Liquid-crystalline rod-coil polymers based on poly(ethy1ene oxide) s and the influence of the complexation of LiCF,SO, on the liquid-crystalline assembly Myongsoo Lee* and Nam-Keun Oh Department of Chemistry, Yonsei University, Sinchon 134, Seoul 120-749, Korea The synthesis and characterization of rod-coil polymers of ethyl 4'-( 4'-oxy-4-biphenylcarbonyloxy)-4-biphenylcarboxylatewith poly(ethy1ene oxide) of three (3-4), seven (7-4), twelve (12-4) and sixteen (16-4) ethylene oxide units, of 4'-(4'-oxy-4- phenylcarbonyloxy)-4-biphenylcarboxylatewith poly(ethy1ene oxide) of sixteen ethylene oxide units (16-3), and of ethyl 4'-oxy-4- biphenylcarboxylate with poly(ethy1ene oxide) of sixteen ethylene oxide units (16-2) are described. All the rod-coil polymers except 16-2 display a layered smectic mesophase and, in particular, 16-4 shows a microphase-separated morphology.The complexes of 16-4 with up to 0.3 mol of LiCF,S03 per mol of ethylene oxide units are also prepared. The rod-coil polymer 16-4 exhibits an enantiotropic smectic B (S,) mesophase. The complexes with 0.05 and 0.1 mol of LiCF,S03, however, exhibit an enantiotropic smectic A (S,) mesophase in addition to an S, phase. In contrast with the complexes with 0.0-0.1 mol of LiCF,S03, the complexes with 0.2 and 0.3mol do not exhibit any smectic mesophases; however, they display a cylindrical micellar mesophase. Rod-coil diblock systems consisting of a flexible coil and a rigid rod, offer the opportunity to study new aspects of liquid-crystalline behavi~ur.l-~ The extent of immiscibility of these systems is expected to be large because of the large chemical differences between the stiff rod and the flexible coil segments, which allows block segregation to occur at relatively short chain-lengths, compared to that in typical flexible block copoly- mers.As a result of microphase segregation, the rod-coil diblock molecules self-assemble into well defined microstruc- tures such as lamellar and cylindrical structures in the melt state, depending on the block composition. Theoretical have shown that various supramolecu- lar structures such as nematic, layered smectic and cylindrical phases may be induced in these systems, depending on the relative volume fraction of blocks. For example, rod-coil molecules containing a short enough flexible coil exhibit a nematic phase. With increasing coil length, block micro- segregation occurs, resulting in a layered smectic assembly.As the coil length is increased further, molecular layers may collapse into discrete cylindrical micellar structures. Experimental works have also been performed on systems consisting of molecular rod blocks and poly(i~oprene)~*~ or poly(isoprene-block-styrene) coil blocks7 to form linear diblocks. The supramolecular structure in the rod-coil systems was observed to change from lamellar to micellar microphase- separated domains as the coil volume fraction was increased, although the molecular packing of the rod blocks was not described.6 To date, little work has been carried out on rod-coil diblock systems containing poly(ethy1ene oxide) coils. Liquid-crystal- line materials containing poly(ethy1ene oxide) have several advantages, in particular due to their complexation capability with alkali-metal cations, which can induce various liquid- crystalline supramolecular structures.A recent publication on poly(ethy1ene oxide) with flexible side groups has shown that complexation of polymer backbones with LiClO, induces a smectic lamellar mesophase from crystalline, uncomplexed polymer.8 Experimental works on taper-shaped molecules con- taining oligo(ethy1ene oxide) have also demonstrated that complexation of the molecules with alkali-metal trifluorome- thanesulfonates can induce a hexagonal columnar mesophase and can enhance the thermal stability of the rnesopha~e.~-'' It is in this context that we have synthesized rod-coil polymers, consisting of a molecular rigid rod and a poly(ethy1- ene oxide) coil, and we have investigated the influence of the rod and coil lengths, and of complexation with LiCF,SO,, on the molecular organization in the melt state.We describe herein the synthesis and thermal characterization of several rod-coil polymers X-Y, where X is the number of ethylene oxide units of the coil segment and Y is the number of phenyl rings of the rod segment, based on: ethyl 4'-(4-oxy-4-biphe- nylcarbonyloxy)-4-biphenylcarboxylate with poly(ethy1ene oxide) containing three (3-4), seven (7-4), twelve (12-4) and sixteen (16-4) ethylene oxide units; 4-(4-oxy-4-phenylcar- bonyloxy)-4-biphenylcarboxylatewith poly(ethy1ene oxide) containing sixteen ethylene oxide units (16-3); and ethyl 4'-oxy-4-biphenylcarboxylate