J. MATER. CHEM., 1994, 4(9), 1365-1375 Inversion of Chirality-dependent Properties in Helical Liquid Crystals: Effects of Structural Modificationt Peter Styring,* Jelle D. Vuijk, Sharron A. Wright, Kohki Takatoh and Chuchuan Dong Liquid Crystals and Advanced Organic Materials Research Group, School of Chemistry, The University of Hull, Hull, UK HU6 7RX The effects of structural modification on the inversion of chirality-dependent properties in the helical mesophases of a number of terphenyls have been investigated. The occurrence of inversion phenomena are very sensitive to small changes in the molecular structure and subtle changes can often produce profound effects. Structural modification of the non-chiral or chiral terminal chains has been shown to destabilize the inversion of chirality-dependent properties; however, a stable room temperature ferroelectric smectic C*phase has been prepared.The inclusion of lateral fluoro- substituents in the aromatic core causes a large reduction in the melting points of the terphenyl materials and the stabilization of inversion phenomena in the cholesteric phase. Changes in the fluoro-substitution pattern results in a change in the direction of the molecular dipole and hence a change in the sign of the spontaneous polarization. Helical mesophases and therefore also inversion phenomena are suppressed in the non-fluorinated material. The unusual liquid-crystalline phase behaviour of a number of materials that undergo inversions in chirality-dependent properties in their helical mesophases, and in unwound cells at infinite pitch, as a function of temperature has been reported previously.'-'' In particular, work has centred on the changes in the helical twist sense, spontaneous polarization (P,j and 1 apparent optical tilt angle (8,) in the chiral smectic C* (Sc*j phase and changes in the helical twist sense in the cholesteric CQH1QO-10 \/(N*) phase.\/ yIf only the helical twist sense and spontaneous polarization in the (S,* phase and the helical twist sense in the N* phase are considered, it is possible to predict a number of combi- 2 nations that might be expected to be observed (Table 1). In addition, when the polarization inverts it can be accompanied by an inversion of 8, in the (Sc* phase, as the tilt drives the spontaneous polarization.An inversion of helical twist sense in either phase is necessarily accompanied by an inversion in the pitch of the helix because, in order for the handedness to change, the pitch of the helix must first increase and then pass through a region of infinite pitch before a helix of opposite twist sense can reform. Several materials have been identified that exhibit inversion phenomena and these can be classified according to the types listed in Table 1. Compound 1 shows inversion phenomena Table 1 Possible combinations of properties that might show inversion 5 smectic C* cholesteric of type 1, i.e. inversion of helical twist sense in both phases accompanied by an inversion in the spontaneous polerization.' inversion of helical inversion of inversion of helical This is the only material known to display type 1 behaviour.tY Pe twist sense ps twist sense The chiral chloro-ester 2,7 derived from the amino acid (S)-1 J J J 2 J X J 3 X J J 4 X X J 5 J X X 6 X J X 7 J J X ~~~ t This paper was submitted in association with the 1st International Conference on Materials Chemistry, July, 1993.alanine, and the chiral fluoro-ester 3,' show inversions of the helical twist sense in the cholesteric phase only and are therefore classified as type 4 inversions. Compound 4, which has a core structure similar to that of 2, shows an inversion in the helical twist sense but no inversion in spontaneous polarization in the (Sc* phase with no inversion in the N* phase and is therefore classified as type 5.6 Finally, the biphenylcarboxylic acid ester of (Sj-2-methylbutm-1-01 (5) shows only an inversion of spontaneous polarization in the (Sc* phase without any helix inversions (type 6).?Although one compound is known (PACMB),’ which shows only a helix inversion in the Sc* phase (type 5), it is a diastereoisomer and contains two independent chiral centres, which are believed to compensate for each other.Thus, the competition between the chiral centres in the diastereoisomers produces the temperature dependent inversion in the helical pitch. In this work, we have concentrated on materials containing a chiral oxirane ring, which possesses two sequential chiral centres.However, these are constrained in a cyclic structure and therefore behave as a single chiral entity relative to the mesogenic core. These materials produce a number of chiral property combinations, however we do not yet have materials possessing a single chiral centre that show inversions of types 2, 3 and 7. We have explained the origins of these inversion phenomena in materials possessing single chiral centres in terms of interconversions between competing rotational conformers, which involve the chiral centre relative to the rigid mesogenic core of the mole~ule.~-l~ We have proposed that at least two competing rotational conformers exist that possess approxi- mately equal potential energies and that these conformers are separated by an energy barrier to rotation (AE), which is accessible over the temperature regime of the mesophase.Demus and co-workers” have also observed an inversion of spontaneous polarization in a number of chiral phenylpy- rimidine liquid crystals. This has been explained in terms of changes in the interaction between molecular steric dipoles and is based on Pikin and Osipov molecular statistical theory.’, We have recently undertaken an investigation of both our own conformer theory and the molecular statistical theory in relation to our materials in order to clarify the situation. In this paper we report the results of our studies into the effects of changes in the structure of (i) the chiral terminal group, (ii) the non-chiral terminal group and (iii) the meso- genic core on helical inversion phenomena.In order to achieve this we have taken the basic structure of 1 and made the appropriate modifications to the terphenyl core and terminal substituents while retaining the oxirane ring as the chiral