Characterization of the sub-micrometer hierarchy levels in the twist-bend nematic phase with nanometric helices via photopolymerization. Explanation for the sign reversal in the polar response
Vitaly P. Panov, Sithara P. Sreenilayam, Yuri P. Panarin, Jagdish K. Vij, Chris J. Welch, Georg H. Mehl
aa r X i v : . [ c ond - m a t . s o f t ] A ug Characterization of the sub-micrometer hierarchylevels in the twist-bend nematic phase withnanometric helices via photopolymerization.Explanation for the sign reversal in the polarresponse.
Vitaly P. Panov, † Sithara P. Sreenilayam, † Yuri P. Panarin, † , ‡ Jagdish K. Vij, ∗ , † Chris J. Welch, ¶ and Georg H. Mehl ¶ † Department of Electronic and Electrical Engineering, Trinity College, University ofDublin, Dublin 2, Ireland ‡ School of Electrical and Electronic Engineering, Dublin Institute of Technology, Dublin 8,Ireland ¶ Department of Chemistry, University of Hull, HU6 7RX, UK
E-mail: [email protected]
Abstract
Photo-polymerization of a reactive mesogen mixed with a mesogenic dimer, shownto exhibit the twist-bend nematic phase ( N T B ), reveals the complex structure of theself-deformation patterns observed in planar cells. The polymerized reactive mesogenretains the structure formed by liquid crystalline molecules in the twist bend phase,thus enabling observation by Scanning Electron Microscope (SEM). Hierarchical or-dering scales from tens of nanometers to micrometers are imaged in detail. Submicron eatures, anticipated from earlier X-ray experiments, are visualized directly. In theself-deformation stripes formed in the N T B phase, the average director field is foundtilted in the cell plane by an angle of up to 45 ◦ from the cell rubbing direction. Thistilting explains the sign inversion being observed in the electro-optical studies. Keywords twist-bend nematic; hierarchical self-assembly; polymerization; Scanning Electron MicroscopyLiquid crystals (LCs) are known to form complex spatial structures that scale fromthe molecular length via nanometric assemblies such as in the Sm C α *, blue phase, darkconglomerate phase or twist-grain-boundary phase to the sample size (cholesteric phase, N T B ). Beyond stimulating major theoretical advances in topology and mean field theory, LCs have made the stunning success of current information display and photonic devicesindustries possible and are at the forefront of emerging technologies relating to light andmatter. The major basis for applications of LCs in photonic devices so far is the three-dimensional deformation of the LC director on the application of a relatively weak externalfield. Meanwhile, preserving the molecular structure of a phase by polymerisation will notonly extend the range of potential applications, but will enable further structural studies bymethods that are normally unsuitable for liquid samples. Synthesis of helical polyacetylenefor potential nano-sized polymer solenoids is an example. Reactive mesogens or polymerizable LCs are designed to have both LC properties (includ-ing miscibility with other liquid crystalline phases/materials) and polymerisation controlledby light, heat, concentration of oxygen, etc. This class of materials have been shown toaid in the liquid crystalline alignment and may also extend the temperature range of themesophases. The twist-bend nematic phase has recently become one of the most topical areas ofresearch in the field of LCs due to the presence of a complicated hierarchy of periodic2tructures. The phase is observed in wide range of materials and mixtures containing ei-ther odd-hydrocarbon-chain-link dimers or bent-core mesogenic materials. Mixing of thematerials provides freedom for control of operating temperature range, that includes roomtemperature. This, combined with a relatively high tolerance to mixture compositions (upto 40% of added non- N T B components), makes the phase ”ready to use” for potential ap-plications.
