J O U R N A L O F C H E M I S T R Y Materials The synthesis and mesomorphism of di-, tetra- and hexa-catenar liquid crystals based on 2,2¾-bipyridine Kathryn E. Rowe and Duncan W. Bruce* Department of Chemistry, University of Exeter, Stocker Road, Exeter, UK EX4 4QD 2,2¾-Bipyridines are known to coordinate to a wide variety of metal centres. In this paper, liquid-crystalline two-chained (dicatenar), four-chained (tetracatenar) and six-chained (hexacatenar) bipyridines are synthesised and their mesomorphism is described.For the tetracatenar bipyridines, a full homologous series, from tetramethoxy to tetratetradecyloxy, was synthesised, and the phase diagram showed a classic progression from nematic and smectic C phases at short chain length, through a cubic phase to a columnar phase. 2,2¾-Bipyridines are some of the most versatile ligands used in synthesise several more examples and in particular, we examined tetracatenar (four-chain) and hexacatenar (six-chain) the construction of metal coordination complexes,1 having been used to make dendrimers,2 helical structures,3 photoactive derivatives. The syntheses and liquid crystal properties of these new bipyridines are now described.systems,4 and a wide range of other supramolecular structures.5 Their attractiveness comes form their synthetic versatility and from their ability to coordinate to a very wide variety of Synthesis of the bipyridines metal centres. In recent publications, we have reported that by suitable Four diVerent types of bipyridine were synthesised, shown in choice of anisotropic ligand, it is possible to form rod-like Scheme 1, and their synthesis is now described.Reaction of liquid crystals based on metals with octahedral stereochemis- methyl 4-hydroxybenzoate, or ethyl 3,4-dihydroxy- or 3,4,5- try.6 The approach is based on the assumption that such a trihydroxybenzoate with bromoalkane under basic conditions high coordination number metal centre perturbs the anisotropy led to the related mono-, di- or tri-alkoxybenzoic acid, after of the ligand to which it binds and leads to complexes with alkaline hydrolysis and an acidic workup.For the mono- and low aspect ratios. Consequently, it is necessary to use rather di-alkylations, butanone was used as the solvent, while pentahighly anisotropic ligands to screen out the perturbing eVects none was preferred for the trialkylation reaction.These acids of the metal complex. We have demonstrated that this is were then esterified with hydroquinone which was monopropossible with imines orthometallated to MnI 7 and ReI,8 and tected with either tetrahydropyran (THP) or with a benzyl with diazabutadienes coordinated to ReI.9 It is, however, of group.Benzyl protection was initially used and was subinterest to note that low aspect ratios appear not to be so sequently readily removed by hydrogenolysis. However, hydromuch of a disadvantage in the design of columnar mesogens, genolysis of the di- and tri-alkoxy esters was painfully slow, as evidenced by the synthesis by Swager10 of mesomorphic which led us to change to THP-protection, where the deproteccomplexes based on octahedral metal centres bound to poly- tion using oxalic acid in methanol at reflux gave essentially substituted b-diketones.quantitative yields. However, we subsequently found that Subsequently, we used the same approach in an eVort to addition of a small amount of triethylamine during the hydroincorporate bipyridines as ligands in metal-based liquid crys- genolysis of the benzyl-protected system led to complete deprotals. 11 Thus, we reported the synthesis and mesomorphism of tection within minutes for any of the systems we studied, diesters of 2,2¾-bipyridine-5,5¾-dicarboxylic acid,12 simul- indicating a clearly preferential route, especially as mono THPtaneously with reports of mesomorphic bipyridines based on protected hydroquinone is obtained in four steps from hydrounsymmetrically substituted 5,5¾-disubstituted bipyridines.13 quinone. After deprotection, the resulting phenol was then However, when complexed to a whole range of metal centres, esterified with 2,2¾-bipyridine-5,5¾-dicarbonyl dichloride.This the liquid crystallinity of the bipyridines was lost.14 Curious was the lowest-yielding step of the whole procedure and despite to understand the absence of mesomorphism in these systems, our best eVorts which included evaluation of other possible we reasoned that by comparison with our own work and with esterification methods, we could not obtain more than about that of Deschenaux with 1,3-disubstituted ferrocenes,15 it 30% yield for this reaction.However, we would recommend seemed necessary to have, in addition to the core of the making the diacid chloride just before it is required as under molecules, an additional four phenyl rings in order to realise these circumstances, yields were consistently higher.mesomorphic metal complexes. Thus, when complexed, the For the sake of later clarity, the bipyridines (Scheme 1) will 2,2¾-bipyridine unit represented part of the core and, according be abbreviated as follows.The two-chain (or dicatenar) bipyrito our idea, four extra rings were required. It was therefore dines, will be labelled Dn, where n denotes the number of necessary to synthesise a six-ring bipyridine to fulfil this design carbon atoms in the alkoxy chains. Similarly, the four-chain criterion, and subsequent complexation to ReI eventually led (tetracatenar) and six-chain (hexacatenar) bipyridines will be to a liquid crystalline complex.16 abbreviated as Tn and Hn, respectively.Having shown that two-chained, six-ring derivatives of 2,2¾- The two-chained bipyridines were found to be highly insolbipyridine could lead to mesomorphic complexes of ReI, we uble making purification diYcult, and while an analytically were keen to pursue the synthesis of further examples as pure sample was obtained for the octyloxy derivative, D8, the bipyridines have enormous potential for coordination to a other derivatives gave carbon values which were low by very wide range of metal centres.Therefore, we proceeded to approximately 1%, despite numerous recrystallisations; compound R8 behaved in a similar fashion.Further purification involved attempts to convert the dicatenar ligands to the * E-mail: d.bruce@exeter.ac.uk J. Mater. Chem., 1998, 8(2), 331–341 331(OH) OH (OH) RO2C (OCnH2 n+1) OCnH2 n+1 (OCnH2 n+1) HO2C (OCnH2 n+1) OCnH2 n+1 (OCnH2 n+1) O O HO N N O O O O (OCnH2 n+1) OCnH2 n+1 (OCnH2 n+1) O O O (C nH2 n+1O) CnH2 n+1O O (C nH2 n+1O) (OCnH2 n+1) OCnH2 n+1 (OCnH2 n+1) O O BzO (i) (v) (ii) Monoalkoxy Dialkoxy Trialkoxy Monoalkoxy Dialkoxy Trialkoxy 456 Monoalkoxy Dialkoxy Trialkoxy 9 10 11 D n T n H n Monohydroxy R = CH3 Di- or tri-hydroxy R = C2H5 (iii) (iv) Dicatenar Tetracatenar Hexacatenar 123 Scheme 1 Synthesis of the polycatenar