centre as shown in Fig. 1. Experimental Analysis of Materials The structures of all intermediates and final products were elucidated by a variety of analytical techniques. Proton nuclear magnetic resonance (‘H NMR) spectra were recorded using a JEOL JNM-GX 270 FT NMR spectrometer. Infrared (IR) spectra were recorded as either KBr disks or liquid films using a Perkin-Elmer 783 IR spectrometer. Mass spectra were recorded using a Finnigan Matt 1020 Automate GC/MS (gas chromatography/mass spectrometry) spectrometer.Satis- factory analyses were achieved in all instances. The purities of the final products were determined by high-performance liquid chromatography (HPLC) in both normal and reversed modes. Normal phase analysis was performed over silica gel (5 pm pore size, 25 cm x 0.46 cm, Dynamax scout column) and reversed-phase analysis was performed over octadecylsi- loxane (5 pm pore size, 25 cm x 0.46 cm, ODS Microsorb Dynamax 18 column). Methanol was used as eluent in both cases. Optical rotations were measured using an AA-10 auto- Fig. 1 General structure of the materials under investigation J. MATER. CHEM., 1994, VOL. 4 matic polarimeter at the sodium D line.Satisfactory analyses were achieved in all instances. Elemental analyses were per- formed on a Carlo Erba 1106 CHN analyser using cyclohexanone-2,4-dinitrophenylhydrazoneas the reference standard. The mesomorphic phase sequence of each compound, sand- wiched between glass slides, was determined by thermal optical microscopy using a Zeiss Universal polarizing microscope equipped with either a Mettler FP82HT microfurnace and FP90 temperature controller or a Mettler FPX2 microfurnace and FP80 temperature controller. Optical results were confirmed by differcntial scanning calorimetry (DSC) using a Perkin-Elmer DSC‘7-PC equipped with an intracooler. Enthalpies of transition, in J g-l, are shown in parentheses below the transition temperatures.Samples were encapsulated in standard aluminium pans and the mesophase ranges scanned at rates of 10 and 2 K min-’. The accuracy of the data derived from the DSC experiments was confirmed by measuring the enthalpy of fusion and melting temperature of pure indium metal. The melting enthalpy of 29.8 J 8-l and melting temperature of 156.7 C compared well with the literature values of 28.5 J g-’ and 156.6“C, respectively. Electro-optic studies were performed on the materials con- tained in 0.25 cm2 ITO-coated active area test cells which were obtained from the Electronics Chemicals High Tech- nology Group. The inner surfaces of the cells were coated with a polyimide alignment layer that had been unidirection- ally buffed and assembled with parallel rubbing directions.The absence of interference fringes in the unfilled cell indicated a homogeneous cell thickness across the active area. Ac voltages were applied in sine-wave mode using an Advance Electronics AF signal generator J2C and dc voltages using a Farnell Instruments LT30-2 stabilized power supply. Applied voltages were determined accurately using a Beckman Industrial DM78 multimeter. When determining the magni- tude of the spontaneous polarization, the hysteresis loop was observed on a Dartron Instruments dual trace oscilloscope D17 and the P, determined using a Diamant bridge.’, Synthesis Six compounds of the general structure shown in Fig. 1 were prepared according to the procedures detailed in Scheme 1. 4-Bromophenylmethano1 was alkylated under standard con- ditions to give l-bromo-4-(propoxymethyl)benzene, 6 (R’= C,H7, R”=H; n=l), which was transformed into boronic acid 8 by treatment with n-butyllithium followed by triisopro- pyl borate in tetrahydrofuran (THF) at -78 ’C.A similar method was used to prepare l-bromo-4-(2-ethoxyethyl) ben-zene, 9 (R=C,H,; n =l), from 2-( 4-bromopheny1)ethanol. Three chiral derivatives of 4’-bromo-2’-fluorobiphenol were prepared14 from (2S73S)-3-propyloxiranemethanol( R’ =C,H,, R”=H) to give 10, (2S,3S)-3-methyl-3-[ 5-( 2-methylpent-2- enyl)] oxiranemethanol [R’ =CH,CH,CH =C(CH,), , R”= CH,] to give 11 and (2&3R)-3-methyl-3-[ 5-( 2-methylpent-2-enyl)] oxiranemethanol [R” =CH,CH,CH =C( CH,),, R’= CH,]to give 12. In order to examine the effect of changes in the mesogenic core, the (2S,3S)-3-propyloxiranemethanol (R’= C,H,, R”=H) derivatives of 4’-bromo-3-fluorobiphenol and 4’-bromobiphenol were prepared using similar procedures to give 13 and 14, respectively.Intermediates 11-14 were coupled with boronic acid, 8 to give the target compounds 15-17 and 20, respectively. Standard cross-coupling procedure^'^ were employed, using [tetrakis(triphenylphosphine)palladium(o)] as the catalyst and 1,2-dimethoxyethane (DME) as the solvent. Similar pro- cedures were used to obtain cross-coupled products from the J. MATER. CHEM., 1994, VOL. 4 6 I 7 0 10-14 9 15,16,17, 20 10,lQ Scheme 1 reaction between boronic acid 9 and intermediates 10 and 11 to give the target materials 18 and 19, respectively.Each of the final products 15-20 were recrystallized a number of times, until constant transition temperatures and HPLC purit- ies of greater than 99.5% were achieved. 1-Bromo-4-( propoxymethyl )benzene (6) A mixture of sodium hydride (80% dispersion in oil, 1.15 g, 40 mmol) and 1-bromoethane (6.0 g, 40 mmol) was added to a solution of 4-bromophenylmethano1(6.0g, 30 mmol) in dry N,N-dimethylformamide (DMF; 50 cm3), and the resulting mixture stirred at room temperature (20 h) under an atmos- phere of dry nitrogen. Water was added carefully to destroy unreacted sodium hydride and the mixture diluted with water (200 cm3) and extracted with diethyl ether (50 cm3). The ethereal layer was washed with water (50 cm3), dried (MgSO,) and the solvent removed in uucuo. The product was purified by distillation under reduced pressure then passed through a short gravity column (silica gel; dichloromethane) to afford a colourless oil.Yield =4.35 g (61 Yo);bp =80 "C (0.5 mmHg). 'H NMR 6, (CDC13,270 MHz, TMS) 0.95 (3 H, t, CHZCH,), 1.65 (2 H, sex, CH,CH,), 3.45 (2 H, t, OCH,CH,), 4.45 (2 H, s, OCH2Ar), 7.24 (2 H, AA'XX', Ar-H), 7.48 (2 H, AA'XX', Ar-H). IR (liquid film): vrnax/cm-l 2980 (CH), 1485 (Ar), 1100,1010,800. MS: m/z 230/228 (M)+, 202/200(M-C2H4)+, 171/169, 107, 91, 58. 