MATERIALS CHEMISTRY COMMUNICATIONS Supermolecular alignment in a liquid crystal-polymer gel as studied optically and by dielectric relaxation spectroscopy Monica M. Marugan, Sara Shinton and Graham Williams* Department of Chemistry, University of Wales, Swansea, Singleton Park, Swansea, UK SA2 8PP Macroscopically aligned liquid crystal (LC) gels have been prepared by photopolymerization in visible light of a non- mesogenic bifunctional monomer in a conventional LC mixture that was prealigned in an electric field. A high degree of molecular orientation is achieved in the gel, and it is stable with time, as shown by polarized optical microscopy and broad-band dielectric relaxation spectroscopy. In recent years it has been shown that macroscopically aligned liquid crystal/polymer networks may be formed by radical- initiated photopolymerization of mono-and bi-functional monomers dissolved in a liquid crystal.Polymerization leads to the formation of a network in which the branched chains or their aggregates form a continuous network dispersed in the smectic, nematic or chiral nematic LC medium. The polymerization process in an LC medium is guided spatially by the long range ordering of molecules of the mesophase. Once formed the network has a set three-dimensional topogra- phy but this may, in principle, be distorted by application of external fields as a result of the orientation and flow of the LC phase or by application of a shearing force. Since an LC-monomer medium may be aligned with an external field or by surface forces it is thus possible to obtain macroscopically aligned LC-polymer networks, also known as anisotropic LC gels. Hikmet and co-worker~’-~ have used bifunctional mon- omers that are themselves liquid crystals in their bulk phase and have reported the preparation of anisotropic LC gels over the entire composition range, and have made extensive investi- gations of their optical and electrooptical properties.They have shown that these novel hybrid materials show promise as new media for optical data storage and for active and passive optical elements for displays and optical processing (see refs. 1-5 and refs. therein). It has been shown recently that non-mesogenic monomers may be polymerized in an aligned nematic6 or smectic7 LC-monomer phase, to produce macro- scopically aligned LC gels.In the present work we describe the preparation of highly aligned LC gels formed by the polymerization of a non-mesogenic dimethacrylate monomer 1 in a conventional LC mixture E7 (Licryllite, Mixture E7 from Merck Ltd.). 1 Monomer 1 was obtained from Akzo Ltd. It is a feature of the present work that monomer 1 is non-mesogenic, in contrast with the pioneering work of Hikmet who used specially synthesized mesogenic monomers in conjunction with conven- tional LC mixtures to prepare aligned gels. We demonstrate that such a non-mesogenic monomer forms a strain-free gel that conforms with the LC phase, thus when a directing E-field is removed the gel retains its macroscopic alignment since this is held by the overall gel structure.A second and important practical feature of our work is that the photopolymerization of 1 at a concentration of 5% (w/w) in E7 was carried out in visible blue light (470 nm) in contrast to earlier works which used UV light. Thus LC-monomer mixtures contained in indium tin oxide (ITO) glass cells mounted in a microscope hot-stage with glass windows could be readily photopolymer- ized in situ, which is not possible using UV light. Since LC-polymer gels show considerable promise as materials for optical data storage and for optical elements, the practical advantages of using visible light for photopolymerization should be recognised (e.g. ordinary glass is used here in place of quartz, we use an inexpensive commercial visible light source used for dental cements, and we do not have the hazards associated with the use of UV light). Samples were prepared by placing the LC-monomer mix-ture (95:5 w/w) in an ITO-coated glass sandwich cell (1cm x 1cm x 20 pm) kept apart using Kapton spacers.The IT0 cell was mounted in a programmable hot-stage (Linkam TMS600) attached to an Olympus BH2 polarizing microscope. Prealignment of the LC-monomer mixture was achieved using ac or dc voltages applied to the IT0 cell. The optical properties of the LC-monomer, and of the LC-polymer gel subsequently made, were monitored using the polarizing microscope. A further feature of the work reported here is our use of dielectric relaxation spectroscopy (DRS) for the in situ monitor-ing of the nature and extent of macroscopic alignment in the unpolymerized and polymerized materials.A GenRad 1689-9620 DigiBridge ( 10-105 Hz), with associated computer control and software developed in-house, was used to obtain the dielectric permittivity and loss data. While optical microscopy provides us with qualitative