bipyridines.Reagents and conditions: (i) CnH2n+1Br, KHCO3; (ii) KOH/EtOH; (iii) BzOH, DCC, DMAP; (iv) H2, Pd/C; (v) 2,2¾-bipyridine-5,5¾-dicarbonyl dichloride, toluene, Et3N. corresponding hydrochloride salt, in order to achieve higher lower phase (M1) being unidentified.However due to decomposolubility thus aiding purification. Unfortunately this was also sition occurring at these elevated temperatures it was not accompanied by decomposition, and so was unsuitable.possible to obtain a clear texture of the lower mesophase on Compound T1 was also very insoluble and we could not cooling making identification impossible. Furthermore, this obtain good analytical data. However, we are not unduly meant that only data from the first DSC cycle could be used. disturbed by this as neither of the first two homologues was The butoxy bipyridine, D4, showed only a nematic phase, mesomorphic.while the longer chain length derivatives, D8, D12 and D14, showed a smectic C phase before giving way to the nematic Synthesis of bipyridine, R8 phase at higher temperatures. The temperature of the SC–N transition was found to vary with heating rate, which is One example of a bipyridine with one of the ester groups attributed to decomposition occurring in the upper reaches of reversed was synthesised to examine the eVect on mesomorphthe SC phase and rapidly accelerating in the nematic phase.ism. The route is shown in Scheme 2. Thus, benzyl 4-hydroxy- Despite this, the SC phase range was found to increase with benzoate was first protected with 3,4-dihydro-2H-pyran under increasing chain length, as is expected with calamitic materials.acid-catalysed conditions at room temperature in ethyl acetate, On both the heating and cooling cycles of the DSC, D12 before the benzyl group was cleaved, in quantitative yield, with showed another mesophase (M2) between the crystal and SC hydrogen over a palladium-on-charcoal catalyst. This carphase. By microscopy, with careful cooling, this transition was boxylic acid 11 was reacted with 4-octyloxyphenol, using a observed, but persisted for only a degree or so before crystallis- standard DCC–DMAP esterification, and the THP group was ation occurred.Due to decomposition occurring in the nematic subsequently cleaved under acidic conditions, at reflux in phase, it was not possible to obtain a good optical texture for methanol.The resulting phenol 13 was reacted with 2,2¾- the SC phase, and therefore for the M2 phase, too. However, bipyridine-5,5-dicarbonyl dichloride at reflux in toluene conthe relative magnitude of both the entropy and enthalpy taining a few drops of triethylamine, to give the reversed ester, changes (obtained on cooling) leads to a tentative assignment R8. In common with the dicatenar bipyridines above, this as a crystal smectic phase.compound was found to be highly insoluble, again making As this additional phase first appeared in the dodecyloxy purification diYcult. derivative it was hoped that its phase range would increase with increasing alkoxy chain length. Hence, the tetradecyloxy Mesomorphism of the dicatenar bipyridines, Dn derivative, D14, was subsequently synthesised as a direct attempt to elucidate the nature of this mesophase.Five dicatenar 2,2¾-bipyridine ligands were synthesised with Unfortunately, the range of the phase was found to be equally n=1, 4, 8, 12 and 14; the thermal data for these compounds short and once more, a good optical texture could not be are collected in Table 1. The methoxy derivative was found to exhibit two mesophases, the upper one being nematic, but the obtained, leaving the mesophase unidentified. 332 J. Mater. Chem., 1998, 8(2), 331–341HO O OBz O O OBz O O O OH O O O OC8 H17 O O O O OC8 H17 HO N N O O O O O O O O OC8H17 C8H17O 10 11 12 13 R8 (i) (ii) (iii) (iv) (v) Scheme 2 The synthesis of the ‘reversed ester’ ligand. Reagents and conditions: (i) 3,4-dihydro-2H-pyran, HCl(g); (ii ) H2, Pd/C; (iii) 4- octyloxyphenol, DCC, DMAP; (iv) oxalic acid, MeOH; (v) 2,2¾-bipyridine-5,5¾-dicarbonyl dichloride, toluene, Et3N.Table 1 Thermal data for the dicatenar bipyridines conventional discoid molecules. Conventional disc shaped molecules stack on top of each other to form a column, these n Transitiona T °/C DH/kJ mol-1 DS/J K-1 mol-1 columns being packed in a two-dimensional array.However, in the polycatenar systems, between two and four molecules 1 Cr�M1 237 —b —b come together to form what is, in eVect, a disc-like repeat unit M1�N 264 —b —b representing a slice through the columns, which are then N�decomp. >264 — — themselves packed in a two-dimensional array. 4 Cr�N 269 45.8 84 Probably the most interesting of the polycatenar systems N�decomp. 350 — — are the tetracatenar compounds which can display lamellar, 8 Cr�SC 231 52.3 104 cubic and columnar mesophases. Generally, lamellar meso- SC�N 328c —b —b phases are observed at short chain lengths while columnar N�decomp. >328 — — mesophases are observed at longer chain lengths. At intermedi- 12 Cr�M2 210 2.1d 4d ate chain lengths, this competition can result in the formation M2�SC 212 35.5d 75d of a frustrated phase, namely the cubic phase. This polymor- SC�N 341c —b —b phism arises due to the fact that in the mesophase, segregation N�decomp.>341 — — of the aromatic and aliphatic parts of the molecule occurs, and 14 Cr�M2 209 0.5d 1d as such these molecules can be regarded as amphiphilic in M2�SC 211 49.0 102 nature.Indeed, their observed polymorphism is similar to that SC�N 355c 4.2 7 N�I 368c —b —b of lyotropic systems in the sense that two-dimensional oblique or rectangular columnar phases, or three-dimensional cubic aM1 and M2 are unidentified mesophases (see text). phases can be inserted in between lamellar and hexagonal bNot seen by DSC. columnar phases. Also, like lyotropic systems, this segregation cThese temperatures are sensitive to the thermal history of the sample of molecular parts results in curvature at the aromatic–ali- due to decomposition occuring in the SC phase. phatic interface.This helps to explain the correlation between dThe thermodynamic data for these transitions are taken from the cooling cycles on the DSC. mesophase type and the length of the aliphatic chain, for as the chain length is increased so the curvature is increased, resulting in the columnar mesophases being stabilised at the expense of the lamellar mesophases.Mesomorphism of the tetracatenar bipyridines, Tn Because of the potential for such rich mesomorphism in Polycatenar liquid crystals17,18 are those which contain, typi- these systems, we undertook the synthesis of a complete cally, three or more chains and a rather extended core.They homologous series from n=1–14. We believe that this is the are classified both by the number of terminal chains they first time that such an homologous series has been investigated