J. MATER. CHEM., 1991, 1(4), 507-509 Semi-crystalline Alkali-metal Salt Complexes with Poly(o1igooxyethyleneoxy-I,2=phenylene)s Jonathan D. Hague and Peter V. Wright* School of Materials, University of Sheffield, Hadfield Building, Mappin Street, Sheffield S I 3JQ UK A series of poly(oligooxyethyleneoxy-l,2-phenylene)shaving the formula where n=2, 3 and 5, have been synthesised with molecular weights of ca. 10000. Although the pure polymers are fully amorphous, thermal analysis and X-ray scattering show that the polyether with n=2 forms a well organised complex with NaSCN, with a stoichiometry of three oxygens per Na+ ion. The melting temperature of the new complex (187 "C) is similar to that of the analogous complex with poly(ethy1ene oxide) (PEO). Comparison with the PEO-NaSCN complex suggests that the new polymer may enclose the Na+ ions to form a 2, helix with 18 bonds per repeat.A plausible 2, helical structure for this new polyether complex has been computed. A Crystalline complex between the polyether with n= 5 and NaSCN is also observed but no organised complexes with the n=3 system or with Nal, LiCIO, or KSCN were obtained with any of the polyethers. Keywords: Polyether-alkali salt complex; Crystallinity; Helical structure; Computed conformation Salt complexes with high-molecular-weight poly(ethy1ene oxide) (PEO) are well known as amorphous phase materials for ionic conduction in so-called 'polymer electrolyte^"-^ and we have studied the organised phases of these materials with a variety of inorganic and organic anion^.^-^^ How-ever, while there have been several reports of amorphous complexes based on high-polymer ligands other than PE0,14 we are not aware of organised complexes based on other synthetic high-polymer ligands, with the notable exception of the sodium-ion complexes with poly(ethy1ene imine).159" Chatani and co-workers have reported crystallographic structures for PEO-NaI17 and PEO-NaSCN." With Na' (and presumably with Li' and Ag'), PEO forms a 2' helix with a repeating unit in which six EO units having conformations (~tgttgttg)~enclose two cations (ie.the stoichi- ometry of the crystalline phase, x =[ether oxygen]/[M '1 = 3).Here, the gauche conformation (g and gx +120") is adopted by the C-C bonds and the trans (tz0")by the C-0 bonds.In order to accommodate external anions of a range of sizes, which are stacked alongside the helix and which interact with the enclosed cations, this helical 'spring' can adjust its fibre repeat distance from ca. 7.1 A (phen-olate") to ca. 8.4A (perchlorate2'). The thickness of the aromatic nucleus (3.2-3.6 A) is readily accommodated so that radical anions, giving electronic condu~tivity,~,'~ or mesogenic and other aromatic anions8*'0-12~21may be stacked alongside the helix with their planes approximately normal to the helix axis. In order to explore further the possibilities of molecular organisation in these high-polymer complexes with stacked appendages, we have embarked upon a study of the complexation behaviour of polyethers, having substituents attached to, or included in, the sequence of co-ordinating atoms along the chain.In this paper, we report on the preparation and complexation of poly(oligooxyethy1eneoxy-1,2-phenylene) chains having the formulae 0 I where n=2, 3 and 5. These polymers are coded P2EOP, P3EOP and PSEOP, respectively. If the 21 helix of PEO- Na' is the linear analogue of the cyclic ether 18-crown-6, the corresponding structure for P2EOP-Na' may be analogous to dibenzo-18-crown-619 and could also be a 2, helix in which successive aromatic rings protrude on either side at each 180" of turn. For PSEOP, a helix analogous to benzo-18-crown-6 having 1' symmetry with an aromatic ring protruding on one side only within each fibre repeat should be formed by complexation with salts of the smaller cations mentioned above.P3EOP