A combination of the flexoelectric properties and nanometer-scaled helicalpitch provides one of the fastest electro-optic responses (a few microseconds) observedso far in LCs. Though yet not optimized for devices (the switching angle is very low), itdemonstrates in principle the potential of making use of emerging chirality for photonicapplications; and, beyond that, for chiral synthesis of pharmaceutical drugs.The scale of the structures in the N T B phase ranges from the sample size/thickness, i.e.several micrometers to several nanometers. Some features are found to be of the submi-cron size. Meanwhile, optical microscopy, though extremely useful, is restricted to haveits resolution limited by optical wavelength. In this Letter, we demonstrate that the tech-nique of photopolymerization is extremely effective for deciphering the complex hierarchicalstructures in the N T B phase using SEM.The structural formulae of the materials used in the experiment are shown in Figure1. Mixture of mesogenic dimer CB-C7-CB (approximately 70 w/w % of the total mixture)with 4-Cyano-4’-pentylbiphenyl ( ∼ ∼
12 w/w%). Rod-shaped, forming a widenematic phase, molecules of RM 257 provide good miscibility and retain the structure afterpolymerization. The photoinitiator Irgacure R (cid:13)
819 ( ∼ ∼ ◦ C and the nematic to N T B transition at ∼ ◦ C on cooling. Figure 1: Chemical formulae of materials used in the experiment.The mixture containing the LCs, the reactive mesogen and the photoinitiator was filledin commercial sandwich cells (KSXX-0X/XX11P6NSS05 EHC R (cid:13) , Japan) by capillary actionat 95 ◦ C and was immediately cooled to 35-55 ◦ C in the N T B phase to avoid temperature-induced polymerization. A microscope-mounted hot-stage connected to a Eurotherm R (cid:13) ± ◦ C . The textureswere observed using polarising optical microscope (POM) Olympus R (cid:13) BX51 equipped with5X, 10X or 20X objective lens and a filter used for blocking the UV radiation of the lightsource. To induce photopolymerization, an ultraviolet LED with peak emission wavelengthat 390 nm was mounted onto the camera port of the microscope, thus UV light was projectedonto a half-millimeter diameter spot on the sample plane. The optimal exposure time of theorder of 20 seconds could be controlled by varying concentration of the photo-initiator. Thisset-up enabled us to preserve the structures by UV polymerization for several experimentalconditions of interest (e.g. at different temperatures) for each sample cell. The polarizerof the microscope was removed from the path of the UV light during exposure. It is notedthat introduction of a prism polarizer did not show any visible influence on the polymerizedstructures indicating that the absorption of UV by the sample was practically isotropic.Heating the sample to a higher temperature phase can be used to evaluate results of poly-merization process as texture in the irradiated area is preserved and shows a clear contrast4o the surroundings (see supplementary information). When the sample was heated to theisotropic state, areas of the cell polymerized in LC phases, still exhibited birefringence whenthe cell was observed between the crossed-polarizers. On the other hand, areas polymerizedin the isotropic state remained isotropic (i.e. dark texture observed between the crossedpolarizers) on consequent cooling to LC phases.When the polymerization was complete, two glass plates of the cell are pulled apartand the liquid crystalline material was washed away with acetone. The polymer ”sponge”remained attached to the glass surfaces and could be used for investigating the preservedstructures. Although the polymerized RM257 has low conductivity, using thin samples (upto 4 µ m) and the Indium Tin Oxide (ITO)-coated glass plate as the ground electrode enabledimaging by the state-of-the-art scanning electron microscope (SEM) (Carl Zeiss Ultra Plus R (cid:13) )with a resolution of approximately 20 nm.Figure 2 shows SEM images of a planar-aligned sample, polymerized at different tem-peratures. Differences between the isotropic (2a), nematic (2b) and N T B (2c, d) phases areclearly observed. The isotropic (Figure 2a) phase is characterized by the lack of a preferredorientation. The nematic phase (Figure 2b) shows a low-ordered structure; the cell rubbingdirection (shown by the yellow arrow) being clearly observed. The N T B phase exhibits arange of structures that depend on the temperature of polymerisation of the sample andits confinement conditions. Although in the vicinity of the N - N T B phase transition POMtextures of both phases are practically uniform, a detailed SEM image clearly shows a differ-ence as the average direction (red lines) of the polymer filaments governed by the N T B phase(Figure 2c) form an angle α with the rubbing direction (yellow arrow). The most spectacularpatterns are observed in the lower temperature region of the N T B phase (Figure 2d) wherethe characteristic self-deformation stripes are clearly visible. In Figure 2d one observes thepolymer ”sponge” splitting between the two glass plates (the top and the bottom substratesof the cell) in three major ways: (i) all of the polymer remains on the imaged glass plate(marked as 100%), (ii) all of the polymer is attached to the second glass plate leaving the5 TB d) ITO 100% TB c) a) b) θ Figure 2: SEM images of planar sample polymerized in different phases. a) Isotropic phase,b) Nematic phase, c) Higher temperature uniform pattern in N T B phase, d) Striped patternin N T B phase. 6TO surface bare (marked as ITO), and, in most cases, (iii) the polymer structures splitsomewhat equally between the two substrates.Dimensions of the hierarchical structures were found to range from as low as a 8-12 nmhelix to sizes of the pattern comparable to the thickness of the cell, thus spanningover 3 orders of magnitude in size. Figure 3 shows a series of SEM micrographs depictingthe characteristic features of the planar-aligned cells under varying magnification.The polymer filaments have a finite thickness, and, therefore, impose a lower limit onthe resolution of the technique. Figure 3a shows that the filaments as thin as 20 nm can beobserved. Thus the resolution of our experiment is not sufficient to give a direct confirmationof the existence of a duplex double-helical DNA-like chains proposed from resonant X-raystudies.