possess (tri-, tetra-, penta- and hexa-catenar) and also by the for tetracatenar mesogens, and the results are presented as a way these are distributed on the terminal benzene rings.phase diagram in Fig. 1, while the thermal data are collected The hexa- and penta-catenar compounds are found to in Table 2. exhibit columnar phases, while the tricatenar compounds show This series of compounds exhibits a phase behaviour that lamellar and cubic mesophases. The columnar phases result is very typical for tetracatenar species, namely nematic from a strong curvature at the aromatic–aliphatic interface in and lamellar phases at short chain lengths, columnar at long a way similar to that observed in lyotropic liquid crystals.The chain lengths, with the changeover being accompanied by the structure of the columnar mesophase formed from polycatenar observation of a cubic phase at intermediate chain lengths.species has been shown by dilatometry studies and X-ray Until the pentyloxy derivative, T5, only nematic phases were observed at elevated temperatures, with decomposition measurements to be slightly diVerent from those formed by J. Mater. Chem., 1998, 8(2), 331–341 333occurring in the upper regions of these phases, beginning above 300 °C and accelerating rapidly above 350 °C.Thus, for the methoxy (T1) and ethoxy (T2) derivatives, no clearing point was observed. While the propoxy (T3) and butoxy (T4) derivatives did clear to the isotropic, this temperature was not found to be reproducible on subsequent runs as it was accompanied by extensive decomposition. Nevertheless, the clearing point decreased quite markedly as the chain length increased.In initial microscopy studies, it appeared that T2–T4 melted over a large temperature range. DSC data showed that there was a crystal–crystal transition immediately prior to melting into the nematic mesophase. Further microscopy allowed the melting temperature to be ascertained; however by DSC the first transition was almost complete before the compounds melted into the nematic phase; hence, the thermodynamic data associated with these melting points are only N I Colh Cub Sc Cr carbon chain length 0 2 4 6 8 10 12 14 350 300 250 200 150 T/°C approximate.Fig. 1 Phase diagram for the tetracatenar bipyridines (Cr=crystal; An additional phase, namely SC, was introduced at the N=Nematic, SC=smectic C; Cub=cubixy chain length and was readily characterised by its onal); D, Cr–N; 1, Cr–SC ; 2, Cr–Colh; $, SC–N; #, SC–Cub; *, optical texture.Thus, the aliphatic chain was now of suYcient Cub–N; +, Cub–Colh; %, N–I; &, Cub–I; x, Colh–I length to stabilise lamellar mesophases. The derivatives T6–T8 represent the change over from lamellar to hexagonal columnar Table 2 Thermal data for the tetracatenar bipyridines mesophases, and as is often typical in tetracatenar systems this is achieved via the cubic phase.Thus, the cubic phase (charac- n Transition T /°C DH/kJ mol-1 DS/J K-1 mol-1 terised by its optical anisotropy, by the slow formation of such 1 Cr�N 259 62.6 117 a texture via the characteristic appearance of square edges N�decomp. >300 — — growing across the preceding texture, its high viscosity and 2 Cr�Na 262 52.4 99 the accompanying appearance of misshapen air bubbles) first N�decomp.>300 —c —c appears in the hexyloxy derivative, between the SC and N phases, giving one of very few well-authenticated examples of 3 Cr�Na 263 55.7 104 N�Ib 345 —c — a cubic phase below a nematic.19 In the heptyloxy derivative, a monotropic columnar hexagonal phase was observed and 4 Cr�Na 242 55.6 109 by the octyloxy derivative we had a compound that exhibited N�Ib 325 —c —c enantiotropic lamellar, cubic and hexagonal columnar phases. 5 Cr�SC 221 61.8 126 We believe that this is also unique in tetracatenar systems. SC�N 229 3.2 6 From T9 onwards, only columnar mesomorphism was N�I 294 1.0 2 observed, with both the melting and clearing temperatures 6 Cr�SC 196 55.8 120 remaining remarkably constant, irrespective of chain length.SC�Cub 220 1.7 14 The optical textures of these hexagonal columnar phases was Cub�N 239 4.0 8 (Cub�SC) (203) —c —c highly characteristic, exhibiting beautiful focal conic monodo- (SC�Cub) (224) —c —c mains and areas of homeotropic orientation. These columnar (N�SC) (229) —c —c mesophases have been identified as disordered hexagonal N�I 272 0.5 1.0 columnar mesophases by X-ray diVraction studies which will 7 Cr�SC 184 59.2 130 be published in due course as part of a much larger structural SC�Cub 197 2.9 6 study of the phase diagram.20 Thus, spacings in the ratio 1, (Col�Cub) (228) (3.6) (7.0) Ó3, Ó4 were seen and, for example, T12 gave d001=43.8 A ° .(I�Col) (234.5) —c —c This phase diagram is absolutely ‘text-book’ for the behav- Cub�I 239 3.8 7 iour of tetracatenar mesogens and is, to our knowledge, the 8 Cr�SC 173 56.0 126 first time that a full homologous series of tetracatenar mesog- SC�Cub 188 2.4 5 ens, from methoxy upwards, has been synthesised. Thus, the Cub�Colhd 229 — — nematic and smectic C phases observed at short chain lengths Colhd�I 237 4.0 8 pass through an ‘intermediate’ cubic phase before giving way 9 Cr�Colhd 171 35.6 4 to a columnar phase. By contrast, in a related piece of work Colhd�I 238 3.9 8 where we complexed two dicatenar alkoxystilbazoles in a trans 10 Cr�Colhd 166 45.2 103 fashion across a PtCl2 centre, the mesomorphism changed Colhd�I 237 4.3 9 suddenly from smectic C to columnar on passing from the 11 Cr�Colhd 165 48.2 110 dodecyloxy derivative (SC only) to the tridecyloxy derivatice Colhd�I 236 5.31 11 (columnar only), without any sign of a cubic phase and without 12 Cr�Colhd 161 48.6 112 any of the homologues showing both lamellar and columnar Colhd�I 234 5.8 11 phases.21 The sensitivity of polycatenar mesogens to their core 13 Cr�Colhd 161 47.7 110 structure has been commented on previously.22 Colhd�I 229 6.6 12 Close examination of the phase diagram reveals some interesting possibilities.For example, there is a direct nematic–cubic 14 Cr�Colhd 161 48.9 113 Colhd�I 230 6.5 13 transition which ought to allow for the production of a cubic monodomain from an aligned nematic, leading to unequivocal aThe DSC data are only approximate due to an incomplete crystal– assignment of the symmetry of the cubic phase, as reported crystal transition immediately prior to melting.previously by us for mesomorphic complexes of silver(I ).23 bThese temperatures are sensitive to the thermal history of the sample Furthermore, we have in T8, a compound with the enanti- due to decomposition occurring in the upper regions of the otropic phase sequence Colh =Cub=SC, previously found nematic phase.cNot seen by DSC. only monotropically and not before in symmetric, tetracatenar 334 J. Mater. Chem., 1998, 8(2), 331–341systems.20We have recently carried out a detailed