1-Bromo-4-( 2-ethoxyethy1)benzene (7) 2-( 4-Bromopheny1)ethanol (5.6 g, 30 mmol) was alkylated with 1-bromoethane using the method described above to afford a colourless oil.Yield =1.80 g (47%); bp =58 "C (0.5 mmHg). 'H NMR 6, (CDCl,, 270 MHz, TMS) 1.20 (3 H, t, J 7.0 Hz, CH,CH,), 2.83 (2 H, t, J 7.0 Hz, OCH,CH, -Ar) 3.47 (2 H, q, J 7.0 Hz, OCH,CH,), 3.60 (2 H, t, J 7.0 Hz, OCH,CH2-Ar), 7.10 (2 H, AA'XX, Ar-H), 7.4 (2 H, AA'XX', Ar-H). IR (liquid film): vrnax/cm-l 2980 (CH),1485 (Ar), 1100, 1010, 800. MS: m/z 230/228 (M)', 186/184 (M-OC,H,)+, 171/169, 89, 59. 4-Propoxymethylphenylboronicacid (8) A solution of 1-bromo-4-(propoxymethyl) benzene (4.0 g, 17.5 mmol) in THF (40 cm3) was cooled to -78 'C in an atmosphere of dry nitrogen and a solution of n-but yllithium (1.6 mol dme3 in hexane, 11.8 cm3, 17.5 mmol) was added slowly. After stirring the mixture at -78 "C (2 h), triisoprop- ylborate (6.6 g, 35 mmol) was added slowly and stirring was continued (1 h) at -78 "C.The mixture was allowed to come to room temperature and stirred (20 h). Hydrochloric acid (lo%, 20 cm3) was added to the stirred mixture, which was then extracted with diethyl ether (3 x 25 cm3). The Gombined ethereal layers were washed with water (25 cm3) md dried (MgSOJ. The solvent was removed in uucuo, to afford a light- yellow oil that solidified on standing. Yield =3.20 g (94%). 'H NMR 6, (C2H,]DMS0, 270 MHz, TMS) 0.75 (3 H, t, CH,CH,), 1.5 (2 H, sex, CH,CH,CH,), 3.3 (2 H, t, OCH,CH,CH,), 4.4 (2 H, s, OCH,-Ar), 7.3 (2 H, AA'XX', Ar-H), 8.05 (2 H, AA'XX', Ar-H). IR (liquid film): vrnax/cm-' 3350 (br, OH), 1610 (Ar) 1350, 700. MS: m/z 484 (3M-44)+, 176, 116, 91, 58.442-Ethoxyethy1)phenylboronicacid (9) The boronic acid derived from 1-bromo-4-( 2- ethoxyethy1)benzene (7)(4.0 g, 17.5 mmol) was prepared using the method described above to afford a yellow oil that solidified on standing. Yield= 3.58 g (76%). 'H NMR 6, ([2H6]DMS0, 270 MHz, TMS) 1.25 (3 H, t, CH3CH,), 3.00 (2 H, t, Ar-CH,CH,O), 3.50 (2 H, q, OCH,CH3), 3.57 (2 H, t, OCH,CH,Ar), 7.4 (2 H, AAXX, Ar-H), 8.25 (2 11, AA'XX', Ar-H). IR (liquid film): vmax/cm-' 3300 (br, BOH), 2940 (CH), 1610, 1370 (br), 1100 (br). MS: m/z 529 (.JM+H)+, 468, 162, 103. 4-Bromo-2-fluoro-4’-[(2S,3S)-3-propyloxiran-2-ylmethoxy] biphenyl (10) (2S,3S)-3-Propyloxiranemethanol( 1.45 g, 12.5 mmol) and diethyl azodicarboxylate (DEAD) (2.18 g, 12.5 mmol) were added successively to a solution of 4-bromo-2’-fluoro-4-hydroxybiphenyl (3.34 g, 12.5 mmol) and triphenylphosphine (3.28 g, 12.5 mmol) in THF (40 cm3).The reaction mixture was stirred at room temperature (20 h), the solvent removed in uucuo and the crude product purified by flash column chromatography [silica gel; dichloromethane: light petroleum (bp 60-80°C) (4: l)], to afford a white solid. Yield= 3.58 g, (78%); mp= 76-78 “C. ‘H NMR 6, (CDCl,, 270 MHz, TMS) 0.99 (3 H, m, CH,CH,), 1.5 (4 H, m, CH,CH,CH,), 2.98 (1 H, m, CH-O-CH-C,H7) 3.12 (1 H, m, CH-O-CH- C3H7), 4.03 (1 H, ABX, 0-CH,-CH-0-CH), 4.22 ( 1 H, ABX, 0-CH2 -CH-0-CH), 6.99 (2 H, AA’XX’, Ar-H), 7.25 (1 H, m, Ar-H), 7.32 (1 H, m, Ar-H), 7.44 (2 H, AA’XX’, Ar-H).IR (liquid film): v,ax/cm-l 2980 (CH), 1800, 1470 (Ar), 1250 (C-F), 1000, 800. MS: m/z 366/364 (M)’, 268/266, 170, 99. 4-Bromo-2-fluoro-4’[( 2S,3S)-3-methyl-3-[ 542-methylpent-2-en-1-yl)] oxiran-Zylmethoxy] biphenyl (1 1 ) 4-Bromo-2-fluoro-4’-hydroxybiphenyl (0.67 g, 2.5 mmol) was alk yla ted with (2S,3S)-3-met hyl- 3- [5-(2-me t hylpen t-2-en- 1-yl)] oxiranemethanol using the method described above to afford a colourless oil. Yield =0.94 g (90%). ’H NMR 6, (CDCl,, 270MHz, TMS) 1.36 (3 H, s, CH-O-C-CH3), 1.62 (3 H, S, C=C-CH3), 1.69 (3 H, S, C=C-CH3), 1.7 (2 H, m, C=CH-CH2-CH,), 2.13 (2 H, q, C= CH--23,-CH,), 3.17 (1 H, t, CH,-CH-0-CMe), 4.11, ( 1 H, ABX, 0-CCH, -CH-0-CMe), 4.19 ( 1 H, ABX, 0-CH,-CH-0-CMe), 5.11 (1 H, m, CH=CMe,), 7.01 (2 H, AA’XX’, Ar-H), 7.26 (1 H, AA’XX’, Ar-H), 7.33 (2 H, AA’XX’, Ar-H), 7.46 (2 H, AAXX‘, Ar-H).IR (liquid film): vmaX/cm-’ 2960 (CH), 1600, 1480 (Ar), 1240, 870, 810. MS: m/z 420/418 (M)’, 268/266, 170, 109. 4-Bromo-2-fluoro-4-[(2S,3R)-3-methyl-3-[ 5-(2-methylpent-2-en-1-yl )] oxiran-2-ylmethoxy] bipohenyl(l2) 4-Bromo-2-fluoro-4’-hydroxybiphenyl ( 1.34 g, 5 mmol) was oxiranemethanol using the method described above to afford a colourless oil. Yield= 1.75 g (83%). ‘H NMR 6, (CDCl,, 270MHz, TMS) 1.41 (3 H, S, CH-0-C-CH,), 1.59 (3 H, S, C=C-CH+,), 1.69 (3 H, S, C=C-CH,), 1.7 (2 H, m, C=CH-CH,-CCH,), 2.15 (2 H, q, C=CH-CH,-CH,), 3.16 (1 H, t, CH,-CH-0-CMe), 4.10 (1 H, ABX, 0-CH, -CH-0-CMe), 4.18 (1 H, ABX, 0-CH,-CH-0-CMe), 5.12 (1 H, m, CH=CMe,), 7.00 (2 H, AA’XX’, Ar-H), 7.27 (1 H, AA’XX’, Ar-H), 7.34 (2 H, AA’XX’, Ar-H), 7.46 (2 H, AAXX’, Ar-H).IR (liquid film): vmaX/cm-’ 2960 (CH), 1600, 1480 (Ar), 1240, 870, 810. MS: m/z 420/418 (M)’, 268/266, 170, 109. 4-Bromo-2-fluoro-4-[(2S,3S)-3-propyloxiran-2-ylmethoxy] biphenyl (13) 4-Bromo-2’-fluoro-4’-hydroxybiphenyl(1.37 g, 5.0 mmol) was alkylated with (2S,3S)-3-propyloxiranemethanol(0.58 g, 5.0mmol) using the method described above to afford a colourless oil. Yield =0.84 g, (53%); mp =66-68 “C. ‘H NMR 6, (CDC13,270 MHz, TMS) 1.00 (3 H, t, CHZCH,), 1.62 (4 H, m, CH2CH,CH,), 2.97 (1 H, ABX, OCH2 -CH-0-CH- C3H7), 3.13 (1 H, ABX, OCH,--CH-O-CH-C3H7), 4.03 alkylated with (2S,3R)-3-methyl-3-(4-methylpent-3-en-l-y1)-white solid that was recrystallized several times from meth- J.MATER. CHEM., 1994, VOL. 4 (1 H, ABX, OCH,-CH-0-CH-), 4.22 (1 H, ABX, OCH2-CH-0-CH-), 6.98 (2 H, AA’XX’, Ar--H), 7.3 (3 H, m, Ar-H), 7.44 (2 H, AA’XX’, Ar-H). IR (KBr): vmaX/cm-’ 2980 (CH), 1800, 1470 (Ar), 1250 (C-F), 1000, 800 cm-l. MS: m/z 366/364 (M)’, 268/266, 170, 99. 4-Bromo-4-[ (2S,3S)-3-propyloxiran-2-ylmethoxy]biphenyl (14) 4-Bromo-4-biphenol (0.75 g, 3.01 mmol) was alkylated with (2S,3S)-3-propyloxiranemethanol(0.35 g, 3.00 mmol) using the method described above to afford a colourless oil. Yield = 0.80 g (94%); mp= 110.3 “C. ‘H NMR 6, (CDCl,, 270 MHz, TMS) 0.99 (3 H, t, J 8.0Hz, CH,), 1.56 (2 H, m, OCH,CH,CH,), 1.60 (2 H, m, CH-0-CH-CH,CH,CH,), 2.96 (1 H, ABX, J 5.0, 3.0 Hz, OCH2-CH-0-CH-), 3.13 (1 H, ABX, J 5.0, 3.0, 2.5Hz, OCH2-CH-0-CH-), 4.03 [l H, ABX, J 11.5 (gem), 5.0 (anti)Hz, OCH,-CH-0-CH-],4.22 [l H, ABX, J 11.5 (gem), 3.0 (syn)Hz) OCH2-CH-0-CH--1, 6.98 (2 H, AA’XX’, J 8.5 Hz, Ar-H), 7.40 (2 H, m, J 8.5 Hz, Ar-H), 7.48 (2 H, AAXX’, J 8.5 Hz, Ar-H), 7.53 (2 H, AA’XX’, J 8.5 Hz, Ar-H).IR (KBr): vmax/cm-’ 2950, 2920 (CH), 1600, 1480 (Ar), 1285,1250(CH-OCH), 825, 805 cm-’. MS: m/z 348/346 (M)’, 250/248, 99, 55. 