58th Annual Report on Research 2013 Under Sponsorship of the ACS Petroleum Research Fund

The goal of this project is to apply a new NMR based method for the characterization of mechanisms and kinetics of metallocene catalyzed polymer reactions. Using hyperpolarization as provided by dynamic nuclear polarization (DNP), signal intensities in NMR spectra are enhanced by several orders of magnitude. For the application to polymer reactions, this technique enables measurement of NMR spectra during the reaction progress, and can yield information on reaction kinetics and mechanisms. In view of studying catalysis by metallocenes, we have first investigated techniques for the DNP hyperpolarization of propylene, by dissolving the gas in a solvent, as well as by directly freezing the gas. Conditions for synthesis of polypropylene, intended to be ultimately compatible with carrying out the reaction in the NMR spectrometer for real-time observation, were then explored in the glove box. Using bis(cyclopentadienyl)dimethylzirconium as a catalyst, and trityltetra(pentafluorophenyl)borate as a cocatalyst, polymer was readily obtained as indicated by ¹H and ¹³C NMR spectroscopy. Because propylene is a gas at room temperature, we found however that quantification of the amount of propylene introduced into a hyperpolarized sample using current techniques may be subject to a larger error than for typical liquid samples. For this reason, we substituted 1-hexene as the monomer for the reaction. 1-hexene is a commonly used monomer in the research of the kinetics of metallocene catalyzed polymerization. We anticipate that the reaction mechanisms for poly-1-hexene and poly-propylene are similar, but that the extra carbon atoms on the 1-hexene would influence the reaction rate. For later application of the DNP-NMR experiment, we also investigated reaction conditions by performing the reaction with 1-hexene in the glove box. We were able to successfully synthesize poly-1-hexene by using the catalysts as described above. This polymer reaction within 10 minutes converted nearly all of 0.3 M 1-hexene to the polymer, as determined by ¹H NMR. The synthesized poly-1-hexene was confirmed to be in the atactic form by comparing the ¹³C NMR spectrum with the literature. Since tacticity is mainly detemined by the catalyst, for which we used a non-constrained metallocene, this result is expected. Further, we carried out DNP-NMR experiments with 1-hexene (in the absence of a reaction). Sufficient enhancements of ¹³C NMR signals were obtained in order to distribute signals over multiple scans using small-flip angle excitation. Such a DNP-NMR data set indicates the apparent decay constants for the signals stemming from different ¹³C atoms in 1-hexene (Figure 1). Based on the reaction kinetics from the reaction obtained in the glove box, and these observed DNP-NMR signals from 1-hexene, we believe that observable quantities of product will be generated under reaction conditions achievable in a DNP experiment, once this experiment is carried out with addition of catalyst to initiate the reaction.

Collaborating with other researchers within our department, we further applied DNP-NMR measurements to another polymer reaction, the ring-opening polymerization of L-lactide. Polylactide is a target for environmentally friendly plastics, since it can be degraded by microorganisms. We are investigating this polymerization reaction with a metal free catalyst. Metal free catalysts are of interest, because metal residues in polymers can be undesirable in certain cases, for example for medical applications. DNP-NMR spectra of the polymer reaction could readily be obtained, and showed buildup of the signals from the polymer species as a function of time. In particular, the signal from the ester carbon in the polymer is prominently located at 169.5 ppm. Interestingly, in addition to this expected signal, a new signal at 175.2 ppm was observed. In order to identify this peak, a correlation experiment was carried out, where the signal from the ester carbon of the monomer was selectively inverted. In this experiment, the peak at 175.2 ppm also showed negative intensity, indicating that it corresponds to this carbon atom in a product of the reaction. We are using additionl DNP-NMR experiments in order to characterize this species and for determining the reaction mechanism.


Figure 1: Stacked plot of time series of spectra acquired from a hyperpolarized sample of 1-hexene using small-flip angle excitation (without reaction; acquired at time intervals of 400 ms). The two panels show two different spectral regions.