investigation Thus, while on cooling from the nematic phase the cubic phase is thermodynamically preferred, the nematic can supercool due of polycatenar complexes of silver(I )24 using X-ray scattering, freeze–fracture electron microscopy,25 and dilatometry, in to these slow kinetics, allowing the appearance of the SC phase.As this phase is thermodynamically unstable with respect to which we confirmed an epitaxial relationship between the columnar and cubic phase, and have proposed a model for the the cubic, then the latter eventually appears, once more giving way to the (now thermodynamically stable) SC phase on columnar-to-cubic transition.26 We now have the possibility of extending this approach to look at the epitaxy of the columnar- further cooling. A similar situation exists with T7, where the heating and to-cubic-to-smectic C transitions in the same material, which we hope will give further information on transitions to cubic cooling phase sequences are as shown below: phases. These studies are already underway and will be Col�Cub�I Heating reported in due course.The mesomorphism of the hexyloxy and heptyloxy deriva- I�Col�Cub�Col Cooling tives were found to be particularly interesting and are now Very similar arguments can be invoked here, with a thermodiscussed in some detail.Thus, on heating, T6 showed the dynamically unstable columnar phase appearing on cooling following phase sequence: the isotropic due to the slow kinetics of cubic phase formation. Cr�SC�Cub�N�I Mesomorphism of the hexacatenar 2,2¾-bipyridines However, on cooling the nematic phase gives way to a SC phase rather than a cubic phase. A cubic phase then grows As the mesomorphism displayed with the hexacatenar 2,2¾- into the SC phase and persists until the SC phase reappears bipyridine systems was found to be predominantly columnar, once more.Thus, the phase sequence on cooling is: only four ligands were synthesised with n=1, 4, 8 and 12. Thermal data are collected in Table 3. I�N�SC�Cub�SC�Cr The mesomorphism observed in these compounds is typical We were initially perplexed by this behaviour, but following of hexacatenar systems, namely columnar mesophases, with discussions with Dr Antoine Skoulios of the IPCMS in the exception of the methoxy derivative which exhibits a rather Strasbourg, we feel we can oVer an explanation.The behaviour diVerent phase sequence. Thus, the butoxy derivative was is best explained by considering a schematic free energy found to show a columnar phase.The cooling cycle of the diagram for the system as shown in Fig. 2. DSC only gave the isotropic–columnar transition, but the Thus, on increasing the temperature and minimising G, crystal–columnar transition was reproducibly obtained on transitions would be expected (and are observed) from subsequent heating cycles, following a cold crystallisation SC�Cub�N�I.On cooling, the reverse would clearly be exotherm. The octyloxy and dodecyloxy derivatives exhibited expected, but the behaviour is modified due to the very slow similar mesomorphism. In both derivatives, the first heating kinetics generally found for the formation of cubic phases. cycle on the DSC gave two transitions prior to clearing into the isotropic.On cooling, neither sample crystallised, and consequently on subsequent heating cycles, the melting transition was no longer obsero crystallise, with the texture that first appeared from the isotropic remaining until solidification.As a result, the mesophase–mesophase transition (reproducible obtained by DSC), was not initially observed by microscopy. On careful reexamination of the dodecyloxy derivative by microscopy, it was possible on the first heat to see a subtle textural change in the highly birefringent sample, from a smooth to a finely grained, less focused surface. However, the textures observed by the first heat were not suYcient for characterisation.The natural texture of the upper phase observed on cooling was characteristic of a columnar phase, and the small change in enthalpy required in the mesophase–mesophase transition combined with the lack of any observed textural change on cooling, suggests that the lower phase is also columnar. X-Ray Table 3 Thermal data for the hexacatenar bipyridines n Transition T /°C DH/kJ mol-1 DS/J K-1 mol-1 1 Cr�Cub 203 —a —a Cub�SA 211 —a —a SA�N 217 —a —a N�I 292 —a —a 4 Cr�Col 143 26.7 64 Col�I 176 3.5 8 8 Cr�Colb 85 45.2 126 Col�Col¾ 157 0.2 1 Col�I 165 6.4 15 12 Cr�Colb 54 38.6 114 Col�Col¾ 142 0.3 1 Col¾�I 151 5.7 14 Fig. 2 Schematic representation to show the thermodynamic relationship between the mesophases in compounds T6 (a) and T7 (b) aNot seen by DSC.bThis transition is only observed on the first heat. J. Mater. Chem., 1998, 8(2), 331–341 335studies are now in progress in order to ascertain the exact opposing each other, therefore they are now serving to stabilise lamellar interactions. nature of these mesophases (and the other analogues in this series). The methoxy derivative was remarkable in showing the phase sequence: Conclusions Cr�Cub�SA�N�I A number of di-, tetra- and hexa-catenar six-ring 2,2¾-bipyri- While transitions from both SA to cubic and nematic to cubic dines have been synthesised.These were found to be novel are known, we are not aware of the I�N�SA�Cub phase liquid-crystalline materials that exhibited a rich polymorphism. sequence having been observed before.All of the phases were Thus, the dicatenar compounds were found to exhibit readily assigned on the basis of optical microscopy, although nematic and SC phases, with the SC phase being stabilised at bizarrely, no thermal changes could be observed by DSC. We the expense of the nematic phase with increasing chain length. have come across this problem before in other studies and This was the only lamellar phase observed, leading to the have usually found that by changing the experimental method observation reported previously15 that the 2,2¾-bipyridine core slightly, it was possible to obtain reproducible data.However, seems to strongly promote SC phase formation in these types in this case it was to no avail. of compounds. This observation was also borne out when the direction of the terminal ester functionality was reversed, as only a SC phase was observed. Mesomorphism of ligand, R8 The tetracatenar system, in particular displayed the most Thermal data for R8 and D8 are collected in Table 4.It is well interesting mesomorphism as these compounds were found to known that the direction of an ester functionality can have a exhibit nematic and lamellar phases at short chain lengths, profound eVect on the observed mesomorphism.27 The six-ring and hexagonal columnar phases at long chain lengths, with a 2,2¾-bipyridines described so far have their ester groups cubic phase appearing at intermediate chain lengths at the arranged