J. MATER. CHEM., 1994, 4( 5), 747-759 Effect of the Position of Lateral Fluoro Substituents on the Phase Behaviour and Ferroelectric Properties of Chiral I-Methylheptyl 4'-[(2-or 3-Fluoro-4-tetradecyloxyphenyl)propioloyloxy]biphenyl-4-ca rboxylates Christopher J. Booth," David A. Dunmur,b John W. Goodby,*" Jaskaran S. Kangb and Kenneth J. Toyne"" School of Chemistry, The University of Hull, Hull, UKHU6 7RX Centre for Molecular Materials, Department of Chemistry, University of Sheffield, Sheffield, UK S3 7HF The syntheses of four chiral laterally fluoro-substituted propiolate esters are described, along with transition tempera- tures, ferroelectric properties, phase diagrams and related data. The position of the fluoro-substituent was found to influence dramatically the formation of twist grain boundary (TGB A* and TGB C)phases as well as the magnitudes of spontaneous polarization and optical tilt angle in the ferroelectric smectic C*mesophases.Differential scanning calorimetric studies revealed the presence of a diffuse liquid-liquid transition above the clearing point in both of the 3-fluoro enantiomers, but not in the racemate. Circular dichroism and optical rotation measurements, carried out over a temperature range in which the diffuse peak occurs, appear to confirm the presence of a degree of chiral organization within the isotropic liquid. It is suggested that this phenomenon may be due to the presence of cybotactic groups or a network of entangled screw dislocations occurring close to the clearing point.The existence of materials that display the novel twist grain boundary phase (TGB A* phaset), as originally predicted by de Gennes' and developed later by Renn and Lubensky,' have now been well doc~mented.~~ The early structural studies performed by Goodby et al. on chiral 1-methylheptyl 4' -[( 4-alkoxyphenyl)propioloyloxy]biphenyl -4 -carboxylates (1) revealed that TGB A* phases were formed only when certain criteria were satisfied as follow^:^*^ 1 n=8to 16,t= (R)or(S) (i) the presence of a short-range smectic A* (Sz) phase and an Sz-smectic C* (Sz) transition close to the clearing point, thus allowing pretransitional fluctuations to stabilize the resulting twisted TGB A* phase; (ii) the transition from the cholesteric or liquid phase must approach being second order (this was often the case for the isotropic liquid to smectic A transitions for homologues of 1 where n =12-15); and (iii) the system must have strong chirality so as to be able to induce a twist through the smectic A phase.However, little work has been performed on the influence of molecular structure on the formation and phase stability of the TGB A* phase other than for increasing the terminal alkoxy substituent's chain length (although fluoro-substituents have been employed in a series of structurally isomeric tolanes'). Our planned work was to investigate the influence and effect of the positioning of lateral fluoro-substituents in the 4-alkoxyphenylpropiolate core on the formation of TGB A* phases and the ferroelectric properties of the smectic C* phases.The two materials selected for this study proved to have markedly contrasting properties, these were the (R)-, t The asterisk notation represents a phase with reduced symmetry caused by the inclusion of chiral molecules into the system, rather than using it to represent a helical macrostructure, hence smectic A compound of chiral compounds is denoted A*. and (S)-1-methylheptyl 4'-[(2-fluoro-4-tetradecyloxyphenyl) propioloyloxy] biphenyl-4-carboxylates (2) and (R)-and (S)-1-methylheptyl 4-[(3-fluoro-4-tetradecyloxyphenyl)propiol-oyloxy] biphenyl-4-carboxylates (3). The choice of the tetra- decyloxy substituent was deliberate in that it is a chain which is long enough to induce 'second-order' clearing behaviour in related systems, and it allows easy comparison with the parent non-fluoro-substituted series.334 F .O 3 t = (H)or (S) Experimental The phenylpropiolic acid core units were prepared as shown in Scheme 1.The first step in the synthesis is the al- kylation of the appropriately substituted 4-bromo-2/3-fluorophenol (compounds 4 and 5)with 1-bromotetradecane to give compounds 6 and 7, respectively. The individual cam- pounds were then cross-coupled with a protected terminal alkyne using palladium(0) tetrakis(triphenylphosphine), copper(1) iodide in anhydrous diisopropylamine to give the appropriately fluoro-substituted 1-(3-hydroxy-3-methyl-butynyl)-4-tetradecyloxybenzenes(compounds 8 and 9).'-'' Both compounds were subsequently deprotected using potass- ium hydroxide in refluxing toluene to furnish the ethynyl- benzenes (10 and 11) as low melting crystalline solids in J.MATER. CHEM., 1994,VOL. 4 HOsBr 4 X=F,Y=H 5 X=H,Y=F 1. C14H290 6 X=F,Y=H 7 X=H,Y=F 1. Ic t C14H290 10 X=F,Y=H 11 X=H,Y=F XY 12 X=F,Y=H 13 X=H,Y=F Scheme 1 Synthetic route to the (2-or 3-fluoro-4-tetradecyloxy-pheny1)propiolic acids. (a) CI4H2?Br, butanone, reflux; (b)3-methyl-but-1-yn-3-01, Pd( PPh,),, CuI, Pr',NH, N,, reflux; (c) KOH, toluene, N,, reflux; (d) (i) BuLi, THF, N,, -10"C; (ii) CO,(s), THF, -10"C to rt; (iii) HC1 (conc.). moderate yields.' Initial attempts to lithiate and carboxylate these terminal acetylenic compounds at -78 "C with butyl- lithium and then solid C02 failed, the starting terminal al- kyne was recovered. Carboxylation was eventually achieved, at the somewhat higher temperature of ca.-10 "C using the same reagents, to give the appropriate (2-or 3-fluoro-4-tetra- decyloxypheny1)propiolic acid (12 and 13). This approach to the propiolic acids differs slightly from the original article in which use was made of the Corey-Fuchs reaction that employs 4-alkoxy-j3,j3-dibromostyrenes prepared from the 4-alkoxy- benzaldehyde precursor^.^*'^ The (R)-or (S)-1-methylheptyl 4-hydroxybiphenyl-4-carboxylates (19 and 20) were synthesized as shown in Scheme 2.4'-Hydroxybiphenyl-4-carboxylicacid (15)was pre- pared by the acid hydrolysis of 4-cyano-4-hydroxybiphenyl (14, Merck). Compound 15 was then protected by treatment with methyl chloroformate in aqueous sodium hydroxide at ca.0 OC;I2 the 4-methoxycarbonyloxybiphenyl-4'-carboxylic acid (16) obtained was then esterified using (S)-or (R)-14 15 Ib 16 Id 12or 13 1 21 X= F, Y =H, (R) 22 X=F,Y=H,(S) 23 X=H,Y=F,(R) 24 X= H,Y = F, (S) Scheme2 Synthetic route to the (R)-and (S)-1-methylheptyl 4' [(2-or 3-fluoro-4-tetradecyloxyphenyl)propioloyloxy]biphenyl-4 carboxylates. (a) H,S04, AcOH, H,O, reflux; (b) (i) NaOH, CH30COC1,H,0,00C; (ii) 1 :1 HCl: H,O, pH= 5; (c) (R)-or (S)-octan-2-01, diethyl azodicarboxylate, PPh,, THF,N,, rt; (d) EtOH, NH3(aq), rt; (e)dicyclohexylcarbodiimide, 4-N.,Y-dimethylaminopyri-dine, CH2C1,, rt.octan-2-01, diethyl azodicarboxylate and triphenylphosphine (all obtained from Aldrich) to give the (R)-and (S)-protected esters (17 and 18)as colourless liquids after flash chromatogra- ph~;'~the reaction proceeds with inversion of the absolute configuration at the chiral centre, but with no detectable ra~emization.'