was prepared in order to investigate the possibility of forming complexes analogous to the crystalline PEO-K+ complexes.' Although the latter have an unknown structure they have crystalline stoichiometry x =4. P3EOP also incorporates four oxygens per repeat and may form an analogous crystalline complex. Experimental Diethylene glycol, triethylene glycol, pentaethylene glycol (99Y0 purity) and catechol were supplied by Aldrich and the ditoluene-p-sulphonates of the glycols were prepared by reac- tion with twice the molar proportions of toluene-p-sulphonyl chloride in pyridine. All monomers were recrystallised.Poly(o1igooxyet h yleneox y-1,2-phenylene)s were prepared from solvent-free equimolar mixtures of an ethylene glycol ditoluene-p-sulphonate and catechol, with NaOH in slight stoichiometric excess (2.1 molar proportion). These compo- nents were finely ground and intimately mixed under dry nitrogen and transferred to a reaction tube. The mixture was heated and pumped under vacuum at 120 "C for ca. 48h. The 508 reaction mixture was first extracted with chloroform and filtered to separate the polyether products from the sodium toluene-p-sulphonate product. The high-molecular-weight polymer was then separated from low-molecular-weight and cyclic material by extracting the latter and precipitating the polymer in boiling ethanol. The yields of high polymer were 30-50%. All three of the polyether materials were amorphous highly viscous liquids and their molecular structures were character- ised using 'H NMR, transmission IR spectroscopy and gel- permeation chromatography, and the thermal behaviour was investigated using differential thermal analysis.The 'H NMR data are in good agreement with those of the corresponding cyclic ethers,22 with small differences between the chemical shifts for the methylene protons in the ring and in the cyclic molecules. The NMR data suggest that the polymers were more than 96% pure. This was supported by the IR data for the linear polyethers which differed slightly from the corre- sponding cyclic spectra23 only in the band at 1050cm-' which appeared at lOOOcm-' in the cyclic (attributed by Peder~en~~to CH2 wag or twist).Molecular size was investi- gated using gel-permeation chromatography with Ultrastyra- gel columns (Waters Associates) and tetrahydrofuran solvent. Complexes with alkali salts were prepared by dissolving the salt and polymer in acetonitrile. In view of the established stoichiometry of the crystalline complexes of the smaller cations with PEO (x =[ether oxygen]/[M+] =3), mixtures with x=2, 3 and 5 were prepared. The complexes were analysed by WAXS, DTA and hot-stage polarised-light microscopy. Discussion The three poly(oligooxyethy1eneoxy-1,2-phenylene)s were fully amorphous. Their glass-transition temperatures, as deter-mined by DTA, were found to be 6, -13 and -26 "C for P2EOP, P3EOP and PSEOP, respectively.The polyethers eluted in tetrahydrofuran at GPC peak maximum elution volumes corresponding to polystyrene standards of molecular weight 7000- 10 000. However, significant fractions of material having 'polystyrene molecular weights' >20 000 were also present in 'most probable' distributions. A crystalline product was precipitated from acetonitrile solutions of P2EOP in the presence of NaSCN over a period of ca. 1 h. The WAXS spectra of P2EOP-NaSCN (x=3), PEO-NaSCN (x=3) and NaSCN pure salt are shown in Fig. l(a), (b)and (c). These spectra demonstrate that a novel crystalline polymer complex has been formed. Mixtures having a greater NaSCN content (x <3) showed WAXS tracings with reflections characteristic of the pure salt, supporting the assumed stoichiometry x =3 for the P2EOP-NaSCN crystal- line lattice.In Fig. l(d), (e) and (f),DTA traces of P2EOP-NaSCN for x =2, 3 and 5, respectively, are shown. The traces in Fig. l(d), (e) and (f)show well