J. MATER. CHEM., 1994,4(11), 1689-1697 a-Fluoro Esters Incorporating a Cyclohexane Ring: Some New Chiral Dopants for Ferroelectric Mixtures Stephen M. Kelly,* Richard Buchecker and Jurg Funfschilling F. Hoffmann-La Roche Ltd., Dept. RLCR, CH-4002 Bade, Switzerland Several new homologous series of optically active a-fluoro esters incorporating a cyclohexane ring and a number of different cores have been synthesized. Several of these chiral dopants possess an enantiotropic chiral smectic C (Sc*) phase at elevated temperatures. They are characterised by a relatively high spontaneous polarisation, long pitch and low rotational viscosity. In addition they exhibit exceptional chemical, thermal and photo- and electro-chemical stability. It is shown that an ester group in a terminal position of the molecular core of these new chiral dopants leads to an increase in the spontaneous polarisation (compared with the values for the analogous ethers) with only a small increase in the rotational viscosity, as determined in a standard Sc base mixture.An ester group in a central position of the molecular core has the opposite effect on these parameters. It is shown that at least two aromatic rings are normally necessary for S, formation. An optically active a-chloro ester incorporating a cyclohexane ring was also prepared and compared with the analogous a-fluoro ester. A higher spontaneous viscosity were determined far the a-fluoro ester. These new a-fluoro esters are excellent chiral dopants for chiral smectic C mixtures for use in the surface- stabilised ferroelectric liquid-crystal display.Surface-stabilised ferroelectric liquid-crystal devices (SSFLCDs)'-' are being developed for commercial appli- cations with a high information content and/or fast response times (e.g. computer screens, printer heads, spatial light modu- lators). The display devices are characterised by exceptionally fast response times (ps), high contrast, good viewing angle dependence and memory (bistability). Most commercially available chiral smectic C (S,*) mixtures designed for SSFLCDs consist of a non-optically active base S, mixture doped with at least one optically active (chiral) dopant.6-8 The chiral dopant should induce the desired value of spon- taneous polarisation (Ps)and a long helical pitch in the base S, mixture (in order to avoid the necessity of pitch compen- sation) without depressing the S, transition temperature (S,*-S,/N) or increasing the rotational viscosity (y), or the birefringence (An) of the mixture excessively.The dopant can also effect the tilt angle (O), which influences the switching times and the contrast (see Mixture Properties). All the mixture components must be chemically, thermally and photo- and electro-chemically stable. For most applications it is essential to utilise chiral dopants, which induce a large spontaneous polarisation without increasing the rotational viscosity excessively. The resultant induced spontaneous polarisation of the mixture must also not be too large (to avoid charge effects), but a large intrinsic value of P, for the chiral dopant allows smaller amounts to be The induced spontaneous polarisation depends, amongst other factors, on the dipole moment at the optically active centre of the chiral dopant.Hence, at least one of the substituents attached to the optically active centre is usually strongly electronegative (e.g. flu~rine,'~-~~ nitrile26,27) in ~hlorine,~~,~' order to polarise the carbon-substituent bond. Other polar groups in the vicinity of the chiral centre (e.g.oxygen," ester,26 oxirane16-'8) also serve to increase the polarisation. The optically active centre should be as close as possible to the core of the molecule (commensurate with phase stability) in order to optimise dipole and steric interactions and thus minimise the rotational freedom of the dipole moment. Similar effects can also be obtained by using a variety of optically active five- and six-membered rings containing polar groups (e.g.dioxolane,28 la~tone,,~-~~ proline rings36).o~azolidine,~~ We have recently reported a wide variety of new, optically active a-fluoro esters incorporating the aliphatic cyclohexane ring and a number of different cores.15 Several of these chiral dopants possess an enantiotropic Sc* phase at elevated tem- peratures, characterised by high spontaneous polarisation, long pitch and low rotational viscosity. In addition they exhibit exceptional chemical, thermal and photo- and electro- chemical stability. Although the spontaneous polarisation is reduced by about 40% compared with that of fully aromatic analogue^,'^.'^ a significantly lower rotational viscosity leads to much shorter response times.Additionally the substitution of the aliphatic trans-l,4-disubstituted cyclohexane ring for an aromatic (benzene) ring was found to result in lower values for the birefringence and in a substantially longer pitch.I5 We now report on attempts to improve further the physical properties of these chiral dopants as determined in a standard Sc base mixture (e.g. reduced response times, higher S,* transition temperatures) by modification of core structures and terminal chains. It has recently been shown that in contrast to the situation observed for nematics, the introduction of an ester group (CO,) between the terminal chain and the core of the rmiolecule of base S, components can lead to lower viscosity values and thus, response Therefore, it was decided to investi- gate the effects of an ester group in this position of chiral dopants in order to determine whether lower induced viscosit- ies for the Sc* mixture could be obtained.It has also been demonstrated recently that introduction of a carbon carbon double bond of a defined configuration in certain positions of the terminal alkanoyloxy chain of a series of 4454-alkylpyrimidin-2-y1)phenylalkanoates can lead to improve- ments in the transition temperatures and other physical properties of relevance to display device appli~:itions.