However, it is reasonably possible to envisage that such chains may exist in thematerial. Moreover, Figures 3a, b show that axes of such chains (or other nm-scale units)are not straight and, in their turn, form helical patterns on the scale of 100-200 nm. Thoughit is difficult to detect more than a couple of undistorted periods (highlighted by green ovalin Figure 3b) at this scale due to a relatively low order and the influence of larger-scalepatterns. These are likely to correspond to the ∼ . µ m correlation length reported fromX-ray experiments. At the sample scale, the average direction of the polymer fibres forms an angle α withthe rubbing direction (Figure 3b, 3c, 3d). Once the stripes are formed on cooling, α reachessaturated values of 40-45 ◦ . Stripes of double cell gap periodicity are formed with neighboringbands tilted with an opposite sign of α (Figure 2d, 3c, 3d). Thus the self-deformation periodicpatterns can be easily identified and compared with textures observed by other methods.Figures 3b and 3c provide details of the apparent ”rope-like textures” visible under certainconditions in thicker cells by POM and the Fluorescent Confocal Polarizing Microscopy(FCPM). We note that the filaments form ”fish-bone” patterns with submicron to micronperiodicity (Figure 3c), caused most likely by intermodulation of the helical patterns withfiner (100-200 nm) periodicities (Figure 3b). A helical structure with a micrometer-scale7igure 3: Hierarchy of the polymerized N T B patterns in planar cells. Yellow arrow repre-sents cell rubbing direction. a) High-magnification SEM image of polymer filaments. b)Sub-micrometre scale pattern. Part of a self-deformation stripe is visible. Average fibredirection (red dashed line) forms a large angle α with the rubbing direction. Green ovalhighlights submicron helical patterns c) Self-deformation stripes in thicker cells. Elementsproviding impression of ”rope-like textures” are highlighted by orange ovals. Color insert -corresponding FCPM image scaled from Figure 2c in. d) Possible boundary between do-mains with opposite handedness. Most of the polymer adheres to one glass plate. Colorinsert - fragment of corresponding POM texture scaled from Figure 3a in.
110 115 120 125 130-1.0-0.50.00.51.01.52.0
Striped N TB m=5, n=9 Uniform N TB N P ho t ode t e c t o r s i gna l ( m V ) Temperature ( ° C) First harmonic Second harmonic3.9 µ m cell, cooling, 0.2 ° C/min Sine wave, 128 Hz, 2.5 V / µ m Figure 4: Temperature dependence of the electro-optic response of DTC5C9 (chemical struc-ture shown). Sign inversion of the first harmonic signal is visible in the temperature regioncorresponding to the onset of the stripes.As reported earlier, an application of the electric field reveals rather large (up to samplesize) structurally chiral domains switching in the opposite directions at a speed correspondingto the deformation of the nanometer-scale helix. The boundaries between such domainscan be identified in POM images by characteristic discontinuities in the striped patternspresent even in the absence of an electric field (see Figure 3a,b in ). In SEM images suchboundary areas are also identifiable by a reduced value of α . The polymer ”sponge” in thearea tends to stay on one of the glass plates rather than being split. These observationswill help in analysing a variety of domain boundary scenarios discussed by C. Meyer et. al. (Figure 7 in ). However, for complete understanding of the domain boundary formation,information about all levels of hierarchy is required. Knowledge of the local heliconical angleof the nano-scale pitch is needed. This angle may differ from the fish-bone angle α beingdiscussed here.Far away from the domain boundary