Kinetics and Mechanism of Candida antarctica Lipase B Catalyzed Solution Polymerization of ǫ-Caprolactone Ying Mei, Ajay Kumar, and Richard Gross* NSF-I/UCRC Center for Biocatalysis and Bioprocessing of Macromolecules, Polytechnic University, Department of Chemistry and Chemical Engineering, Six Metrotech Center, Brooklyn, New York 11201 Received October 15, 2002; Revised Manuscript Received April 10, 2003 ABSTRACT: Studies of the kinetics and mechanism of Candida antarctica Lipase B (CALB) catalyzed ǫ-caprolactone (ǫ-CL) polymerizations in toluene were performed. The kinetic plot of ln ([M] 0/[M]t) vs time was carried out to 96% ǫ-CL conversion and Mn 11 970. The plot is linear (r 2 ) 0.998), indicating that termination did not occur and the propagation rate is first order with respect to monomer concentration. Changes in the water (e.g., initiator) concentration did not change the polymerization rate but did change the number of chains [R-OH]. Thus, the polymerization is zero order with respect to [R-OH] and initiator concentration. A plot of ln k app vs ln [enzyme] gave 0.7 as the reaction order of the enzyme concentration. The apparent activation energy for Novozyme-435 catalyzed ǫ-CL polymeri- zation in toluene is 2.88 kcal mol -1 . This is well below 10.3 kcal mol -1 , the activation energy for aluminum alkoxide catalyzed ǫ-CL polymerization in toluene. Upward deviation from linearity for Mn vs fractional ǫ-CL conversion and decreases in the number of chains was accentuated by low enzyme water contents and high monomer conversion. These results are consistent with a competition between ring-opening chain-end propagation and chain growth by steplike polycondensations. CALB was irreversibly inhibited by modification with paraoxon at the lipase active site (Ser105). The modified enzyme was no longer active for the polymerization. This supports that the polymerizations studied herein occurred by catalysis at the active serine residue (Ser105) and not by other chemical or nonspecific protein-mediated processes. Introduction Certain lipases have been found to have an extraor- dinary ability to catalyze the synthesis of polyesters by in-vitro reactions with nonnatural substrates. 1-4 Our laboratory and others have begun to explore the kinetics and mechanism(s) of lipase-catalyzed lactone ring- opening polymerization. For porcine pancreatic lipase (PPL) catalyzed ǫ-caprolactone (ǫ-CL) polymerization in heptane, plots of log {[M] 0 /[M] t } vs time and M n vs conversion to about 85% monomer conversion were constructed. The linearity of these two plots indicated termination and chain transfer did not occur. Therefore, the system provided “controlled” polymerizations where the molecular weight was a function of the monomer- to-initiator stoichiometry. 2 The polymerization of ǫ-CL catalyzed by physically immobilized Lipase B from Candida antarctica (Novozyme-435) in bulk at 70 °C was investigated by Deng et al. 3 He reported that increased lipase concentration results in more rapid monomer conversion but decreased product molecular weight. A linear relationship between ln {([M] 0 - [M] i )/ ([M] t - [M] i )} vs reaction time and M n vs monomer conversion (until 80%) was found. A kinetic analysis and study of products at low conversion for the bulk polym- erization of ω-pentadeclactone (PDL) at 50 °C using an immobilized lipase from a pseudomonas sp. (I-PS-30) was also reported. 4 Kobayashi and co-workers analyzed ǫ-CL and 12-dodecanolide (DDL) polymerizations using Michaelis-Menten kinetics. Their studies were per- formed in isopropyl ether, at 60 °C, catalyzed by the lipase from Pseudomonas fluorescens. They reported that enhanced polymerizability of lactones with larger ring size is mainly due to the larger reaction rate (V max ) and not to differences in binding affinity. Furthermore, they reported that the transformation of the lipase- lactone complex to the acyl-enzyme intermediate is the key step for the lipase-catalyzed lactone polymeriza- tion. 5 Work was performed to determine whether the known active site of a poly(-hydroxybutyrate) depolymerase was also the site that catalyzed lactone ring-opening polymerizations in organic media. To this end, Doi and co-workers prepared three site-specific mutants and the wild-type poly(-hydroxybutyrate) depolymerase from A. Faecalis T1. 6 The relative activities of the wild-type and mutant enzymes were compared for -butyrolactone ring-opening polymerization in organic media. They found that none of the mutant enzymes showed polym- erization activity whereas the wild-type enzyme actively catalyzed the polymerization. Their results proved that the active site, and not one or more remote sites, is responsible for the catalysis of -butyrolactone polym- erization. The following defines the objectives of this study relative to previous reports: 1. Lipases from various origins including porcine pancreatic lipase (PPL), Lipase PF from Pseudomonas fluorescens, Lipase SPF from Pseudomonas sp., and Lipase B from Candida antarctica (CALB) have been used as model enzymes. At this stage of our under- standing, each lipase must be considered dissimilar in the mechanism of lactone polymerization until proven otherwise. Furthermore, CALB immobilized on Lewatit (Novozyme-435) is particularly attractive for more intense research due to its extraordinary activity for lactone polymerizations in addition to its ability to perform these polymerizations with high regioselectivity during initiation. 1,7 2. The existence of transesterification and polycon- densation between diacids and diols has been reported by others and our laboratory. 1,8 Although discussed elsewhere, 9 the effect of the concurrent processes of 5530 Macromolecules 2003, 36, 5530-5536 10.1021/ma025741u CCC: $25.00 © 2003 American Chemical Society Published on Web 06/24/2003