~~ Therefore, it was decided to extend this system to the chiral dopants as described above.Only (E)-alk-2-enoyloxy esters were synthesized as this has been shown to be the most advantageous position and configuration of the carbon-carbon double bond.39 A preliminary screening of one homologue each of the trans-4-(2',3'-difluoro-4'-n-alkoxybiphenyl-4-yl)cyclohexyl(R)-2-fluorohexanoates and trans-4-[4-(5-alkylpyrimid1n-2-y1)-phenyl] cyclohexyl (R)-2-fluorohexanoates had indicated that such chiral dopants were promising candidates for commercial SSFLCD mixt~res.'~ Therefore, it was decided to synthesize a homologous series of each type, in order to be able to identify the esters with the most advantageous comhination of physical properties.Large variations are often observed within a homologous series (e.g. a factor of two in switching times) and trends are often impossible to predict despite the large body of information already available for such sys- tem~.~~-~~It was also hoped that by substituting an alkoxy chain for the alkyl chain of the trans-4-[ 4-(5-alkylpyrimidin-2-yl)phenyl]cyclohexyl (R)-2-fluorohexanoates Sc* phases could be induced. One optically active (R)-2-chlorohexanoate analogue of J. MATER. CHEM., 1994, VOL. 4 (i.e. no racemisation occurred during esterificat~on). The struc- tural and isomeric purity was determined by differential thermal analysis and capillary gas chromatography as usual and, where necessary, on liquid-crystal-packed columns.48 The transition temperatures of the esters prepared, recorded in Tables 1-5, were determined by optical microscopy using a Leitz Otholux I1 POL BK microscope in conjunction with a Mettler FP 82 heating stage and FP 80 control unit.All the monotropic liquid-crystal phases could be observed using the tra~zs-4-(2’,3’-difluoro-4’-n-alkoxybiphenyl-4-yl)cyclohexyl (R)-2-fluorohexanoates was also prepared in order to deter- mine the relative merits of these two widely used polar groups in otherwise identical chiral dopant^.^^,^^ A small series of biphenylyl- and phenylpyrimidinyl-cyclo-hexyl (R)-2-fluorohexanoates with alkyl and alkoxy chains were prepared in order to determine the necessity of hetero- atoms (e.g. nitrogen or oxygen) for the formation of Sc phases in such systems.46 Experimental Synthesis The trans-4-(4-n-alkanoyloxyphenyl)cyclohexyl(R)-2-fluoro-hexanoates (l-ll) and trans-4-{4-[(E)-alk-2-enoyloxy]phen-yl} cyclohexyl (R)-2-fluorohexanoates ( 12-18) were prepared by selective esterification of 4-(trans-4-hydroxycyclohexyl)-phen01’~ at the phenolic hydroxy group with either alkanoic or (E)-a1 k-2-enoic acids to produce the corresponding trans-4-(alkanoyloxy-or (E)-alk-2-enoyloxy-phenyl)cyclohexan-l-01s.These cyclohexanols were then esterified a second time with (R)-2-fluorohexanoic a~id’~,~~ to yield the desired diesters (l-ll) and (12-18). The general method of synthesis of the trans-4-(2’,3’-difluoro-4’-n-alkoxybiphenyl-4-yl)cyclohexyl (R)-2-fluoro-hexanoates ( 19-30) and trans-4-[ 4-(5-alkylpyrimidin-2-y1) phenyl] cyclohexyl (R)-2-fluorohexanoates (32-36) has already been described for one homologue of each series.” The (R)-2-chlorohexanoate (31) was prepared as described for a microscope and no virtual values (extrapolated) had to be determined.When necessary the Mettler stage could be cooled (-50°C) by allowing N2 gas, cooled by liquid N2, to pass through the stage at a controlled rate. The liquid-crystal transition temperatures were also determined using a Mettler DTA TA 2000. The purity of the compounds was determined by a thin- layer chromatography (TLC), gas chromatography and differential thermal analysis (DTA). A Perkin-Elmer 8310 capillary gas chromatograph and GP-100 graphics printer were used.Precoated TLC plates, 4cm x 8 cm, SiOz SIL G/UV254, layer thickness 0.25 mm (Machery -Nagel, Duren, Germany), were utilised. Column chromatography was carried out using silica gel 60 (230-400 mesh ASTM). Reaction solvents and liquid reagents were purified by either distillation or drying shortly before use. Reactions were carried out under N, unless water was present as a reagent or solvent. All temperatures were Table 1 Transition temperatures for the trans-4-( 4-n-alkanoyloxy- pheny1)cyclohexyl (R)-2-fluorohexanoates (1-1 1 ) ~ the trans-4-(2’,3’-difluoro-4’-n-alkoxybiphenyl-4-yl)cyclohexyl 1 1 48 (R)-2-fluorohexanoates (19-30) using (R)-2-chlorohexanoic acid14 instead of (R)-2-fluorohexanoic acid. The trans-4-[4-(5-alkoxypyrimidin-2-yl) phenyl] cyclohexyl (R)-2-fluorohexanoates (37-41) were synthesized by condensation of 4-(trans-4-h y droxycyclohexyl)benzamidine hydrochloridelS with the aldehyde prepared in situ from benzyloxyacetaldehydediethylacetal to yield trans-4-[ 4-( 5-benzyloxypyrimidin-2-yl)phenyl]cyclohexanol.Removal of the benzyl protection group by catalytic hydrogenation resulted in the corresponding phenol, which could be selec- tively alkylated in a Williamson ether synthesis to produce trans-4-[ 4-(5-alkoxypyrimidin-2-yl)phenyl]cyclohexanol. Ester- ification with (R)-2-fluorohexanoic acid as above yielded the desired esters (37-41).The methods of synthesis and structural analysis of the new esters (1-42) are described in detail below. The configuration of the carbon-carbon double bond in the alkenyl chain of the new esters (12-18) was confirmed by ‘H nuclear magnetic resonance (NMR) spectroscopy (the trans-olefinic coupling constants, ca.12-18 Hz, are larger than those of the corre- sponding cis-olefinic coupling constants, ca. 7-11 Hz) and by infrared (IR) spectroscopy (the trans-absorption bands are narrow and exact ca. 970-960 cm -I, whereas the cis-absorp- tion bands are observed at distinctly different wavelengths ca. 730-675 cm-’). The optical purity (90.6% ee) of the optically active (R)-2-fluorohexanoic acid (prepared and purified as described in the 1iterat~re.I~~~~) was determined according to literature rnethod~.”