# Fileset

[Manuscript.docx](https://mdr.nims.go.jp/filesets/d21cb0f7-6645-472f-b1db-0749fcb5c2ad/download)

## Creator

Mohamed Alaasar, Ahmed F. Darweesh, Yu Cao, [Konstantin Iakoubovskii](https://orcid.org/0000-0002-7610-4717), [Masafumi Yoshio](https://orcid.org/0000-0002-1442-4352)

## Rights

[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Electric field- and light-responsive oxadiazole bent-core polycatenar liquid crystals](https://mdr.nims.go.jp/datasets/bf82d135-5679-4752-ba6e-0854187906b0)

## Fulltext

Electric Field- and Light-Responsive Oxadiazole-Bent-Core Polycatenar Liquid CrystalsMohamed Alaasar,*a Ahmed F. Darweesh,b Yu Cao,c Konstantin Iakoubovskii,d Masafumi Yoshiod,e aInstitute of Chemistry, Martin Luther University Halle-Wittenberg, 06120 Halle, GermanybDepartment of Chemistry, Faculty of Science, Cairo University, 12613 Giza, EgyptcShaanxi International Research Center for Soft Matter, State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an 710049, P. R. ChinadResearch Center for Macromolecules & Biomaterials, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, JapaneJapan Science and Technology Agency, PRESTO, 4-1-8 Honcho, Kawaguchi, Saitama, 332-0012, JapanAbstractAlkyl chain engineering of polycatenar liquid crystals (LCs), which consist of a rigid aromatic core with flexible chains at both ends, serves as an effective method for controlling their self-assembled nanostructures. Incorporating stimuli-responsive molecular units into the design of polycatenars represents a promising strategy for developing novel functional LC materials. In this study, we present the design and synthesis of new oxadiazole-based bent-core tetracatenar LCs with alkoxy azobenzene terminals. These compounds exhibit the formation of 3D bicontinuous cubic (Cubbi) or columnar phases, with variations observed by altering the chain length. The shortest homologue displays a Cubbi phase of the double gyroid type (Cubbi/Iad) and a triple-network phase with I23 space group (Cubbi/I23). Longer homologues form a hexagonal columnar LC phase. Upon introducing oxadiazole into the molecular core, we observed the alignment of columnar LC phases during cooling from isotropic liquids under an applied electric field. Furthermore, the incorporation of fluorinated azobenzene facilitated orientation switching within the columnar LC phases. Through UV irradiation, we successfully modified the LC phase structure via the reversible trans-to-cis photoisomerization of the azobenzene units. These findings underscore the potential applications of the reported materials in information-storage devices.Keywords: Bent-core liquid crystals; polycatenars; oxadiazole; ferroelectricity; azobenzene; columnar phases1. IntroductionThe search for new materials that can be used for optoelectronic and electronic devices, such as field-effect transistors, light-emitting diodes and solar cells, is a hot topic that attracts a wide range of materials scientists.[[endnoteRef:1],[endnoteRef:2],[endnoteRef:3],[endnoteRef:4],[endnoteRef:5]] Organic materials for such applications have several advantages compared to inorganic ones, because they are relatively abundant, non-toxic, inexpensive and can be easily modified.[3,[endnoteRef:6]] Organic liquid crystals (LCs) represent one important class of such promising materials in different renewable technological applications.[[endnoteRef:7]] LCs are characterized by their ability to respond to external stimuli, such as electric field, resulting in tailor-made nanostructures. Photo-responsive LCs derived from azobenzene-based materials exhibit reversible trans-to-cis photoisomerization. They have been studied for several decades [[endnoteRef:8],[endnoteRef:9]], and still attract researchers as they can exhibit macroscopic changes upon atomic-scale modifications.