so that their dipoles oppose one another (see Fig. 3), transferral point between the lamellar and columnar leading to net lateral dipoles which can be regarded as being mesophases.mutually repulsive and therefore, promoting nematic phases, Finally, the mesomorphism of hexacatenar species synas observed. However, if the outer ester groups are reversed, thesised was found to be essentially columnar, with the excepthen the dipoles would be arranged as a kind of ‘outboard’ tion of the methoxy derivative which behaved as a calamitic dipole which should promote smectic phases.Similar eVects mesogen. are seen on the introduction of fluoro substituents into mesomorphic systems.28 Reversal of the terminal ester group was found to promote SC phase formation by destabilising both the crystal and the Experimental nematic phase. Nematic phase formation was suppressed to Elemental analysis were determined by the University of the extent that it was no longer observed and the melting SheYeld Microanalysis Service.Mass spectra were recorded point was lowered by some 34 °C. This is unsurprising as the using the Fast Atom Bombardment technique (FAB), at the dipole moments of the ester functionalities are now no longer University of SheYeld.Infrared spectra were measured using a Nicolet MAGNA 550 FTIR infrared spectrometer; UV–VIS spectroscopy was carried out using a ATI Unicam UV4 Table 4 Thermal data for D8 and the ‘reversed’ ester, R8 machine. NMR spectra were recorded on either a Bruker Comp. Transition T /°C DH/kJ mol-1 DS/J K-1 mol-1 ACF-300 or a Bruker DRX-400 spectrometer, where the chemical shifts are reported relative to the internal standard R8 Cr�SC 197 37.3 80 of the deuterated solvent used.J values are in Hz. Analysis by SC�decomp. >280 — — DSC was carried out on a Perkin-Elmer DSC7 instrument D8 Cr�SC 231 52.3 104 using heating and cooling rates of either 5 or 10 K min-1. SC�Na 328 — — Analysis by hot stage microscopy was carried out using a Zeiss N�decomp.a >328 — — Labpol, or Olympus BH40 microscope equipped with a Link- Am HFS91 hot stage, TMS92 controller and LNP2 colling aThese temperatures are sensitive to the thermal history of the sample due to decomposition occurring in the SC phase.unit. Silica gel particle size was 40–63 mm. N N O O O O OCnH2 n+1 O O O O CnH2 n+1O 2 3 4 7 8 9 10 5 6 11 12 13 14 16 15 D8 N N O O O O O OC8 H17 O C8H17O O O 12 13 14 15 16 3 4 9 10 11 8 6 7 5 R8 Fig. 3 Diagram to show the relative dispositions of dipoles in the dicatenar compounds and the related ‘reversed ester’ 336 J. Mater. Chem., 1998, 8(2), 331–341Synthesis of the 3,4-dialkoxybenzoic acids All derivatives were prepared similarly and one example is given. All other derivatives were obtained in yields ranging from 80–95%.OCH2CH2(CH2)5CH3 O O O 1 3 4 5 6 7 8 9 I0 2 11 12 14 13 15 16 22 4c 4-(4-Octyloxybenzoyloxy)-1-benzyloxybenzene 4c. 4-Octyloxybenzoic acid (5 g, 0.02 mol), 4-benzyloxyphenol (4 g 0.02 mol), and dicyclohexylcarbodiimide (4.1 g, 0.02 mol), were dissolved OCH2CH2(CH2)5CH3 OCH2CH2(CH2)5CH3 HO2C 1 2 3 4 7b 8b 5 6 7a 8a 14a 14b 2h in dichloromethane. To this N,N-dimethylaminopyridine (0.25 g 0.002 mol) was added and the reaction was stirred at 3,4-Bis(octyloxy)benzoic acid 2h.Ethyl-3,4-dihydroxy- room temp. for 24 h. The colourless precipitate was removed benzoate (5 g, 0.027 mol), potassium carbonate (15.15 g, by filtration and the solvent was evaporated. Crystallisation 0.1 mol) and 1-bromooctane (10.6 g, 0.055 mol) were placed in from ethanol (2×) gave the product as a colourless solid.butanone (150 cm3), and the reaction heated at reflux for 84 h. Yield: 6.8 g (80%); dH (CDCl3): 8.05 (2H, AA¾XX¾, H12, JAA¾XX¾ Water (100 cm3) was added and the aqueous phase extracted 9), 7.35 (5H, m, H1–3), 7.05 (2H, AA¾XX¾, H8, JAA¾XX¾ 9), 7.05 against dichloromethane (3×100 cm3). The organic extracts (2H, AA¾XX¾, H13, JAA¾XX¾ 9), 6.87 (2H, AA¾XX¾, H7, JAA¾XX¾ were combined, dried over MgSO4, filtered and evaporated to 9), 5.00 (2H, s, H5), 3.95 (2H, t, H15, 3JHH 6.5), 1.75 (2H, qt, give a brown solid.A solution of potassium hydroxide (3.07 g, H16), 1.40 (2H, m, H17), 1.25 (8H, m, H18–21), 0.80 (3H, t, H22); 0.055 mol) in ethanol (95%, 100 cm3) was added and the dC (CDCl3): 165.3(C10), 163.5(C14), 156.4(C6), 144.8(C9), reaction heated at9(C4), 132.3(C12), 128.6, 128.0, 127.5(C1–3), 122.6(C8), and the solution acidified with HCl (conc., 10 cm3). The 121.6(C11), 115.5(C7), 114.3(C13), 70.5(C5), 68.3(C15), 31.8, resulting colourless precipitate was collected and crystallised 29.4, 29.3, 29.1, 26.0, 22.7(C16–21), 14.1(C22). twice from ethanol to give 3,4-bis(octyloxy)benzoic acid as a colourless solid.Yield: 9.12 g (88%); dH (CDCl3): 7.72 (1H, dd, Synthesis of the 4-(3,4-dialkoxybenzoyloxy)-1-benzyloxybenzenes H6, 3JHH 8.5, 4JHH 2), 7.60 (1H, d, H2, 4JHH 2), 6.89 (1H, d, H5, 3JHH 8.5), 4.07 and 4.05 (4H, t, H7a and b), 1.85 (4H, qt, H8), All derivatives were prepared similarly and one example is 1.30 (20H, m, H8–13), 0.90 (6H, t, H14) given.All other derivatives were obtained in yields ranging from 58% to quantitative. Synthesis of the 3,4,5-trialkoxybenzoic acids All derivatives were prepared similarly and one example is given. All other derivatives were obtained in yields ranging from 55–70%. OCH2CH2(CH2)5CH3 OCH2CH2(CH2)5CH3 O O O 6 7 8 9 10 11 12 13 14 15 16 17b 18b 24b 1 2 3 4 5 17a 18a 24a 5h 4-[3,4-Bis (octyloxy)benzoyloxy]-1-benzyloxybenzene 5h.This was prepared from 3,4-bis(octyloxy)benzoic acid using OCH2CH2(CH2)5CH3 OCH2CH2(CH2)5CH3 OCH2CH2(CH2)5CH3 HO2C 1 2 3 4 5b 6b 12b 5a 6a 12a 3c the procedure given for 4c. This gave the product as a colourless solid. Yield: 3 g (80%); dH (CDCl3): 7.82 (1H, dd, 3,4,5-Tris(octyloxy)benzoic acid 3c. Methyl 3,4,5-trihydroxy- H16, 3JHH 8.5, 4JHH 2), 7.68 (1H, d, H12, 4JHH 2), 7.40 (5H, m, benzoate (5 g, 0.027 mol), potassium carbonate (22.5 g, H1–3), 7.13 (2H, AA¾XX¾, H8, JAA¾XX¾ 9), 7.01 (2H, AA¾XX¾, 0.16 mol) and 1-bromooctane (15.65 g, 0.082 mol), were placed H7, JAA¾XX¾ 9), 6.93 (1H, d, H15, 3JHH 8.5), 5.09 (2H, s, H5), in pentan-3-one (150 cm3), and the reaction heated at reflux 4.09 and 4.08 (4H, t, H17a and b), 1.87 and 1.86 (4H, qt, H18a for 84 h.Water (100 cm3) was added and the aqueous phase and b), 1.51 (4H, m, H19), 1.35 (16H, m, H20–23), 0.91 (6H, t, extracted against dichloromethane (3×150 cm3). The organic H24); dC (CDCl3): 165.4(C10), 156.4(C6), 153.8(C14), 148.7(C13), extracts were combined dried over MgSO4, filtered and evapor- 144.8(C9), 136.9(C4), 128.6, 128.0, 127.5(C1–3), 124.3(C8), ated to give a brown oil.A solution of potassium hydroxide 122.6(16C), 121.8(C11), 115.5(C7), 114.7(C12), 112.1(C15), (3.04 g, 0.054 mol) in 95% ethanol (150 cm3) was added and 70.5(C5), 69.4&69.1(C17a and b), 31.8, 29.4, 29.3, 29.2, 29.1, 26.0, the reaction heated at reflux for 2.5 h. Water (100 cm3) was 22.7(C18–23), 14.1(C24). added and the solution acidified with conc.hydrochloric acid (20 cm3). The resulting colourless precipitate was collected and Synthesis of the 4-(3,4,5-trialkoxybenzoyloxy)- crystallised (2×) from ethanol to give 3,4,5-tris(octyloxy)ben- 1-benzyloxybenzenes zoic acid as a colourless solid. Yield 11 g (54%); dH (CDCl3): 7.30 (2H, s, H2), 4.00 and 4.03 (6H, t, H5a and b), 1.84 (6H, qt, All derivatives were prepared similarly and one example is H6), 1.50 (6H, m, H7), 1.30 (24H, m, H8–11), 0.85 (9H, t, H12); given.All other derivatives were obtained in yields ranging dC (CDCl3): 172.0(CO2H), 152.8(C3), 143.1(C4), 123.7(C1), from 65–89%. 