~~~ The optical purity of the (R)-2-fluoro- hexanoates was determined similarly and found to be identical 2 2 51 3 3 38 - 4 4 25 51 5 5 42 59 6 6 52 59 7 7 42 64 8 8 41 67 9 9 45 68 10 10 52 71 11 11 50 72 Table 2 Transition temperatures for the tran.s-4-{4-[(E)-alk-2-enoyloxy] pheny1)cyclohexyl (R)-2-fluorohexanoates (12-18)” 12 1 13 2 14 3 15 4 16 5 17 6 18 7 ‘Values given in temperature.81 (611 -76 78-92 -(46)< -80 74-66 77 57 -77 75 -78 parentheses represent a monotropic transition J. MATER. CHEM., 1994, VOL. 4 Table 3 Transition temperatures and enthalpies of fusion for the trans-4-(2’,3’-difluoro-4‘-n-alkoxybiphenyl-4-yl)cyclohexyl(R)-Zfluoro-hexanoates ( 19-30) and the trans-4-( 2’,3’-difluoro-4‘-decyloxybiphenyl-4-y1)cyclohexyl (R)-2-chlorohexanoate (31) EF ester n X (C-SA/N*)/OC (SA-N*/I)/”C (N*-I)/”C AH/kJ mol-’ 19 1 F 70 -116 22.0 20 2 F 73 108 138 23.5 21 3 F 59 122 127 25.2 22 4 F 44 133 -26.5 23 5 F 33 131 -16.6 24 6 F 38 133 -22.7 25 7 F 37 129 -22.4 26 8F 33 134 -18.1 27 9F 41 127 -23.6 28 10 F 35 128 -22.3 29 11 F 42 125 30 12 F 38 134 -29.8 -31 10 c1 34 112 -measured externally unless otherwise stated.The ‘H NMR spectra were recorded at 60 MHz (Varian T-60), 80 MHz (Bruker WP-80) or 250 MHz (Bruker HX-270). Tetra-methylsilane was used as the internal standard. Mass spectra were recorded on a MS9 (AEZ Manchester) spectrometer. The S,* mixture SC9-1219 consists of 5-(5-heptyl-1,3-di-oxan-2-y1)-2-(4-octyloxypheny1)pyridine(7.0 wt.%), 5-( 5-octyl-1,3-dioxan -2 -yl) -2 -(4-octyloxyphenyl) pyridine (7.0 wt.%), 5-(5-decyl- 1,3 -dioxan-2-yl)-2-(4-octyloxyphenyl)pyridine (6.0 wt.%), 4-[ 2-(trans-4-pentylcyclohexyl)ethyl]phenyl 4-decyloxybenzoate ( 15.9 wt.%), 4-[ 2-(trans-4-pentylcyclohex-yl)ethyl] phenyl 4-dodecyloxybenzoate (7.1 wt.%), 4-[ 2-(trans-4-pentylcyclohexyl)ethyl]phenyl 2,3-difluoro-4-(und~cyloxy)-benzoate (7.0 wt.%), 2-(4-hexyloxyphenyl)-5-non!lpyrimi-dine ( 14.9 wt.Yo), 5-nonyl-2-( 4-nonyloxyphenyl )pyrimi- dine ( 19.9 wt.%),5-heptyl-2-( 4-octyloxypheny1)py rimidine (5.0 wt.%) and 2-(4-hexyloxyphenyl)-5-ocytlpyrimidine (10.1 wt.%). The determination of the physical properties of the chiral mixtures containing the new esters was carried out as pre- viously des~ribed.~’,~’ Synthesis of trans-4-( CAcetoxyphen yl )cyclohexanol A solution of N,N-dicyclohexylcarbodiimide (0.78 g, 0.0038 mol) in dichloromethane (50 cm3) was added slowly to a solution of 4-(trans-4-hydroxycyclohexyl)pheno11~(0.60 g, 0.0031mol), acetic acid (Fluka) (0.18 g, 0.0031mol), 4- (dimethy1amino)pyridine (0.04 g) and dichloromethane (25 cm3) at 0 “C, stirred at room temperature overnight, filtered and the filtrate evaporated down under reduced pressure.The residue was purified by column chromatography on silica gel using a 1: 1 hexane:ethyl acetate mixture as eluent followed by recrystallisation from ethanol to yield 0.65 g (90%) of the pure ester.v,,,/cm-’: 3409, 3329, 2927, 2852, 1756, 1626, 1575, 1223, 835. Mass spectrometry (MS) m/z: 234 (M+), 192 (C12H1602), 174 (C12H14O). Table 4 Transition temperatures for the trans-4-[ 4-( 5-n-alkylpyrimidin-2-yl)phenyl]cyclohexyl (R)-2-fluorohexanoates (32-36) and tr6zns-4-14-(5-n-alkoxypyrimidin-2-yl) phenyl]cyclohexyl (R)-2-fluorohexanoates (37-41) -.32 34 98 -137 -.33 53 101 -143 -.34 41 109 -143 -.35 49 113 -145 -.36 48 116 -145 37 58 -81 161 105 38 44 78 95 162 103 -.39 53 88 102 162 -.40 60 -92 106 163 -.41 35 70 98 108 165 Table 5 Comparison of the transition temperatures for the trans-4-(4’-decylbiphenyl-4-yl)cyclohexyl(R)-2-fluorohexanoate (42), trans-4-[4-( 5- nonylpyrimidin-2-yl) phenyl]cyclohexyl (R)-2-fluorohexanoates (36) and trans-4-[ 4-( 5-nonyloxypyrimidin-2-yl)phenyl]cyclohexyl (R)-Xuoro-hexanoates (40) --42 145 36 48 116 -145 40 60 90 108 165 Synthesisof trans-4( 4-Acetoxypheny1)cyclohexyl(R)-2-Fluorohexanoate, 1 A solution of N,N-dicyclohexylcarbodiimide (0.68 g, 0.0033 mol) in dichloromethane (50 cm3) was added slowly to a solution of trans-4-(4-acetoxyphenyl)cyclohexanol(O.65g, 0.0028 mol), (R)-2-fluorohexanoic acid15 (0.37 g, 0.0028 mol), 4-(dimethy1amino)pyridine (0.04 g) and dichloromethane (25 cm3) at 0 "C and then stirred at room temperature over- night.The reaction mixture was worked up and purified as described above to yield 0.85 g (42%) of the pure ester. 'H NMR 8, (CDCI,; standard TMS; 250MHz): 0.90-0.95 (3 H, t), 1.25-2.14 (16 H, overlapping peaks), 2.29 (3 H, s), 2.62 (1 H, overlapping peaks), 4.77-4.81 (1 H, t), 4.96-5.01 ( 1 H, overlapping peaks), 6.99-7.02 (2 H, d), 7.18-7.26 (2 H, d).v,,,/cm-': 2942, 2862, 1751, 1509, 1373, 1200, 840. MS m/z: 350 (M'), 308 (C18H2503).Microanalysis found (expected): C 68.4 (68.5), H 7.8 (7.7), F 5.5 (5.4)%. [a],,=+8.3 (c. 0.0060 g cm-,; CHC1,). The transition temperatures of this ester (1) and similar esters (2-11 and 12-18) prepared using this general method are collated in Tables 1 and 2. Synthesis of trans-4-[ 4-( 5-Benzyloxypyrimidin-2-yl )phenyl] cyclohexanol N,N-Dimethylformamide (8.9 cm3, 115 mmol) was added dropwise to phosporyl chloride (8.6 cm3, 94 mmol) at 0 "C and then stirred for 15 min.A solution of benzyloxyacetal- dehyde diethyl acetal (14.0 g, 62 mmol) in N,N-dimethylfor- mamide (30 cm3) was added dropwise to the reaction mixture, which was heated at 50°C for 18 h. A solution of 4-(trans- 4-hydroxycyclohexyl) benzamidine hydrochloride ( 15.9 g, 62 mmol) in N,N-dimethylformamide (60 cm3) was added dropwise to the cooled reaction mixture (room temperature) and then stirred for 30 min. Triethylamine (69 cm3) was added dropwise and the reaction mixture heated at 50°C for 2 h, poured onto water (500 cm3), cooled to 0 "C, acidified with 36% hydrochloric acid (pH 3-4), stirred for 20 min at this temperature and then extracted into ethyl acetate (3 x 300 cm'). The combined organic layers were washed with water (3 x 300 cm3), dried (Na,SO,), filtered and evaporated down.The residue was purified by column chromatography on silica gel using a 1:1 ethyl acetate: toluene mixture as eluent and recrystallised from tert-butyl methyl ether to yield 6.3 g (33%) of the desired alcohol; mp 220-222 "C. 