~Compounds 17 and 18 were then deprotected using an ethanol-ammonia solution at room temperature to give the corresponding (R)-or (S)-1-methylheptyl 4'-hydroxy- biphenyl-4-carboxylate esters (19 and 20) in good yield.12 Each of these optically active biphenol esters were then esterified with the appropriate (2-or 3-fluoro-4-tetradecyloxy- J. MATER. CHEM., 1994, VOL. 4 pheny1)propiolic acid using dicyclohexylcarbodiimide and 4-N,N-dimethylaminopyridine at room temperature to give the four liquid-crystalline target compounds (21-24).15 Procedures The chemical structures of all intermediates and final products, were confirmed by a combination of the following techniques: 'H nuclear magnetic resonance (NMR) spectroscopy (JEOL GX NM270 FT-NMR spectrometer using tetramethylsilane as the internal standard), infrared spectroscopy (Perkin-Elmer 783 spectrometer) and mass spectrometry (MS; Finnigan 1020 GC-MS spectrometer).Specific optical rotations were per- formed using a Bendix-NPL Automatic Polarimeter Type 143 optical unit and controller; chloroform (SpectrosoL) was used as the solvent. At each stage of preparation the materials were purified by column chromatography (Fisons 60-120 mesh silica gel) or flash chromatography (May & Baker Sorbsil c60 40-60H pm silica gel) as described by Clark Still6 The purity of all intermediates and final compounds was checked by thin-layer chromatography (TLC; using Merck 60 F254 preformed aluminium- backed plates) and by normal- and reversed-phase high-performance liquid chromatography (HPLC) using Microsorb C18 or Si columns and acetonitrile (May & Baker, Chromanorm) as the mobile phase.Each of the final products was found to have a chemical purity in excess of 99.5%. All solvents used in reactions were dried, distilled and stored as described in Perrin and Armarego." The initial phase assignments and transition temperatures were determined by thermal polarized light microscopy using a polarizing microscope (Zeiss Universal) in conjunction with a hot-stage and controller (Mettler FP82 microfurnace and FP8O control unit).Differential scanning calorimetry (DSC; Perkin-Elmer DSC7 calorimeter, TAC7/PC controller and IBM system/2 Model 50Z computer) was used to determine both the transition temperatures and the heats of transitions. The instrument was calibrated against an indium standard (measured AH =28.35 J g-', literature value 28.45 J g-')'' and all enthalpies are quoted in kJ mol-l. Tilt angle and ferroelectric polarization measurements were performed in planar aligned 3.5 or 3.6 pm thick cells (Electronics Chemicals High Technology Group) with an active area of 0.25 cm2 indium-tin oxide electrodes, which had been previously treated with unidirectionally buffed poly- imide alignment layers.The electrical contacts were made directly to the internal surfaces. The cells were filled by capillary action while the materials were in their isotropic liquid states. Reasonably good alignment for optical and polarization studies was achieved by slowly cooling each material from the isotropic liquid into their respective smectic states. Cooling rates were of the order of ca. 0.2 "C min-l. Once filled, the cell was connected to an ac frequency generator (6 or 15 V peak to peak and 60 Hz), a dual trace oscilloscope and a Diamant bridge.lg Tilt angles (6) in the smectic C* phase were obtained by measuring the angle (26) between the optical extinction positions for the ferroelectrically switched states observed in the polarizing microscope.The ferroelectric polarization (P,) was measured using the Diamant bridge, while observing a characteristic hysteresis loop for switching at 60 Hz; polarization values are quoted in nC cm-2. A sensitive modulation technique using a photoelastic modulator was used to make precision optical rotation measurements in the presence of linear birefringence. Compound 22 was studied in a 100 pm thick glass cell with planar aligned walls over the temperature range from ca. 80 to ca. 110 "C.The full details of the technique will be described elsewhere. Before taking each measurement, the sample was 749 kept at constant temperature (i.e.kO.1 "C) for ca. 5 min. Circular dichroism (CD) measurements were carried out using the spectropolarimeter JASCO J-4OCS. Compound 22 was studied in a quartz cell of thickness 18 pm with homeotrop- ically aligned walls. The sample was maintained at constant temperature (ie. k0.2 "C) for ca. 5 min before running a spectrum. l-Bromo-3-fluoro-4-tetradecyloxybenzene,6 1-Bromotetradecane (9.15 g, 33.0 mmol) in butanone (20 ml) was added dropwise to a stirred, refluxing mixture of 4-bromo-2-fluorophenol (4) (6.02 g, 31.5 mmol), potassium carbonate (5.62 g; 40.7 mmol) and butanone (80ml). The resulting reaction mixture was heated under reflux for a further 20 h, TLC showed virtually complete reaction.The cooled reaction mixture was then filtered to remove the excess of potassium carbonate and precipitated potassium bromide. The filtrate was washed with 5% v/v sodium hydroxide (2 x 50 ml) then water (50 ml) and the organic layer was dried (MgSO,), filtered and the solvent evaporated off under reduced pressure to give a colourless oil, which was purified by column chromatography [silica gel; 1: 1 dichloro-methane-light petroleum (bp 40-60 "C)] to give a colourless liquid. This was then dried in U~CUO(P,O,, 0.20 mmHg, room temperature, 5 h). Yield, 11.82 g, 97%; mp 28-29 "C; 6, (270 MHz, solvent CDCl,, standard TMS); 0.87 (3 H, m), 1.35 (22 H, mI, 1.80 (2 H, quintet), 4.00 (2 H, t), 6.82 (1 H, t), 7.20 (2 13, m). v/cm-' (KBr disc): 2930, 2860, 1505, 1470, 1305, 1265. 1210, 1130, 875, 860, 800, 635 and 570.m/z 388 (M', 3%), 386 (M+, 5%), 191 (98%), 190 (loo%), 135 (lo%), 125 (5O/0), 111 (?&yo), 98 (21%). l-Bromo-2-~uoro-4-tetradecyloxybenzene,7 This was prepared using a similar method to that described for compound 6. Quantities used: compound 5 (4.03 g, 21.1 mmol), 1-bromotetradecane (6.41 g, 23.1 mmol), potass- ium carbonate (5.59 g, 40.5 mmol) and butanone (115 nil). Yield, 7.48 g, 92%; mp 30-31 "C; 6H (270 MHz, sdvent CDC13, standard TMS): 0.85 (3 H, t), 1.35 (22 H, m). 1.75 (2 H, quintet), 3.90 (2 H, t), 6.60 (1 H, d), 6.70 (1 H, d), 7.40 (1 H, t). v/cm-' (KBr disc): 2920, 2850, 1605, 1580, 1485, 1465,1320,1290,1170,1140, 1020,830 and 640. m/z 388 (M', 20%), 386 (M', 20?40), 191 (29%), 189 (28%), 134 (200/1,).3-Fluoro- 1-( 3-hydroxy-3-methylbut- l-ynyl)-4-tetradecyl oxybenzene, 8 A stream of dry nitrogen was bubbled through a stirred, dark- green mixture of compound 6 (5.40 g, 13.9 mmol), pal- ladium(0) tetrakis( triphenylphosphine) ( 1.05 g, 0.91 minol), copper(1) iodide (0.13 g, 0.68 mmol) and dry diisopropyl- amine (30ml) for a period of 10min. A solution of 3-methylbut-1-yn-3-01 (2.54 g, 30.2 mmol) in dry diisopro- pylamine ( 10 ml) was added dropwise at room temperature; the reaction mixture turned deep orange-brown. The reaction was then heated for 4 h under gentle reflux and under nitrogen. The cooled reaction was filtered through a pad of Hyflo- supercel II(BDH) and water (50 ml) added to the filtrate.The crude product was then extracted into diethyl ether (3 x 50 ml); the combined ether extracts were washed with brine (50 ml), dried (MgS04), filtered and the solvent was evaporated off to give a brown oil. The crude product was then purified by flash chromatography [fine mesh silica gel; 10% v/v ethyl acetate in light petroleum (bp 40-60 "C)] to give a white solid, which was recrystallized (cyclohexane) and dried in uucuo ( P205,0.40 mmHg, room temperature, 24 h). Yield, 4.16 g, 76%; mp 45-46 "C; 6, (270 MHz, CDCl,, standard TMS): 0.80 (3 H, t), 1.30 (22 H, m), 1.60 (6 H, s), 750 1.80(2H,quintet), 1.90(1 H, s, broad), 4.00 (2 H, t), 6.85 (1 H, t), 7.12 (2 H, d). v/cm-' (KBr disc): 3400,2990, 2960, 2920, 2850, 1615, 1570, 1520, 1470, 1310,1275, 1240, 1160, 1120, 950, 880, 815, 720 and 615.m/z 390 (M', lo%), 375 (5%), 194 (30%), 179 (100%).2-Fluoro-1-ynyl)-4-tetradecyl-1-(3-hydroxy-3-methylbut-oxybenzene,9 This compound was prepared using a similar method to that described for compound 8. Quantities used: compound 7 (5.40g, 13.95mmol), palladium (0) tetrakis(tripheny1phos-phine) (1.03g, 