Monomer Sequence Distribution in Ethylene—Propylene Rubber Measured by 13C NMR. 3. Use of Reaction Probability Model

Abstract

Abstract One can measure r1r2 for an ethylene—propylene rubber by determining composition and χ, the ratio of contiguous to isolated propylene units. Our previous determination of χ depended on measuring methine carbon resonances in a 13C NMR spectrum. The χ method based solely on methine peak area measurements is inaccurate. This is not because the χ method is theoretically unsound, but because it is impossible to accurately extract the methine areas, necessary to calculate χ from the complex 13C NMR spectra of ethylene—propylene rubbers. The presence of inverted propylene produces ambiguity in a portion of the methine region. This means the methine areas can be used only to calculate a maximum and minimum boundary for r1r2. Another difficulty in using only methine areas to determine χ arises at propylene levels less than 35 wt%, because the methine areas for contiguous propylene triads become very small and are overlapped by strong resonances from long methylene sequences. We have developed a reaction probability method which accurately accounts for all the resonance areas in a 13C NMR spectrum of an ethylene—propylene rubber instead of restricting the analysis to the tertiary carbon areas. The reaction probability method produces a complete calculated 13C NMR spectrum as a best fit to the observed spectrum. The deduced probabilities provide the following derived quantities: (1) “r1r2”, a measure of monomer sequence randomness, (2) the distribution of methylene sequence lengths, (3) composition, (4) amount of propylene inversion. The probabilities can be used to directly calculate “r1r2”. Or alternatively, the probabilities can be used to calculate the methine carbon areas including the separation of the ambiguous methine area into contiguous and isolated contributions. These areas calculated from the probabilities can then be used to determine χ and the resultant r1r2. The two values for r1r2 are in good agreement, lending credence to the probability analysis. The composition derived from probabilities agrees with other analytical methods such as proton NMR and infrared. The limitation of the method is that we cannot accurately determine the percent propylene inversion. We can only conclude that in the majority of the copolymers of 20 to 60 wt% propylene that we have studied, the concentration of inverted propylene is significant and is between 10 and 40% of the total propylene present. Propylene enriched in carbon-13 could be used to prepare copolymers and perhaps resolve this uncertainty.