGA (entry 2). The electron withdrawing nature of the nitrile group is a significant factor in preventing ligand degradation given that a sterically similar methyl substituent at R3 (3) also leads to an inactive catalyst (entry 3). In addition to the dramatic impact that ligand electronics has on activity, there is also a significant steric effect of the aryl substituents (entries 2, 4, 5). Complex 4, with a ligand that is intermediate in steric bulk relative to those of 2 and 5, proved to be the most active catalyst. We also attempted the polymerization in the absence of catalyst, which gave no conversion. With a competent catalyst in hand, we explored the copolymerization of a series of epoxides with DGA for the synthesis of a variety of aliphatic polyesters (Table 2, entries 1-6).12 Under optimized conditions, the CHO/DGA copolymerization afforded poly(cyclohexene diglycolate) with high Mn and narrow MWD (entry 1). Vinyl cyclohexene oxide (VCHO) reacted with DGA under the same conditions as the CHO/DGA copolymerization (entry 2). The comonomer trans-(R)-limonene oxide13 (LO) also copolymerized with DGA; however, higher temperature and a longer reaction time are required (entry 3). Notably, polyesters containing LO and VCHO subunits have the potential to be useful precursors to more elaborate polymers through postpolymerization modification of the pendant vinyl groups. Aliphatic epoxides, Scheme 1. Alternating Copolymerization of Epoxides and Cyclic Anhydrides