0.89mmol), copper(1) iodide (0.14g, 0.74mmol), 3-methylbut-1-yn-3-01 (2.55g, 30.4mmol) and diisopropylamine (50ml). The crude product was purified by flash chromatography [fine mesh silica gel; 10% v/v ethyl acetate in light petroleum (bp 40-60 "C)] to give a yellow solid, which was recrystallized from cyclohexane and dried in uucuo (P,O,, 0.20mmHg, room temperature, 18h).Yield, 2.34g, 43%, mp 44-45 "C; SH (270MHz, solvent CDCl,, standard TMS): 0.90(3 H, t), 1.40(22H, m), 1.65 (6H, s), 1.75 (2 H, quintet), 2.00(1 H, s), 3.95 (2 H,t), 6.60 (2H, m), 7.40(1 H, t). v/cm-' (KBr disc): 3460, 2920, 2850, 1620, 1500, 1465, 1285, 1230, 1160, 1115,960, 900 and 835. m/z 390(M+, 24%), 375 (loo%), 262(45%), 183 (23%), 179 (IOOYo),71 (45%). l-Ethynyl-3-Juoro-4-tetradecyloxybenzene,10 Compound 8 (4.06g, 10.4mmol), potassium hydroxide (0.62g, 11.1 mmol) and toluene (100ml) were stirred and heated under reflux and under nitrogen for 2h (TLC showed complete reaction). The acetone and toluene azeotrope was removed periodically using a Dean and Stark receiver and replaced with an equal volume of toluene.The cooled reaction mixture was poured into water (100ml) and the organic phase separated off. The aqueous phase was then washed with diethyl ether (3x 50ml) and recombined with the organic layer before being washed with brine (50ml), dried (MgSO,), filtered and the solvent evaporated off to give an orange oil. The crude product was purified by flash chromatography [fine mesh silica gel; 9: 1 dichloromethane-light petroleum (bp 40-60"C)]. The yellow solid obtained was dried in uacuo ( P205,0.35mmHg, room temperature, 6h). Yield, 3.08g, 89%; mp 34-35 "C; SH (270MHz, solvent CDCl,, standard TMS): 0.85(3 H, t), 1.35(22H, m), 1.83 (2H, quintet), 3.00(1 H, s), 4.03 (2 H,t), 6.85(1 H, t), 7.20 (2H, d). v/cm-' (KBr disc): 3300,2960, 2920, 2850, 1615, 1575, 1470, 1310, 1225, 1110,945, 880, 815, 720,625 and 600.WI/Z 332(M', 17%), 136 (loo%), 83(8%), 69 (19%), 55(45%). l-Ethynyl-2-$uoro-4-tetradecyloxybenzene,11 This compound was prepared using a similar method to that described for compound 10. Quantities used: compound 9 (2.35g, 6.02mmol), potassium hydroxide (0.39g, 6.95mmol) and toluene (70 ml). The crude product was purified by flash chromatography (fine mesh silica gel; dichloromethane). The yellow solid obtained was recrystallized (cyclohexane) and dried in uucuo (P,O,, 0.1mmHg, 40"C, 3h). Yield, 1.28g, 64%; mp 41-42"C; 6, (270MHz, solvent CDCl,, standard TMS): 0.85 (3 H, t), 1.30 (22H, m), 1.75 (2 H, quintet), 3.20(1 H, s), 3.95 (2 H, t), 6.65 (2 H, m), 7.35 (1 H, t).v/cm-l (KBr disc): 3300,2950, 2920, 2850, 1610, 1565,1500,1470, 1290, 1160, 1110,1010,850, 825, 720, 675 and 630.m/z 332(M+,40%), 165(15%), 136(100%). ( 3-Fluoro-4-tetradecyloxypheny1)propiolic Acid, 12 Butyllithium (3.8ml, 2.5mol I-' in hexanes) was added drop- wise to a stirred, cooled (ca. -10 to -7 "C) solution of compound 10 (3.03g, 9.1mmol) in dry tetrahydrofuran J. MATER. CHEM., 1994, VOL. 4 (35ml) under nitrogen. The resulting solution was kept at ca. -7"C for a further 10min before being poured onto a stirred slurry of crushed solid C02 and dry tetrahydrofuran (5ml) and allowed to warm to room temperature overnight. The solution was acidified with concentrated hydrochloric acid and water (100ml) was added. The product was extracted into diethyl ether (3x 50ml), the combined ether extracts were then washed with brine (50ml), dried (MgSO,), filtered and the solvent evaporated off to give a yellow solid.The product was purified by flash chromatography [fine mesh silica gel; dichloromethane (initially) and 9: 1 dichloro-methane-methanol (finally)], recrystallized (cyclohexane-ethyl acetate), washed [light petroleum (bp 40-60"C)] and dried in uucuo ( Pz05,0.30mmHg, room temperature, 10h). Yield, 2.47g, 72%; mp 91-93 "C; SH (270MHz, solvent CDCl,, standard TMS): 0.85(3 H, t), 1.25 (22 H,m), 1.85 (2H, quintet), 4.05(2H, t), 6.93(1 H, t), 7.35 (2 H, m), no carboxylic proton detected. v/cm-' (KBr disc): 2920, 2850, 2205, 1685, 1610, 1455, 1310,1275, 1250, 1230, 1125, 870, 810 and 610.m/z 376(M', trace), 136(100%).(2-Fluoro-4-tetradecyloxyphenyl)propiolicAcid,13 This compound was prepared using a similar method to that described for compound 12. Quantities used: compound 11 (4.50g, 13.6mmol), butyllithium (1.4ml, 10.0mol 1-' in hex- anes) and dry tetrahydrofuran (60ml). The crude product was purified by flash chromatography [fine mesh silica gel; dichloromethane (initially) and 9:1 dichloromethane-meth-anol (finally)] to give a yellow solid, which was recrystallized (cyclohexane) and dried in uucuo (P205,0.20mmHg, 40"C, 12h). Yield, 3.58g, 70%; mp 98-101 "C; 6, (270MHz, solvent CDCl,, standard TMS): 0.90(3 H, t), 1.30(22H, m), 1.65 (2H, m), 3.90 (2 H, t), 6.55 (2 H, t), 7.45(1 H, t), no carboxyl proton detected.v/cm-' (KBr disc): 3440, 2920, 2850, 2200, 1690, 1615, 1505,1465, 1235, 1215, 1170, 1115,850and 610. m/z 376 (M+, trace), 332 (M+-CO,, 6O/h), 253 (5%), 180 (16%), 136 (100%). 4'-Hydroxybiphenyl-4-carboxylicAcid,15 A mixture of concentrated sulfuric acid ( 115ml) and water (115ml) was added dropwise to a stirred suspension of 4-cyano-4'-hydroxybiphenyl,14 (25.62g, 131.4mmol), in gla- cial acetic acid (400ml). The mixture was heated under reflux for 48h; the cooled reaction mixture was then poured into water (600ml) with stirring and the white precipitate filtered off. The aqueous filtrate was washed with diethyl ether (4x 70ml); the combined extracts were then washed with water (50ml), dried (MgSO,).filtered and the solvent evapor- ated off to yield a white solid. Both crops of product wert combined, dried thoroughly and recrystallized (glacial acetic acid) and dried in uucuo (P,O,, 0.30mmHg, 50"C, 5h). Yield, 23.49g, 84%; mp 197-200"C; SH(270MHz, solvent CDCl,, standard TMS): 6.95 (2 H, d), 7.50(2 H, d), 7.60 (2H, d), 8.05 (2 H, d), 9.15(1 H, s, broad), no carboxylic proton was observed. v/cm-' (KBr disc): 3400, 1680, 1605, 1590, 1425, 1385, 1300,1195, 830, 770 and 720.m/z 214(Mi,loo%), 197 (48%), 168 (7%), 139 (15%), 115 (12%). 4'-Methoxycarbonyloxybiphenyl-4-carboxylicAcid,16 Compound 15 (15.02g, 70.2mmol) was added slowly to a vigorously stirred solution of sodium hydroxide (8.15g, 203.8mmol) in water (300ml) at -4 C. Methyl chlorc- formate (10.82g, 114.5mmol) was added dropwise and the temperature maintained at 0 "C.The resulting white slurry was then stirred under these conditions for a further 4h. The pH was then adjusted to 5 using concentrated hydrochloric acid solution (1 : 1,concentrated HC1-water) and the volumiii- J. MATER. CHEM., 1994, VOL. 4 ous white precipitate filtered off and washed with water. The white solid was dried and recrystallized (glacial acetic acid) and dried again in vucuo (P,O,, 0.20 mmHg, 40 "C, 4 h). Yield, 14.83 g, 78%; mp 256-260 "C, lit. transition temp. K 233-235 N > 300 ISO ("C)20;6, (270 MHz, solvent CDCl,, standard TMS): 3.90 (3 H, s), 7.37 (2 H, d), 7.60 (4 H, m), 8.10 (2 H, d), no carboxylic proton detected.v/cm-l (KBr disc): 2970, 2840, 2680, 2525, 1870, 1675, 1610, 1430, 1325, 1295, 1220, 1185, 1065, 1005,945, 930, 870, 825 and 775. m/z 272 (M', 60%), 228 (loo%), 213 (82%), 184 (43%), 138 (28%), 114 (13%). (R)-(-)-l-Methylheptyl-4'-Methoxycarbonyloxybiphenyl-4-carboxylate, 17 Triphenylphosphine (6.80 g, 25.9 mmol) and (S)-octan-2-01 (5.03 g, 38.6 mmol) in dry tetrahydrofuran (35 ml) were added dropwise to a stirred mixture of compound 16 (7.03 g, 25.8 mmol) and diethyl azodicarboxylate (4.50 g, 25.9 mmol) in dry tetrahydrofuran (40 ml) under nitrogen. The