defined endotherms at 169, 177 and 187 "C, respectively; hot-stage polarised-light microscopy indi- cated that these endotherms correspond to order-isotropic transitions. (Dibenzo- 18-crown-6-NaSCN melts at 230 0C.23) These results thus confirm the formation of a novel complex, and it is of interest that these melting temperatures are well within the temperature range observed for PEO-NaSCN and PEO-NaI complexes.6 In the latter cases also the complexes with slight salt deficiencies gave higher melting temperatures which were attributed6 to a pronounced propensity for lamel- lar thickening in the salt-deficient samples.The failure of the P2EOP-NaSCN sample with x=2 to undergo lamellar thick- ening owing to salt-stabilised crystal surfaces may also account for the lower melting endotherm (169 "C) in this sample. A J. MATER. CHEM., 1991, VOL. 1 I I I I I 1 I I30 0 30 60 90 120 150 180 Tl°C Fig. 1 (a) Wide-angle X-ray scattering (WAXS) spectrum for P2EOP- NaSCN, x =3; (b)WAXS spectrum for PEO-NaSCN, x =3; (c)WAXS spectrum for NaSCN. (d)Differential thermal analysis (DTA) tracing for P2EOP-NaSCN, x =2; DTA tracing for P2EOP-NaSCN, x =3; DTA tracing for P2EOP-NaSCN, x =5. (e)DTA tracing for PSEOP- NaSCN, x=4.(h) and (i) are DTA tracings for pure P5EOP and P2EOP, respectively distribution of lamellar dimensions is suggested by the broader endotherms of the lower melting compositions. Fig. l(d) also reveals an endotherm at 28 "C indicating the presence of excess of NaSCN when x=2, and supporting the stoichi- ometry x =3 for P2EOP-NaSCN. The lower-temperature region of the DTA traces for P2EOP-NaSCN reveals a glass- transition temperature at 25-30 "C and a first-cycle endo- therm at ca. 60 "C. The latter is reminiscent of a similar endotherm in PEO-salt complexes, but in P2EOP-NaSCN this cannot be ascribed to the uncomplexed polyether which is amorphous. This endotherm, which is removed by thermal cycling, may perhaps correspond to disorganised, less stable regions in the solution-deposited sample.However, whereas P2EOP-NaSCN x =3 powdered samples are difficult to shape, the salt-deficient x=5 material is malleable in the presence of some solvent and may be rolled into oriented films or filaments. However, we have been unable as yet to obtain fibre photographs of sufficient quality for structural determination. The similar stoichiometry and melting temperatures of the PEO-NaSCN and P2EOP-NaSCN systems strongly suggest that the two complexes may have similar molecular and unit- cell structure, in accord with our molecular-design strategy J. MATER. CHEM., 1991, VOL. 1 Fig. 2 A computed helical model for P2EOP-Na' being a plausible structure for P2EOP-NaSCN, after the structure for PEO-NaSCN (Form I) determined by Chatani and co-worker~.'~ (This proposed structure for P2EOP-NaSCN is not based on any crystallographic experimentation.) Filled spheres are carbon atoms; small open, shaded spheres are oxygen atoms; (hydrogen atoms are not shown).The cations would be located near the helix axis for development of the 2, helix of the PEO-Na' systems. Using the computational procedures for helical parameters of Miya~awa,~~a plausible 21 helix for P2EOP-NaSCN with a similar fibre repeat distance (7.32 A) to that of PEO-NaSCN is readily computed, as shown in Fig. 2. The oxygen-to-axis distances (2.3-2.5 A) are within the range observed for PEO- Na' systems. The conformations of the nine skeletal bonds of the half-repeat are ttgttgttcis, the last being the fixed C-C bond of the 1,2-phenylene ring.In this sequence, small tor- sional deviations from conformational-energy minima are required to compensate for the fixed, planar configuration of the aromatic C-C bond. This fixed C-C bond would exert a constraint on the ability of the helix to expand along its axis. This may be a contributory factor in the failure which