with poly(ethy1ene oxide) contain- ing sixteen ethylene oxide units (16-2).We also describe the mesomorphic behaviour of the resulting rod-coil polymers and the mesophase change of 16-4 upon lithium complexation. Experimental Materials 4-Hydroxy-4'-biphenylcarboxylic acid (98YO),1,3-diisopro-pylcarbodiimide (DIPC, 99'%0), bromoethane (98%), toluene-p-sulfonyl chloride (98 %), 4-dimethylaminopyridine (99%), tetrabutylammonium hydrogen sulfate (TBAH, 97%), lithium trifluoromethanesulfonate (lithium triflate, 97%) (all from Aldrich) and the other conventional reagents were used as received.Poly(ethy1ene oxide) monomethyl ethers of average molecular masses ((M,}) 350,550 and 750 (Aldrich) were dried under vacuum at 50 "C for 40 h. Dichloromethane was washed initially with concentrated sulfuric acid, then with water, before being dried over magnesium sulfate, heated to reflux over calcium hydride and then freshly distilled under argon. Pyridine was heated to reflux over calcium hydride and was then distilled. Dimethyl sulfoxide (DMSO) was stirred with calcium hydride for 24 h at 100"C and then distilled under reduced pressure. Lithium triflate was dried at 120°C under vacuum for 24 h. 4-Dimethylaminopyridinium toluene-p-sulfonate (DPTS) was prepared as described previously.'2 Techniques 'HNMR spectra were recorded from CDCl, solutions on a Bruker AM 300 spectrometer operating at a proton frequency of 300 MHz or a Bruker AM 500 spectrometer, using Me4% as an internal standard.A Perkin Elmer DSC-7 differential scanning calorimeter equipped with a 1020 thermal analysis J. Muter. Chem., 1996, 6(7), 1079-1086 1079 controller was used to determine the thermal transitions which were reported as the maxima and minima of their endothermic or exothermic peaks, respectively. In all cases, the heating and cooling rates were 10°C min-' unless otherwise specified. A Nikon Optiphot 2-pol optical polarized microscope (magnifi- cation: 100 x) equipped with a Mettler FP 82 hot-stage and a Mettler FP 90 central processor was used to observe the thermal transitions and to analyse the anisotropic texture^.'^,'^ Molecular mass distributions were determined by size exclusion chromatography (SEC) with a Waters R401 instrument equipped with a US HR5E-500-H22 column and a Millenum data station. The measurements were made at 40°C using the UV detector with THF as solvent (1 ml min-l) and a cali- bration plot constructed with polystyrene standards was used to determine the molecular mass distributions.Microanalyses were performed with a Perkin Elmer 240 elemental analyser. All final rod-coil molecules were purified by repeated recrys- tallization from a mixture of CH2Cl, and hexane until their transition temperatures remain constant. The purity of the products was checked by thin layer chromatography (TLC; Merk, silica gel 60).Methylene chloride-methanol mixtures were used as eluents and the spots were detected by either UV irradiation or exposure in an iodine chamber. Synthesis The synthesis of rod-coil polymers with different coil lengths n is outlined in Scheme 1. The rod-coil polymers of poly(ethy1- ene oxide) monomethyl ether ({M,} =750) with two (16-2) and three phenyl rings (16-3) systems were synthesized as outlined in Schemes 2 and 3, respectively. Synthesis of compounds 5-8. Methyloxy di(ethy1eneoxy)ethyl tosylate 5 and methyloxy poly(ethy1eneoxy)ethyl tosylates 6-8 were all synthesized using the same procedure. A representative example is described for 8. Poly(ethy1ene oxide) monomethyl ether ({M,} =750, 15.7 g, 20.9 mmol) was dissolved in 5 ml dry pyridine under argon.A solution of toluene-p-sulfonyl chloride (4.4 g, 23 mmol) in 5 ml dry pyridine was then added dropwise to the mixture. The reaction mixture was stirred at room temperature under argon overnight. The resulting 1 n=3 5 n=3 2 n=7 6 n=7 3 n=12 7 n=12 8 n=16 KOH, EtOH H-COOH + CH3CH2Br 14 EtOH KOH t l3 I 15 0 n=3 10 n=7 11 n=12 DIPC, DPTS 12 n=16 CH2C12 n-4 Scheme 1 Synthesis of the rod-coil polymers 3-4,7-4, 12-4 and 16-4 1080 J. Muter. Chem., 1996, 6(7), 1079-1086 4 8 I 16-2 Scheme 2 Synthesis of the rod-coil polymer 16-2 I l6 15 I 16-3 Scheme 3 Synthesis of the rod-coil polymer 16-3 solution was poured into water and extracted with methylene chloride.The methylene chloride solution was washed with water, dried over anhydrous magnesium sulfate, and filtered. The solvent was removed in a rotary evaporator, and the crude product was purified by column chromatography (silica gel, methylene chloride eluent) to yield 14.0 g (74.1%) of a colourless oil. Compound 5. Yield, 95%. 