reaction was stirred for a further 24 h at room temperature; TLC indicated that the reaction had reached completion. The white precipitate was removed by filtration through a pad of Hyflo- supercel, the filtrate was then washed with brine (50 ml), dried (MgSO,), filtered and evaporated to give a white solid. This was purified by flash chromatography [fine mesh silica gel; 5% v/v ethyl acetate in light petroleum (bp 40-60 "C)] to give a colourless oil, which was dried in uucuo (P,OS, 0.20 mmHg, room temperature, 5 h).Yield, 8.06 g, 81%; 6, (270 MHz, solvent CDCl,, standard TMS): 0.85 (3 H, t), 1.25 (13 H, m), 3.90 (3 H, s), 5.16 (1 H, sextet), 7.25 (2 H, d), 7.60 (4 H, d), 8.10 (2 H, d). v/cm-' (KBr disc): 2960, 2930, 2860, 1760, 1720, 1610, 1500, 1440, 1380, 1265, 1185, 1110, 1010,945, 830 and 775. m/z 384 (M', 27%), 272 (lOOYo), 255 (32%), 228 (55%), 213 (~OYO), 139 (20%). = -12.5" (0.0409 g ml-', CHCl,). (S)-(+)-1-Methylheptyl4'-Methoxycarbonyloxybiphenyl-4-carboxylate, 18 This compound was prepared using a similar method to that described for compound 17.Quantities used: triphenylphos- phine (6.74 g, 25.7 mmol), (R)-octan-2-01 (5.03 g, 38.7 mmol), compound 16 (7.00 g, 25.7 mmol), diethyl azodicarboxylate (4.50 g, 25.9 mmol) and dry tetrahydrofuran (110 ml). The crude product was purified by flash chromatography [fine mesh silica gel; 5% v/v ethyl acetate in light petroleum (bp 40-60 "C)] to give a colourless liquid, which was dried in uucuo (P,O,, 0.20 mmHg, room temperature, 5 h). Yield, 7.98 g, 81%; SH (270 MHz, solvent CDCl,, standard TMS): 0.85 (3 H, t), 1.35 (13 H, m), 3.95 (3 H, s), 5.15 (1 H, sextet), 7.25 (2 H, d), 7.60 (4 H, d), 8.10 (2 H, d).v/cm-l (KBr disc): 2960, 2930, 2860, 1765, 1715, 1610, 1440, 1260, 11 10,1005,930,830 and 770. m/z 384 (M+,23YO),272 ( loo%), 255 (38%), 228 (64%), 213 (23%), 196 (lo%), 185 (14%), 168 (15%), 152 (12%), 139 (20%). +30.6" (0.0425 g ml -';CHC1,) (R)-(-)-l-Methylheptyl4-Hydroxybiphenyl-4-carboxylate,19 A solution of compound 17 (8.06 g, 20.9 mmol) in ethanol (30 ml) was added dropwise to a stirred mixture of ammonia (105 ml, 35% solution) and ethanol (180 ml) at room tempera- ture. TLC analysis showed complete reaction after a period of 30 min. The reaction was then poured into water (300 ml) and cooled in solid CO,; the precipitated product was then filtered off, dried and recrystallized (cyclohexane-ethyl acetate, 4: 1). The colourless crystals were then dried in vacuo (P,O,, 0.10 mmHg, 40 "C, 5 h).Yield, 5.52 g, 81%; mp 84-87 "C; 6, (270 MHz, solvent CDCl,, standard TMS): 0.85 (3 H, t), 1.35 (11 H, m), 1.65 751 (2 H, m), 5.00 (1 H, sextet), 5.20 (1 H, sextet), 6.95 (2 H, d), 7.55 (2 H, d), 7.60 (2 H, d), 8.10 (2 H, d). v/cm-' (KBr disc): 3350, 2960, 2910, 2840, 1680, 1600, 1585, 1560, 1530, 1495, 1350, 1290, 1270, 1195, 1110, 1060, 920, 830 and 770. VZ/Z 326 (M', 26%), 214 (loo%), 197 (21Yo), := -35.1" (0.0342 g ml-'; CHCl,). (S)-(+)-1-Methylheptyl4'-Hydroxybiphenyl-4-carbox).Iate,20 This compound was prepared using a similar method to that described for compound 19. Quantities used: compound 18 (7.51 g, 19.6 mmol), ethanol (210 ml) and ammonia (105 ml, 35% solution).Yield, 4.89g, 77%; mp 87-89 "C; 6, (270MHz. solvent CDCl,, standard TMS): 0.85 (3 H, t), 1.30 (11 H, m), 1.70 (2 H, m), 5.17 (1 H, sextet), 5.45 (1 H, s, broad), 6.95 (2 H, d), 7.55 (2 H, d), 7.60 (2 H, d), 8.20 (2 H, d). v/cm-' (KBr disc): 3360, 2960, 2920, 2850, 1685, 1605, 1590, 1535, 1500, 1407, 1300, 1280, 1200, 1115, 1060, 835, 770 and 730. m/z 326 M', 12%), 214 (100%), 197 (20%). [a]ET=7 +38.8" (0.0232 g m1-I; CHC1,) (R)-(-)-1-Methylheptyl4-[( 3-Fluoro-4-tetradecyloxyphenyl) propioloyloxy] biphenyl-4-carboxylate, 21 Dicyclohexylcarbodiimide (0.35 g, 1.7 mmol) was added to a stirred mixture of compound 12 (0.59 g, 1.6 mmol), compound 19 (0.50 g, 1.5 mmol), 4-N,N-dimethylaminopyridine(0.04 g, 0.3 mmol) and dry dichloromethane (10 ml) at room tempera- ture.The reaction mixture was stirred for a further 18 h before the precipitated 'urea' was removed by filtration through Hyflo-supercel and the filtrate washed successively with water (50 ml), 5% v/v acetic acid solution (2 x 50 ml) and water (50 ml). The organic phase was then dried (MgSO,), filtered and the solvent evaporated off to yield a brown solid. The crude product was purified twice by flash chromatography [fine mesh silica gel; 5% v/v ethyl acetate in light petroleum (bp 40-60 "C) and then fine mesh silica gel; 9: 1 dichloro-methane-light petroleum (bp 40-60 "C)].The colourless solid was recrystallized (cyclohexane x 3) and dried in uucuo(P,05, 0.10 mmHg, 40 "C, 6 h). Yield, 0.15 g, 14%; 6, (270 MHz, solvent CDCl,, standard TMS): 0.90 (6 H, t), 1.30 (33 H, m), 1.85 (4 H, quintet), 4.05 (2 H, t), 5.15 (1 H, sextet), 6.95 (1 H, t), 7.35 (4 H, m), 7.65 (4 H, m), 8.20 (2 H, d).v/cm-l (KBr disc): 2920, 2850, 2220, 1735, 1720, 1710, 1610, 1510, 1280, 1235, 1205, 1180. 1115, 880, 815 and 770. m/z 684 (M+, trace), 359 (55%), 345 I 15%), 214 (14%), 163 (loo%), 139 (11%). [a]:*=--21.1" (0.0095 g ml-'; CHC1,). Calc. for C4,H5,F0,: C, 77.16; H, 8.38. Found: C, 77.09; H, 8.27. (S)-(+)-l-Methylheptyl4-[( 3-Fluoro-4-tetradecyloxypheriy1)-propioloyloxy] biphenyl-4-carboxylate, 22 This compound was prepared using a similar method to that described for compound 21. Quantities used: dicyclohexylcar- bodiimide (0.37 g, 1.8 mmol), compound 12 (0.58 g, 1.5 mmol), compound 20 (0.50 g, 1.5 mmol), 4-N,N-dimethylaminopyrid-ine (0.03 g, 0.25 mmol) and dry dichloromethane (10 ml 1.The crude product was purified twice by flash chromatography [fine mesh silica gel; 5% v/v ethyl acetate in light petroleum (bp 40-60 "C) and then fine mesh silica gel; 9: 1 dichloro-methane-light petroleum (bp 40-60 "C)]. The colourless solid was recrystallized (cyclohexane) and dried in uucuo ( P,05, 0.30 mmHg, 40 "C, 17 h). Yield, 0.18 g, 17%; aH(270 MHz, solvent CDCl,, standard TMS): 0.90 (6 H, t), 1.30 (33 H, m), 1.75 (4 H, m), 4.05 (2 H, t), 5.20 (1 H, sextet), 6.95 (1 H, t), 7.30 (4H, m). 7.65 (4 H, m), 8.10 (2 H, d). v/cm-' (KBr disc): 2910, 2850, 2210, 1735, 1720, 1705, 1610, 1510, 1280, 1235, 1205, 1180, 1115, 880, 815 and 770.m/z 684 (M', trace), 359 (70%), 335 J. MATER. CHEM.. 1994,VOL.4 (9%), 214 (lo%), 163 (loo%), 139 (8%). [a]kT= +26.9" 1185, 1110, 1005, 920, 850, 815, 765, 730 and 605. rn/z 684 (0.0112 g ml-'; CHCl,). Calc. for C44H,7F05: C, 77.16; (M', lo%), 359 (lOOo/o), 335 (9%), 214 (lo%), 163 (70%). H, 8.38. Found: C, 77.31; H, 8.13. = +25.7" (0.0257 g ml-'; CHC1,). Calc. for C4,H,,F05: C, 77.16; H, 8.38. Found: C, 77.04; H, 8.32. (R)-(-)-l-Methylheptyl4'-[( 2-E;luoro-4-tetradecyloxyphenyl)-propioloyloxy] biphenyl-4-carboxylate, 23 This compound was prepared using a similar method to that Results described for compound 21. Quantities used: dicyclohexylcar- bodiimide (0.78 g, 3.8 mmol), compound 13 (1.09 g, 2.9 mmol), compound 19 (0.94 g, 2.8 mmol), 4-N,N-dimethylaminopyrid-ine (0.11 g, 0.9 mmol).The crude product was purified twice by flash chromatography [fine mesh silica gel; 4% v/v ethyl acetate in light petroleum (bp 40-60 "C) and then fine mesh silica gel; 5% v/v light petroleum (bp 40-60% "C) in dichloro- methane]. The off-white solid was recrystallized [ethanol (-72 "C) x 2, cyclohexane x 11 and dried in uacuo (P,05, 0.3 mmHg, 50 "C, 5 h). Yield, 0.80 g, 41YO;hH (270 MHz, solvent CDCl,, standard TMS): 0.88 (6 H, t), 1.27 (33 H, m), 1.80 (4 H, m), 3.98 (2 H, t), 5.18 (1 H, sextet), 6.70 (2 H, t), 7.29 (2 H, d), 7.55 (1 H, t), 7.64 (4 H, m), 8.12 (2 H, d). v/cm-' (KBr disc): 3420, 2960, 2920, 2850, 2200, 1710, 1615, 