'H NMR BH (CDCl,; standard TMS; 250 MHz): 1.50 (5 H, overlapping peaks), 1.94 (4 H, overlapping peaks), 2.62 (1 H, overlapping peaks), 3.60, (1 H, overlapping peaks), 4.14-4.21 (2 H, q), 7.26-7.29 (2 H, overlapping peaks), 8.23-8.27 (2 H, d), 8.51 (2 H, s). v,,,/cm-': 3430,2927,2854, 1604, 1542, 1436, 1269, 1063, 993, 781 cm-'. MS m/z: 360 (M'), 342 (C23H2zN20). Synthesis of trans-4-[ 4-( 5-Hydroxypyrimidin-2-yl )phenyl] cyclohexanol A mixture of trans-4-[ 4-( 5-benzyloxypyrimidin-2-yljphenyll-cyclohexanol (1.0 g, 2.7 mmol), ethyl acetate (20 cm3), and 10% palladium on active charcoal (0.3 g) were hydrogenated until no more hydrogen was taken up.The catalyst was filtered off and the filtrate evaporated down. The residue was purified by column chromatography on silica gel using a 20 :1 dichloromet hane-met hanol mixture as eluent and recrys tal- lised from ethanol to yield 0.5 g (60%) of the desired alcohol. 'H NMR 6, (CDCI,; standard TMS; 250MHz): 1.89 (8 H, overlapping peaks), 2.49-2.51 (1 H, s), 3.34 (1 H, s), 7.29-7.33 (2 H, d), 8.14-8.18 (2 H, d), 8.40 (2 H, s). vrnax/cm-': 3424, 3257,2930,2855,2725,1611,1555,1429,1284,1050,793 cm-'. MS m/Z: 270 (M'), 252 (C16H16N20). J. MATER.CHEM., 1994, VOL. 4 Synthesis of trans-4-[ 44 5-Decyloxypyrimidin-2-yl )phenyl] cyclohexanol A mixture of 1-bromodecane (Fluka; 0.7 g, 0.0031 mol), trans-4-[ 4-( 5-hydroxypyrimidin-2-yl)phenyl]cyclohexanol (0.5 g, 0.0026 mol), potassium carbonate (0.14 g, 0.0104 mol) and butan-2-one (50 cm3) was heated under gentle reflux over- night, then filtered to remove inorganic material. The filtrate was diluted with water (1000 cm3) and then extracted into diethyl ether (3 x 100 cm3). The combined organic extracts were washed with water (2 x 500 cm3), dried (MgSO,), filtered and then evaporated down. The residue was purified by column chromatography on silica gel using a 9: 1 hexane- ethyl acetate mixture as eluent and recrystallised from ethanol to yield 0.5 g (47%) of the desired alcohol; mp, 160-162°C.v,,,/cm-': 3421, 2925, 2853, 1609, 1541, 1438, 1273, 1064, 840, 782 Cm-'. MS m/Z: 410 (M'), 392 (C26H36N20). Synthesis of trans-4-[ 44 5-Decyloxypyrimidin-2-yl )phenyl] cyclohexyl (R)-2-fluorohexanoate, 41 A solution of N,N-dicyclohexylcarbodiimide (0.3 g, 1.2 mmol) in dichloromethane (10 cm3) was added slowly to a solution of trans-4-[ 4-( 5-decyloxypryrimidin-2-yl )phenyl ]cyclohex-anol (0.5 g, 1.0mmol), (R)-2-fluorohexanoic acid15 (0.2 g, 1.0 mmol), 4-(dimethy1amino)pyridine (0.04g t and dichloro- methane (25 cm3) at 0 "Cand then stirred at room temperature overnight. The reaction mixture was worked up and purified, as described above, to yield 0.4 g (62%) of the desired ester.'H NMR 8, (CDC1,; standard TMS; 250 MHz): 0.88-0.96 (6 H, overlapping peaks), 1.28-2.00 (34 H, overlapping peaks), 2.62 (1 H, overlapping peaks), 4.06-4.11 (2 H, t), 4.76-5.11 (2 H, t), 7.26-7.29 (2 H, overlapping peaks), 8.25-8.28 (2 H, d), 8.44 (2 H, s). vmax/cm-': 2925, 2855, 1733, 1608, 1540, 1436, 1278, 1082, 854, 778. MS m/z: 526 (M'), 392 (C&35N20'). Microanalysis found (expected): C 75.2 (75.31, H 9.2 (9.3), N 5.5 (5.5), F 3.6 (3.7)%. +4.3 (C 0.0080 g cm-,; CHCl,). The transition temperatures of ester 41 and similar esters 37-40, prepared using this general method, are collated in Table 4. Mesomorphic Properties The transition temperatures of an homologous series of trans-4-(4-n-alkanoyloxyphenyl)cyclohexyl(R)-2-fluorohexanoates (1-11) are recorded in Table 1.The first three homologues (n= 1-3) do not exhibit mesomorphic behaviour. The other members of the series only exhibit an SB mesophase above the crystalline state. The plots of the S, transition temperature against the number of carbon atoms (n) in the alkanoyloxy chain rise with increasing chain length and show the normal pattern of alternation. No other mesophases could be observed. Table 2 contains the transition temperatures of the corre- sponding trans-4- (4-[(E)-alk-2-enoyloxy] phenyl} cyclohexyl (R)-2-fluorohexanoates (12-18) with an additional carbon- carbon double bond. The introduction of the trans double bond into diesters 1-11 to yield diesters 12-18 results in a small increase in the SB transition temperature (+3"C, on average, comparing only homologues of equal chain length); this is unusual. In all previous investigations of this effect in esters39,45$46and ordered mesophases were partially or totally suppressed, Sc or N phases were induced and the width of the Sc phase was extended.A melting point for several homologues could not be determined due to the low tendency for crystallisation of the ordered SB phase. The (E)-but-2-enoyloxy substituted ester (12) exhibits an N phase instead of the SB phase observed for the other homologues. J. MATER. CHEM., 1994, VOL. 4 This unusual behaviour underlies the strong nematic tendenc- ies of the (E)-but-2-enoyloxy function recently described for aromatic esters39 and aliphatic cyclohexyl ester^.^^,^^ The transition temperatures of an homologous series of the trans-4-(2',3'-difluoro-4'-n-alkoxybiphenyl-4-yl)cyclohexyl(R)-2-fluorohexanoates (19-30) are listed in Table 3.The first three members of the series exhibit an N* phase at relatively elevated temperatures (127 "C, on average). Eleven homol- ogues possess an enantiotropic SA phase at similar tempera- tures (126"C, on average). The plots of the s, transition temperature uersus the number of carbon atoms in the alk- anoyloxy chain show an increase with short chain lengths and then remain relatively flat for longer chain lengths. The plots show the normal pattern of alternation. The melting point is higher for short chain lengths than for longer chains.However, the melting point does not vary greatly with chain length (45 "C, on average). No other mesophases could be determined. This is unusual considering that the esters contain three-rings and a combined chain length of up to 18 carbon atoms. This must be in part