'HNMR, 6 2.37 (s, 3H, CH,-phenyl), 3.38 (s, 3 H, CH,O), 3.55-4.10 (m, 12 H, OCH,), 7.33 (d, 2 Ar-H o to CH,, J =8.2 Hz), 7.74 (d, 4 Ar-H o to SO,, J = 8.3 Hz). Elemental analysis for C,,H,,O,S: Calc. C, 52.81; H, 6.96. Found C, 52.54; H, 6.83%. Compound 6. Yield, 80%. 'H NMR, 6 2.37 (s, 3 H, CH3- phenyl), 3.35 (s, 3 H, CH,O), 3.55-4.07 (m, 28 H, OCH,), 7.28 (d, 2 Ar-H, o to CH,, J= 8.2 Hz), 7.72 (d, 2 Ar-H, 0 to SO,, J= 8.3 Hz).Elemental analysis for C22H38010S: Calc. C, 53.43; H, 7.74. Found C, 53.28; H, 7.71%. Compound 7. Yield, 86%. 'H NMR, 6 2.37 (s, 3 H, CH,- phenyl), 3.34 (s, 3 H, CH,O), 3.55-4.10 (m, 48 H, OCH,), 7.30 (d, 2 Ar-H, o to CH,, J= 8.2 Hz), 7.74 (d, 2 Ar-H, o to SO,, J =8.3 Hz). Elemental analysis for C32H58015S: Calc. C, 53.77; H, 8.18. Found C, 53.45; H, 7.93%. Compound 8. Yield, 74%. 'H NMR, 6 2.37 (s, 3 H, CH,- phenyl), 3.31 (s, 3 H, CH,O), 3.55-4.10 (m, 64 H, OCH,), 7.28 (d, 2 Ar-H, o to CH,, J=8.2 Hz), 7.72 (d, 2 Ar-H, o to SO,, J =8.3 Hz). Elemental analysis for C,,H,,O,,S: Calc. C, 53.92; H, 8.37. Found C, 53.58; H, 8.44%. Synthesis of compounds 9-12.4-[ Methyloxy di(ethy1eneoxy)- ethyloxy]-4-biphenylcarboxylicacid 9 and 4'-[methyloxy poly- (ethyleneoxy)ethyloxy-4-biphenylcarboxylicacids 10-12 were all synthesized using the same procedure.A representative example is described for 12. 4'-Hydroxy-4-biphenylcarboxylic acid (1.5 g, 7.00 mmol) and KOH (0.9 g, 16.1 mmol) were dissolved in 100 ml methanol. The mixture was heated at reflux for 1 h, and compound 8 (6.33 g, 7.00mmol) was added dropwise. The resulting solution was heated at reflux for 24 h and then cooled to room temperature and acidified with 1 mol dm-, HC1. The resulting solution was poured into water and extracted with chloroform. The chloroform solution was washed with water, dried over anhydrous magnesium sulfate, and filtered.The solvent was removed in a rotary evaporator, and the crude product was then purified by column chromatog- raphy [silica gel, methylene chloride-methanol (15: 1) eluent] to yield 2.7 g (41%) of a waxy solid. Compound 9. Yield, 62%. Mp, 156°C; TSAPN162"C, TN-, 179°C. 'H NMR, 6 3.37 (s, 3 H, CH,O), 3.50-4.20 (m, 12 H, OCH,), 6.97 (d, 2 Ar-H, o to CH20, J=8.7 Hz), 7.55 (2 Ar-H, m to CH,O, J=8.6 Hz), 7.63 (d, 2 Ar-H, m to COOH, J = 8.3 Hz), 8.10 (d, 2 Ar-H, o to COOH, J= 8.3 Hz). Elemental analysis for C2,H2,06: Calc. c, 66.65; H, 6.71. Found C, 67.02; H, 6.83%. Compound 10. Yield, 54%. Mp, 113.7 "C. 'H NMR, 6 3.35 (s, 3 H, CH,O), 3.50-4.20 (m, 28 H, OCH2), 6.96 (d, 2 Ar-H, o to CH,O, J=8.7 Hz) 7.54 (2 Ar-H, m to CH,O, J=8.5 Hz), 7.62 (d, 2 Ar-H, m to COOH, J=8.1 Hz), 8.11 (d, 2 Ar-H, o to COOH, J =8.2 Hz).Elemental analysis for C,,H,,O,,: Calc. C, 62.71; H, 7.51. Found C, 63.11; H, 7.41%. Compound 11. Yield, 50%. Mp, 96.6"C. 'H NMR, 6 3.34 (s, 3 H, CH,O), 3.50-4.20 (m, 48 H, OCH,), 6.98 (d, 2 Ar-H, o to CH20, J=8.7 Hz) 7.55 (2 Ar-H, rn to CH20, J=8.6 Hz), 7.63 (d, 2 Ar-H, m to COOH, J=8.1 Hz), 8.11 (d, 2 Ar-H, o to COOH, J =8.2 Hz). Elemental analysis for C38H60015: Calc. C, 60.30; H, 7.99. Found C, 60.71; H, 7.84%. Compound 12. Yield, 42%. Mp, 44 "C. 'H NMR, 6 3.38 (s, 3 H, CH,O), 3.50-4.20 (m, 64 H, OCH,), 7.02 (d, 2 Ar-H, o to CH20, J=8.7 Hz) 7.56 (2 Ar-H, m to CH20, J=8.8 Hz), 7.64 (d, 2 Ar-H, m to COOH, J=8.4 Hz), 8.1 1 (d, 2 Ar-H, o to COOH, J =8.3 Hz). Elemental analysis for C46H76019: Calc. C, 59.21; H, 8.21.Found C, 59.69; H, 7.78%. Synthesis of ethyl 4-hydroxy-4-biphenyl carboxylate 15. 4'-Hydroxy-4-biphenylcarboxylic acid ( 5.0 g, 23.3 mmol) and KOH (1.3 g, 23.3 mmol) were added to 30 ml dry DMSO under nitrogen. The resulting solution was stirred at 120 "C for 1 h, then cooled to 70°C. To the resulting solution were added TBAH (1.3 g) and bromoethane (2.54 g, 23.3 mmol). After being stirred at 70 "C for 24 h, the light-yellow solution was poured into water. The resulting precipitate was recrys- tallized from toluene to yield white crystals (2.0 g, 35.4%). Mp, 142'C. 'H NMR, 6 1.40 (t, 3 H, CH,CH,, J=7.15 Hz), 4.37 (9, 2 H. CH2CH3, J=7.1 Hz), 6.96 (d, 2 Ar-H, o to OH, J= 8.65 Hz), 7.15 (d, 2 Ar-H, m to OH, J=8.61 Hz), 7.62 (d, 2 Ar- H, m to COO, J=8.36 Hz), 8.05 (2 Ar-H, o to COO, J = 8.42 Hz).Synthesis of polymers 3-4, 7-4, 12-4 and 16-4. Ethyl 4'-(4'- [methyloxy di(ethyleneoxy)ethyloxy]-4-biphenylcarbonyloxy} 4-biphenylcarboxylate 3-4 and ethyl 4'-(4'-[methyloxy poly(- eth yleneoxy) e th ylox y] -4- biphen ylcarbon ylox y } -4- bip henylcar- boxylates 7-4, 12-4 and 16-4 were all synthesized using the same procedure. A representative example is described for 16-4. Compound 12 (3.9 g, 4.12 mmol), compound 15 (1.0 g, 4.12 mmol) and DPTS (1.2 g, 4.12 mmol) were dissolved in 50 ml dry CH,C1, under argon. The resulting mixture was stirred for 1 h and neat DIPC (1.48 ml) was then added. The reaction mixture was stirred at room temperature overnight and was then poured into methanol, the resulting precipitate was filtered off and dried under vacuum.The product was purified by column chromatography using silica gel [methylene chloride-methanol (15: 1) eluent] to yield 1.95 g (39%) of a white solid, which was further purified by recrystallization from a mixture of methylene chloride and hexane. Polymer 3-4. Yield, 60.4%. 