information on the alignment of LC-monomer and LC-polymer gel samples, the dielectric data presented here provide direct unambiguous quantitative information of the extent of alignment of such samples (1cm2 x 20 pm) in the IT0 cells. Note that such information is not easy to obtain by other methods. For example, FTIR measurements are inappropriate owing to the opacity of the glass electrodes to IR light.NMR spectroscopy normally requires sophisticated instrumentation and a larger sample than that under investigation here and, importantly, uses high magnetic fields that will induce further director orientation in an LC sample so that it is not studied in its field-free state, unlike the DRS studies reported here. For an LC-monomer sample that was not deliberately prealigned using external voltages, the material appeared trans- parent in the microscope, but on illumination as the gel formed a turbid strongly scattering texture was immediately observed. The origins of the scattering are a combination of two factors: (i) a nanophase scale separation of network polymer may occur, as discussed by Braun et a1.* leading to scattering from J.Muter. Chem., 1996, 6(4), 667-669 667 the polymer network itself, and (ii) the network acts as a large internal surface at which the LC molecules interact and may be leading to strong scattering from the anisotropic LC polydomain itself. A further LC-monomer sample was aligned homeotropically using 30 V (1 kHz) and illumination applied for 10 min [using an Elipar-2 Dental Lamp (A,,, = 470 nm)] and maintained for a further 15 min after illumination to stabilize the alignment of the LC-polymer film. Plate l(a) shows a microscope picture of the resultant gel sample which includes the edge between the region to which the field was applied during gel formation (right-hand side) and that where no field was applied (left-hand side).The scattering texture of the nominally unaligned region is seen while the excellent transparency of the aligned region is apparent (appears black between crossed polarizers). Plate 1(b) shows the conoscopic image obtained for the transparent, homeotropically aligned region of this sample. A Maltese cross is seen, indicating a high degree of homeotropic alignment (nllz), against a scat- tering background. The optical transparency of the aligned gel is not of the quality of the pre-aligned LC-monomer mixture, showing that scattering of light occurs from the network in the aligned gel. The transparency of the aligned gel is, neverthe- less, very high and is sufficient to allow this hybrid material to act as an information storage medium.Fig. 1 shows the dielectric loss spectrum as a function of frequency for the liquid-crystal mixture E7 at -20 "C.In the off condition no directing voltage is applied to the sample and only a small loss is observed in this frequency range. Application of a dc biasing voltage (from an internal source in the DigiBridge) leads to a development of the loss peak centred on 20 kHz at this sample temperature. Under these conditions the sample is homeotropically (H) aligned. The data clearly show that the 'voltage-untreated' sample in Fig. 1 is planarly aligned and becomes H-aligned upon application Plate 1 (a) Nominally unaligned (left) and homeotropically aligned (right) regons of the LC gel viewed through crossed polarizers (magnification 100 x).(b)Conoscopic image of the homeotropically aligned LC gel. OD 00 L -0 w4 0 0 0 0 1 2 3 4 5 log (frequency/Hz) Fig. 1 Dielectric loss factor E" vs. log (frequency/Hz) for nominally unaligned (0)and homeotropically aligned (0)liquid-crystal mixture E7 at -20 "C 20 . 15 0 0 OO 0 w 10 0 0 "1 2 3 4 5 log (frequency/Hz) Fig. 2 Dielectric permittivity E' and loss factor E" us. log (frequency/Hz) at -20 "C for LC gels prepared in the absence (0)and presence (0) of a directing E field. The enhancement of the loss peak on alignment (0 to 0) demonstrates the alignment for a gel prepared from the aligned LC-monomer mixture. of 20V. The data of Fig. 1 act as a reference for the LC gels described below.Fig. 2 shows permittivity (E') and loss (E") data for gel samples made in the absence and presence of a directing voltage. The sample prepared in the absence of the voltage has a small permittivity (E'E5.8) and small loss factor in this range at -20 "C, and is planarly aligned. The aligned gel made from an LC-monomer mixture in the presence of 30 V at 1 kHz gives the well defined loss peak whose frequency location is essentially the same as that for E7 in its H-aligned state (see Fig. 1). The relaxation strength AE of the loss peak for both samples is readily estimated from the product of peak height (E",~,) and full-width at half-height (A+).The loss peak for the aligned gel is ca. 6% broader than that for E7, indicating a greater complexity for this process in the gel.The ratio of