2’-Fluoro-4-[( 2S,3S)-3-met h yl-345-( 2-methylpen t-2-en- 1-yl )] oxiran-2-ylmethoxy [-4”-propoxymethyl-l ,1‘:4’,1’’- terphenyl(l5) An aqueous solution of sodium carbonate (2 mol dm-3, 8 cm3) and the coupling catalyst [tetrakis( triphenylphos- phine)palladium(o)] (30 mg, 30 pmol) were added to a solution of 4-bromo-2-fluoro-4’-(2S,3S)-3-methyl-3-[5-(2-methylpent-2-en- 1-yl)] oxiran-2-ylmethoxy] biphenyl (0.45 g, 1.07mmol) in DME (8 cm3).4-Propoxymethylphenylboronic acid (0.19 g, 1.07 mmol) was added and the mixture stirred at 90 “C (24 h). The dark-brown mixture was diluted with diethyl ether (100 cm3) and water (50 cm3). The separated aqueous layer was washed with diethyl ether (2 x 25 cm3) and the combined organic layers were washed with water (50 cm3) and dried (MgSO,). The solvent was removed in uucuo and the crude product was purified by flash column chromatography (silica gel; dichloromethane) to afford a anol. Yield =140 mg (30%); mp = -8.0 ”C: for meso-morphism see Results section.[x]”’~ = -9.8 O; HPLC > 99.5%. Calc. for C32H37F03; C, 78.58; H, 7.62%. Found: C, 78.63; H, 7.68%. ‘H NMR 6, (CDCl,, 270MHz, TMS) 0.97 (3 H, t, J 7.2 Hz, CH,CH,), 1.33 (3 H, s, CH-0-C-CH,), 1.55 (2 H, m, C=CH-CH, -CH,), 1.65 (3 H, s, C=C-CH,), 1.68 (2 H, m, CH,CH,CH,), 1.69 (3 H, S, C=C-CH3), 2.14 (2 H, q, J 7.1 Hz, C=CH-CCH2-CH2), 3.18 (1 H, t, J 5.4Hz, CH,-CH-0-CMe), 3.48 (2 H, t, J 7.2 Hz, -OCH,CH,CH,), 4.13 [l H, ABX, J 11.0 (gem), 5.6 (anti)Hz, 0-CH,-CH-0-CMe], 4.19 [l H, ABX, J 11.0 (gem), 5.0 (syn)Hz, 0-CH,-CH-0-CMe], 4.55 (2 H, s, Ar-CH,-0), 5.11 (1 H, m, J 9.2, 1.8 Hz, CH=CMe,), 7.05 (2 H, AA’XX’, J 9.0Hz, Ar-H), 7.34-7.63 (9 H, m, Ar-H). IR (liquid film): vmax/cm-l 2980 (CH), 1600, 1485 (Ar), 1230 (C-F), 1240, 800.MS: m/z 488 (M)’, 336, 277. 248. 2’-Fluoro-4-[( 2S,3R)-3-methyl-3- [5-(2-methylpen t-2-en-l- yl )] oxiran-2-ylmethoxy]-4”-propoxymethyI-l,1’:4‘,l’’-terphenyl(16) 4-Bromo-2-fluoro-4’-[( 2S,3R)-3-methyl-3-[ 5-( 2-methylpent-2-en- 1-yl-)] oxiran-2-ylmethoxy] biphenyl (0.45 g, 1.07 mmol ) J. MATER. CHEM., 1994, VOL. 4 was coupled with 4-propoxymethylphenylboronic acid (0.19 g, 1.07mmol) using the method described above to afford a white solid, which was recrystallized several times from light petroleum (bp 30-40 “C). Yield =100 mg (19”/0); mp =56.6 “C; for mesomorphism see Results section. = -3.1”; HPLC>99.5%. Calc. for C,,H,,FO,: C, 78.58; H, 7.62%. Found: C, 78.70; H, 7.65%. ‘H NMR 6, (CDCl,, 270 MHz, TMS) 0.96 (3 H, t, J 7.4 Hz, CH2CH3), 1.41 (3 H, s, CH-0-C-CH,), 1.57 (2 H, m, C=CH-CH2-CH2), 1.62 (3 H, s, C=C-CH,), 1.68 (2 H, m, CH,CH,CH,), 1.70 (3 H, S, C=C-CH,), 2.19 (2 H, q, J 7.7 Hz, C=CH-CH,-CH,), 3.18 (1 H, t, J 5.2Hz, CH,-CH-0-CMe), 3.48 (2 H, t, J 6.7 Hz, -OCH,CH,CH,), 4.12 [1 H, ABX, J 10.8 (gem), 6.0 (anti)Hz, 0-CH,-CH-0-CMe], 4.21 [l H, ABX, J 10.8 (gem), 5.2 (syn)Hz, 0-CH,-CH-0-CMe], 4.56 (2 H, s, Ar-CH,O), 5.13 (1 H, m, J 7.1, 1.6 Hz, CH=CMe,), 7.02 (2 H, AA’XX’, J 8.8 Hz, Ar-H), 7.35-7.61 (2 H, m, Ar-H).IR (liquid film): vma,/cm-l 2980 (CH), 1600, 1485 (Ar), 1235 (C-F). 1240, 800. MS: m/z 488 (M)+, 336, 277, 248. 2’-Fluoro-4-[(2S,3S)-3-propyloxiran-2-ylmethoxy]-4-propoxymethy1-1,1’:4’1”-terpheny1(17) 4-Bromo-3-fluoro-4’-[(2S, 3S)-3-propyloxiran-2-ylmethoxy]-biphenyl (0.80 g, 2.2 mmol) was coupled with 4-propoxy- methylphenylboronic acid (0.5 g, 2.8 mmol) using the method described above to afford a white solid, which was recrystallized several times from methanol.Yield =270 mg (28%); mp =66.5 “C; for mesomorphism see Results section. [cY],~= -10.5”; HPLC >99.5%. Calc. for C28H,1F0,: C, 77.39; H, 7.19%. Found: C, 77.37; H, 7.23%. ‘H NMR 6, (CDCI,, 270 MHz, TMS) 0.96 (3 H, t, J 7.2 Hz, CH,CH,), 0.99 (3 H, t, J 7.2Hz, CH,CH,), 1.4-1.7 (6 H, m, 3xCH,), 2.98 (2 H, ABX, J 7.2Hz, OCH,CHOCH), 3.13 (1 H, m, J 5.5, 3.0 Hz, OCH,CHOCH), 3.48 (2 H, t, J 7.1 Hz, ArCH,OCH,CH,), 4.01 (2 H, ABX, J 11.0, 5.5 Hz, CH2-CH-0-CH), 4.23 (2 H, ABX, J 11.0, 3.0Hz, CH,-CH-0-CH), 4.56 (2 H, s, OCH,Ar), 7.0 (2 H, AA‘XX‘, J 8.9Hz, Ar-H), 6.78 (2 H, m, Ar-H), 7.35 (3 H, m, Ar- H), 7.62 (4 H, m, Ar-H).IR (KBr): vmax/cm-l 2940 (C-H), 1620 (C-C), 1485 (C-H), 1310 (C-0), 1230 (C-F), 1170, 1100, 820. MS: m/z 434 (M)’, 277, 149, 99, 43. 2’-Fluoro-4-[( 2S,3S)-3-propyloxiran-2-ylmethoxy]-4”-ethoxyethyl-l,l’:4‘,1’’-terphenyl(18) 4-Bromo-2-fluoro-4’-[( 2S, 3S)-3-propyloxiran-2-ylmethoxy]-biphenyl (0.45 g, 1.07 mmol) was coupled with 4-ethoxy- ethylphenylboronic acid (0.19 g, 1.07 mmol) using the method described above to afford a white solid, which was recryst- allized several times from methanol. Yield= 108 mg ( 18%); mp =101.5 “C; for mesomorphism see Results section. = -13.0’; HPLC >99.5YO.Calc. for C,,H,,FO, : c, 77.39; H, 7.19%. Found: C, 77.45; H, 7.25%. ‘H NMR hH (CDCI,, 270 MHz, TMS) 0.99 (3 H, t, J 7.2 Hz, CH,CH,), 1.22 (3 H, t, J 7.0 Hz, OCH,CH,), 1.57 (4 H, m, CH,CH,CH,), 2.95 (2 H, t, J 7.3Hz, Ar-CH,CH,O), 3.00 (1 H, ABX, J 7.2Hz, OCH,CH-0-CCH-), 3.14 [l H, ABX, J 5.4 (anti), 3.8 (syn)Hz, OCH,CH-0-CH-1, 3.53 (2 H, q, J 7.0Hz, OCH,CH,), 3.68 (3 H, t, J 7.3 Hz, Ar-CH,CH,O), 4.05 [1 H, ABX, J 11.1 (gem), 5.4 (anti)Hz, OCH,CH-0-CH-1, 4.22 [l H, ABX, J 11.1 (gem), 3.8 (syn)Hz, OCH,CH-O-CH-],7.01 (2 H, AA’XX’, J 8.9 Hz, Ar-H), 7.32-7.45 (5 H, m, Ar-H), 7.54 (4 H, m, Ar-H). IR (liquid film): vmax/cm-’ 2’380 (Ar), 1600, 1490 (Ar), 1250 (C-F), 1060, 810. MS: m/z 434 (M)+,375, 335, 277, 246. 