1510, 1465, 1300, 1280, 1230, 1190,1115,1000,860, 830 and 775.rn/z 684 (M+, trace), 359 (loo%), 214 (20?40), 163 (95%). [cx]:~= -13.8" (0.0217 g m1-I; CHC1,). Calc. for C4,H,,F05: C, 77.16; H, 8.38. Found: C, 77.09; H, 8.23. (S)-(+)-1-Methylheptyl4-[( 2-Fluoro-4-tetradecyloxyphenyl)-propioloyloxy] biphenyl-4-carboxyEate, 24 This compound was prepared using a similar method to that described for compound 21. Quantities used: dicyclohexyl- carbodiimide (0.60 g, 2.9 mmol), compound 13 (0.89 g, 2.36 mmol), compound 20 (0.73 g, 2.2 mmol), 4-N,N-dimethylaminopyridine (0.10 g, 0.8 mmol) and dry dichloro- methane (20ml). The crude product was purified twice by flash chromatography [fine mesh silica gel; 4% v/v ethyl acetate in light petroleum (bp 40-60 "C)and then fine mesh silica gel; 5% v/v light petroleum (bp 40-60 "C) in dichloro- methane].The colourless solid obtained was recrystallized [ethanol (-72 "C) x 2, cyclohexane x 11 and dried in ~UECMO (P,O,, 0.3 mmHg, 40 "C, 6 h). Yield, 0.46 g, 30%; hH (270 MHz, solvent CDCl,, standard TMS): 0.88 (6 H, t), 1.32 (33 H, m), 1.79 (4 H, quintet), 3.98 (2 H, t), 5.18 (1 H, sextet), 6.69 (2 H, m), 7.30 (2 H, d), 7.52 (1 H, t), 7.63 (2 H, d), 7.66 (2 H, d), 8.15 (2 H, d). v/cm-' (KBr disc): 2920, 2850, 2210, 1710, 1615, 1510, 1300, 1280, Optical Microscopy The phase sequences for the enantiomeric compounds 21-24 in addition to their racemates 25 and 26, as determined by optical microscopy, are listed in Table 1. On cooling from the isotropic liquid, the 3-fluoro enanti-omers (21 and 22) both form similar iridescent planar smectic C* textures at 75.3 and 75.5 "C, respectively, before crystalliz- ation occurs at much lower temperatures.This behaviour differs noticeably with the racemate, compound 25, which cools from the isotropic liquid to a focal-conic smectic. A texture (81.5 "C) before giving way directly to a Grandjean smectic C texture at 79.3 "C. No TGB A* phase behaviour was observed for the 3-fluoro enantiomers (21 and 22). It is interesting to note that the enantiomers do not display identical transition temperatures and they also have noticeably lower transition temperatures than the racemate. The 2-fluoro enantiomers (23 and 24), on cooling from the isotropic liquid, form the characteristic verrnis (filament) tex- ture of the TGB A* phase, before the filaments coalesce to give a platelet texture.34 Examples of the verrnis texture of the TGB A* phase is given in Fig.1. Further cooling resulted in the formation of homeotropic regions bordered by areas of focal-conic smectic A texture, which in turn gave a conventional schlieren smectic C* texture before finally recrystallizing. Can0 wedge cells were employed to investigate further the TGB A*-Sf transition in compounds 23 and 24, the surface orienting coating coupled with the separation of the glass plates giving improved textures of the mesophases. On heating, at the transition from S,* to TGB A*, the focal-conic fans of the smectic A texture develop transition lines and the texture gives way to the texture of a TGB A* phase, which is characterized by zones of filaments and Grandjean plane disclination lines.Rotation of the upper polarizer of the microscope clearly showed a colour dispersion for white light caused by the helicity and hence the optically active nature of the phase (unlike the lower smectic A* phase which showed no dispersion effect). The presence of Grandjean disclination Table 1 Transition temperatures for the (R)-, (S)-and (R,S)-l-methylheptyl 4'-[( 2-or 3-fluoro-4-tetradecyloxyphenyl)propioloyloxy] biphenyl-4-carboxylates compound number 21 22 23 24 25 26 X Y absolute transition temperatures/"C" X Y configuration Is0 TGB A* SAP** sc*/sc Kb F F H H F H H H F F H F (R)-(0 (R)-(SI-(R,s)-c (R,W .. . . . . 75.3 75.5 85.8 87.1 81.5 87.7 . . 80.8 82.9 . . 72.6 73.5 79.3 76.8 49.6 62.7 33.9 39.2 44.8 33.6 . . . . . . "Recorded at a cooling rate of 2"C min-'. bRecrystallization temperatures determined by thermal optical microscopy. 'Racemic samples were prepared by weighing equal amounts of (R)-and (S)-enantiomers into clean glass vials and mixing whilst in the isotropic state. J. MATER. CHEM., 1994, VOL. 4 Fig. 1 TGB A* filaments in an isotropic background coalescing into a TGB A* platelet texture (compound 23) at 85.6 "C(100x) lines confirms the presence of a helical structure in the TGB A* phase; however, the absence of any effect that is related to helicity or optical activity in the smectic A* phase confirms the structure of this phase as being non-helical and conse- quently proves the presence of a TGB A* to Sz transition. Once again the racemate (26) displays different mesogenic behaviour to that shown by the enantiomers, and here the TGB A*-ST-S,X sequence is replaced by an SA-Sc sequence; the phase thermal stabilities of the racemate (26) are slightly higher than the phases shown by the enantiomers (23 and 24).Differential Thermal Analysis All of the enantiomers (21-24) and the racemates (25 and 26) were analysed by DSC at heating and cooling rates of 5 and 10 'C min-'; the transition temperatures and enthalpies are given in Table 2. The 3-flUOrO enantiomers, 21 and 22, clearly show sharp isotropic (Iso)liquid (Liq)-smectic C* transitions, as first observed by optical microscopy. However, both of the cooling thermograms of 21 and 22 show bizarre, diffuse transitions above the isotropic liquid-smectic C* peak, pos- sibly indicating some event taking place in the isotropic liquid.If this behaviour is compared with the racemate (25),in which no diffuse peak is observed (only an Iso-SA-S~ phase sequence), the 1s0-S~transition occurs at essentially the same temperature as the diffuse peak in thermograms for com-pounds 21 and 22. This is clearly demonstrated in Fig. 2, which depicts superimposed cooling thermograms of 21, 22 and 25 (all recorded at a cooling rate of 5°C min-l). WII I I I I I I1 I0 20 30 40 50 60 70 80 90 1 77°C Fig.2 DSC thermograms of the (R)-, (S)-and (R,S)-forms of 1-methylheptyl 4-[(3-Auoro-4-tetradecyloxyphenyl)propioioyloxyJ -4-carboxylates [compounds 21(b), 22(4 and 25(c) recorded at a cooling rate of 5 Oc min-'1 Summation of the enthalpies of the isotropic liquid-qmectic C* and liquid-liquid transition for 21 and 22 gives values that are roughly equivalent to the isotropic liquid-smectic A transition in the racemate (25).The 2-fluoro enantiomers (23 and 24) give normal thermo- grams in that no diffuse liquid-liquid transitions are observed; here only the Iso-TGB A* and S2-S; transitions arc seen, the S2-S: transition being second order in nature (AH=ca. 0.14 kJ mol-I). The TGB A*-S,* transition was not detected by calorimetry.However, as this transition simply corres'ponds to the expulsion of the twist from the phase, it would be expected to be accompanied by a relatively small cnergy change. The racemate (26) again shows similar phase events with enthalpies comparable to those of the enantiomers. Spontaneous Polarization and Tilt-angle Studies The spontaneous polarization (P,) of compounds 21-24 were measured as a function of temperature on cooling from the Curie point (Ts,* or TA*<*).The results for the (R)-enanti- omers 21 and 23 are shown in Fig. 3, and were recorded at 6 V (open symbols) and 15 V (filled symbols) at 60 €IT. The magnitude of the spontaneous polarization of the 3-flUOrO compound (21) was found to be slightly more than double that obtained for the 2-fluoro compound (23), i.e.for 21. P, = ca. 90 nC cmP2 and for 23, P,= ca. 40 nC cm-2 at 4 "C helow the Curie point. Similar behaviour was noted for the (S)-enantiomers, compounds 