'H NMR, 6 1.40 (t, 3 H, CH,CH,, J=7.2 Hz), 3.37 (s, 3 H, CH,O), 3.50-3.80 (m, 8 H, OCH,), 3.88 (t, 2 H, CH,CH,O-phenyl, J=5.1 Hz), 4.18 (t, 2 H, CH,CH,O-phenyl, J =5.1 Hz), 4.40 (4, 2 H, OCH2CH3, J= 7.1 Hz), 7.02 (d, 2 Ar-H, o to CH20, J=8.7 Hz), 7.34 (d, 2 Ar- H, o to biphenylcarboxylate, J= 8.8 Hz), 7.60 (2 Ar-H, m to CH20, J=8.8 Hz), 7.63-7.80 (m, 6 Ar-H, m to C00-phenyl, m to biphenylcarboxylate and m to COOCH,), 8.10 (d, 2 Ar- H, o to COOCH,, J=8.5 Hz), 8.23 (d, 2 Ar-H, o to COO-phenyl, J =8.4 Hz).Elemental analysis for C,,H3,08: Calc. C, 71.90; H, 6.21. Found C, 71.82; H, 6.33%. Polymer 7-4. Yield, 30%. 'H NMR, 6 1.39 (t, 3 H, CH,CH,, J=7.0 Hz), 3.35 (s, 3 H, CH,O), 3.50-3.82 (m, 26 H, OCH,), 3.87 (t, 2H, CH,CH,O-pheny, J=4.9 Hz), 4.17 (t, 2 H, CH,CH,O-phenyl, J =4.8 Hz), 4.38 (4, 2 H, OCH,CH,, J = 7.1 Hz), 7.02 (d, 2 Ar-H, o to CH20, J=8.7 Hz), 7.33 (d, 2 Ar-H, o to biphenylcarboxylate, J =8.5 Hz) 7.59 (2 Ar-H, m to CH,O, J =8.6 Hz), 7.63-7.80 (m, 6 Ar-H, tn to C00-phenyl, rn to biphenylcarboxylate and in to COOCH,), 8.09 (d, 2 Ar-H, o to COOCH,, J =8.4 Hz), 8.22 (d, 2 Ar-H, o to COO- phenyl, J =8.4 Hz).Elemental analysis for C43H52012: Calc. C, 67.88; H, 6.89. Found C, 67.45; H, 6.83%. Polymer 12-4. Yield, 30%. 'H NMR, 6 1.42(t, 3 H, CH,CH3, J=7.1 Hz), 3.38 (s, 3 H, CH30), 3.50-4.20 (m, 44 H, OCH,), 3.87 (t, 2 H, CH,CH,O-phenyl, J=4.9 Hz), 4.17 (t, 2 H, CH,CH,O-phenyl, J =4.8 Hz), 4.40 (9, 2 H, OCH2CH,, J = 7.1 Hz), 7.04 (d, 2 Ar-H, o to CH20, J=8.7 Hz), 7.3 (d, 2Ar- H, o to biphenylcarboxylate, J= 8.6 Hz) 7.60 (2 Ar-H, m to CH20, J=8.7 Hz), 7.63-7.80 (m, 6 Ar-H, tn to C00-phenyl, m to biphenylcarboxylate and m to COOCH,), 8.12 (d, 2 Ar- H, o to COOCH,, J=8.4 Hz), 8.25 (d, 2 Ar-H, o to COO- phenyl, J =8.4 Hz). Elemental analysis for C53H72017: Calc. C, 64.88; H, 7.40. Found C, 64.44; H, 7.42%.Polymer 16-4. Yield, 39%. 'H NMR, 6 1.39 (t, 3 H, CH2CH,, J=7.2 Hz), 3.35 (s, 3 H, CH,O), 3.50-3.80 (m, 60 H, OCH,), 3.87 (t, 2 H, CH,CH,O-pheny, J= 5.0 Hz), 4.17 (t, 2 H, CH,CH,O-phenyl, J =5.1 Hz), 4.37 (9, 2 H, OCH,CH,, J = 7.2 Hz), 7.00 (d, 2 Ar-H, o to CH20, J=8.7 Hz), 7.3 (d, 2 Ar-H, o to biphenylcarboxylate, J=8.6 Hz) 7.60 (2 Ar-H, m to CH20, J= 8.7 Hz), 7.63-7.80 (m, 6 Ar-H, m to C00-phenyl, m to biphenylcarboxylate and rn to COOCH,), 8.13 (d, 2 Ar-H, o to COOCH,, J=8.4 Hz), 8.25 (d, 2 Ar-H, o to COO- phenyl, J =8.4 Hz). Elemental analysis for C,lH8,O21: Calc. C, 63.31; H, 7.66. Found C, 63.10; H, 7.75%. Synthesis of ethyl 4'-[methyloxy poly (ethy1eneoxy)ethyloxy 1-4-biphenylcarboxylate 16-2. Ethyl 4'-hydroxy-4-biphenylcarb-oxylate (1.87 g, 7.74 mmol) and KOH (0.46 g, 7.74 mmol) were dissolved in 100 ml methanol.The mixture was heated at reflux for 1 h, and compound 8 (6.54 g, 7.74 mmol) was added dropwise. The resulting solution was heated at reflux for 24 h, then cooled to room temperature, poured into water and extracted with chloroform. The chloroform solution was washed with water, dried over anhydrous magnesium sulfate, and filtered. The solvent was removed in a rotary evaporator, and the crude product was then purified by column chromatog- raphy (silica gel, methylene chloride eluent) to yield 2.9 g J. Muter. Chem., 1996, 6(7), 1079-1086 1081 (38 8%) of a waxy solid 'H NMR, 6 1 34 (t, 3 H, CH2CH3, J=7 2 Hz), 3 31 (s, 3 H, CH30), 3 50-3 80 (m, 60 H, OCH,), 3 82 (t, 2 H, CH2CH20-phenyl, J=5 1 Hz), 4 12 (t, 2 H, CH2CH20-phenyl, J=5 0 Hz), 4 32 (9, 2 H, CH2CH3, J= 7 1 Hz), 6 96 (d, 2 Ar-H, o to CH20,J =8 8 Hz), 7 48 (2 Ar-H, rn to CH20, J=8 8 Hz), 7 56 (d, 2 Ar-H, m to COOCH2, J= 8 3 Hz), 8 00 (d, 2 Ar-H, o to COOCH,, J =8 4 Hz) Synthesis of 4-[methyloxy poly (ethy1eneoxy)ethyloxy ]ben-zoic acid 16.Compound 16 was synthesized according to the procedure described for compound 12, but starting from 4-hydroxybenzoic acid (2 3 g, 16 6 mmol) and compound 8 (15 g, 16 6 mmol) Yield, 6 1 g (42 2%) 'H NMR, 6 3 37 (s, 3 H, CH30), 350-420 (m, 64H, OCH,), 697 (d, 2Ar-H, o to CH,O, J =8 7 Hz), 7 55 (2 Ar-H, rn to CH,O, J =8 6 Hz), 7 63 (d, 2 Ar-H, m to COOH, J=8 3 Hz), 8 10 (d, 2 Ar-H, o to COOH, J=8 3 Hz) Synthesis of ethyl 4-[4-[methyloxy poly(ethy1eneoxy)ethy- loxy]benzoyloxy]-4-biphenylcarboxylate 16-3.Compound 16-3 was synthesized according to the procedure described for compound 16-4, but starting from