due to the two lateral fluorine atoms, which have been shown to decrease the tendency for ordered phase A comparison between the thermal data of esters 19-30 with a direct linkage between the two phenyl rings and the corresponding die~ters'~ with an additional carboxy group (CO,) between the same two rings reveal that the diesters are superior with respect to the liquid-crystal transition temperatures. The diesters exhibit the desired order of phases for SSFLCDs (i.e.Sc*, SA and N* phases). However, the monoesters (19-30) do not depress the Sc* phase transition temperature of the base mixture excess- ively (see Mixture Properties). The clearing point (S,-I) of the chloro-substituted ester (31)is significantly lower (-14 "C) than that of the corresponding r-fluoro-substituted ester (28) with the same chain length (n= 10). This is almost certainly due to the larger van der Waals radius of the chlorine atom. Table 4 contains the thermal data for the two homologous series of trans-4-[ 4-( 5-alkylpyrimidin-2-y1)phenyllcyclo-hexyl (R)-2-fluorohexanoates (32-36) and trans-4-[ 4-( Salk- oxypyrimidin-2-yl) phenyl] cyclohexyl (R)-2-fluorohexanoates (37-41). The alkyl substituted series (32-36) only possesses orthogonal smectic phases (B and A) at elevated tempera- tures (107 "C and 143 "C, respectively).The melting point is relatively low and unusually uniform (45 "C, on average). The alkoxy-substituted series also exhibits SB and SA phases. However, the transition temperatures for the SB phase are lower (-1SoC, on average), whereas those of the SA phase are higher (+2O"C, on average) to an almost equal extent. An N* phase is observed for two homologues with short chains (37 and 38). All the homologues prepared possess an Sc* phase (98 "C, on average). An ordered (as yet unidentified) smectic phase is observed for homologue 41 with the longest chain studied. The melting point of the alkyl and alkoxy- substituted series are very similar (45 and 50"C, on average, respectively).This is unusual and is probably due to the presence of the S, phase. The thermal data collated in Table 5 show clearly that, even for three-ring systems, compounds containing the biphenyl moiety (e.g. 42) do not exhibit an S, phase. Indeed the alkyl- substituted phenylpyrimidine (36) also only exhibits ortho- gonal phases. Only the combination of the phenylpyrimidine core and an alkoxy chain in the ester (40) gives rise to an Sc* phase. The dependence of the Sc+ phase on the nature and position of dipoles in the cyclohexyl-phenyl-pyrimidine mesogenic system will be discussed in detail elsewhere.47 The liquid-crystal transition temperatures of the three-ring pyrimidines (37-41) show clearly that substances containing the trans- 1,4-disubstituted cyclohexane ring can exhibit an enantiotropic Sc* phase.This is in accord with previous results for a variety of three-ring phenyl ben~oates~"'~ with various linking units (e.g. single bond, epoxymethano, ethyl, carboxy, four-unit-linking group) and phenylpyrim-idines.52,55,56The results in Tables 1-5 show clearly that two- ring systems incorporating a 1,4-disubstituted cyclohexane ring are not sufficient for S, formation (exceptions containing a strong lateral dipole are known) and the results also show that a minimum of two aromatic rings per aliphatic; ring is required for S, formation (phenyl benzoates with 1,4-&substi- tuted bicyclo C2.2.21 octane ring in place of the 1,4-disubsti- tuted cyclohexane ring of the diesters also exhibit an Sc pha~e.~~,~~This is in contrast to statistical theories of the Sc phase; the theories normally assume a fully arom;ttic core str~cture.~'+~~ Mixture Properties Since most of the new chiral dopants (1-42) do not themselves possess an Sc* phase, parameters such as the spontaneous polarisation or switching time have to be determined in mixtures.Therefore, a small amount (7 wt.%) of the dopant is added to a standard non-chiral base mixture (S('9-1219) with the phase sequence Sc-SA =76 "C, SA-N =81 "C and N-I =103 "C. The transition temperatures, the spontaneous polarisation and the switching time of the resultant mixtures are then determined under standard conditions (T: 15 Vpp/p square wave, time to maximum current; P,: 10 Vpp/p triangu- lar wave form).In order to discuss the differences in response tinies, it is necessary to define more exactly the parameters involved. The spontaneous polarisation, the effective viscosity ;md the switching time, all depend on the Sc tilt angle 8. However, it is difficult to measure 6' reliably as details of the surface alignment influence the result. Therefore, attempts were made to estimate and then eliminate the influence of variations of 0 without actually measuring it. The spontaneous polxisation can be related to the tilt angle by62 P,= Po sin 8 (1) where Po is a constant characteristic for the dopant. An effective viscosity yeff can be defined63 via yeff dyl/dt +P, sin ylE =0 (2) where q is the angle of rotation on the Sc* cone and E the applied electric field.This equation allows the definition of a characteristic time, z (3) The experimental switching time is proportional to :. Based on geometrical considerations it follows63 that yeff is related to 6' by yeff=yo sin2 8 (4) where yo is independent of 8 and represents the rotational viscosity of a hypothetical nematic-like Sc structure with H = 90°C. Combining eqn. (l), (2)and (4) finally leads to zE =(yo/Po)sin f3 From these equations it follows: (i) In contrast to the expec- tation that a higher P, necessarily means a shorter switching time, an increase of P,, which is due to an increase in 8, leads to longer switching times.