108.5(C2), 73.6(C5a), 6.2(C5b), 31.9, 31.8, 30.3, 29.5, 29.4, 29.3, 26.1, 22.7 (C6–11), 14.1(C12). Synthesis of the 4-(4-alkoxybenzoyloxy)-1-benzyloxybenzenes All derivatives were prepared similarly and one example is given.All other derivatives were obtained in yields ranging from 70–83%. OCH2CH2(CH2)5CH3 OCH2CH2(CH2)5CH3 O O O 6 7 8 9 10 11 12 13 14 15b 16b 22b 1 2 3 4 5 15a 16a 22a OCH2CH2(CH2)5CH3 6c J. Mater. Chem., 1998, 8(2), 331–341 3374-[3,4,5-Tris(octyloxy)benzoyloxy]-1-benzyloxybenzene 6c. 116.2(C2), 114.6(C7), 112.0(C10), 69.4 and 69.1(C12a and b), 31.8, 29.4, 29.3, 29.2, 29.0, 26.0, 22.7(C13–18), 14.1(C19). This was prepared from 3,4,5-tris(octyloxy)benzoic acid using the procedure described for 4c. The crude product was purified by crystallisation from ethanol (250 cm3), to give the pure Synthesis of the 4-(3,4,5-trialkoxybenzoyloxy)phenols product as a colourless solid.Yield: 0.88 g (65%); dH (CDCl3): All derivatives were prepared similarly and one example is 7.40 (5H, m, H1–3), 7.38 (2H, s, H12), 7.10 (2H, AA¾XX¾, H8, given. All other derivatives were obtained in yields ranging JAA¾XX¾ 9), 7.00 (2H, AA¾XX¾, H7, JAA¾XX¾ 9), 5.06 (2H, s, H5), from 80–95%. 4.05 (6H, m, H15), 1.82 and 1.80 (6H, qt, H16a and b), 1.50 (6H, m, H17), 1.30 (24H, m, H18–21), 0.85 (9H, t, H22).Synthesis of the 4-(4-alkoxybenzoyloxy)phenols All derivatives were prepared similarly and one example is given. All other derivatives were obtained in yields ranging from 76% to quantitative. O HO OCH2(CH2)6CH3 OCH2(CH2)6CH3 OCH2(CH2)6CH3 8 9 O 10a 21a 1 2 3 4 5 6 7 10b 21b 11c 4-[3,4,5-Tris(octyloxy)benzoyloxy]phenol 11c. Compound 6c (3.9 g, 5.7 mmol) was dissolved in freshly distilled THF (150 cm3) and triethylamine (1 cm3) and 10% wet Degassu OCH2CH2(CH2)5CH3 O O HO 1 2 3 4 5 6 7 8 9 10 11 21 9c Pd/C catalyst (0.05 g) was added.The reaction flask was evacuated and placed under hydrogen (three times), before 4-(4-Octyloxybenzoyloxy)phenol 9c. 4-(4-Octyloxybenzoy- being stirred at room temp. under an atmosphere of hydrogen.loxy)-1-benzyloxybenzene (6.6 g, 0.015 mol) was dissolved in After the calculated amount of hydrogen had been taken up freshly distilled THF (150 cm3) and triethylamine (1 cm3), and the catalyst was removed by filtration through Celite and the 10% wet Degassu Pd/C catalyst (0.05 g) was added. The solvent was evaporated. The crude product was purified by reaction flask was evacuated and placed under hydrogen flash chromatography on silica gel using THF as the eluent.(repeated three times), before being stirred at room temp. Yield: 3.37 g (quantitative); dH (CDCl3) 7.38 (2H, s, H7), 7.03 under an atmosphere of hydrogen. After 223 cm3 of hydrogen (2H, AA¾XX¾, H3, JAA¾XX¾ 9), 6.83 (2H, AA¾XX¾, H2, JAA¾XX¾ 9), had been taken up the catalyst was removed by filtration 5.18 (1H, s, OH), 4.02 (6H, t, H10), 1.80 and 1.75 (6H, qt, through Celite and the solvent was evaporated, to give a H11a and b), 1.45 (6H, m, H12), 1.30 (24H, m, H13–16), 0.85 (9H, colourless solid as the product. Yield: 4.7 g (90%); dH (CDCl3): t, H17); dC (CDCl3): 166.2(C5), 153.8(C1), 152.9(C8), 144.1(C4), 8.12 (2H, AA¾XX¾, H7, JAA¾XX¾ 9), 7.02 (2H, AA¾XX¾, H3, 142.8(C9), 123.9(C6), 122.5(C3), 116.2(C2), 108.5(C7), JAA¾XX¾ 9), 6.95 (2H, AA¾XX¾, H8, JAA¾XX¾ 9), 6.80 (2H, AA¾XX¾, 73.7(C10a), 69.3(C10b), 31.9, 31.8, 31.6, 30.3, 29.5, 29.4, 29.3, H2, JAA¾XX¾ 9), 5.38 (1H, s, OH), 4.02 (2H, t, H10, 3JHH 6.5), 26.9, 26.1, 22.7 (C11–16), 14.1(C17). 1.80 (2H, qt, H11), 1.45 (2H, m, H12), 1.30 (8H, m, H13–16), 0.85 (3H, t, H17); dC (CDCl3): 166.2(C5), 163.7(C9), 153.7(C1), Synthesis of the bis[4-(4-alkoxybenzoyloxy)phenyl] 2,2¾- 144.2(C4), 132.4(C7) 122.5(C3), 121.4(C6), 116.3(C2), 114.3(C8), bipyridine-5,5¾-dicarboxylates 68.4(C10), 31.8, 29.3, 29.2, 29.1, 26.0, 22.7(C11–16), 14.1(C17). All derivatives were prepared similarly and one example is given.Yields and elemental analyses are collected in Table 5. Synthesis of the 4-(3,4-dialkoxybenzoyloxy)phenols No 13C NMR data could be obtained for any of these materials due to poor product solubility.All derivatives were prepared similarly and one example is given. All other derivatives were obtained in yields ranging from 80–95%. Table 5 Yields and analytical data (Calc. %) Found % n Yield (%) C H N D1 20 (69.0) 67.6 (4.1) 4.0 (4.0) 4.2 D4 15 (70.8) 69.5 (5.2) 5.2 (3.6) 3.5 OCH2CH2(CH2)5CH3 OCH2CH2(CH2)5CH3 O O HO 1 2 3 4 5 6 7 8 9 10 11 12b 13b 23b 12a 13a 23a 10h D8 18 (72.6) 72.4 (6.3) 6.3 (3.1) 3.0 D12 12 (74.1) 73.2 (7.2) 7.2 (2.8) 2.8 D14 12 (74.7) 74.3 (7.6) 7.5 (2.6) 2.7 T1 20 (66.7) 64.4 (4.3) 4.2 (3.7) 3.6 4-[3,4-Bis(octyloxy)benzoyloxy]phenol 10h.Compound 5h T2 15 (68.0) 67.4 (5.0) 5.0 (3.5) 3.3 (6.6 g, 0.0118 mol) was dissolved in freshly distilled THF T3 8 (69.1) 68.9 (5.6) 5.6 (3.2) 3.2 (150 cm3) and triethylamine (1 cm3) and 10% wet Degassu T4 18 (70.1) 69.6 (6.1) 6.1 (3.0) 3.0 Pd/C catalyst (0.05 g) was added.The reaction flask was T5 8 (71.0) 71.0 (6.6) 6.7 (2.9) 2.8 evacuated and placed under hydrogen (three times), before T6 22 (71.8) 71.8 (6.9) 7.1 (2.7) 2.6 being stirred at room temp. under an atmosphere of hydrogen. T7 22 (72.5) 72.3 (7.4) 7.2 (2.5) 2.5 T8 25 (73.1) 73.3 7.7 7.9 (2.4) 2.5 After the calculated amount of hydrogen had been taken up T9 19 (73.7) 73.4 (8.0) 7.9 (2.3) 2.3 the catalyst was removed by filtration through Celite and the T10 10 (74.3) 74.0 (8.3) 8.2 (2.2) 2.2 solvent was evaporated.The crude product was purified by T11 32 (74.9) 74.6 (8.6) 8.5 (2.1) 2.1 crystallisation from ethanol, to give a colourless solid as the T12 5 (75.2) 75.0 (8.8) 8.7 (2.0) 2.0 product.Yield: 5.25 g (95%); dH (CDCl3): 7.74 (1H, dd, H11, T13 5 (75.2) 75.2 (9.2) 9.0 (2.0) 2.0 3JHH 8.5, 4JHH 2), 7.57 (1H, d, H7, 4JHH 2), 6.95 (2H, AA¾XX¾, T14 18 (76.0) 75.7 (9.2) 9.0 (1.9) 1.8 H1 20 (64.7) 64.4 (4.4) 4.5 (3.4) 3.5 H3, JAA¾XX¾ 9), 6.85 (1H, d, H10, 3JHH 8.5), 6.75 (2H AA¾XX¾, H4 12 (69.6) 69.4 (6.8) 6.9 (2.6) 2.6 H2, JAA¾XX¾ 9), 5.15 (1H, s, OH), 4.01 and 4.00 (4H, t, H12), H8 18 (72.6) 72.4 (6.3) 6.3 (3.1) 3.0 1.80 (4H, qt, H13), 1.40 (4H, m, H14), 1.25 (16H, m, H15–18), H12 12 (73.5) 73.2 (8.6) 8.9 (2.0) 2.0 0.83 (6H, t, H19); dC (CDCl3): 166.1(C5), 153.9(C1), 153.7(C9), R8 24 (72.6) 71.6 (6.3) 6.1 (3.1) 3.4 148.6(C8), 144.2(C4), 124.5(C3), 122.5(C11), 121.5(C6), 338 J.Mater. Chem., 1998, 8(2), 331–341N N O O O O OCnH2 n+1 O O O O CnH2 n+1O 2 3 4 7 8 9 10 5 6 11 12 13 14 16 15 D8 N N O O O O OCnH2 n+1 O O O O CnH2 n+1O OCnH2 n+1 CnH2 n+1O 2 3 4 7 8 9 10 5 6 11 12 13 14 16 15 17 18 T8 Bis[4-(4-octyloxybenzoyloxy)phenyl] 2,2¾-bipyridine-5,5¾- hot through Celite. The colourless solution was evaporated and the solid recrystallised (×2) from 1,4-dioxane, giving the dicarboxylate D8.The apparatus was flame dried prior to use. 2,2¾-Bipyridine-5,5¾-dicarboxylic acid dichloride (2.4 g, product as a cream solid. Yield: 0.62 g (17%); dH (CDCl3): 9.43 (2H, dd, H6, 4JHH 2, 5JHH 1), 8.65 (2H, dd, H3, 3JHH 8.5, 5JHH 3.6 mmol), and 4-(4-octyloxybenzoyloxy)phenol (1.74 g 7.1 mmol) were placed in freshly distilled toluene (50 cm3). 