2’-Fluoro-4-[( 2S,3S)-3-methyl-3-[ 5-(2-methylpent-2-en-1-yl )]oxiran-2-ylmethoxy]-4”-ethoxyethyl-l,1’:4,1“-terphen yl (19) 4-Bromo-2-fluoro-4’[(2S, 3S)-3-methyl-3-[ 5-( 2-methj Ipent-2- en-1-yl)] oxiran-2-ylmethoxy] biphenyl (0.45 g, l.07 mmol) was coupled with 4-(2-ethoxyethyl)phenylboronicacid (0.19 g, 1.07mmol) using the method described above to .ifford a white solid, which was recrystallized several times from meth- anol.Yield =200 mg (55%); mp =45.7 “C; for mesomarphism see Results section. = -10.7”; HPLC >99.5%. Calc. for C,,H3,F0,: C, 78.58; H, 7.62%. Found: C, 78.71; H 7.66%. ‘H NMR 6, (CDC1, 270 MHz, TMS) 1.23 (3 H, t, J 7.2 Hz, OCH,CH,), 1.37 (3 H, s, CH-0-C-CH,), 1.58 (2 H, m, C=CH-CH,-CH,), 1.63 (3 H, S, C=C-CH,), 1.70 (3 H, S, C=C--CH,), 2.13 (2 H, q, J 7.1 Hz, C=CH-CH, -CH,), 2.95 (2 H, t, J 7.2 Hz, Ar-CH,CH,O), 3.18 (1 H, t, f 5.4 Hz, CH,-CH-0-CMe), 3.54 (2 H, q, J 7.1 Hz, -OCH,CH,), 3.68 (2 H, t, J, 7.2 Hz, Ar-CH,CH,O), 4.13 [l 31, ABX, J 11.0 (gem), 5.6 (anti)Hz, 0-CCH, -CH-0--CMe], 4.19 [l H, ABX, J 11.0 (gem), 5.0 lsyn) Hz, 0-CH,-CH-0-CMe], 5.12 (1 H, m, J 9.2, 1.8Hz, CH=CMe2), 7.03 (2 H, AA’XX’, J 8.9 Hz, Ar-H), 7.32-7.46 (5 H, m, Ar-H), 7.54 (4 H, AA’XX’, J 8.1 Hz, Ar--H).IR (liquid film): vmax/cm-’ 2980 (CH), 1600, 1490 (Ar), 1250 (C-F), 1060, 810. MS: m/z 488 (M)+,336, 277, 153 4-[(2S,3S)-3-propyloxiran-2-ylmethoxy]-4-propoxymethyl-l,1’:4,l”-terphenyl (20) 4-Bromo-4‘- [(2S,3S)- 3 -propyloxiran-2-ylmethoxy]biphenyl (97.5 mg, 0.282 mmol) was coupled with 4-ethoxyethyl-phenylboronic acid (80.0 mg, 0.412 mmol) using the method described above to afford a white solid, which was recryst- allized several times from methanol.Yield =69.2 mg (59%); mp =2 10 “C; for mesomorphism see Results section. [a]24= -28.0”; HPLC>99.5%. Calc. for C28H3203: C, 80.73; H, 7.74%. Found. C, 80.88; H, 7.86%. ‘H NMR 6, (CDCI,, 270 MHz, TMS) 0.96 (3 H, t, J 8.8 Hz, CH,CII,), 0.99 (3 H, t, J 7.6 Hz, CH,CH,), 1.56 (4 H, m, 2 x CH,CH,CH,), 1.66 (2 H, m, CH,CH,-CH-0-CH-), 2.99 (1 H, m, J 6.0, 4.8, 3.7Hz, CH,-CH-0-CH-CH,-0), 3.14 (1 H, m, J 6.0, 3.7 Hz, CH,-CH-OO--CH-CI~~-O]), 3.47 (2 H, t, J 6.6Hz, OCH2CH2-), 4.05 [l H, dd, J 11.4 (gem), 6.0 (anti)Hz, CH-0-CH-CH2-01, 4.23 [l H, dd, J 11.4 (gem), 3.8 (syn)Hz, CH-O-CH- CH,-01, 4.55 (2 H, s, 0-CH,-Ar), 7.01 (2 H, AA’XX’, J 8.5 Hz, Ar-H), 7.43 (2 H, AA‘XX‘, .I 8.2 Hz, Ar-If), 7.57 (2 H, AA’XX’, J 8.5Hz, Ar-H), 7.59 (2 H, AA’XX’, J 8.2Hz, Ar-H), 7.73 (4 H, AA’XX’, Ar-H).IR (liquid film): v,a,/cm-l 2980 (Ar), 1600, 1490 (Ar), 1250 (C-F), 1060, 810. MS: m/z 434 (M)’, 375, 335, 27’7,246. Results Optical and Electro-optical Studies EfSeect of Structural Changes in the Chiral Terminal Chain While retaining the oxirane ring as the source of chirality, the nature of the hydrocarbon component of the chain N as varied. This was achieved by using the chiral oxirane.; derived from (2E)-3,7-dimethylocta-2,6-dien-l-ol(geraniol) and (22)- 3,7-dimethylocta-2,6-dien-l-o1 (nerol)14 to give 15 and 16, respectively. The remaining molecular structure was main- tained as for 1, which has been reported previously as showing the following phase sequence.8 J. MATER.CHEM., 1994, VOL. 4 F H on a Silicon Graphics XS24 4000 UNIX-based workstation. foqJqJ+o&-LAll structures were minimized using CHARMm (MSI) rou-H 1 K 59 Sc: 46.2 S,--46.8 S,: 103.3 Ng 106.3 NZ 112.1 NZ 158.8 BPI 162.9 BPI1 164.6 "C Is0 It should be noted that we have recently reassessed16 the helical twist sense in the Sc* and cholesteric phases of 1. We have found that the higher temperature Sc* and N* phases both possess left-handed helices whereas the lower tempera- ture phases have right-handed helices. However, the inversion temperatures are confirmed to be as reported previously. The problem experienced in the initial assignment arose because measurements were made close to the inversion points and this made it difficult to assign the twist sense correctly owing to the subtle colour changes observed. All the compounds reported here will be discussed relative to 1, which will act as the reference standard.Compound 15 is a low melting solid (mp= -8 "C) that possesses a room temperature Sc* phase. The full phase sequence has been determined as: F ti 15 K -8 S,: 44.3 Ng 76.9 Is0 "C (4.2) (0.13) (0.62) (J g-7 No inversions are observed in either phase, nor is there an inversion in the helical twist sense at the phase transition, both the Sc* and N* phases possess right-handed helices (Zaevorotation of plane-polarized light).In both phases, the helix direction is the same as that observed in the lower temperature regions of each of the corresponding helical phases observed for 1. Compound 16 on the other hand shows no mesomorphism, simply melting from the crystal to the isotropic liquid (K 60 "C Iso, AH=34.0 J g-'). Although supercooling occurs down to room temperature, there is no evidence of a monotropic phase and shearing the sample only induces crystallization. H 16 The inclusion of a methyl group (R") and long hydrocarbon group (R') at the chiral C-3 carbon in 15 produces a substantial decrease in the transition temperatures, particularly with respect to the melting temperature. Furthermore, the increased steric bulk of the methyl group at the chiral centre is detrimen- tal to helix inversion.In addition, the long hydrocarbon substituent (R') would be expected to provide considerable rotational damping, which also leads to a reduction in liquid- crystalline properties because of the increased steric repulsive effects. Modelling studies have been carried out to determine the effect of the methyl substituent and R' on the torsional energy profile of 15 relative to that observed in 1 (Fig. 2). The initial calculations on 1 were performed using Microsoft EXCEL 4.0 interfaced via Dynamic Data Exchange (DDE) to the molecular modelling package HYPERCHEM 3, operating in the Microsoft Windows environment. These studies were extended to the more sophisticated package QUANTA 3.3 [Molecular Simulations Inc.