(ii) If both z and P, increase (or decrease) upon changing the side chain within a homologous series of dopants, then this is probably a change of 8 of the mixture induced by the dopant. (iii) If T decreases and P, increases, this is probably due to a change of Po, especially for low dopant concentrations, where changes of viscosity are small. The transition temperatures (Sc*-SA, SA-N*and N*-I) of a series of mixtures of the trans-4-(4-n-alkanoyloxypheny1)cy-clohexyl (R)-2-fluorohexanoates (l-ll)are plotted versus the numbers of carbon atoms in the terminal chain of the esters in Fig. 1. The Sc* and S, transition temperatures both increase with increasing chain length, whereas the clearing point remains basically constant. This is unusual as the pure esters only exhibit an SBmesophase (see Table 1).The spontaneous polarisation and the observed switching time of the same mixtures are plotted uersus the number of carbon atoms in the terminal chain of the esters in Fig. 2. There are significant variations from one homologue to another, but no systematic dependence on the chain length is observed. The high value of the spontaneous polarisation and short response time of the mixture containing ester 10 are particularly interesting. Similar trends for the transition temperatures, spontaneous polarisation and response times of the trans-4-{4-[(E)-alk-2-enoyloxy] phenyl} cyclohexyl (R)-2-fluorohexanoates (12-18) as those of the trans-4-(4-n-alkanoyloxyphenyl)cyclohexyl(R)-2-fluorohexanoates (1-11) are shown in Fig. 3 and 4.The presence of the trans-carbon-carbon double bond does not seem to lead to any significant improvement. The transition temperatures of a series of mixtures of the trans-4-(2’,3’-difluoro-4’-n-alkoxybiphenyl-4-yl)cyclohexyl(R)-2-fluorohexanoates (19-30) are plotted uersus the numbers of carbon atoms in the terminal chain of the esters in Fig. 5. The Sc* and S, transition temperatures both rise with increas- t N* 1 3 5 7 9 11 n Fig. 1 Chiral nematic-isotropic (N*-I), smectic A-chiral nematic (SA-N*)and chiral smectic C-smectic A (Sc*-SA) transition tempera- tures versus the number of carbon atoms (n)in the alkanoyloxy chain of the trans-4-(4-n-alkanoyloxyphenyl)cyclohexyl(R)-2-fluorohexa-noates (l-llj mixtures Fig.2 Spontaneous polarisation (W, P,) and switching time (0,T) uersus the number of carbon atoms (n)in the alkanoyloxy chain of the trans-4-(4-n-alkanoyloxyphenyl)cyclohexyl (R)-2-fluorohexa-noates (l-ll)mixtures J.MATER. CHEM., 1994, VOL. 4 9 \t-4 5 6 7 8 9 10 n Fig. 3 Chiral nematic-isotropic (N*-I), smectic A-chiral nematic (SA-N*)and chiral smectic C-smectic A (Sc*-SA)transition tempera- tures versus the number of carbon atoms (nj in the alkanoyloxy chain of the trans-4-{ 4-[(E)-alk-2-enoyloxy] phenyl) cyclohexyl (R)-2-fluoro- hexanoate (12-18) mixtures 7.0 ’500 t 4.0I 13001 I I 4 6 8 10 n Fig. 4 Spontaneous polarisation (W, P,) and switching time (0,z) versus the number of carbon atoms (n) in the alkanoyloxy chain of the trans-4-{4-[(E)-alk-2-enoyloxy] pheny1)cyclohexyl (R)-2-fluoro-hexanoate (12-18) mixtures 6ot 501 , J 2 4 6 8 10 12 n Fig.5 Chiral nematic-isotropic (N*-I), smectic A--chiral nematic (SA-N*)and chiral smectic C-smectic A (Sc*-SA)transition tempera- tures versus the number of carbon atoms (n)in the alkoxy chain of the trans-4-( 2‘,3’-diflouro-4’-n-alkoxybiphenyl-4-yljcyclohexyl (R)-2-fluorohexanoate (19-30) mixtures ing chain length, they appear to reach a maximum for intermediate chain lengths, and then appear to stabilise. The clearing point (N*-I) is remarkably independent of chain length. The absolute values for each of the three transitions are higher than those observed for the corresponding mixtures incorporating an equal amount of any of the two-ring esters J.MATER. CHEM., 1994, VOL. 4 9.0 I 1600 n Fig. 6 Spontaneous polarisation (a,P,) and switching time (0,5) versus the number of carbon atoms (n) in the alkoxy chain of the trans-4-(2',3'-diflouro-4'-n-alkoxybiphenyl-4-yl)cyclohexyl (R)-2-fluoro- hexanoate (19-30) mixtures (l-ll and 12-18). The spontaneous polarisation and response times of the same mixtures are plotted versus the numbers of carbon atoms in the terminal chain of the esters in Fig. 6. Although the plots exhibit considerable scatter, the switching times increase with increasing chain length. The switching times are in general longer than those of the two-ring esters (l-ll).Hence, the two-ring esters (l-ll and 12-18) and the three-ring difluoro esters (19-30) offer the possibility of choos-ing between higher transition temperatures and shorter response times according to application specifications. The difluoro esters (19-30) are of especial interest owing to the negative value of the dielectric anisotropy attributable to the two fluorine atoms in a lateral position.This facilitates a good orientation by electric field effects. The transition temperatures of mixtures incorporating the trans-4-[ 4-( 5-alkylpyrimidin-2-yl)phenylIcyclohexyl (R)-2-fluorohexanoates (32-36) and trans-4-[4-(5-alkoxypyrimidin-2-yl )phenyl]cyclohexyl (R)-2-fluorohexanoates (37-41) differing only in the presence of an additional oxygen atom attached to the pyrimidine ring of the esters (37-41) are plotted versus the number of (methylene and oxygen) units in the terminal alkyl and alkoxy chains in Fig.7. The data were plotted in this way in order to reveal possible odd-even effects for esters with alkyl and alkoxy chains of the same total 50 6 7 a 9 10 11 12 chain length Fig. 7 Chiral nematic-isotropic (N*-I), smectic A-chiral nematic (SA-N*)and chiral smectic C-smectic A (Sc*-SA) transition tempera- tures uersus the total length (oxygen and methylene units) of the alkyl/alkoxy chain of the trans-4-[4-( 5-alkylpyrimidin-2-y1)phenyll-cyclohexyl (R)-2-fluorohexanoate (32-36) and trans-4-[4-( 5-alkoxy- p yrimidin-2-yl ) phenyl] cyclohex yl (R)-2-fluorohexanoate (37-41 ) mixtures. (Note that the x-axis is the total number of atoms in the side chains including oxygen.) 