compound 16 (3 1 g, 3 56 mmol) and compound 15 (0 86 g, 3 56 mmol) Yield, 107 g (28%) 'H NMR, S 139 (t, 3 H, CH2CH3,J=7 2 Hz), 3 35 (s, 3H, CH,O), 350-375 (m, 60H, OCH,), 388 (t, 2H, CH2CH,0-phenyl, J =5 1 Hz), 4 20 (t, 2 H, CH,CH,O-phenyl, J =5 1 Hz), 4 39 (9, 2 H, OCH2CH3, J =7 1 Hz), 7 00 (d, 2 Ar-H, o to CH20, J =8 7 Hz), 7 30 (d, 2 Ar-H, o to benzoate, J = 8 8 Hz), 7 60-7 70 (m, 4 Ar-H, m to benzoate and rn to COOEt), 7 63-7 80 (m, 6 Ar-H, rn to C00-phenyl, rn to biphenylcarb- oxylate and m to COOCH,), 8 10 (d, 2 Ar-H, o to COOCH,, J=8 5 Hz), 8 15 (d, 2 Ar-H, o to C00-phenyl, J=8 4 Hz) Preparation of complexes of 16-4.Complexes of 16-4 with lithium triflate were prepared by mixing solutions of 16-4 (10 mg ml-') in dry methylene chloride with an appropriate volume of 0 724 mmol ml-' salt in dry acetonitrile solution, followed by slow evaporation of the solvent under reduced pressure at room temperature and subsequent drying in a vacuum oven at 60°C to maintain constant mass Addition of Table 1 Characterization of the rod-coil polymers (data on first line are from first heating and cooling scans, data on second line are from second heating scan) compound M,/M," 3-4 7-4 106 12-4 1 08 16-4 1 05 16-2 106 16-3 105 ~ ~~ ~~~~~ phase transitionsb (in "C) and corresponding enthalpy changes (in kJ mol ') heating g 1101 K 1606(102)SA 2852(18)N 3027(054)I g 982 K 1569 (96) SA 2801 (1 5) N 301 6 (048) I g 823 K 1557 (127) SA 231 8 (1 29) I g 73 5 K 1506 (11 7) SA 2322 (1 18) I g 68 5 K 135 2 ( 15 6) SA 148 1 (1 81) I g 603 K 1344 (149) SA 1479 (1 95) I g 32 0 K 55 7 (74 9) K 120 3 (19 4) SB 123 0' (-) I K347(444)K 1205(197)SB 1230'(-)I K432(673)1 K 349 (557) I K 325(703)SX352'(-)1 K 266 (51 6) S, 344 (1029) I cooling 12944(041)N 2713(165)SA 1473(100)K 852 g I2239 (1 01) SA 139 3 (7 5) K 643 g I 1440(192)SA 1289(154)K554g I ll94'(-)SB 1160(189)K 115(696)K I103 (602) K I258 (935) S, 5 l(53 1) K 'From SEC data bN, nematic, SA, smectic A, S,, smectic B, S,, unidentified smectic phases, K, crystalline, I isotropic 'Overlapped peak, data obtained from optical polarized microscopy d e 9 1 eg fdI la ll I I I 8.5 8-0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 6 Fig.1 300 MHz 'H NMR spectrum of 16-4 1082 J Muter Chern, 1996, 6(7), 1079-1086 3-4& (b 1 3-4 /'//" 16-21 7-4 12-4 10 90 170 250 330 10 90 170 250 330 TI" C Fig. 2 DSC traces (10"C min-') recorded during the second heating scan (a) and the first cooling scan (b)of 3-4, 7-4, 12-4 and 16-4 K 01 1 I I I90 5 10 15 20 c; JJU 300 250 200 150 100 A IrA5010 9 K 0 5 10 15 20 no. of EO units Fig. 3 Dependence of the phase-transition temperatures of rod-coil polymers 3-4, 7-4, 12-4 and 16-4 on the number of EO repeat units in the coil.(a) Data from second heating scan: A, q;0,T,; 0,TsA-TN;0, ?;. (b) Data from first cooling scan: A,q;U, TM;+, TsA-Tl; .,T. I 16-4 SB I K/ -10 30 70 110 150 7°C Fig. 4 DSC traces (10"C min-') recorded during the second heating scan (upper traces) and the first cooling scan (lower traces) of 16-2, 16-3 and 16-4 LMI I 1-10 30 70 110 150 -10 30 70 110 150 TI"C Fig. 5 DSC traces (10 "C min-') recorded during the second heating scan (a) and the first cooling scan (b)of the complexes of 16-4 with lithium triflate. [Li+]/[EO]=(i) 0, (ii) 0.05, (iii) 0.1, (iv) 0.2, (v) 0.3. a solution of the salt in acetonitrile to methylene chloride gave a precipitate free of 16-4.The absence of the free salt was verified by DSC (absence of the melting point of the free salt in the heating and cooling scans) and optical polarized microscopy. J. Mater. Chem., 1996, 6(7), 1079-1086 1083 Table 2 Thermal transitions of the complexes of rod-coil polymer 16-4 with lithium triflate (data on first line are from first heating and cooling scans, data on second line are from second heating scan) phase transitions‘ (in “C) and corresponding enthalpy changes (in kJ mol-’) heating cooling 0 00 0 05 0 10 0 20 0 30 g 32 0 K 55 7 (74 9) K 120 3 (19 4) SB 123 Ob (-) I K 34 7 (44 4) K 120 5 (19 7) SB 123 Ob (p)I K 22 4 (34 8) K 1104 (9 5) SB 117 3’ (-) SA 141 8 (1 04) I K 15 6 (22 5) K 113 3 (8 7) SB 116 7’ (-) SA 142 5 (097) I K 1007 (1 01) SB 108 2 (24) SA 147 5 (045) I K 994(123)SB 1076(195)S, 1467(038)1 K 825 (124)M 1342(024)I K 723 (O45)M 1334(018)1 K 128 4 (54 3) M 130 2’ (-) I Cr 63 4 (38 3) K 128 2 (38 4) M 130 0’ (-) I Ill94’(-)SB 1160(189)K 115(696)K I 138 9 (092) SA 116 3’ (-) S, 109 6 (645) K -6 I 141 0 (035) SA 1109’ (-) S, 107 3 (4 5)K I 130 7 (0 32) M 59 6 (7 14) K I 1074 (0 13) M 71 2 (154) K 5 (20 3) K “SA, smectic A, SB, smectic B, M, cylindrical mesophase, Cr, recrystallization, K, crystalline, I, isotropic bOverlapped peak, data obtained from optical polarized microscopy 180 13 0 80 3 0 9-20 R 00 0.1 0.2 0.3 1 13 n1 0 M 8 K 3 0 K K.