areas, the angle α is onset to an initial value of95-30 ◦ (Figure 2c) at the N to N T B phase transition and then increases rapidly on coolingto a saturated value of 40-45 ◦ which seems to be limited by the geometry of planar samples.This is in agreement with observations for pure CBC7CB (Figure 4b in ).Since in a conventional nematic phase the molecular director is parallel to the rubbingdirection (i.e. α = 0), one could expect existence of materials with a small ( < ◦ ) initialvalue of α in the N T B phase. If the average direction of the nanometer-scale helix axis followsthe trend defined by the angle α , then, in such materials, an increase in this angle towards itssaturation value will cause sign inversion in the first harmonic of the electro-optical response.(See Section 2.2 of Ref. for details of the experiment).Though the sign inversion is not detected in the mixture of CBC7CB with 5CB usedfor SEM imaging presented above, this hypothesis is confirmed for at least two materialsof the di-fluoroterphenyl dimer series (Figure 4, inset) with m = 5, n = 7 and m = 5, n= 9 (the compounds DTC5C7 and DTC5C9). When a planar cell is positioned betweenthe crossed polarizers with the rubbing direction at 22.5 ◦ to the polarizer direction, theoptical response at the fundamental frequency (1 f ) of the applied field is proportional to sin (4(22 . ◦ ± α )) and its second harmonic (2 f ) is proportional to cos (4(22 . ◦ ± α )) (seesupplementary information). Therefore, at some point close to the onset of self-deformationstriped pattern, the first harmonic will cross the zero value. This sign reversal is clearlyvisible in the data set presented in Figure 4. Therefore the initial value of α at the transitionto the N T B phase on cooling is below 22.5 ◦ . This is in perfect agreement with the values forthe director tilt obtained by NMR (Figure 11 in ). The second harmonic signals generatedby the change in α in the bands with opposite tilt will be canceled out, thus the 2 f signalobserved is mainly generated by the change in birefringence. This indicates that not only thein-plane position of the optical axis, but also the birefringence is varying with the appliedfield, i.e. the axis of the nano-scale helix may significantly deviate from the plane of the cell.The absence of sign reversal of the response in the 20/80% w/w mixture of 5CB andCB-C7-CB confirms that the value of α stays between 22.5 ◦ and 45 ◦ over the entire N T B ∅ N T B phase scaling from 20 nmto several µ m and larger are revealed.We rule out the presence of the ”rope-like” micrometre-scale helix in favour of the ”fish-bone” structure. The structure is formed by a more dense irregular helix-like pattern(s) of ∼
100 nm size. Those, in turn, are formed by ∼
10 times smaller helical arrangements ofmolecules reported earlier.
We also note that the angle between the rubbing direction and the average filament axisis close to 40-45 ◦ in most cases. This finding has led us to propose the existence of materialswith sign reversal of the electro-optic response. An example of such a material is given.Apart from the clear potential in revealing of complex liquid crystalline structures, thephoto polymerization technique in combination with the unique properties of the twist-bendphase will open a novel platform for applications of micro- and nano-technology of advancedfunctional materials. Supporting Information Available
The following files are available free of charge. Additional high-resolution images andmathematical details are available in supplementary information.
Acknowledgement