1695 length.The corresponding values for the spontaneous polaris- ation and switching time are plotted in Fig. 8. In contrast to the alkoxy-substituted esters (37-41) the alkyl-substituted esters (32-36) do not possess an S,* phase as single compo- nents. Thus, the mixtures containing them exhibit lower Sc* transition temperatures. The mixtures containing the alkyl- substituted esters possess significantly higher values for spon-taneous polarisation and shorter switching times. As discussed above this indicates a significantly larger value for Po. This suggests that the dipole moment of the oxygen atom in the alkoxy chain compensates the dipole moments attached to the chiral centre.This is quite remarkable and implies a rather strong correlation of the orientation of the dipoles over several freely rotatable bonds. In the alkoxy-substituted esters a distinct odd-even effect for both the spontaneous po1,irisation and switching time is observed. This is attributed to the odd-even effect of Po. This is another indication that the dipole moment resulting from the oxygen atom in the alkoxy chain is correlated with the dipole moment of the chir;il centre. In Fig. 1-8 clear odd-even effects for the phase tiansition temperatures or the spontaneous polarisation and tht: switch- ing times are only observed for the alkoxy-substituted esters. Odd-even effects are most pronounced if the bond:, for the even (all trans) positions are closely aligned to the effective molecular axis.This seems to be the case for the alkoxy chains, but not for the alkyl chains or the esters. The data collated in Table 6 allow the effect of an ester group in a terminal position in the core of a chiral dopant to be determined. The ether15 and diester 9 differ only in the 61 1aoo chain length Fig. 8 Spontaneous polarisation (a,P,) and switching tiine (0,5) versus the total length (oxygen and methylene units) of [he alkyl/ alkoxy chain of the trans-4-[ 4-( 5-alkylpyrimidin-2-y I )phenyl] cyclohexyl (R)-2-fluorohexanoates (32-36) and trans-4-[ 4-( 5-alkoxy-pyrimidin-2-yl)phenyl] cyclohexyl (R)-2-fluorohexanoate (37-41 ) mixtures. (Note that the x-axis is the total number of atoms in the side chains including oxygen.) Table 6 Comparison of the transition temperatures, spc mtaneous polarisation and response times for two mixtures consisting of 7 wt.% the reference ether trans-4-( 4-decyloxyphenyl)cyclohexyl (R)-2-fluor~hexanoate'~and trans-4-(4-decanoyloxyphenyl)cyclohexyl(R)-2-fluorohexanoate (9) and 93 wt.% of the base Sc mixture 929-1219 X (S,,-S,)/OC (S,N*)/"C (N*-I)/"C PJnC cm -2 ~/ps CH, 64.5 82.6 97.8 3.8 460 co 68.0 81.5 97.1 5.7 400 presence of a carbonyl group (CO) instead of a methylene unit (CH,), i.e.the chain lengths are the same. The S,* transition temperatures for the mixture containing 7 wt.% of the ester (X =CO) is significantly higher (+3.5 “C) than that incorporating an equal amount of the ether (X =CH,).The spontaneous polarisation is significantly higher (+1.9 nC cmP2) for the ester than for the ether, indicating a greater value for Po, as shown by the significantly shorter response time. Shorter response times for ester mixtures compared with related ether mixtures have already been observed for non-optically active esters exhibiting an Sc phase, where the carbonyl function was also in a terminal instead of a central position in the core of the m~lecule.~’-~~ However, this was probably owing to a lower tilt angle rather than to a larger value for Po as the chiral dopant is the same. The effect of an ester group in a central position of a chiral dopant can be elucidated from the data in Table 7.The Sc* transition temperature for the mixture containing the reference a-fluoro (di-)ester15 with a second ester group in a central position is higher (+1.2 “C) than that incorporating an equal amount of the a-fluoro (mono-)ester (28).This is not surprising as the reference a-fluoro (di-)ester15 exhibits an enantiotropic Sc phase at elevated temperatures in the pure state. The SA transition temperature for the mixture containing ester 28 is higher (+3.8 “C) reflecting the high SAtransition temperature of the pure material (see Table 3). The clearing points are almost equal, which is surprising considering the absence of an N* phase for ester 28 and the high N* phase transition temperature for the diester. Spontaneous polarisation is sig- nificantly higher (+ 1.9 nC cmP2) for ester 28 than for the diester.This is probably owing to a change of Po as well as of y, because of presence of the second carboxy (ester) group. Even if Po were constant and the tilt angle were fully respon- sible for the increase of P,, the rotational viscosity of mono-ester 28 would still be substantially lower. The data collated in Table 8 allow a valid comparison of the relative effects of either a fluorine or a chlorine atom attached directly to the optically active centre of the chiral dopant. The transition temperatures of the mixture containing 7 wt.% of a-fluoro ester 28 are all higher than those observed for the corresponding mixture containing an equal amount of the otherwise identical a-chloro ester 31.The spontaneous polarisation for 3-fluoro ester 28 is greater (88.5 nC cmP2, extrapolated to 100%) than that of the analogous a-chloro ester 31 (64 nC cmP2, extrapolated to 100%).This could be because of the stronger electronegativity of the fluorine atom. In addition, the rotational viscosity of the a-fluoro ester must also be lower than that of the a-chloro ester as shown by the significantly lower response time, which is only partially due Table 7 Comparison of the transition temperatures, spontaneous polarisation and response times for two mixtures consisting of 7 wt.% reference ester trans-4-[ 4-(2’,3’-difluoro-4-decyloxybenzoyloxy)phen-yl]cyclohexyl (R)-2-fl~orohexanoate’~and trans-4-(2’,3’-difluoro-4-decyloxybiphenyl-4-y1)cyclohexyl (R)-2-fluorohexanoate (28) and 93 wt.% of the base Sc mixture SC9-1219 (S,*-SA)/”C (SA-N*)/3C (N*-I)/”C PJnC cm-’ Z/~S CO, 76.7 83.5 103.9 5.4 650 ~ 75.5 88.3 104.3 6.2 415 J.MATER. CHEILI., 1994, VOL. 4 Table 8 Comparison of the transition temperatures, spontaneous polarisation and response times for two mixtures consisting of 7 wt.