1), 8.55 (2H, dd, H4, 3JHH 8.5, 4JHH 2), 7.78 (2H, dd, H18, 3JHH 8.5, 4JHH 2), 7.60 (2H, d, H14, 4JHH 2), 7.29 and 7.22 (8H, Triethylamine (1 cm3) was added and the reaction heated at reflux, under nitrogen overnight. The solvent was evaporated AA¾XX¾, H9 and 10, JAA¾XX¾ 9), 6.88 (2H, d, H17, 3JHH 8.5), 4.03 and 4.02 (8H, t, H19a and b), 1.60 (8H, m, H20), 1.43 (8H, m, and dichloromethane added. This was extracted against 10% ammonia solution.The aqueous phase was then extracted with H21), 1.30 (32H, m, H22–25), 0.95 (12H, t, H26); dC (CDCl3): 164.9(C12), 163.7(C7), 158.7(C2), 153.9(C16), 151.2(C6), 148.9 dichloromethane (2×100 cm3), the organic extracts combined and the solvent evaporated to give a crude dark brown solid.and 148.6, 147.8(C8,11 and 15), 138.8(C4), 125.8(C5), 124.4(C3), 122.9 and 122.5(C9 and 10), 121.6(C18), 121.2(C13), 114.5(C14), The solid was placed in ethyl acetate, heated to reflux, allowed to cool to room temp. and the solid collected by centrifugation 111.9(C17), 69.3 and 69.1(C19a and b), 31.8, 29.4, 29.3, 29.1, 29.0, 26.0, 22.7 (C20–25), 14.1(C26). (×2). The solid was then heated to reflux in 1,4-dioxane and collected (×2), giving the product as a cream solid.dH (CDCl3) 9.40 (2H, dd, H6, 4JHH 2.5, 5JHH 1), 8.64 (2H, dd, H3, 3JHH 8.5, The synthesis of the bis[4-(3,4,5-trialkyloxybenzoyloxy)- phenyl] 2,2¾-bipyridine-5,5¾-dicarboxylates 5JHH 1), 8.54 (2H, dd, H4, 3JHH 8.5, 4JHH 2.5), 8.06 (4H, AA¾XX¾, H14, JAA¾XX¾ 9), 7.22 (8H, AA¾XX¾, H9 and 10), 6.90 (4H, AA¾XX¾, All derivatives were prepared similarly and one example is H15, JAA¾XX¾ 9); The alkyl region was unresolvable.given. Yields and elemental analyses are collected in Table 5. Synthesis of the bis[4-(3,4-dialkoxybenzoyloxy)phenyl] 2,2¾- Bis{4-[3,4,5-tris(octyloxy)benzoyloxy]phenyl} 2,2¾-bipyribipyridine- 5,5¾-dicarboxylates dine-5,5¾-dicarboxylate H8. The apparatus was flame dried prior to use. 2,2¾-Bipyridine-5,5¾-dicarboxylic acid dichloride All derivatives were prepared similarly and one example is given. Yields and elemental analyses are collected in Table 5. (0.25 g, 0.9 mmol), and 4-[3,4,5-tris(octyloxy)benzoyloxy]- phenol (10.6 g 1.8 mmol) were placed in freshly distilled toluene (50 cm3). Triethylamine (1 cm3) was added and the reaction Bis{4-[3,4-bis(octyloxy)benzoyloxy]phenyl} 2,2¾-bipyridine- 5,5¾-dicarboxylate T8.The apparatus was flame dried prior to heated at reflux, under nitrogen overnight. The solvent was evaporated and dichloromethane added. This was extracted use. 2,2¾-Bipyridine-5,5¾-dicarboxylic acid dichloride (0.9 g, 3.2 mmol), and 4-[3,4-bis(octyloxy)benzoyloxy]phenol (3 g, against 10% ammonia solution. The aqueous phase was then extracted with dichloromethane (2×100 cm3), the organic 6.4 mmol) were placed in freshly distilled toluene (50 cm3).Triethylamine (1 cm3) was added and the reaction heated at extracts combined and the solvent evaporated to give a crude dark brown solid. The solid was placed in ethyl acetate, heated reflux, under nitrogen overnight. The solvent was evaporated and dichloromethane added.This was extracted against 10% to reflux, allowed to cool to room temp. and the solid collected by centrifugation (×2), before being crystallised from 1,4- ammonia solution. The aqueous phase was then extracted with dichloromethane (2×100 cm3), the organic extracts combined dioxane. The crude product was placed in chloroform, heated to reflux and filtered hot through Celite.The colourless solution and the solvent evaporated to give a crude dark brown solid. The solid was placed in ethyl acetate, heated to reflux, allowed was evaporated and the solid recrystallised (×2) from 1,4- dioxane, giving the product as a cream solid. Yield: 0.23 g to cool to room temp. and the solid collected by centrifugation (×2), before being crystallised from 1,4-dioxane. The crude (19%); dH (CDCl3): 9.50 (2H, dd, H6, 4JHH 2, 5JHH 1), 8.73 (2H, dd, H3, 3JHH 8.5, 5JHH 1), 8.63 (2H, dd, H4, 3JHH 8.5, 4JHH product was placed in chloroform, heated to reflux and filtered N N O O O O OCnH2 n+1 O O O O CnH2 n+1O OCnH2 n+1 CnH2 n+1O CnH2 n+1O OCnH2 n+1 2 3 4 7 8 9 10 5 6 11 12 13 14 16 15 H8 J.Mater. Chem., 1998, 8(2), 331–341 3392), 7.42 (4H, s, H14), 7.35 and 7.30 (8H, AA¾XX¾, H9 and 10, JAA¾XX¾ 9), 4.08 and 4.06 (12H, t, H17a and b), 1.86 and 1.80 (12H, m, H18a and b), 1.49 (12H, m, H19), 1.30 (48H, m, H20–23), 0.90 (18H, t, H24); dC (CDCl3): 164.9(C12), 163.6(C7), 158.8(C2), 153.0(C15), 151.1(C6), 148.8 and 148.0(C8 and 11), 143.3(C16), 138.7(C4), 125.9(C5), 123.6(C13), 122.9 and 122.5(C9 and 10), 121.7(C3), 108.7(C14), 73.6(C17a), 69.3(C17b), 31.9, 31.8, 30.4, O O 1 2 3 4 5 6 7 8 9 O 10 O OCH2CH2(CH2)5CH3 11 12 13 14 15 16 22 12 29.5, 29.4, 29.3, 26.1, 22.7 (C18–23), 14.1(C24). 4-Octyloxyphenyl 4¾-(tetrahydropyran-2-yloxy)benzoate 12.Synthesis of the reversed ester, R8 Compound 11 (2.5 g, 0.012 mol), 4-octyloxyphenol (2.5 g 0.012 mol), and dicyclohexylcarbodiimide (DCC) (2.32 g 0.012 mol) were dissolved in dichloromethane (80 cm3). 4- (N,N-Dimethylamino)pyridine (0.14 g 1.2 mmol) was then added and the reaction stirred at room temp. overnight. The colourless precipitate was removed by filtration and the solvent evaporated. Crystallisation from ethanol (×2) gave the product as a colourless solid. Yield: 4 g (84%); mp 81 °C (sublimes); dH (CDCl3): 8.13 (2H, AA¾XX¾, H8, JAA¾XX¾ 9), 7.10 (2H, AA¾XX¾, H7, JAA¾XX¾ 9), 7.15 (2H, AA¾XX¾, H13, JAA¾XX¾ 9), 6.90 (2H, O O 1 2 3 4 5 6 7 8 9 O 10 O 11 12 13 15 14 10 AA¾XX¾, H12, JAA¾XX¾ 9), 5.55 (1H, t, H5, 3JHH 3), 3.95 (2H, t, H15, 3JHH 6.5), 3.88 (1H, m, H1eq), 3.62 (1H, m, H1ax), 2.05, Benzyl 4-(tetrahydropyran-2-yloxy)benzoate 10. 3,4- 1.90 and 1.70 (6H, m, H2,3 and 4), 1.78 (2H, m, H16), 1.45 (2H, Dihydro-2H-pyran (20 cm3, 0.22 mol) and ethyl acetate satum, H17), 1.30 (8H, m, H18–21), 0.90 (3H, t, H22); dC (CDCl3): rated with HCl(g) (3.5 cm3) was added to a solution of benzyl 165.3(C10), 161.4(C6), 156.8(C14), 144.4(C11), 132.1(C8), 4-hydroxybenzoate (10 g, 0.044 mol) in ethyl acetate (100 cm3). 