22 and 24. The factors determining the magnitude of P, are the tilt angle, transverse molecular dipole (pt) and the chirality-induced broken symmetry for rotation about the long molecular axis. In terms of a simple mean-field theory2' P, can be written as: P, = Po sin 0 =Npt(cos 4) where (cos 4) is a ferroelectric order parameter and N is the number density. Quadrupolar ordering of molecules about their long axes in tilted smectic phases is driven by the tilt angle, which generates macroscopic biaxiality, and in dis- cussing the effects of molecular structure it is better to consider the reduced polarization Po.Chirality in tilted smectic phases causes non-centrosymmetric phase symmetry perpendicular to the tilt plane, thus a molecule rotating about its tilted long 754 J. MATER. CHEM., 1994, VOL. 4 Table 2 DSC data for the (R)-,(S)-and (R,S)-1-methylheptyl 4'-[( 2-or 3-fluoro-4-tetradecyloxyphenyl)propioloyloxy]biphenyl-4-carboxylates transition temperatures/"C and enthalpies/kJ mol-la compoundnumber absolute configuration Liq-Liq' Is0 SAP2 s,*/sc K 21 22 (R)-(9- 79.7 79.5 [1.631 [1.403 74.5 74.3 [1.231 [1.11] 36.7 46.0 C22.181 [24.641 . . 23 (R)- C5.411 84.9 [0.17] 75.3 [34.521 33.9 . 24 25 26 (S)-(R,S)-(R,S)- [5.641 [3.521 [5.921 85.2 79.5 86.2 c0.121 [0.461 [0.061 75.7 77.6 75.4 [34.1 51 [22.461 [27.793 33.6 33.7 23.7 .. . "Data recorded at a cooling rate of 5 "C min-l. 'These values were taken from maximum peak temperatures not as onset temperatures. -10 -8 -6 -4 -2 0 reduced temperaturel'c Fig.3 Magnitude of the spontaneous polarization plotted as a function of reduced temperature of (R)-1-methylheptyl 4'-[( 3-fluoro and 2-fluoro-4-tetradecyloxyphenylj propioloyloxy] biphenyl-4-car-boxylates [compounds (a) 21 and (bj 23, respectively] axis experiences a potential of dipolar symmetry. This affects both the lateral steric dipole of the molecule and the transverse electric dipole responsible for P,. A maximal contribution to P, is achieved if the steric dipole is parallel to the electric dipole.Switching studies performed in polyimide aligned cells with a 9V dc source revealed that both the (R)-enantiomers, 21 and 23, have positive polarizations [Ps(+)] and conversely the (S)-enantiomers (22 and 24) have negative polarizations [P,(-)]. The electronically switched optical tilt angle charac-teristics for the (R)-enantiomers of both systems are given in Fig. 4. Here the 2-fluoro (R)-enantiomer (23), shows a much 030-0. 25 --* ._ 20 -c Q 15-0 oooooo 00 1 .10 . , . , . , 0, . , . , . , , 0,-16 -12 -8 -4 0 reduced temperature/"C larger optical tilt angle that appears to saturate in the region of ca. 30";in contrast, the 3-flUOrO (R)-enantiomer (21) satu-rates at a value of ca. 15" which, surprisingly, is a very low value for a smectic phase cooled directly from the isotropic liquid.This finding is the opposite to the trend encountered for the polarization properties. Standardizing the spontaneous polarization with respect to tilt angle (P,/Q)for the two systems shows that the 3-fluoro system's effective spontaneous polarization is approximately four times larger in comparison with the 2-fluoro system. This is shown in Fig. 5; the open and filled symbols denote data for the 3-fluoro and 2-fluoro compounds, respectively, at 15 V (squares) and 6 V (circles), respectively. Neglecting short-range interactions, we assume that the chiral centre in the 3-(21) and 2-(23) fluoro compounds causes a similar broken rotational symmetry potential for both materials.The average transverse electric dipole moment of 21 may be greater than that of 23 because of hindered rotation of the terminal C14H290-group caused by the 3-substituted fluorine atom; this might also lead to an increased steric dipole. Biaxial ordering in the tilted smectic C phase will reduce free rotation about the triple bond in the core of the molecule and 2-fluoro substitution increases the bulk of the molecule core. In order to accommodate this increased bulk at a similar packing density, the tilt angle for 23 increases in comparison with that for 21. The P, of 23 is less than that of 21, so that either there is a compensation of transverse dipole components in the core of 23, which is absent in the 3-flUOrO-materials, or the angle between the 9, Fig.4 Optical tilt angle measured as a function of reduced tempera-Fig. 5Standardized spontaneous polarization (Ps/O)as a function ture of (R)-l-methylheptyl4'-[(2-and 3-fluoro-4-tetradecyloxyphenyl) of reduced temperature of the (R)-1-methylheptyl 4-[(2-and 3-propioloyloxy] biphenyl-4-carboxylates [compounds 23 ( 0)and 21 fluoro-4-tetradecyloxyphenyl)propioloyloxy]biphenyl-4-carboxylates ( 0), respectively] [compounds (a) 23 and (b)21, respectively] J. MATER. CHEM., 1994, VOL. 4 electric dipole and the steric dipole is greater for 23 than for 21. Since biaxial ordering will tend to align the molecular core with the transverse component of the F-dipole perpen- dicular to the tilt plane, it is more likely that dipole compen- sation in 23 results in a reduced Ps.These speculations can only be substantiated by detailed molecular modelling. Optical Purity Studies The dependence of the diffuse peak, TGB A* and smectic C* phases on optical purity were studied by construction of full phase diagrams of both the 1-methylheptyl 4'-[( 3-fluoro-4-tetradecyloxypheny1)propioloyloxy] biphenyl -4 -carboxylates and 4'-[( 2-fluoro-4-tetradecyloxyphenyl)propioloyloxy]bi-phenyl-4-carboxylates. This was carried out by accurate weighing of the desired amounts of the (R)-and (S)-enanti-omers into a clean vial and then heating to the isotropic liquid and mixing thoroughly. The binary mixtures were then analysed by polarized-light microscopy of DSC (when required).The miscibility phase diagram for the 1-methylheptyl 4-[(3-fluoro-4-tetradecyloxyphenyl ) propioloyloxy] biphenyl- 4-carboxylates is shown in Fig. 6; all data points were obtained by optical microscopy studies, with the exception of the liquid-liquid transitions, which originate from DSC mixture studies. There are a number of striking features of the system. First, the dramatic and steep increase in the clearing point temperatures (ca. 7 "C)as the optical purity decreases. Second, although no TGB A* phases were noted for the pure enanti- omers, two distinctly different regions of induced TGB A* phase and what is believed to be TGB C* phase were observed by optical microscopy, occurring at compositions of between 10 and 25wt.% of either (R)-or (S)-enantiomer.The TGB A* phase was characterized by the usual filament texture, as exemplified earlier in Fig. 1. Furthermore, on cooling from liquid-liquid transitiont -isotropic liquid 741Ii 0 20 40 60 80 1 0 (R )-enantiomer (wt.%) Fig. 6 Miscibility phase diagram of binary mixtures of the (R)-and (S)-forms of 1-methylheptyl 4'-[( 3-fluoro-4-tetradecyloxyphenyl) propioloyloxy] biphenyl-4-carboxylates (compounds 21 and 22) (all data points obtained by optical microscopy unless otherwise indicated; +data points from DSC studies) the isotropic liquid the proposed TGB C* phase appears as a shimmering, grey schlieren texture; pitch bands being observed in certain domains.22 Other areas of the texture appeared as highly coloured (selectively reflecting) planar texture with clearly visible s= k1 brush defects and are believed to be ferroelectric S,* phase forming.This demon- strates that the two different helical axis orientations are associated with the TGB C* to smectic C* transition. Examples of the TGB C* phase texture are given in Fig. 7-9; Fig. 9 shows the disclination lines described earlier. Thermal studies on suitable mixtures failed to locate any phase transition between the TGB A* and TGB C* regions in the phase diagram, possibly for two reasons: first, because of the near-vertical nature of the transition line between the two phases (shown as a dotted line in Fig. 6):second, because of the low enthalpy expected to accompany such a transition in a mixed system.However, the vertical nature of a line separating the TGB A* and TGB C* phases is somewhat expected as the TGB A* phase appears above the smectic A* phase on heating and similarly the TGB C* phase is formed on heating the Sz phase. The relationship between the TGB-phase type and the structure of the ensuing smectic phase formed on cooling has been found to occur in other systems for individual, pure materials.