% trans-4-(2’,3’-difluoro-4-decyloxybiphenyl-4-yl)cyclohexyl (R)-2-fluorohexanoate (28) and trans-4-(2’,3’-difluoro-4‘-decyloxybiphenyl-4-y1)cyclohexyl (R)-2-~hlorohexanoate (31) and 93 R t.% of the base S, mixture SC9-1219 F.F ..-c10H210 o* ox X (SC*-SA)/T (SA-N*)/”C (N*-I)/”C PJnC cmp2 Z/~S F 75.5 88.3 104.3 6.2 41 5 c1 74.5 86.3 102.3 4.5 630 to the higher value of Po.This is a reflection of the smaller size of the fluorine atom and the shorter carbon-fluorine bond compared with that of chlorine. These results highlight the attractiveness of fluorine as a substituent attached to the optically active centre of chiral dopants. The authors express their gratitude to Mr. C. Haby, Mr. W. Janz and Mr. J. Reichardt for technical assistance in the preparation and evaluation of the compounds. Dr. W. Arnold (NMR), Mr. W. Meister (MS), Dr. M. Grosjean (IR), Mr. F. Wild and Mr. B. Halm (DTA) are thanked for the measure- ment and interpretation of the required spectra. References 1 N. A. Clark and S. T. Lagerwall, Appl. Phys. Lett., 1989,36, 899. 2 N. A. Clark, M. A. Hanschy and S. T. Lagerwall. Mol.Cryst. Liq. Cryst., 1983,94,213. 3 K. Skarp and M. A. Hanschy, Mol. Cryst. Liq. Cryst., 1988, 165, 439. 4 M. A. Hanschy and N. A. Clark, Ferroelectrics, 1984,59, 69. 5 S. T. Lagerwall, N. A. Clark, J. Dijon and J. F. Clerck, Ferroelectrics, 1989,94, 3. 6 M. Brunet, J. Phys. Colloq., 1975,36, C1-321. 7 R. Eidenschink, R. Hopf, B. S. Scheuble and A. E. F. Wachtler, Proc. 16th Freiburger Arbeitstagung Fliissigkristalle, Freiburg, 1986. 8 T. Geelhaar, T. Escher and E. Bohm, Proc. 17th Freihurger Arbeitstagung Flussigkristalle, Freiburg, 1987. 9 J. W. Goodby and T. Leslie, in Liquid Crystals and Ordered Fluids, ed. A. C. Griffin and J. F. Johnson, Plenum, New York, 1984, vol. 4, p. 1. 10 C. Escher, Kontakte, 1986,2,3. 11 D.M. Walba, S. C. Slater, W. N. Thurme5, N. A. Clark, M. A. Hanschy and F. Supon, J. Am. Chem. Soc., 1986,108,521 1. 12 J. P. le Pesant, B. Mourney, M. Hareng, G. Decobert and J. C.Dubois, Paris Display ’84,1984, 217. 13 J. W. Goodby, Science, 1986,231, 350. 14 J. Bomelburg, G. Heppke and A. Ranft, Z. Nuturforsch. B: Chem. Sci., 1989,44, 1127. 15 R. Buchecker, S. M. Kelly and J. Funfschilling, Iiq. Crpst., 1990, 8,217, and references cited therein. 16 D. M. Walba, R. Vohra, N. A. Clark, M. A. Hanschy, J. Xue, D. S. Parma, S. T. Lagerwall and K. Skarp, J. 4m. Chrrn. Soc., 1986,108,7424. 17 D. M. Walba and N. A. Clark, Ferroelectrics, 1988,84, 65. 18 D. M. Walba, H. A. Razawi, N. A. Clark and D. S. Parma, J. Am. Chern. SOC., 1988,110,8686.19 D. M. Walba, K. F. Eidman and R. C. Haltiwanger, J. Org. Chem., 1989,54,4943. 20 M. D. Wand, R. Vohra, D. M. Walba, N. A. Clark and R. Shao, Mol. Cryst. Liq. Cryst., 1991,202, 183. 21 S. Nakamura and H. Nohira, Mol. Crpst. Liq. Cryst., 1990, 185, 199. J. MATER. CHEM., 1994, VOL. 4 1697 22 H. Nohira, S. Nakamura and M. Kamei, Mol. Cryst. Liq. Cryst., 42 S. M. Kelly, Liq. Cryst., 1993, 14, 675. 1990,180,379. 43 S. M. Kelly, J. Funfschilling and A. Villiger, Liq. Crvst., 1993, 23 S. Arakawa, K. Nit0 and J. Seto, Mol. Cryst. Liq. Cryst., 1991, 14, 813. 204, 15. 44 S. M. Kelly, Liq. Cryst., 1993, 16, 67. 24 C. Bahr and G. Heppke, Mof. Cryst. Liq. Cryst., 1987, 148, 29. 45 S. M. Kelly, A. Germann and M. Schadt, Liq. Cryst., 1994,16,491.25 T. Sakurai, N. Mikami, R. Higurchi, M. Honma, M. Ozaki and 46 S. M. Kelly and J. Funfschilling, J. Muter. Chem., 1994, 4, 1673. K. Yoshino, J. Chem. SOC.,Chem. Commun., 1986,978. 47 A. Focella, F. Bizzarro and C. Exon, Synth. Commun.. 1991, 21, 26 T. Geelhaar, H. A. Kurmeier and A. E. F. Wachtler, Liq. Cryst., 2165. 1989,5, 1269. 48 G. Osterheld, P. Marrug, R. Ruhrer and A. Germann, 27 L. K. M. Chan, G. W. Gray, D. Lacey, R. M. Scrowston, I. G. J. Chromatogr., 1982,234,99. Shenouda and K. J. Toyne, Mol. Cryst. Liq. Cryst., 1989, 172, 125. 49 S. M. Kelly, R. Buchecker, J. Fromm and M Schadt, 28 G. Scherowsky and M. Sefkow, Mol. Cryst. Liq. Cryst., 1991, Ferroelectrics, 1988,85, 385. 202,207. 50 S. M. Kelly and R. Buchecker, Helv. Chim. Acta, 1983, 71, 451 29 G. Scherowsky and J.Gay, Liq. Cryst., 1989,5,1253. and 461. 30 G. Scherowsky, J. Gay and M. Gunararte, Liq. Cryst., 1992, 11, 51 S. M. Kelly, R. Buchecker and M. Schadt, Liq. Cryst, 1988, 3, 745. 1115 and 1125. 31 G. Scherowsky and M. Sefkow, Liq. Cryst., 1992,12,355. 52 S. M. Kelly and A. Villiger, Liq. Cryst., 1988,3, 1173. 32 T. Kusumoto, A. Nakayama, Chem. Lett., 1992,2047. 53 S. M. Kelly, Liq. Cryst., 1989,5, 171. 33 K. Sakaguchi and T. Kitamura, Ferroelectrics, 1991, 114,265. 54 S. M. Kelly, Helv. Chim. Acta, 1989,72, 594. 34 K. Sakaguchi, T. Kitamura, Y. Shiomi, M. Koden and 55 S. M. Kelly, J. Funfschilling and F. Leenhouts, Liq. Crvst., 1991, T. Kuratate, Chem. Lett., 1991, 1383. 10,243. 35 T. Kusumoto, K. Sato, T. Hiyama, S. Takehara, M. Osawa, 56 S. M. Kelly, Mol. Cryst. Liq. Cryst., 1991,204,27. A. Nakayama and T. Fujisawa, Chem. Lett., 1991, 1623. 57 R. Dabrowski, J. Dziaduszek, B. Sosnowska and J. Przedmojski, 36 H. R. Dubal, C. Escher, D. Gunter, W. Hemmerling, Y.Inoguchi, Ferroelectrics, 1991, 114, 229. I. Muller, M. Murakami, D. Ohlendorf and R. Wingen, Jpn. 58 R. Dabrowski, J. Dziaduszek, J. Szulc, K. Czuprjnski and J. Appl. Phys., 1988,27, L2241. B. Sosnowska, Mol. Cryst. Liq. Cryst., 1991,209,201. 37 J. Funfschilling and S. M. Kelly, Proc. 20th Freiburger 59 W. W. McMillan, Phys. Rev. A: Gen. Phys., 1973,8, 1921. Arheitstugung Fliissigkristalle, Freiburg, 199 1. 60 A. Wulf, Phys. Rev. A: Gen. Phys., 1975, 11, 365. 38 S. M. Kelly, J. Funfschilling and A. Villiger, Liq. Cryst., 1993, 61 R. G. Priest, J. Chem. Phys., 1976,65,408. 14,699. 62 K. Siemensmeier and H. Stegemeier, Chem. Phys. Lett., 1988, 148, 39 S. M. Kelly and J. Funfschilling, J. Muter. Chem., 1993,3,953. 409. 40 J. Funfschilling, S. M. Kelly and A. Villiger, Liq. Cryst., 1993, 63 C. Escher, T. Geelhaar and E. Bohm, Liq. Cryst., 1988,3,469. 14, 713. 41 F. Leenhouts, S. M. Kelly and A. Villiger, Displays, 1990,41. Paper 4/01809E; Received 25th March, 1994