relaxation strengths, : for the aligned samples is ca. 0.84. Since 5% of the gel material is the polymer network, which has a much smaller contribution per repeat unit than that from the highly polar molecules comprising E7, this ratio would be ca. 0.95 if the LC gel has full homeotropic alignment. The dielectric properties of low molar mass and polymeric liquid crystals in the absence and presence of electric and magnetic fields and surface forces are well documented (see ref. 9 for a recent review). For a partially aligned LC sample having axial symmetry with respect to the measuring field direction (z axis) the measured dielectric permittivity &',(a)and loss factor E",(o) may be written as linear functions of the macroscopic director order parameter, &, of the material.2(l+2Sd)&'r,,(0)+j(1-Sd)E"(U)) (1)&"z(0)=-3 668 J. Muter. Chem., 1996, 6(4), 667-669 where 11 and I refer to measurements made for fully homeotropic (nllE)and fully planar (nlE)samples respectively. Thus, as Sd is changed from -3 (planar) through 0 (unaligned) to 1 for the H-aligned material, then E”,(u)) is changed system- atically. E”II (0)and E”~(U)) are functions of (i) the local order parameter, S, which is unchanged on changing Sd, (ii) of p1 and pt where p1 and pt are the longitudinal and transverse components of the effective dipole moment of the mesogenic groups, and (iii) Fourier transforms of four orthogonal relax- ation functions (see ref.11 for details). For the E7 mixture the 00 mode (also known as a 6 process) dominates the dielectric loss peak in the aligned samples in Fig. 1 and 2. Using eqn. (l), remembering the first term on the right- hand side refers to the 6 process of Fig. 1 and 2 and assuming that the E7 sample was fully aligned homeotropically, then we calculate Sd=0.83 for the aligned gel sample. This value falls below that for full H-alignment and suggests that a part of the LC material is bound to the internal surfaces of the LC-gel network, removing its contribution to the 6 process in the dielectric spectrum. We have made measurements of the low frequency permittivity, E,, for the aligned LC gel at room temperature over a period of time.We find no decrease (<0.1% change) in E, over a period of several weeks, indicating that the alignment is preserved quantitatively and that the network is in a strain-free condition in the aligned liquid crystal. Finally, note that Stannarius et aL6 used NMR spectroscopy, with a 4.7 T magnet, to study the director distribution in LC gels formed from 4,4-bis-acryloxylbiphenyl, as a non-mesogenic monomer, and a 50: 50 mixture of n-pentyl and n-pentyloxy cyanobiphenyls. In this case the strong magnetic field used to study a material will perturb the director orientation when the field direction makes an angle with the C, symmetry axis for the sample, thus such measurements determine director distri- butions in the presence of the B-field.6 In contrast, the weak measuring E-fields used in our dielectric relaxation experiments do not influence the director distribution in the gel.In a future publication” we shall show that strong directing E-fields may be applied to gels having different initial macroscopic align- ments and the resultant changes in director orientation may be measured simultaneously using our dielectric method. We shall demonstrate12 that dielectric studies provide a means of determining Sd in LC gels subjected to additional E-fields and that saturation of orientation can be achieved for gels that were initially only partially aligned or were planarly aligned, but only at fields far higher than that for an unpolymerized LC mixture.The authors gratefully acknowledge grant support from the EPSRC to S.S. and from the EC Human Capital Mobility Programme to M.M., and equipment support from the EPSRC. References 1 D. J. Broer, R. A. M. Hikmet and G. Challa, Makromol. Chem., 1989,190,3202. R. A. M. Hikmet, J. Lub and D. J. Broer, Adu. Mater., 1991,3,392. R. A. M. Hikmet, Liq. Cryst., 1991,9,405. R. A. M. Hikmet, Adu. Mater., 1992,4679. R. A. M. Hikmet and B. H. Zwerver, Liq. Cryst., 1993,13, 561. R. Stannarius, G. P. Crawford, L. C. Chien and J. W. Doane, J. Appl. Phys., 1991,79, 135. 7 A. Jakli, L. Rosta and L. Noirez, Liq. Cryst., 1995, 18,601. 8 D. Braun, G. Frick, M. Grell, M. Klimes and J. H. Wendorff, Liq. Cryst., 1992,11,929, and references therein. 9 G. Williams, in The Molecular Dynamics of Liquid Crystals, ed. G. R. Luckhurst and C. S. Veracini, Kluwer, Dordrecht, 1994, p. 431. 10 G. S. Attard, K. Araki and G. Williams, Brit. Polym. J., 1987, 19, 119. 11 K. Araki, G. S. Attard, A. Kozak, G. Williams, G. W. Gray, D. Lacey and G. Nestor, J. Chem. SOC., Faraday Trans. 2, 1988, 84, 1067. 12 G. Williams, M. Marugan and S. Shinton, manuscript in preparation. Communication 5/06571B; Received 5th October, 1995 J. Mater. Chem., 1996, 6(4), 667-669 669