we have encountered to prepare P2EOP-Na' complexes with the larger I -anion. However, preliminary DTA investi- gations indicate that P2EOP-Na' complexes with various phenolate anions. SiddiquiIg observed that PEO-sodium phenolate has a short fibre repeat distance of 7.1 A which is similar to that of PEO-NaSCN. All attempts to prepare organised alkali-metal salt com- plexes with P3EOP mixtures were unsuccessful.For Na' and Li', this is in accord with expectation for the 2, helix having x =3. However, mixtures with KSCN (which forms crystalline complexes with PEO having x=4) were also prepared with P3EOP. The structure of PEO-KSCN is unknown but if PEO-K+ simply forms a 2, helix, analogous to PEO-Na' but with greater radius and fibre repeat distance, 8 A,20,25the failure of P3EOP-KSCN to crystallise may, at first sight, be surprising and offer some support to other models.20 A limited quantity (x>4) of KSCN dissolved in P3EOP so as to raise the glass-transition temperature from -13 to 25 "C. However, an organised phase in the PSEOP-NaSCN sys- tem may be readily observed by polarised-light microscopy.DTA evidence for this phase, which melts at 134 "C, is shown in Fig. l(g) and a distinctive WAXS pattern may be observed. Although it seems likely that the organised phase may have a stoichiometry x =3 with a PEO-Na+-like helix, salt separ- ation was observed in mixtures having x<4. Further work is required to establish the stoichiometry of this phase or to obtain evidence for a 'lop-sided' 1, helical adduct. Further investigations of organised, linear high-polymer- salt adducts having structures analogous to cyclic ('crown') oligomers are in progress. We are grateful to the University of Sheffield and Unilever for a research grant (J. D. H.), and to Mr. P. Hempstead, Department of Chemistry, University of Sheffield for assist- ance in computer drawing.References 1 D. E. Fenton, J. M. Parker and P. V. Wright, Polymer, 1973, 14, 589. 2 P. V. Wright, Br. Polym. J., 1975, 7, 319. 3 M. B. Armand, J. M. Chabagno and M. Duclot, in Fast Zon Transport in Solids, ed. P. Vashisha, J. N. Mundy and G. K. Shenoy, North Holland, New York, 1979, pp. 131-136. 4 Polymer Electrolyte Reuiews, ed. J. R. MacCallum and C. A. Vincent, Elsevier, London, 1987, vol. 1. 5 Polymer Electrolyte Reviews, ed. J. R. MacCallum and C. A. Vincent, Elsevier, London, 1989, vol. 2. 6 C. C. Lee and P. V. Wright, Polymer, 1982, 23, 681. 7 D. R. Payne and P. V. Wright, Polymer, 1982, 23,690. 8 J. A. Siddiqui and P. V. Wright, Polym. Commun., 1987,28, 7.9 J. A. Siddiqui and P. V. Wright, Polym. Commun., 1987, 28, 89. 10 B. Mussarat, K. Conheeney, J. A. Siddiqui and P. V. Wright, Br. Polym. J., 1988, 20,293. 11 P. V. Wright, Polymer, 1989, 30,1179. 12 P. V. Wright, in Polymer Electrolyte Reviews, ed. J. R. MacCallum and C. A. Vincent, Elsevier, London, 1989, vol. 2, ch. 2. 13 J. A. Siddiqui and P. V. Wright, Faraday Discuss. Chem. SOC., 1989, 88, 113. 14 J. M. G. Cowie, ref. 4, p. 69. 15 C. S. Harris, D. F. Shriver and M. A. Ratner, Macromolecules, 1986, 19, 987. 16 C. K. Chiang, G.T.Davies, C.A. Harding and T.Takahashi, Macromolecules, 1985, 18, 827. 17 Y. Chatani and S. Okamura, Polymer, 1987, 28, 1815. 18 Y. Chatani, S. Okamura and Y. Fujii, Polym. Preprints, 1989, 30(1), 404. 19 J. A. Siddiqui, PhD Thesis, University of Sheffield, 1989. 20 T. Hibma, Solid State Zonics, 1983, 9, 10, 1101. 21 J. Patel, P. V. Wright, R. Orr, D. W. Bruce, D. A. Dunmur and P. M. Maitlis, Handbook of British Liquid Crystal Society Annual Conference, Bristol, April 1990, poster P23. 22 P. Live and S. I. Chan, J. Am. Chem. SOC., 1976,98, 3769. 23 C. J. Pedersen, J. Am. Chem. SOC.,1967, 89, 7017. 24 H. Sugeta and T. Miyazawa, Biopolymers, 1967, 5, 673. 25 J. M. Parker, P. V. Wright and C. C. Lee, Polymer, 1981, 22, 1305. Paper 0/04608F; Received 15th October, 1990