Some observations on the structure of Durham polyacetylene C. S. Brown and M. E. Vickers BP Research Centre, Sunbury-on-Thames, Middlesex, TW16 7LN, UK and P. J. S. Foot, N. C. Billingham and P. D. Calvert School of Chemistry and Molecular Sciences, University of Sussex, Brighton, BN1 9QJ, UK (Received 10 March 1986; revised 16 June 1986) The structure of polyacetylene produced from a precursor polymer is compared with that of polymer produced by a soluble Ziegler-Natta catalyst. X-ray diffraction shows that the precursor polyacetylene can be produced with a wide range of ordering from apparently amorphous to crystalline, although the ordering is never as high as that of polymer produced by the Ziegler-Natta route. An increase in ordering, as measured by X-ray peak width, has been observed during isomerization; this contrasts with normal polyacetylene where no change in order is found. Both forms of polyacetylene show a gradual shift in the interchain d- spacing from 395 pm (cis/trans mixed) to 365 pm (trans). Since only a single peak is found in this region the c/s- and trans-forms are not phase-separated on a scale of 3 nm or more. Small-angle X-ray scattering also indicates that discrete crystalline and amorphous regions, as found in polyethylene, are not present (Keywords: polyacetylene; precursor polymer; X-ray scattering; morphology; chain packing) INTRODUCTION Polyacetylene (PA) produced by the Shirakawa route, using a soluble Ziegler-Natta catalyst, crystallizes during polymerization to the cis-form. Subsequent heating converts this to the trans-form but the original morphology is preserved. Polyacetylene may also be prepared by the 'Durham' route 1 involving transfor- mation of a soluble precursor polymer as shown in Figure 1. This PA has variously been described as amorphous 1 and crystalline 2 but certainly it is normally less ordered than Shirakawa polymer. Preparation of an oriented, highly crystalline form of Durham PA has been described 3'4 and an X-ray study of the structure of this material is in an accompanying paper 4. In the absence of good, highly oriented fibres it has been difficult to determine unambiguously the crystal structures of PA. The cis-form has been reported to be orthorhombic with unit cell dimensions of 761, 447, 439pm. The trans-form is monoclinic or orthorhombic. Table I summarizes the unit cells that have been proposed for PA. The trans-conformation is extended 12 Yoalong the chain when compared to cis- but is contracted perpendicular to the chain by about the same amount such that the cis-form has a density close to that of the trans-isomer. The uncertainty over the crystal structure leads to an uncertainty in the density for trans- PA. A similar orthorhombic structure to trans-PA is observed for polyethylene (742, 495, 254 pro). It is close to a hexagonal packing of the chains such that it can be viewed as a slight deviation from close packing of smooth cylinders. Robin et al. 5 find that the PA structure really only deviates from pseudo-hexagonal packing at greater than 80 ~o conversion to the trans-isomer. The isomerization occurs with a slow shift of the X-ray reflections, suggesting that it is an isomorphous change with no loss of the crystal identity. Using the Scherrer formula with the width of the hkO reflections it is possible to estimate the extent of order from the X-ray coherence length perpendicular to the chain axis. This Robin et al. s find to be 15 + 3 nm and unchanged during isomerization. An equivalent measurement on polyethylene yields a 'mosaic block' size of 25-60 nm, which is much less than the actual extent of the crystals perpendicular to the chains 6. After cold drawing the polyethylene mosaic size drops to about 10 nm. During isomerization the reflections from the chain periodicity disappear, suggesting a loss of longitudinal coherence between neighbouring chains. These reappear when more than 80~o of the polymer is in the trans-state at the same time as the packing becomes sufficiently regular to show splitting of the (110) and (200) reflections at 370 pro. The 'Durham' polymer (Figure 1) has two bonds which should allow free rotation and two rigid bonds per repeat unit. Comparison of g.p.c, data 7 with light scattering studies show that the precursor polymer, in solution in tetrahydrofuran, is more tightly coiled than polystyrene and that it has a chain expansion factor of about 6.8. This compares with a chain expansion of 6.3 in polyethylene measured as the radius of gyration of the random coil compared to that of an equivalent freely-jointed chain. Hence the precursor polymer is a randomly coiled, amorphous polymer. This Durham polymer can be converted to a predominantly cis-PA (typically 75 ~o cis, 25 ~o trans) by heating at 60°C for 45 rain. At higher temperatures the material isomerizes to trans-PA. We have discussed the kinetics of transformation and isomerization processes elsewhere s'9. Infra-red studies show that the isomerization is from 3 to 10 times faster than in Shirakawa polymer but with a similar activation 0032-3861/86/111719503.00 © 1986Butterworth& Co. (Publishers) Ltd. POLYMER, 1986, VOI 27, November 1719