(MSI)] running tines (conjugate gradients method) prior to the calculations. The loss of mesomorphism in 16 can be explained in terms of the unfavourable geometry about the C-3 chiral centre. Whereas the long hydrocarbon chain in 15 lies within a rotational cone of fairly small diameter relative to the meso- genic core, the diameter of the cone is much larger in 16. This is because the restricted geometry at C-3, imposed by the cyclic chiral moiety, results in lowest energy conformations where the long substituent lies well off-axis. This effective molecular broadening disrupts the packing of the molecules and destabilizes mesophase formation. The large degree of disorder resulting from the molecular broadening is manifested in the large cooling hysteresis effect observed on crystalliz- ation.Plate 1 shows equivalent lowest energy conformers of 1, 15 and 16 in order to emphasize the broadening effect. EfSect of Structural Changes in the Non-chiral Tev-mind Chain Compounds 18 and 19, which possess ethoxyethyl terminal substituents, were investigated in order to determine if the position of the oxygen atom in the non-chiral chain had an effect on helix inversion. It has been demonstrated previously* that increasing the length of the terminal hydrocarbon sub- stituent from C,H, to C5H,, and C7H,, while retaining an oxymethylene link to the aromatic core destabilizes inversion phenomena. This may be attributable to an increase in rotational damping brought about by the increased chain length or to a lengthening of the 'zig-zag' shape.Compounds 18 and 19 both show orthogonal smectic A* (SA*)and N* phases but no Sc* phases. Compound 19 shows an additional blue phase I (BPI). The helical twist sense in the N* phases was investigated in both materials, however no inversions were observed. The full phase sequences are shown below: F H4Yx50 0m-L\ / \ / \ / , H 18 K 101.5 S,* 123.2 Ng 152.4 Is0 C (28.0) (0.43) (0.66) (J g I) F H 19 K 46 SA*85.3N; 87.1 BPI 87.2 Is0 C (52.0) (2.3) (0.18) (-) (J g-') The orthogonal SA* phase is characterized by its typical focal conic texture and high degree of fluidity. The change in position of the oxygen link therefore suppresses helix inver- sions and tilted smectic phases while stabilizing the ortho- gonal SA*phase. The melting points are increased significantly relative to the oxymethylene-linked analogues, while the clear- ing temperatures are reduced.Overall, the introduction of an oxygen in the 3-position relative to the core serves to destabil- ize mesophase formation relative to the oxymethylene deriva- tives. This is possibly due to an odd-even effect relating to the position of the oxygen relative to the core and it is therefore conceivable that an oxytrimethylene spacer would stabilize tilted mesophase formation and inversions in helical properties. J. MATER. CHEM.. 1994. VOL. 4 0 36 108 180 252 324 0 36 108 180 252 324 torsion angle/degrees Fig. 2 Calculated relative potential energies as a function of torsion angle [O-C, -C,-C,] for 1 and 15 Plate 1 Lowest-energy conformers (QUANTAICHARMm) for 1, 15 and 16 Efect of Structural Changes in the Mesogenic Core The possibility that the lateral aromatic fluoro-substituent aids the inversion of helical twist sense and chirality-dependent properties is of major interest. We have looked at the effect of changing the substitution pattern whilst keeping the overall structure as similar as possible.In this respect 17, a structural isomer of 1, and 20, which does not possess lateral fluorine atoms, were prepared. For 17, the fluorine atom was simply omitted from the 2'-position in the terphenyl core but a fluorine atom was included at the 2-position on the phenyl ring carrying the chiral substituent.Only the chiral products derived from (2S,3S)-3-propyloxiranemethanolwere synthe- sized as this chiral moiety had given inversion properties in 1. In this way the overall molecular dipole would be shifted while still retaining an approximately equal dihedral angle between the first and second phenyl rings, relative to the chiral substituent. The dihedral angles in question have been calculated as 42" for 1 and 43" for 17 (Fig. 3) using QUANTA/ CHARMm. All structures were minimized using CHARMm routines (conjugate gradient methods) prior to the calculations. Compound 17 shows an inversion in the helical twist sense of the N* phase but not in the Sc* phase.The full phase sequence is given below: H 17 K 53 K' 66.5 Sq 97.5 NR104.5NZ 106.0Nt150.21~0 "C (26.1) (37.1) (0.7) (-) (-) (1.6) (J g-') Optical microscopy shows that in the upper temperature N* phase the pitch increases as the sample is cooled. At 106.0"C the pitch diverges suddenly to infinity. This is charac- terized by a change in the optical texture from a characteristic cholesteric Grandjean plane texture in the N*L phase to a schlieren texture containing two- and four-brushed singularit- ies (i.e. a normal, non-helical nematic phase) in the N*, phase. On further cooling, a Grandjean plane texture reforms. The helical twist sense in the mesophases of 17 was determined optically relative to fixed, crossed polarizers.The upper tem- perature cholesteric phase was determined to have a left-handed helix as it caused a dextro (D) rotation of plane polarized light. However, as the chiral group is composed of two sequential chiral centres within a three-membered hetero- cycle, it is not possible to relate the Gray-McDonnell rules17 to this material. In the lower temperature cholesteric phase a laevo (L) rotation of light was observed, thereby confirming the helix to be right-handed. No inversion of helical twist sense was observed in the Sc* phase, the phase causing D-F += 42" 1 #= 43" 17 Fig. 3 Calculated torsion angles (QUANTA/CHARMm) for 1 and 17 J. MATER. CHEM., 1994, VOL. 4 rotation of plane polarized light throughout the complete mesophase range, thereby proving the helix to be left-handed. In order to further investigate the Sc* phase, we determined the temperature dependence of 6, and P, of 17.When all lateral fluorines were removed. to give 20, a dramatic increase in the melting point was observed. Whereas the monofluoro derivatives 1 and 17 melted at 59 and 66.5 "C, respectively, 20 did not melt until 213 "C, giving a chiral, orthogonal E* phase which was characterized by its optical texture. A transition to an orthogonal SA*phase was observed at 231 "C, which persisted to 257 "C when clearing was observed. H 20 K 210 E* 227.8 SA*257.3 Is0 "C (22.7) (26.7) (-) (kJ rnol-' Electro-optic Studies on Compound 17 Temperature Dependence of the Apparent Opticrrl Tilt Angle The apparent optical tilt angle was measured using a field reversal method.A positive dc field of 3.2 V pm -' was applied across the electrode area, which had a cell spacing of 4.8 pm, and the crossed polarizers were rotated until extinction was achieved. Homogeneous planar alignment was achieved by overlaying a 1 kHz, 30 V (peak-to-peak) ac signal on the dc field to cause molecular fribulation. Once alignment was satisfactory, and prior to the switching studies, the ac