[[endnoteRef:10],[endnoteRef:11],[endnoteRef:12],[endnoteRef:13],[endnoteRef:14],[endnoteRef:15]] Columnar LCs can provide high and anisotropic charge-carrier mobility, which is required for  electronic applications. This is because in a columnar phase, each column provides a one-dimensional path for charge transport through its central part, which is composed of rigid molecular cores. Meanwhile, the flexible terminal chains of the molecules isolate those columns as a result of the nano-segregation of incompatible molecular subunits.[[endnoteRef:16],[endnoteRef:17]] Different molecular structures exhibit columnar LC phases that have been tested for optoelectronic applications, such as organic light emitting diodes [[endnoteRef:18],[endnoteRef:19],[endnoteRef:20],[endnoteRef:21]] and field-transistors.[[endnoteRef:22],[endnoteRef:23]]   B. Roy, N. De and K. C. Majumdar, Chem. - A Eur. J., 2012, 18, 14560–14588. S. Laschat, A. Baro, N. Steinke, F. Giesselmann, C. Hägele, G. Scalia, R. Judele, E. Kapatsina, S. Sauer, A. Schreivogel and M. Tosoni, Angew. Chemie - Int. Ed., 2007, 46, 4832–4887. U. H. F. Bunz and J. Freudenberg, Acc. Chem. Res., 2019, 52, 1575–1587. S. Kumar, Chemistry of discotic liquid crystals: From monomers to polymers, CRC Press, 2016. N. Tober, T. Rieth, M. Lehmann and H. Detert, Chem. – A Eur. J., 2019, 25, 15295–15304 V. Balzani, G. Bergamini and P. Ceroni, Angew. Chemie - Int. Ed., 2015, 54, 11320–11337. T. Kato, J. Uchida, T. Ichikawa and T. Sakamoto, Angew. Chem. Int. Ed., 2018, 57, 4355. A. Natansohn and P. Rochon, Chem. Rev., 2002, 102, 4139–4175. H. M. D. Bandara and S. C. Burdette, Chem. Soc. Rev., 2012, 41, 1809–1825. A. L. Leistner and Z. L. Pianowski, European J. Org. Chem., 2022, e202101271.  G. A. Leith, C. R. Martin, A. Mathur, P. Kittikhunnatham, K. C. Park and N. B. Shustova, Adv. Energy Mater., 2022, 12, 2100441. H. Sugiyama, S. Sato and K. Nagai, Polym. Adv. Technol., 2022, 33, 2113–2122. B. Zhang, Y. Feng and W. Feng, Nano-Micro Lett., 2022, 14, 138. I I. Dominguez-Candela, I. Zulkhairi, I. Pintre, N. F. K. Aripin, J. Lora-Garcia, V. Fombuena, M. B. Ros and A. Martinez-Felipe, J. Mater. Chem. C, 2022, 10, 18200–18212. M. Alaasar, Liq. Cryst., 2016, 43, 2208–2243. D. Adam, F. Closs, T. Frey, D. Funhoff, D. Haarer, P. Schuhmacher and K. Siemensmeyer, Phys. Rev. Lett., 1993, 70, 457–460. D. Adam, P. Schuhmacher, J. Simmerer, L. Häussling, K. Siemensmeyer, K. H. Etzbachi, H. Ringsdorf and D. Haarer, Nature, 1994, 371, 141–143. O. Kasdorf, J. Vollbrecht, B. Ohms, U. Hilleringmann, H. Bock and H. S. Kitzerow, Int. J. Energy Res., 2014, 38, 452–458. J.-F. Wang, Y. Kawabe, S. E. Shaheen, M. M. Morrell, G. E. Jabbour, P. A. Lee, J. Anderson, N. R. Armstrong, B. Kippelen, E. A. Mash and N. Peyghambarian, Adv. Mater., 1998, 10, 230–233. A. K. Yadav, B. Pradhan, H. Ulla, S. Nath, J. De, S. K. Pal, M. N. Satyanarayan and A. S. Achalkumar, J. Mater. Chem. C, 2017, 5, 9345–9358. K. Kotwica, A. S. Kostyuchenko, P. Data, T. Marszalek, L. Skorka, T. Jaroch, S. Kacka, M. Zagorska, R. Nowakowski, A. P. Monkman, A. S. Fisyuk, W. Pisula and A. Pron, Chem. - A Eur. J., 2016, 22, 11795–11806. A. M. Van De Craats, N. Stutzmann, O. Bunk, M. M. Nielsen, M. Watson, K. Müllen, H. D. Chanzy, H. Sirringhaus and R. H. Friend, Adv. Mater., 2003, 15, 495–499. W. Pisula, A. Menon, M. Stepputat, I. Lieberwirth, U. Kolb, A. Tracz, H. Sirringhaus, T. Pakula and K. Müllen, Adv. Mater., 2005, 17, 684–689.A large variety of unconventional molecular designs that differ from traditional calamitic or discotic LCs systems have been designed and reported.[[endnoteRef:24]] Polycatenar LCs having multiple flexible terminal chains are an important class of such unconventional structures.[24] Recently, it was found that polycatenar LCs with asymmetrical terminal chains with respect to the rod-like aromatic core could exhibit interesting nano-structured LC phases. Those include bicontinuous cubic (Cubbi) phases with I23 symmetry, which lack a mirror symmetry and were initially assigned as Imm [[endnoteRef:25]], as well as an achiral version with Iad space group.