-2 UI I I 00 0.1 0.2 03 [Li+]/[EO] Fig. 6 Dependence of the phase-transitlon temperatures of the com- plexes of 16-4 with lithium triflate on [LiCF,S03]/[EO] (a) Data from second heating scan 0,T,, A, TSB 0, T, (b)Data from first sA, cooling scan W, T,, A,TSBsA, 0,T, Results and discussion The synthesis of rod-coil polymers with different coil lengths is outlined in Scheme 1 Commercially available triethylene glycol monomethyl ether and poly(ethy1ene oxide) monomethyl ethers (normal average molecular masses {Mw},350, 550 and 750) were used as starting materials for 3-4, 7-4, 12-4 and 16-4, respectively To investigate the effect of the rod length, the rod-coil molecules of poly(ethy1ene oxide) monomethyl ether ((M,} = 750) with two (16-2) and three (16-3) phenyl 1084 J Muter Chem, 1996, 6(7),1079-1086 ring systems are synthesized as outlined in Schemes 2 and 3, respectively Remarkably, all rod-coil compounds could be isolated by column chromatography (silica gel) from the resulting mixture of each esterification reaction using a mixture of CH,Cl, and methanol (15 1 v/v) as eluent The rod-coil polymers were then recrystallized from methylene chloride and hexane to obtain highly monodisperse polymers It is well known that polydispersity influences the phase behaviour of polymers, especially of those with low molecular masses l5 Therefore, it is essential that the rod-coil polymer is highly monodisperse in order to investigate its accurate phase behavi our Table 1 shows the characterization of the rod-coil polymers All rod-coil molecules showed polydispersity values <1 1, as determined from SEC, indicative of high purity Fig 1 presents a typical ‘H NMR spectrum of 16-4 with its protonic assign- ments The resonances of the expected methoxy chain-end of the coil and the ethyl chain-end of the rod can be observed easily at 6 3 35 and 4 37, respectively No other signals indica- tive of impurities are observed The DSC traces obtained during the first and subsequent heating scans are identical for the rod-coil polymers The experimental data collected from both scans are summarized in Table 1 However, only the second heating and first cooling scans will be presented in more detail Fig 2 presents the DSC traces of 3-4, 7-4, 12-4 and 16-4 Rod-coil compound 3-4 exhibits enantiotropic nematic (N) and smectic A (S,) meso-phases On the optical polarizing microscope, the nematic mesophase of 3-4 exhibits a schlieren texture while the SA mesophase exhibits a focal conic texture Both 7-4 and 12-4 display a crystalline (K) melting and an enantiotropic SA mesophase Interestingly, 16-4, which contains the longest coil length in this study, exhibits two crystalline melting transitions in addition to an enantiotropic smectic B (S,) phase This is not unexpected because microphase separation between stiff rod and flexible coil segments occurs and consequently, each crystalline melting transition corresponding to the rod and coil segments is exhibited The dependence of the transition temperatures of the rod- coil molecules (n-4 series) determined by DSC as a function of the number of ethylene oxide units is plotted in Fig 3(a) (for the second heating scans) and Fig 3(b)(for the cooling scans) Both TMand 7;decrease with increasing number of ethylene oxide units However, the gradient of the 7;variation is much steeper than that of TM This is similar to the usual trend for liquid crystals containing flexible tails or spacer moieties l6 This plot demonstrates that the higher crystalline melting transition of 16-4 corresponds to the rod block because this agrees with the continuous character of the dependence of TM on the number of ethylene oxide units In the case of rod-coil molecules with short chain-lengths, Plate 1 Representative optical polarized micrographs (100x) of the texture exhibited by: (a) the SB mesophase of 16-4 at 120°C on the cooling scan; (b) the SA mesophase of the complex of 16-4 with 0.1 mol LiCF,SO, at 138°C on the cooling scan; (c) the cylindrical micellar mesophase of the complex of 16-4 with 0.2 mol of LiCF,SO, at 110 C on the cooling scan the coil may couple with the anisotropic rod owing to the relatively high miscibility between rods and coils with short length, which can induce a nematic phase as exhibited by rod-coil compound 3-4.However, as the length of hydrophilic poly(ethy1ene oxide) coil increases or the temperature decreases, the immiscibility between the hydrophobic rigid rods and the hydrophilic flexible coils increases. This allows for the increasing lateral intermolecular interactions of aro- matic rods.As a result, a layered S, phase can be induced, as exhibited by 3-4 (the lower-temperature mesophase), 7-4 and 12-4. As the length of the coil is increased further, the sharper interdomains between microphase-separated domains may exist to form a more ordered layered smectic phase,I7 as exhibited by rod-coil polymer 16-4 which displays an SB phase. This is well established by theoretical prediction^.'