signal was removed. The angle on the Vernier scale of the microscope stage was noted and the polarity of the applied voltage was reversed to give the transmission state. The microscope stage was then rotated to regain extinction and the angle on the Vernier scale was again noted.The difference between the two readings corresponds to the angle 28,. The results of these studies are shown graphically in Fig. 4. Compound 1 has 8, approaching 0' close to the S,*-N* transition, which increases with decreasing temperature to a maximum of 22.5". On further cooling, 8, falls to 0" and then increases in a negative sense to a maximum of -22.5" prior to crystallization. In contrast, 17 has 6, approaching 23" just below the Sc*-N* transition and this decreases almost linearly when the temperature is reduced. An increase in 8, is observed prior to crystallization in a 1.4 pm cell, however, this could be attributed to pre-transitional or surface latching effects. It should be noted that the transition temperatures in the 4.8 pm cell are approximately 3 "C lower (e.g.Sc* 95.2 N*) than as a thick film on an untreated glass slide. This is an artifact of the reduced cell thickness and interactions with the polyimide surface coating and is particularly evident in a 1.4 pm cell in ** 0. t -I 1OL 45 55 65 75 a5 95 TI'C Fig. 4 Temperature dependence of the apparent optical tilt angle in 17 J. MATER. CHEM., 1994, VOL. 4 ..*** *.*. ** ** . 0 40 50 60 70 80 90 100 TIT Fig. 5 Temperature dependence of the spontaneous polarization in 17 which the transition is observed at 94.1 "C. A similar linear decrease in 8, has been observed in the Sc* phase of another material,' which exhibits an inversion of helical twist sense in the N* phase.Temperature Dependence of the Spontaneous Polarization The spontaneous polarisation of 17 was determined using an applied ac voltage of 68 V (14.2 V pm-') peak-to-peak at a frequency of 100 Hz in a 4.8 pm cell from a temperature just below the Sc,*-N*Rphase transition. The results are shown graphically in Fig. 5. In contrast to 1, the maximum value of P,observed is almost double, peaking at 31.2 nC cm-2. The direction of the spontaneous polarization was determined according to reported procedures." A dc voltage of -20 V was applied, making the top plate of the cell the cathode. The polarizers were rotated to give extinction. The applied field was reversed to make the top plate of the cell the anode and the direction in which the optical stage had to be rotated to regain extinction gave the polarization direction.The ScT phase has a negative polarization, Ps(-), in contrast with 1 which shows the opposite polarization direction in the S,,* phase, formed likewise from the cholesteric phase. The optical tilt angle was found to vary almost linearly with the spontaneous polarization throughout the temperature range of the study. This is in good agreement with theory" and shows the material to be behaving normally. It would appear from the trend in the curve of spontaneous polarization uersus temperature that the chirality-dependent properties in the S,* phase are beginning to diverge towards inversion and that the inversion phenomena are not observed simply because the material crystallizes.This would suggest that if crystalliz- ation could be suppressed then inversion would occur. That is to say, there is a 'virtual' inversion point. Discussion A number of structural modifications have been made to the basic terphenyl structure shown in Fig. 1. The effect of these changes on the chirality-dependent properties of the resulting materials has been investigated. It has been shown previously' that inversion phenomena are sensitive to the length of the terminal non-chiral hydrocarbon chain (R). We have extended this by demonstrating that the position of the oxygen link in the five-atom non-chiral substituent, relative to the mesogenic core, is also important in the stabilization of inversion phen- omena.While inversions are observed in materials of the same general structure when n= 1, they are absent when n=2. We intend to extend this study further by examining selected examples where n =0 or 3, while maintaining the total number of atoms in the chain at five. Compounds with n=2 possess stable orthogonal SA* and cholesteric phases but Sc* phases are destabilized and the melting point is increased significantly; typically by 50 "C relative to materials with n =1. The nature of the substituents around the chiral centre was also found to be very important. Inversion phenomena were only observed in compounds where R" =H and R' is ;i simple unsubsituted hydrocarbon. When a long, substituted hydro- carbon is introduced as R' and a methyl group as R", the mesophase stability is considerably decreased but resiilts in a material exhibiting a room temperature Sc* phase If the substituents R' and R" are inverted however, all mesomorph- ism is lost; this has been attributed to the unfavour:ible off-axis arrangement of the long hydrocarbon chain.which destabilizes the packing of the molecules and consthquently mesophase formation. For 15 the broadening effect i < princi-pally due to the methyl substituent, which serves to destroy a degree of the molecular packing, resulting in a decrease in the melting point. However, in this isomer there is only one high-energy conformation (see Fig. 2) with a torsion mgle for C-C, -C2 -C, approaching 0".This gives an insurmount- able energy barrier to rotation so that the available confor- mations are confined within a potential energy valley. Fig. 6 shows three conformers that are contained within that valley and would be expected to be readily accessible over the temperature regime of the experiment. In terms of mcsophase formation, conformer (a) is the most favoured as the terminal chain lies along the molecular long axis. However, the other two conformers, (b) and (c), are not conducive to mcsophase formation even though they are low-energy structure <.Hence, conversion between such conformers would not be expected to give inversion of liquid-crystalline properties but would lead to destabilization of the mesophase, as was Observed.The most interesting effects are observed when the fluorine- substitution pattern