122.7(C9), 122.4(C12), 116.1(C7), 115.1(C13), 96.1(C5), The reaction was stirred at room temp.for 12 h before the 68.5(C15), 62.0(C1), 30.1(C4), 25.1(C2), 18.5(C3), 31.8, 29.4, solvent was removed. Flash chromatography on neutral alum- 29.3, 29.2, 26.1, 22.7(C16–21), 14.1(C22). ina using dichloromethane as the eluent gave the product as a colourless oil which solidified on standing. Yield: 12 g (88%); mp 68–71 °C (decomp.); dH (CDCl3): 8.03 (2H, AA¾XX¾, H8, JAA¾XX¾ 9), 7.40 (5H, m, H13–17), 7.06 (2H, AA¾XX¾, 7H, JAA¾XX¾ 9), 5.50 (1H, t, H5, 3JHH 3), 3.88 (1H, m, H1eq), 3.62 (1H, m, H1ax), 2.01, 1.90 and 1.70 (6H, m, H2,3 and 4); dC (CDCl3): HO 1 2 3 4 O 5 O OCH2CH2(CH2)5CH3 6 7 8 9 10 11 17 13 166.2(C10), 161.0(C6), 136.4(C12), 131.6(C8), 128.6, 128.12 and 128.07(C13–15), 123.3(C9), 115.9(C7), 96.1(C5), 66.4(C11), 4-Octyloxyphenyl 4-hydroxybenzoate 13.Compound 12 (4 g, 62.0(C1), 30.1(C4), 25.1(C2), 18.5(C3). 9.4 mmol) and oxalic acid (50 mg) were placed in methanol– water (100 cm3; 951), and the reaction was heated at reflux for 72 h. The reaction was cooled to room temp. and the resulting colourless needles collected and washed with ethanol– water (151), to give the pure product. The mother liquor was evaporated and the resulting colourless solid crystallised from ethanol–water to give more product as colourless needles.O O 1 2 3 4 5 6 7 8 9 OH 10 O 11 Yield: 2.95 g (95%); mp 159 °C; dH (CDCl3): 8.10 (2H, AA¾XX¾, H3, JAA¾XX¾ 9), 7.10 (2H, AA¾XX¾, H7, JAA¾XX¾ 9), 6.89 and 6.93 (4H, AA¾XX¾, H2 and 8, JAA¾XX¾ 9), 5.80 (1H, s, OH), 3.95 ( 2H, 4-(Tetrahydropyran-2-yloxy)benzoic acid 11.To a solution t, H10, 3JHH 6.5), 1.78 (2H, qt, H11), 1.45 (2H, m, H12), 1.30 of compound 10 (11.5 g, 0.037 mol) in freshly distilled THF, (8H, m, H13–16), 0.90 (3H, t, H17); dC (CDCl3): 165.5(C5), wet Degassu Pd/C catalyst was added (10%, 50 mg). The 162.3(C1), 156.7(C9), 144.5(C6), 132.3(C3), 122.5(C7), reaction was placed under vacuum and hydrogen (×3) before 120.6(C4), 115.5, 115.0(C2 and 8), 68.4(C10), 31.8, 29.3, 29.2, being left to stir under an atmosphere of hydrogen at room 26.0, 22.6(C11–16), 14.1(C17).temp. After hydrogen (900 cm3) had been used the reaction was filtered through Celite and the solvent evaporated. The crude product was crystallised from ethanol to give a colourless Bis[(4-octyloxybenzoyloxy)phenyl] 2,2¾-bipyridine-5,5- dicarboxylate R8.This was prepared from compound 13 using solid as the product. Yield: 6 g (73%); mp 159 °C (decomp.); dH (CDCl3): 8.05 (2H, AA¾XX¾, H8, JAA¾XX¾ 9), 7.10 (2H, the procedures for esterification with 2,2¾-bipyridine-5,5¾-dicarboxylic acid dichloride described above. This gave the product AA¾XX¾, 7H, JAA¾XX¾ 9), 5.55 (1H, t, H5, 3JHH 3), 3.88 (1H, dt, H1eq), 3.62 (1H, m, H1ax), 2.01, 1.90 and 1.70 (6H, m, as a cream solid.Analytical data are found in Table 5. dH (CDCl3): 9.51 (2H, dd, H6, 4JHH 2, 5JHH 1), 8.75 (2H, dd, H3, H2,3 and 4); dC (CDCl3): 172.0(C10), 161.6(C6), 132.2(C8), 122.4(C9), 116.0(C7), 96.1(C5), 66.4(C11), 62.1(C1), 30.1(C4), 3JHH 8.5, 5JHH 1), 8.64 (2H, dd, H4, 3JHH 8.5, 4JHH 2), 8.33 (4H, AA¾XX¾, H10, JAA¾XX¾ 9), 7.44 (4H, AA¾XX¾, H9, JAA¾XX¾ 9), 25.1(C2), 18.5(C3).N N O O O O O OC8 H17 O C8H17O O O 12 13 14 15 16 3 4 9 10 11 8 6 7 5 R8 340 J. Mater. Chem., 1998, 8(2), 331–34112 D. W. Bruce and K. E. Rowe, L iq. Cryst., 1995, 18, 161. 7.14 (4H, AA¾XX¾, H14, JAA¾XX¾ 9), 6.95 (4H, AA¾XX¾, H15, 13 L. Douce, R. Ziessel, R. Seghrouchni, E. Campillos, A. Skoulios JAA¾XX¾ 9), 3.98 (4H, t, H17, 3JHH 6.5), 1.85 (4H, qt, H18), 1.50 and R.Deschenaux, L iq. Cryst., 1995, 18, 157. (4H, m, H19), 1.27 (16H, m, H20–23), 0.90 (12H, t, H24); MS 14 K. E. Rowe and D. W. Bruce, L iq. Cryst., 1996, 20, 183. m/z: [M+] 892.39; no 13C NMR data could be obtained due 15 See e.g. R. Deschenaux and J. W. Goodby, in Ferrocenes, ed to the product’s insolubility. A. Togni and T. Hayashi, VCH, Weinheim, 1995, ch. 9. 16 K. E. Rowe and D. W. Bruce, J. Chem. Soc., Dalton T rans., 1996, 3913. Support from the EPSRC and the University of Exeter is 17 J. Mathe�te, H.-T. Nguyen and C. Destrade, L iq. Cryst., 1993, 13, gratefully acknowledged. 171. 18 H.-T. Nguyen, C. Destrade and J. Malthe�te, Adv. Mater., 1997, 9, 375. References 19 D. W. Bruce and S. A. Hudson, J. Mater. Chem., 1994, 4, 479; W. Weissflog, G. Pelzl, I. Letko and S. Diele, Mol. Cryst., L iq. 1 E. C. Constable, Adv. Inorg. Chem. Radiochem., 1984, 30, 69. Cryst., 1995, 260, 157. 2 G. X. Liu and R. J. Puddephatt, Organometallics, 1996, 15, 5257. 20 B. Heinrich and D. Guillon, unpublished results. 3 J.-M. Lehn and A. Rigault, Angew. Chem., Int. Ed. Engl., 1988, 21 B. Donnio and D. W. Bruce, J. Chem. Soc., Dalton T rans., 1997, 27, 1095. 2745. 4 V. Balzani, A. Juris, M. Venturi, S. Campagna and S. Serroni, 22 W. Weissflog, I. Letko, S. Diele and G. Pelzl, Adv. Mater., 1996, Chem. Rev., 1996, 96, 759. 8, 76. 5 See e.g. J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 23 D. W. Bruce, B. Donnio, S. A. Hudson, A.-M. Levelut, S. Megtert, 1995. D. Petermann and M. Veber, J. Phys. II Fr., 1995, 5, 289. 6 D. W. Bruce, Adv.Mater., 1994, 6, 699. 24 D. W. Bruce, B. Donnio, D. Guillon, B. Heinrich and M. Ibn- 7 D. W. Bruce and X.-H. Liu, J. Chem. Soc., Chem. Commun., 1994, Elhaj, L iq. Cryst., 1995, 19, 537. 729. 25 B. Donnio, D. W. Bruce, H. Delacroix and T. Gulik-Krzywicki, 8 D. W. Bruce and X.-H. Liu, L iq. K., 1995, 18, 165. L iq. Cryst., 1997, 23, 147. 9 S. Morrone, G. Harrison and D. W. Bruce, Adv. Mater., 1995, 7, 26 B. Donnio, D. W. Bruce, B. Heinrich, D. Guillon, H. Delacroix 665; S. Morrone, D. Guillon and D. W. Bruce, Inorg. Chem., 1996, and T. Gulik-Krzywicki, Chem. Mater., 1997, in press, B. Donnio, 35, 7041. PhD T hesis, University of SheYeld, 1996. 10 H. Zheng and T. M. Swager, J. Am. Chem. Soc., 1994, 116, 27 G. W. Gray, T hermotropic L iquid Crystals, Wiley, Chichester, 761 T. M. Swager and H. Zheng, Mol. Cryst., L iq. Cryst., 1995, 1987. 260, 301. 28 M. A. Osman, Mol Cryst, L iq. Cryst., 1985, 128, 45; G. Nestor, 11 See also: K. Hanabusa, J.-I. Higashi, T. Koyama, H. Shira, N. Hojo G. W. Gray, D. Lacey and K. J. Toyne, L iq. Cryst., 1990, 7, 669; and A. Kurose, Makromol. Chem., 1989 190, 1; T. Kuboki, D. W. Bruce and S. A. Hudson, J. Mater. Chem., 1994, 4, 479; K. Araki, M. Yamada and S. Shiraishi, Bull. Chem. Soc. Jpn., 1994, G. W. Gray, M. Hird and K. J. Toyne, Mol. Cryst., L iq. Cryst., 67, 984; D. W. Bruce, J. D. Holbrey, A. R. Tajbakhsh and 1991, 204, 91. G. J. T. Tiddy, J. Mater. Chem., 1993, 3, 905; A. Elgayoury, L. Douce, R. Ziessel, R. Seghrouchni and A. Skoulios, L iq. Cryst., 1996, 21, 143. Paper 7/06400D; Received 2nd September, 1997 J. Mater. Chem., 1998, 8(2), 331–341