*' Therefore the postulation that a vertical line separates the two TGB phases above their respective smectic phases is possible if not probable. As the system's optical purity decreases the TGB A* regions give way to a conventional smectic A* phase (ie. toward Fig. 7 Texture of the TGB C* phase observed for a binary mixture of compounds 21 and 22 (composition 14 wt.%, compound 21) at 78.6 "C(100 x) Fig.8 TGB C* phase just below the transition to the isotropic liquid for a binary mixture of compounds 21 and 22 (compsition 12 wt.%, compound 22) at 78.4 "C (100x) racemic composition). Furthermore, the transition tempera- tures of the diffuse liquid-liquid event (obtained from DSC studies on the mixtures) drop as the optical purity decreases, eventually disappearing at ca. 20% wt.wt. compositions. Underlying the whole combined TGB C*/TGB A*/smectic A* region was a continous smectic C* or smectic C phase region (depending on optical purity). From the above findings it is evident that the strong molecular chirality of this system not only stabilizes the TGB C* and TGB A* phases but appears to be capable of suppressing the clearing point and other phase transitions.The phase diagram for the (R)-and (S)-forms of 1-methylheptyl 4'-[(2-fluoro-4-tetradecyloxyphenyl)propiol-oyloxy] biphenyl-4-carboxylate is shown in Fig. 10. Again, the clearing point rises as the optical purity of the binary mixture decreases; however, it is not quite so marked as for the 3-flUO1-0 system (see Fig. 6). Two 'wing' like regions of the TGB A* phase appear between compositions of 0 and 5% wt.wt. of either (R)-or (S)-enantiomers, this contrasts to the pure 3-fluoro systems. It is interesting to note that the temperature ranges for these TGB A* 'wings' differ noticeably. As the enantiomers were determined to be in excess of 99.5% pure by HPLC, the differences in the phase range can only be attributed to very small differences in the optical purities of the two starting alcohols [(R)-and (S)-octan-2-01], indicat- ing the sensitivity of the system to optical purity, particularly at high enantiomeric excess.The presence of these TGB A* 'wings' also depresses the appearance of the SZ-Sz transition J. MATER. CHEM.. 1994, VOL. 4 Fig. 9 TGB C* schlieren texture for a binary mixture of compounds 21 and 22 (composition as for Fig. 8) at 77.4 "C ( LOO x) by a proportional amount; beyond compositions of 5% wt.wt. the Si-S,*/S, transition begins to rise to a higher and almost constant value. No induced TGB C* phases were observed for binary mixtures of the 2-fluoro compounds (23 and 24).Once again the high 'molecular' chirality associated with the 1-methylheptyl moiety appears to stabilize the formation of TGB A* phases and to some extent suppresses the clearing point. Optical Rotation Studies Compound 22 was found to exhibit strong optical activity just above the S$-Is0 phase transition. Fig. 11 shows that in the isotropic phase, the optical rotatory power of 22 increases from ca.-0.5" mm-' at ca. 108 "C to a maximum of ca.-5.2" mm-' at ca. 87 "C. Below ca. 80 'C the optical rotatory power changes sharply indicating the entrance to another phase, i.e. the S$ phase. This suggests the existence of strong chiral interactions in the isotropic phase, but this behaviour is unlike normal pretransitional behaviour.For example, the optical rotatory power of 22 in the pretransition region is found to be ca. 5 times larger than that of the pretransitional behaviour of CE6 (a common cholesteric liquid-crystal material). Furthermore, the reciprocal optical rotation plots of 22 and CE6 are very different near the liquid crystal to isotropic phase transition. The onset of strong pretransitional chiral interactions in the isotropic phase coincides in temperature with the broad peak detected bj DSC. Similar behaviour has been reported by Frame et al." J. MATER. CHEM., 1994, VOL. 4 TGB A' isotropic liquid woQ88i lo---.. ,TGB A* I 78 08ol 0 20 40 60 80 100 (R)-enantiomer (wt.%) Fig.10 Miscibility phase diagram of binary mixtures of the (R)-and (S)-forms of l-methylheptyl4-[( 2-fluoro-4-tetradecyloxyphenyl) propioloyloxy] biphenyl-4-carboxylates (compounds 23 and 24) (all data points obtained by optical microscopy) t.1 .... I.,..l....l....I....1....1 80 85 90 95 100 105 T/"C Fig. 11 Optical rotatory power as a function of temperature in the isotropic phase of compound 22 However, in their case, enhanced optical activity was observed just above the Sz-Is0 phase transition. Circular Dichroism Studies Optical rotation provides a useful monitor of the bulk chiral interactions whereas CD probes the local molecular chirality. The CD absorption observed in the region 300-400 nm in 22 can be attributed to the group X-CO-OR*, where X is the biphenyl ring (see Fig.12). A strong enhancement of the CD signal is observed just above the S,*-Iso phase transition. This is illustrated in Fig. 12, which shows that a peak that is related 77.8 3 x -3 x lo4 500 . 13 h/nm Fig. 12 Circular dichroism as a function of temperature in the isotropic phase of compound 22 r """""""""""''1 10000-8000-vd 1 6000-4000 -1 . 1 .... I....,....I....,..I 78 79 80 81 82 77°C Fig. 13 Reciprocal circular dichroism in the wavelength region 355-370 nm as a function of temperature of compound 22 to the intrinsic CD absorption develops at Lax=355-1370 nm on cooling from the 'true' isotropic to just above the S,*phase. The CD diverges as the S$-Iso transition is approached from above according to T-Tc*-Iso,as illustrated in Fig.13. On entering the S,* phase the CD spectrum changes completely (as shown in Fig. 14) to a broad band associated with selective reflection. The negative intrinsic CD measurements correlate well with the negative optical rotation results in the Optical Rotation Studies section and occur in the same region as the DSC anomaly. Discussion The diffuse transition observed in cooling DSC thermograms of 21 and 22, as well as being reproducible, have been noted in other homologues of this series,24 in the original parent compounds (1, n= 14)3and in related highly chiral systems.25 The authenticity of this event in related systems has been reinforced by X-ray diffraction st~dies.~ However, the diffuse transition is usually associated with the appearance of frus-trated phases, such as the TGB A* phase.For 21 and 22 (the outer fluoro systems) no frustrated phases apparently occur, and so this is the first time that this type of transition is seen with what hitherto appears to be normal phase behaviour. Optical rotation and circular dichroism spectral studies have 5 x 0 <4 I -5 x lo4 6007L 3 Wnm Fig. 14 Circular dichroism at 76.8 “C in the smectic C* phase of compound 22 also confirmed the presence of a change in the liquid phase, and the results clearly show the presence of strong local chiral interactions. A tentative explanation for the liquid-liquid transition in other systems is related to the ensuing disordering of the TGB A* phase str~cture.