in the aromatic core is chan;ed. The torsion angles between the first and second rings rrdative to the chiral centre (see Fig. 3) in 1 and 17 have been cdlculated and have been found to be essentially the same. The torsion angle has the same relative sign in both cases and the only difference between the two materials is the direction of the overall molecular dipole. The properties observed fix 17 are in many ways similar to those of 1 although a nJmber of significant differences have been observed. In particular, the loss of the inversion phenomena in the Sc* phase. However, it is proposed that this is related to the increa:,e in the crystallization temperature of 17 relative to 1 and it would appear that the trend is towards inversion, if the spcmtaneous polarization is extrapolated to lower temperature. rherefore one might consider that 17 has a 'virtual' inversion point at ca.35 "C. It has been noted that the direction of spontaneous polarization in the Sc* phase of 17 is opposite to triat in the higher temperature S,: phase of 1. This can be explained in terms of the shift in molecular dipole caused by changing the fluorination position from 2'-(ring two) in 1 to 3-(ring one) in 17. This has been demonstrated in computer simulations using the QUANTA/CHARMm molecular model1 ing pack- age. The inversion of helical twist sense in the csholesteric phase of 17 is identical in nature to that observed in 1, with a slight decrease in the transition temperatures.Fig. 7 shows a schematic representation of the overall molecul-ir dipoles calculated for 1, 17 and 20. By using conventional notation to predict the direction of the spontaneous polarization of 1 and 17, relative to the molecular shape and dipole direction, we predict that 1 should have a positive spontaneous polariz- ation, P,(+j, whereas 17 should have a negative value, Ps(-), at the same temperature. We have found from our studies that this is the case and we are now looking into prediction of spontaneous polarization in other systems using modelling techniques. Compound 20 shows orthogonal, non-helical E* and SA* phases at vastly increased temperatures relative to 1 and 17, and substantial decomposition is observed.This shows the importance of lateral fluoro-substitution on the formation of J. MATER. CHEM., 1994, VOL. 4 R OAr predict Ps(+) R on-axis: pro-mesogenic 0 F H R R highly off-axis: disfavoured H F H R R highly off-axis: disfavoured Fig. 6 Examination of the structures of the three lowest-energy conformers of compound 15.For explanation see text. compound side elevation front elevation 1 /flcalcS = 2.15 D predict f,(+) 17 kal$= 1.61 D predict fs(-) peal: = 2.15 D predict fs(-) Fig. 7 Calculated molecular dipoles (using CINDO calculations in QUANTAICHARMm) and their relationship to the spontaneous polarization in the S,, phase for 1, 17 and 20 thermally stable liquid-crystal phases.Whereas lateral fluor- ination has been shown to reduce dramatically the melting point relative to a non-fluorinated system, the melting point of 1 is more than 150"C less than that of 20. Further studies are being carried out to investigate the effects of lateral fluorination and the direction of the overall dipole on the occurrence of inversion phenomena. Conclusions A number of structural changes have been related to the occurrence of inversion phenomena in chiral liquid-crystal systems. We have shown that the occurrence of inversion phenomena are very sensitive to small changes in the molecu- lar structure and that subtle changes can often produce profound effects. We are grateful to Thorn EM1 Central Research Laboratories and Bell Northern Research (Europe) for funding the lec- tureship to P.S.and the Erasmus Scheme for support in the form of a studentship to J.D.V. We also thank Mr. R. Knight, Mr. A. D. Roberts and Mrs. B. Worthington for the spectro- scopic analysis of the materials, and Mrs. J. Haley for assist- ance in the electro-optical studies. References 1 Ph. Martinot-Lagarde, R. Duke and D. Durand, Mol. Cryst. Liq. Cryst., 1981,75,249. J. MATER. CHEM., 1994, VOL. 4 1375 2 3 4 L. Komitov, S. T. Lagerwall, B. Stebler, G. Andersson and K. Flatischler, Ferroelectrics, 1991, 114, 167. (a)J. S. Patel and J. W. Goodby, Philos. Mug. Lett., 1987,55, 283; (b)J. S. Patel and J. W. Goodby, J. Phys. Chern., 1987,91, 5838.H. Stegemeyer, K. Siemensmeyer, W. Sucrow and L. Appel, 2.Nuturforsck, A. Phys. Sci., 1989,44A, 1127. 13 14 Cryst. Liq. Cryst., 1988,158,3; M. A. Osipov and S. A. Pibin, Mol. Cryst. Liq. Cryst., 1983, 103, 57. H. Diamant, K. Drenck and R. Pepinsky, Rec. Sci. Instrurn.,1957, 28, 30. The chiral oxiranes were prepared according to the procedure described in: J. G. Hill, K. B. Sharples, C. M. Exon and 5 L. Komitov, K. Flatischler, G. Andersson, S. T. Lagerwall and R. Regenye, Org. Synth., 1984,63,66. B. Stebler, Ferroelectrics, 1991, 114, 151. 15 S. Gronowitz, A.-B. Hornfeldt and Y.-H. Yang, Chem. Scr., 1986, 6 I. Dierking, F. GeiDelmann, P. Zugenmaier, W. Kuczynski, 26, 311. 7 S. T. Lagerwall and B. Stebler, Liq. Cryst., 1993,13,45. A. J. Slaney, I. Nishiyama, P. Styring and J. W. Goodby, J. Muter. 16 J. W. Goodby, P. Styring, A. J. Slaney, J. D. Vuijk, J. ‘5. Patel, C. Loubser and P. L. Wessels, Ferroelectrics, 1994, 147-9 291. 8 Chem., 1992,2, 805. P. Styring, J. D. Vuijk, I. Nishiyama, A. J. Slaney and 17 G. W. Gray and D. G. McDonnell, Mol. Cryst. Liq. Cry t., 1977, 34, 21 1. 9 10 11 J. W. Goodby, J. Muter. Chem., 1993,3, 399. C. Loubser, P. L. Wessels, P. Styring and J. W. Goodby, J. Muter. Chem., 1994, 4, 71. J. W. Goodby, A. J. Slaney, C. J. Booth, I. Nishiyama, J. D. Vuijk, P. Styring and K. J. Toyne, Mol. Cryst. Liq. Cryst., 1994,243,231. S. Saito, K. Murashiro, M. Kikuchi, T. Inukai, D. Demus, 18 19 J. W. Goodby, E. Chin, T. M. Leslie, J. M. Geary and J. 4.Patel, J. Am. Chem. SOC., 1986,108,4729. J. W. Goodby, R. Blinc, N. A. Clark, S. T. Lagerwal,, M. A. Osipov, S. A. Pikin, T. Sakura, K. Yoshino and 13. Zeks, Ferroelectric Liquid Crystals, Gordon and Breach, Philadelphia, PA. 1991. M. Neundorf and S. Diele, Ferroelectrics, submitted for 12 publication. L. A. Beresnev, L. M. Blinov, M. A. Osipov and S. A. Pikin, Mol. Paper 4/02014F; Receiued 5th April, 1994