.^ This result indicates that the length of the poly(ethy1ene oxide) coil plays an important role in the transition temperature and the nature of the mesophase of the rod-coil polymer. To investigate the effect of the length of the molecular rod, the thermal behaviours of 16-2, 16-3 and 16-4 are compared in Fig.4 which presents the second heating and first cooling DSC scans. As can be seen, 16-2 exhibits only a crystalline melting, while 16-3 displays a crystalline phase followed by an enantiotropic smectic X (S,) phase and 16-4 exhibits two crystalline melting transitions in addition to an SB phase as described already. Upon increasing the length of the rod, 7; increases rapidly. However, within experimental error, TM is relatively constant and independent of the rod length. This indicates that the lower crystalline melting transition corre-sponds to the poly(ethy1ene oxide) coil segments. Rod-coil systems containing poly(ethy1ene oxide) can com- plex with alkali-metal cations. When alkali-metal cations are added to the host rod-coil polymer molecule, they will be dissolved selectively in the poly(ethy1ene oxide) coil segments of the rod-coil diblock system via ion-dipolar interactions.It has already been reported that alkali-metal salts are selectively soluble in the poly(ethy1ene oxide) segments of a block copoly- mer containing poly(ethy1ene oxide)." Therefore, complexation of the rod-coil polymer with an alkali-metal salt results in an increase of the relative volume fraction of coil compared to that of rod, which may give rise to a novel supramolecular architecture of the rod-coil system. In this context, we have investigated the mesomorphic behaviour of complexes of 16-4 with LiCF,SO,. The phase behaviour of the rod-coil polymer and its complexes with lithium triflate was characterized by a combination of DSC and thermal optical polarized microscopy.Fig. 5 shows DSC traces of the second heating and the first cooling scans of the rod-coil polymer and of its complexes with 0.05-0.3 mol lithium triflate per ethylene oxide unit. The first and subsequent heating scans and the first and subsequent cooling scans are identical except for the complex with 0.3 mol LiCF,S03, which undergoes recrystallization on the second heating scan. The phase transitions from Fig. 5 are summarized in Table 2 and are plotted in Fig. 6(a) (data from the second heating scan) and Fig. 6(b) (data from the cooling scan) as a function of the concentration of LiCF,SO, in the rod-coil polymer. As shown in Fig. 5, 16-4 exhibits an enantiotropic SBphase in addition to crystalline melting transitions.A representative texture of an SBphase exhibited by 16-4 is shown in Plate l(a). The complexes of 16-4 with 0.05 and 0.1 mol LiCF,SO, per ethylene oxide unit exhibit an enantiotropic S, mesophase in addition to the S, and K phases. Upon complexation of 16-4 with up to 0.1 mol LiCF,SO,, increases and both crystalliza- tions are suppressed. This trend agrees well with previous results" and is predicted by theory." Plate l(b) presents a representative texture displayed by the S, phase exhibited by the complex with 0.1 mol LiCF,SO,. However, in contrast to the thermal behaviour of the complexes with up to 0.1 mol LiCF,S03, the complexes with 0.2 and 0.3 mol do not exhibit smectic layered mesophases, but they display an enantiotropic cylindrical micellar mesophase (M) as their highest-tempera- ture mesophase.On cooling from the isotropic (I) phase, first a platelet-like growth of the texture can be observed with a final development of pseudo-focal conic domains which is characteristic of a conventional disordered columnar meso-phase, as shown in Plate ~(c).