~ It is believed that the TGB A* phase is stabilized by the presence of screw dislocations (hence, it is frequently termed ‘a lattice of screw dislocations’).A situation might then arise in which the screw dislocations ‘melt’ at the clearing point leaving cybotactic clusters of smectic A regions in an isotropic background, the smectic A regions may then melt completely to give an amorphous isotropic liquid. Conversely, the smectic A clusters may first melt leaving a network or entanglement of material that originally formed the dislocation ‘cores’ in an isotropic background, this network then eventually melts to give the amorphous isotropic liquid. The results of optical rotation and CD measurements suggest that the latter is a more likely explanation.It should be noted that other workers have observed the persistence of long- range order above the clearing points of a number of unrelated ionic amphiphilic materials;26 however, whether this is a related phenomenon is not at all clear. Similarly, in ceramic systems both entangled and disentangled flux phases have been found to occur.27 However, this model cannot account for the appearance of the isotropic liquid-isotropic liquid-smectic C* sequence observed in the thermal studies of the enantiomers 21 and 22 because of the lack of a TGB A* phase. Consequently, we may have to question the actual nature of the so-called S,* phase. The optical purity and miscibility studies show that the clearing points and other transitions become depressed as the optical purity of the system increases, this is not only in agreement with earlier unrelated TGB A* system studie~,~ but also with de Gennes’ original hypothesis.’ Thus, the physical studies and physical nature of the material suggest that a frustrated phase must be present in 21 and 22.However, as no smectic A* phase was detected, this indicates that the smectic C* phase cannot be ‘normal’ and may possibly have a frustrated structure. Consequently, the ferroelectric C* phase may actually be a TGB C* phase, for which two possible structures are suggested: one in which there is a twist in the tilt direction within the blocks of smectic layers making up J.MATER. CHEM., 1994, VOL. 4 the macroscopic helical structure, and one in which this twist is absent.28 Detailed structural investigations are currently underway in order to unravel this mystery, the results of which will be reported later. We now turn to the transition occurring in the liquid state. As the thermogram shows a very broad diffuse peak, this event may not constitute a real phase transition, but may in fact be related to a structural change occurring in the liquid. This could be a conformational or a packing change rather than the presence of an entangled or disentangled fog phase, as found at TGB A* to isotropic transitions. Thus, this gives us two possibilities, either the C* phase is normal and the liquid-liquid transition is related to a structural change, or else the entangled/disentangled fog phase is present and the C* phase is in fact a frustrated phase (ie.a TGB C* phase).28 Turning now to the peculiarities found in the ferroelectric properties of the two related systems. It is clear that the 3-flUOrO (outer) compounds have about twice the value of the spontaneous polarization in comparison with the 2-fluoro (inner) systems, whereas they have half the value of the tilt angle. Normalizing with respect to tilt angle (PJO) shows that the effective spontaneous polarization is four times larger for the outer fluoro systems in relation to the inner fluoro compounds. This is remarkable as the change in the position of the fluoro-substituent is taking place at the opposite end of the molecular to the ‘chiral moiety’, which strongly influ- ences the polarization. Conclusions The position of fluoro-substituents in the 1-methylheptyl 4-[(4-tetradecyloxyphenyl)propioloyloxy]biphenyl-4-carb-oxylate system has clearly been shown to influence the forma- tion of TGB A*, TGB C* and S,* phases in the enantiomeric materials and in addition was found to affect markedly the ferroelectric and optical tilt angle behaviour in the S,* phases. The calorimetric and circular dichroism data have further confirmed the authenticity and reproducibility of the diffuse liquid-liquid event in the isotropic liquid.Finally, the two types of material synthesized provide further examples of the influence of optical purity (enantiomeric excess) on the trans- ition temperatures of phase transformations.The authors thank the SERC for their financial support (to C.J.B. and J.S.K.) and the MOD (DRA Malvern). They also thank Mrs. B. Worthington, Mr. R. Knight and Mr. A. D. Roberts for their assistance in the various spectroscopic measurements and Dr. A. J. Slaney for advice on polarization and tilt angle measurements. Circular dich roism measure-ments were made at the SERC National CD Service, University College London, and the assistance of Dr. Alex Drake in making these measurements is gratefully acknowl- edged. Finally, our thanks are expressed to Dr. R. Pindak (AT&T Bell Laboratories, USA) for his insights into the nature of the TGB C* phase.References 1 P. G. de Gennes, Solid State Commun., 1972, 10,753. 2 S. R. Renn and T. C. Lubensky, Phys. Rev.A., 1988,38,2132. 3 J. W. Goodby, M. A. Waugh, S. M. Stein, E. Chin, R. Pindak and J. S. Patel, J. Am. Chem. SOC.,1989, 111, 8119. 4 J. W. Goodby, M. A. Waugh, S. M. Stein, E. Chin, R. Pindak and J. S. Patel, Nature (London), 1987,337,449. 5 A. J. Slaney and J. W. Goodby, Liq. Cryst., 1991,9, 849. 6 A. J. Slaney and J. W. Goodby, J.Muter. Chem., 1991, 1, 5. J. MATER. CHEM., 1994, VOL. 4 7 A. Bouchta, H. T. Nguyen, M. F. Achard, F. Hardouin, C. Destrade, R. J. Twieg, A. Maaroufi and N. Isaert, Liq. Cryst., 1992,12, 575. 8 T. X. Neenan and G. M. Whitesides,J. Org. Chem., 1988,53,2489. 9 G. Just and R. Singh, Tetrahedron Lett., 1987,48,5981. 10 A. Carpita, A. Lessi and R. Rossi, Synth. Commun., 1983,571. 11 E. J. Corey and P. L. Fuchs, Tetrahedron Lett., 1972,36,3769. 12 E. Chin and J. W. Goodby, Mol. Cryst. Liq. Cryst., 1986,141,3 11. 13 0.Mitsunobu, Synthesis, 1981, 1. 14 I. Nishiyama and J. W. Goodby, J. Muter. Chem., 1993,3, 149. 15 B. Nieses and W. Steglich, Angew. Chem., Int. Ed. Engl., 1978, 17, 522. 16 W. Clark Still, M. Kahn and A. Mitra, J. Org. Chem., 1978, 43, 2923. 17 D. D. Perrin and W. L. F. Arrnarego, Purification of LaboratoryChemicals, Pergamon Press, Oxford, 3rd edn., 1988. 18 CRC Handbook of Physics and Chemistry, ed. R. C. Weast, CRC Press, Boca Raton, 68th edn., 1988. 19 H. Diarnant, Rev. Sci. Instrum., 1957,28,30. 20 A. J. Slaney, Ph.D. Thesis, University of Hull, 1992. 21 B. Zeks, Ferroelectrics, 1988,84,3. 22 L. Navailles, H. T. Nguyen, P. Barois, C. Destrade and N. Isaert, Liq. Cryst., 1993,15,479; J. Phys. II (France), 1992,2, t889; Phys. Rev. Lett., 1993,71, 545. 23 K. C. Frame, J. L. Walker and P. J. Collings, Mol. Cryst. Liq. Cryst., 1991, 198,91. 24 C.J. Booth, J. W. Goodby and K. J. Toyne, unpublished results. 25 I. Nishiyama and J. W. Goodby, unpublished results. 26 V. Busico, A. Ferraro and M. Vacatello, Mol. Cryst. Liq. Cryst., 1985,128,243. 27 P. L. Gammel, D. J. Bishop, G. J. Dolan, J. R. Kwo, C. A. Murray, L. F. Schneemeyer and J. Waszaczak, Phys. Rev. Lett.. 1987, 59, 2592. 28 S. R. Renn and T. C. Lubensky, Mol. Cryst. Liq. Crvst., 1991, 209, 349. Paper 3/04521H; Received 28th July, 1993