Journal of Polymer Science Part B: Polymer Physics, Vol.40, No.5, 424-433, 2002
Microphase separation in poly (acrylonitrile-butadiene-styrene) (ABS) studied with paramagnetic spin probes. II. Simulation of electron spin resonance spectra
We recently presented electron spin resonance spectra of poly(acrylonitrile-butadiene-styrene) (ABS) doped with 10-doxylnonadecane (10DND) and 5-doxyldecane (5DD) as spin probes. The spectra were measured in three types of ABS that differed in their butadiene contents and methods of preparation. Results for the ABS polymers were evaluated by comparison with similar studies on the homopolymers polybutadiene (PB) and polystyrene (PS) and the copolymers poly(styrene-co-acrylonitrile) (SAN) and poly(styrene-co-butadiene) (SB). Only one spectral component was detected for 10DND in PB, PS, SAN, and SB. In contrast, two spectral components differing in their dynamic properties were detected in the ABS samples and were assigned to spin probes located in butadiene-rich domains (the fast component) and SAN-rich domains (the slow component). The presence of two spectral components was taken as an indication of microphase separation. In this study, we present details on the dynamics and microphase separation by simulating spectra of 10DND in ABS, PB, PS, and SAN. The simulations are based on a dynamic model defined by the components of the rotational diffusion tensor and the diffusion tilt angle between the symmetry axis of the rotational diffusion tensor and the direction of the nitrogen 2p(z) atomic orbital. The jump diffusion model led to good agreement with experimental spectra. In this model, the spin probe has a fixed orientation for a given time and then jumps instantaneously to a new orientation. The temperature variation of the rotational correlation time in PB and PS consisted of two dynamic regimes, with different activation energies. The transition temperature at which the change in dynamics occurs (T-tr) is 380 K for PS and 205 K for PB, essentially the same as the corresponding glass-transition temperatures measured by differential scanning calorimetry. We suggest that T-tr is a better indicator of the glass transition than the temperature at which the total spectral width is 50 G, especially for large probes. The simulation program allowed the determination of the relative intensities of the fast and slow spectral components as a function of temperature; this information was used to clarify the redistribution of the probe above the glass transition of the SAN-rich component in ABS systems.
Keywords:poly(acrylonitrile-butadiene-styrene) (ABS);microphase separation;electron spin resonance spectroscopy;nitroxide spin probes;nitrogen hyperfine splitting;jump diffusion motional model;rotational correlation time;spectral simulations;transition temperature