[[endnoteRef:26]] These Cubbi phases are composed of three-dimensional networks with three-way junctions, where the double gyroid (Cubbi/Iad) contains two of such networks with opposite chiralities, while three helical networks are present in the Cubbi/I23 phase. Chirality was even observed in the isotropic liquid networks of these molecules.[[endnoteRef:27]] Different aromatic core units were used to design such asymmetric polycatenars, including π-conjugated 5,5’-diphenyl-2,2’-bithiophene,[[endnoteRef:28],[endnoteRef:29],[endnoteRef:30],[endnoteRef:31]] π-conjugated [1]benzothieno[3,2-b]benzothiophene [[endnoteRef:32],[endnoteRef:33]] and azobenzene  moiety.[[endnoteRef:34],[endnoteRef:35],[endnoteRef:36],[endnoteRef:37],[endnoteRef:38]] Only in a few studies asymmetric hockey-stick polycatenars were designed using 2,5-disbstituted thiophene [[endnoteRef:39]] or 4-cyanoresorcinol as bent-core units.[[endnoteRef:40],[endnoteRef:41]] None of these materials had polar LC phases, which are usually exhibited by bent-core mesogens; instead, non-polar lamellar smectic A, Cubbi or columnar phases were observed. A promising unit for designing bent-core (banana) LCs is the ~135° bent 1,3,4-oxadiazole core, which can form biaxial nematic,[[endnoteRef:42],[endnoteRef:43],[endnoteRef:44],[endnoteRef:45],[endnoteRef:46]] polar and chiral nematic and smectic phases.[[endnoteRef:47],[endnoteRef:48],[endnoteRef:49],[endnoteRef:50],[endnoteRef:51],[endnoteRef:52],[endnoteRef:53],[endnoteRef:54],[endnoteRef:55]] It has also been used for designing star-shaped fluorescent columnar LC phases.[[endnoteRef:56],[endnoteRef:57],[endnoteRef:58],[endnoteRef:59],[endnoteRef:60],[endnoteRef:61]]  Textbook X. Zeng and G. Ungar, J. Mater. Chem. C., 2020, 8, 5389–5398. Y. Cao, M. Alaasar, A. Nallapaneni, M. Salamończyk, P. Marinko, E. Gorecka, C. Tschierske, F. Liu, N. Vaupotič and C. Zhu, Phys. Rev. Lett., 2020, 125, 027801. C. Dressel, F. Liu, M. Prehm, X. B. Zeng, G. Ungar and C. Tschierske, Angew. Chem. Int. Ed. 2014, 53, 13115–13120. C. Dressel, T. Reppe, S. Poppe, M. Prehm, H. Lu, X. Zeng, G. Ungar and C. Tschierske, Adv. Funct. Mater., 2020, 30, 2004353. T. Reppe, S. Poppe, X. Cai, Y. Cao, F. Liu and C. Tschierske, Chem. Sci., 2020, 11, 5902–5908. M. Alaasar, A. F. Darweesh, X. Cai, Y. Cao, F. Liu and C. Tschierske, Chem. Eur. J., 2021, 27, 14921–14930. M. Alaasar, A. F. Darweesh, C. Anders, K. Iakoubovskii, M. Yoshio, Mater. Adv., 2023, in press. O. Kwon, X. Cai, W. Qu, F. Liu, J. Szydlowska, E. Gorecka, M. J. Han, D. K. Yoon, S. Poppe and C. Tschierske, Adv. Funct. Mater., 2021, 31, 2102271. O. Kwon, X. Cai, A. Saeed, F. Liu, S. Poppe and C. Tschierske, Chem. Commun., 2021, 57, 6491–6494. M. Alaasar, S. Poppe, Q. Dong, F. Liu and C. Tschierske, Angew. Chem. Int. Ed., 2017, 56, 10801-10805. M. Alaasar, S. Poppe, Y. Cao, C. Chen, F. Liu, C. Zhu and C. Tschierske, J. Mater. Chem. C., 2020, 8, 12902–12916.  M. Alaasar, M. Prehm, Y. Cao, F. Liu and C. Tschierske, Angew. Chem. Int. Ed., 2016, 128, 320–324.  M. Alaasar, Y. Cao, Y. Liu, F. Liu and C. Tschierske, Chem.–A Eur. J., 2022, 28, e202201857. Y. Cao, M. Alaasar, L. Zhang, C. Zhu, C. Tschierske and F. Liu, J. Am. Chem. Soc., 2022, 144, 6936-6945. J. Matraszek, D. Pociecha, N. Vaupotič, M. Salamończyk, M. Vogrin and E. Gorecka, Soft Matter, 2020, 16, 3882–3885. M. Alaasar and S. Poppe, J. Mol. Liq., 2022, 351, 118613. M. Alaasar, X. Cai, Y. Cao and F. Liu, New J. Chem., 2022, 46, 15871–15881. T. J. Dingemans and E. T. Samulski, Liq. Cryst., 2000, 27, 131–136. L. A. Madsen, T. J. Dingemans, M. Nakata and E. T. Samulski, Phys. Rev. Lett., 2004, 92, 145505. B. R. Acharya, A. Primak and S. Kumar, Phys. Rev. Lett., 2004, 92, 145506. F. Vita, T. Tauscher, F. Speetjens, E. T. Samulski, E. Scharrer and O. Francescangeli, Chem. Mater., 2014, 26, 