Pergamon 0014-3057(94)E0046-7 Eur. Poly'm. J. Vol. 30, No. 9, pp. 1089-1095, 1994 Copyright @ 1994 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0014-3057/94 $7.00 + 0.00 NMR STUDY OF PLASTICIZED PVC MARIA I. B. TAVARES,1 ELISABETH E. C. MONTEIRO, m* ROBIN K. HARRIS 2 and ALAN M. KENWRIGHT2 qnstituto de Macromol6culas, Universidade Federal do Rio de Janeiro, P.O. Box 68525, Rio de Janeiro, RJ 21945 Brazil, and qRC in Polymer Science and Technology, University of Durham, South Road. Durham DH1 3LE, England (Received 8 June 1993; accepted in final form 23 September 1993) Abstract--Samples of PVC plasticized by di-isobutylphthalate and di-2-ethylhexylphthalate, with plasti- cizer content varying up to 180 parts per hundred, have been examined using solid-state tH- and t3C-NMR. Proton spin-lattice relaxation times in both the laboratory and the rotating frames were measured, with analysis into multi-exponential components where necessary. Cross-polarisation magic- angle spinning ~3C spectra have also been obtained, including variation with contact time. The data are discussed in terms of the domain structure of the samples at the microscopic level and of the role of the plasticizer. INTRODUCTION A knowledge of the domain structure and of the dynamics at the molecular level of polymers is of considerable importance to the understanding of their physical properties, yet the complex nature of most systems is hard to unravel. This is true even for homopolymers, partly because of the general existence of crystalline and amorphous domains and partly because of the spread of molecular weights. Variations with sample history often exist. Usually there are further complications because the systems of interest may be copolymers or there may be a practical requirement for additives. A case in point of the latter type is the widespread use of plasticizers, which lead to changes in the glass transition tempera- ture and in mechanical properties. In recent years NMR has come to the fore in the investigation of polymer structure and dynamics, partly because of the ability of the cross-polarisation (CP)/high-power proton decoupling (HPHD)/magic- angle spinning (MAS) suite of techniques to generate high-resolution ~3C spectra [1-3]. However, it is nei- ther sensible nor feasible to rely on such experiments alone when the samples examined are microscopically heterogeneous [1], since the behaviour of the 13C spins under such circumstances depends on proton relax- ation characteristics. The latter are unlikely to be represented by single-exponential behaviour except in extreme circumstances. Direct measurement (in the ~H resonance) of spin-lattice relaxation times in both the laboratory and rotating frames (TIn and T~, respectively) is essential, including analysis where necessary into multi-exponential components. Spin- diffusion [4] will tend to cause averaging of such *To whom all correspondence should be addressed. components when domain sizes are small, and T~ is more readily averaged in this way than T~,. Peaks in the 13C CP/HPHD/MAS spectrum will vary in their intensity depending on the participation of the rel- evant molecular fragment in the different domains and on the relationship of the variables (contact time and recycle time) of the experiment to the effective domain values of T~ and T~,. Various discriminat- ing experiments [5] are also possible--for example, (a) omission of the CP technique (giving direct single- pulse excitation, SPE, of the carbons) leads to spectra depending on T c and the recycle time, and (b) long contact (or delayed contact) in the CP experiment will tend to show peaks from domains with long T~p. Variable-contact CP experiments yield intensities which can be analysed [6] to give characteristic cross-polarisation times, Tco, and values of T~,, though the complexities of domain existence must in principle be considered. Poly(vinylchloride), PVC, is one of the most indus- trially-important polymers, yet, in spite of many investigations by a range of techniques, including solid-state NMR [7, 8] its detailed microstructure is not fully understood. Although the commercial polymer is generally amorphous in character [9], there are at least two types of domain, as is shown by the analysis of the proton relaxation behaviour in the rotating frame (see below), and varying degrees of crystallinity have been found [9 12]. The domains will, in any case, consist of locally more-ordered and less-ordered regions, the order correlating with differences in molecular motion. Plasticizers are normally used with PVC, and com- monly various alkyl-phthalates have found favour for this purpose [13]. It is not yet understood how the plasticizers operate at the molecular level, although a number of relevant studies have been reported [14-16]. (Scheme 1 overleaf). 1089