The work as supported by 13/US/I12866 from the Science Foundation Ireland as part of11he US-Ireland Research and Development Partnership program jointly administered withthe United States National Science Foundation under grant number NSF-DMR-1410649.Financial support for the Hull group was from FP7 EU Grant No. 216025 and the EPSRCproject EP/M015726.
References (1) Goodby, J. W.; Collings, P. J.; Kato, T.;Tschierske, C.; Gleeson, H. F.; Raynes. P.;Handbook of Liquid Crystals, 2nd Edition, (Wiley-VCH, Weinheim 2014) ISBN: 978-3-527-32773-7.(2) Ackerman P. J.;Smalyukh, I. I.;
Phys. Rev. X , , 7, 011006.(3) Shamid, S. M.; Dhakal, S.; Selinger, J. V.; Phys. Rev. E , , 87 052503.(4) Dierking, I; Archer, P; RSC Adv., , , 3, 26433.(5) Akagi, K.; Chem. Rev. , 109, 5354-5401.(6) Lee Y.-J., Yu, C.-J.; Kim, Y.-K.; Jo, S. I.; Kim, J.-H.; Appl. Phys. Lett , , 98,033106.(7) Kang, S. W.; Sprunt, S.; Chien, L.-C.; Appl. Phys. Lett. , , 76, 24, 3516.(8) Varanytsia, A.; Chien, L.-C.; Scientific Reports , , 7, 41333.(9) Crawford, G. P.; ˘Zumer, S; Liquid Crystals in Complex Geometries formed by polymerand porous networks, (Taylor & Francis Ltd., 1996) ISBN:0-7484-0464-3.(10) Labeeb, A.; Gleeson, H. F.; Hegmann, T.; Appl. Phys. Lett. , , 107, 232903.(11) Panov, V. P.; Varney, M. C.; Smalyukh, I. I.; Vij, J. K.; Tamba, M. G.; Mehl, G. H.; Mol. Cryst. Liq. Cryst. , , 611, 180-185.1212) Tripathi, C. S. P.; Losada-Perez, P.; Glorieux, C.; Kohlmeier, A.; Tamba, M.-G.; MehlG. H.; Leys, J.; Phys. Rev. E , , 84, 041707.(13) Tamba, M.-G.; Yu, C. H.; Tang, B. J.; Welch, C.; Kohlmeier, A.; Schubert C. P.; Mehl,G. H.; Materials , , 7, 3494-3511.(14) Panov, V. P.; Balachandran, R.; Nagaraj, M.; Vij, J. K.; Tamba, M. G.; Kohlmeier,A.; Mehl, G. H.; Appl. Phys. Lett. , , 99, 261903.(15) Meyer, C.; Luckhurst, G. R.; Dozov, I.; Phys. Rev. Lett. , , 111, 067801.(16) Panov, V. P.; Balachandran, R.; Vij, J. K.; Tamba, M. G.; Kohlmeier, A.; Mehl, G.H.; Appl. Phys. Lett. , , 101, 234106.(17) Sreenilayam, S. P.; Panov, V. P.; Vij, J. K.; Shanker, G.; Liquid Crystals , , 44, 1,244-253.(18) Borshch, V.; Kim, Y.-K.; Xiang, J.; Gao, M.; Jakli, A.; Panov, V. P.; Vij, J. K.; Imrie,C. T.; Tamba, M. G.; Mehl, G. H.; Lavrentovich, O. D.; Nature Communications , , 4, 2635-2643.(19) Chen, D.; Porada, J. H.; Hooper, J. B.; Klittnick, A.; Shena, Y.; Tuchband, M. R.;Korblova, E.; Bedrov, D.; Walba, D. M.; Glaser M. A.;, Maclennan J. E.; Clark N. A.;, PNAS , , 15931-15936.(20) Panov, V. P.; Nagaraj, M.; Vij, J. K.; Panarin, Yu. P.; Kohlmeier, A.; Tamba, M. G.;Lewis, R. A.; Mehl, G. H.; Phys. Rev. Lett. , , 105, (16):167801.(21) Tuchband, M. R.; Shuai, M.; Graber, K. A.; Chen, D.; Zhu, C.; Radzihovsky, L.;Klittnick, A.; Foley, L. M.; Scarbrough, A.; Porada, J. H.; Moran, M.; Yelk, J.; Bedrov,D.; Korblova, E.; Walba, D. M.; Hexemer, A.; Maclennan, J. E.; Glaser, M. A.; Clark,N. A.; arXiv:1703.10787 [cond-mat.soft].1322) Salamo´nczyk, M.; Vaupotiˇc, N.; Pociecha, D.; Wang, C.; Zhu, C.; Gorecka. E.;arXiv:1703.08007 [cond-mat.soft].(23) Cestari, M.; Diez-Berart, S.; Dunmur, D. A.; Ferrarini, A.; de le Fuenta, M. R.; Jackson,D. J. B.; Lopez, D. O.; Luckhurst, G. R.; Perez-Jubindo, M. A.; Richardson, R. M.;Salud, J.; Timmi, B. A.; Zimmermann, H.; Phys. Rev. E , , 84 031704.(24) Agra-Kooijman,D. M.; Singh, G.; Fisch, M. R.; Vengatesan, M. R.; Song, J.-K.; Kumar,S.; Liquid Crystals , , 44, 1, 191-203.(25) Panov, V. P.; Vij, J. K.; Balachandran, R.; Borsch, V.; Lavrentovich, O. D.; Tamba,M. G.; Mehl, G. H.; Proc. SPIE 8828, Liquid Crystals XVII, , , 88280X.(26) Meyer, C.; Luckhurst G. R.; Dozov, I.; J. Mater. Chem. C , , 3, 318.(27) Stevenson, W. D.; Ahmed, Z.; Zeng, X. B.; Welch, C.; Ungar, G.; Mehl, G. H.; PCCP , , 19, 13449-13454(28) Emsley, J. W.; Lelli, M.; Joy, H.; Tamba M.-G.; Mehl, G. H.; Phys.Chem.Chem.Phys. , , 18, 9419. 14 raphical TOC Entryraphical TOC Entry