~O The complex with 0.3 mol LiCF3S03 undergoes salt-induced crystallization through endothermic and exothermic peaks on the second heating scan followed by the transition from M to I, indicating that com- plexation of 16-4 with LiCF,SO, induces an S, phase or a J. Mater. Chem., 1996,6(7), 1079-1086 1085 cylindrical micellar mesophase from its layered SB phase, depending on the salt concentration The existence of a cylindrical micellar mesophase is in contrast with the normal behaviour of rigid-rod calamitic mesogens which show lamellar smectic and/or nematic phases Lamellar and cylindrical micellar mesophases for a given compound are commonly found in lyotropic liquid crystals2' 22 and also contribute to the thermal behaviour of some amphi- philic molecules such as silver thi~lates,~~ and biforked mol- ecules2425 On the other hand, columnar mesophases are usually induced by disk-shaped mesogens and are consequently prone to this kind of stacking 26 Therefore, the system discussed here is an unusual case of a calamitic mesogenic system owing to the ability of the rod-coil polymer to change from lamellar to micellar upon complexation Apparently the lamellar smectic structure observed in the uncomplexed rod-coil polymer and the complexes with up to 0 1 mol LiCF,SO, is still the most efficient packing of melt coils because the volume fraction of the coil parts is not large enough For higher concentrations of LiCF,SO,, however, the volume fraction of coil segments increases by insertion of the salt into the coil domain through ionic interactions between Li' and electron-donor oxygen atoms, and the system becomes unstable owing to space crowding of the coil segments, conse- quently, the lamellar structure of the rod-coil polymer will break apart into cylindrical micelles, as predicted by theory The main advantage of these micelles relative to lamellae is that the coils grafted onto their top and bottom surfaces are able to fan out into a larger region of space, presumably to reduce the thermodynamic coil-stretching penalty This might explain qualitatively the phase behaviour of the complex system, although more distinctive morphological experiments, such as X-ray scattering methods, are required to support this speculative explanation, such studies are in progress In conclusion, a series of rod-coil molecules containing poly(ethy1ene oxide)s and the complexes of 16-4 with LiCF,SO, have been prepared The lengths of the coils and rods in the rod-coil molecule influence significantly the nature and the thermal stability of the mesophase The complexation of 16-4 with lithium cations induces either a thermo-dynamically stable smectic A phase or a cylindrical micellar mesophase, depending on the salt concentration In particular, transformation of a layered smectic phase into a cylindrical micellar assembly by simple complexation is promising These results provide access to a large variety of fundamental investi- gations and technological applications Financial support of this work by the Korea Science and Engineering Foundation ( 1995) is gratefully acknowledged References 1 A Halperin, Macromolecules, 1990,23, 2724 2 A N Semenov and S V Vasilenko, Sou Phys JETP (Engl Trans1 ), 1986,63,70 3 A N Semenov, Mol Cryst Liq Cryst, 1991,209,191 4 D R M Williams and G H Fredrikcson, Macromolecules, 1992, 25,3561 5 L H Radzilowski, J Wu and S I Stupp, Macromolecules, 1993, 26,879 6 L H Radzilowski and S I Stupp, Macromolecules, 1994,27,7747 7 S I Stupp, M Lee, S Son, L S Li and M Kesser, Polym Prepr, 1993,43,184 8 F B Dias, J P Voss, S V Batty, P V Wright and G Ungar, Makromol Chem , Rapid Commun , 1994,15,961 9 V Percec, J Heck, D Tomazos, F Falkenberg, H Blackwell and G Ungar, J Chem Soc Perkin Trans I, 1993,2799 10 V Percec, D Tomazos, J Heck, H Blackwell and G Ungar, J Chem SOC Perkin Trans 2,1994,31 11 V Percec, J Heck, G Johansson and D Tomazos, Makromol Chem Macromol Symp , 1994,77,237 and references therein 12 J S Moor and S I Stupp, Macromolecules, 1990,23,65 13 D Demus and L Richter, Textures of Liquid Crystals, Verlag Chemie, Weinheim, 1978 14 G W Gray and J W Goodby, Smectic Liquid Crystals Textures and Structures, Leonard Hill, Glasgow, 1984 15 V Percec and M Lee, Macromolecules, 1991,24, 1017 16 G W Gray, Molecular Structure and the Properties of Liquid Crystals, Academic Press, London, 1962, ch 9 17 J Adams and W Gronski, Makromol Chem Rapid Commun, 1989,10,553 18 M Watanabe, S Oohashi, K Sanui, N Ogata, T Kobayashi and Z Ohataki, Macromolecules, 1985, 18, 1945, M Watanabe, K Nagaoka, M Kanba and I Shinohara, Polym J , 1982,14,877 19 A Keller, G Ungar and V Percec, in Advances in Liquid Crystalline Polymers, ed C K Ober and R A Weiss, ACS Symp Ser 435, American Chemical Society, Washington DC, 1990, p 308 20 C Destrade, P Foucher, H Gasparoux and N H Tinh, Mol Cryst Liq Cryst, 1984,106,121 21 B Luemann and H Finkelmann, Colloid Polym Scz, 1987, 265, 506 22 H Hoffmann, Adu Mater , 1994,6,116 23 M J Baena, P Espinet, M C Lequerica and A M Levulet, J Am Chem SOC,192,114,4182 24 Y Fang, A M Levulet and C Destrade, Liq Cryst, 1990,7,265 25 C Destrade, N H Tinh, A Roubineau and A M Levulet, Mol Cryst Liq Cryst, 1988,159,163 26 K Praefcke, P Marquardt, B Kohne, Z Luz and R Poupko, Liq Cryst, 1991,9,711 Paper 5/05652G, Received 25th August, 1995 1086 J Mater Chem , 1996, 6(7),1079-1086