4671–4674. H. F. Gleeson, S. Kaur, V. Görtz, A. Belaissaoui, S. Cowling and J. W. Goodby, ChemPhysChem, 2014, 15, 1251–1260. V. Görtz and J. W. Goodby, Chem. Commun., 2005, 3262–3264. M. Nagaraj, K. Usami, Z. Zhang, V. Görtz, J. W. Goodby and H. F. Gleeson, Liq. Cryst., 2014, 41, 800–811. M. Nagaraj, J. C. Jones, V. P. Panov, H. Liu, G. Portale, W. Bras and H. F. Gleeson, Phys. Rev. E - Stat. Nonlinear, Soft Matter Phys., 2015, 91, 042504. A. Belaissaoui, S. J. Cowling and J. W. Goodby, Liq. Cryst., 2013, 40, 822–830. N. A. Zafiropoulos, W. Lin, E. T. Samulski, T. J. Dingemans and S. J. Picken, Liq. Cryst., 2009, 36, 649–656. S. W. Choi, S. Kang, Y. Takanishi, K. Ishikawa, J. Watanabe and H. Takezoe, Chirality, 2007, 19, 250–254. S. Kang, Y. Saito, N. Watanabe, M. Tokita, Y. Takanishi, H. Takezoe and J. Watanabe, J. Phys. Chem. B, 2006, 110, 5205–5214. I. H. Chiang, C. J. Long, H. C. Lin, W. T. Chuang, J. J. Lee and H. C. Lin, ACS Appl. Mater. Interfaces, 2014, 6, 228–235. E. Westphal, H. Gallardo, N. Sebastián, A. Eremin, M. Prehm, M. Alaasar and C. Tschierske, J. Mater. Chem. C, 2019, 7, 3064–3081 J. Han, J. Mater. Chem. C, 2013, 1, 7779–7797. A. Paun, N. D. Hadade, C. C. Paraschivescu and M. Matache, J. Mater. Chem. C, 2016, 4, 8596–8610. S. K. Pathak, S. Nath, J. De, S. K. Pal and A. S. Achalkumar, New J. Chem., 2017, 41, 4680–4688. B. G. Kim, S. Kim, J. Seo, N. K. Oh, W. C. Zin and S. Y. Park, Chem. Commun., 2003, 3, 2306–2307. A. A. Vieira, E. Cavero, P. Romero, H. Gallardo, J. L. Serrano and T. Sierra, J. Mater. Chem. C, 2014, 2, 7029–7038. E. Westphal, M. Prehm, I. H. Bechtold, C. Tschierske and H. Gallardo, J. Mater. Chem. C, 2013, 1, 8011–8022.In this work, we have combined azobenzene and 1,3,4-oxadiazole to design and synthesize hockey-stick asymmetric polycatenars (Scheme 1). These molecules have the same mesogenic core terminated with three hexadecyloxy chains at one end and a variable alkoxy chain at the other end. The corresponding LCs show a transition from two different types of Cubbi to columnar phases, depending on the terminal chain length attached to the azobenzene-based core. The addition of oxadiazole to the molecular core allowed to polarize it by the application of external electric field to the isotropic phase. Additional fluorination of the aromatic core induced ferroelectricity in the columnar LC phase. The presence of azobenzene in the core allowed us to exploit its trans-to-cis photoisomerization, which resulted in reversible structural changes in the LC phase. All these mechanisms, combined in one molecule, lead to a versatile tunable nanostructure with potential optoelectronic applications. 2. Results and discussion2.1. Synthesis The synthesis of the oxadiazole-derived polycatenars (An, A12F3 and A12F23) is illustrated in Scheme 1 and is detailed in the Supporting Information. It started with the esterification of commercially available 4-benzyloxybenzoic acid (1) and pentafluorophenol (2), followed by reaction of the obtained intermediate 3 with 4-hydroxybenzoic acid hydrazide (4) to give compound 5. Compound 5 was cyclized to give the 1,3,4-oxadiazole heterocyclic protected intermediate 6 using thionyl chloride (SOCl2).[[endnoteRef:62]] The highly reactive acyl chloride of 3,4,5-trihexadeclyoxybenzoic acid (7)[[endnoteRef:63]], generated using SOCl2 and a catalytic amount of N,N-dimethylformamide (DMF), was used in acylation reaction of the phenol 6 to obtain the protected compound 8, which was unprotected via hydrogenation in presence of palladium on activated carbon. At the last synthesis stage, another acylation reaction was performed between the acyl chlorides of the different azobenzene-based benzoic acid derivatives 10n [36] with the phenol 9 to yield the final polycatenars. The crude materials were purified by column chromatography using CHCl3/ethylacetate (4:1) as an eluent, followed by recrystallization from chloroform/ethanol (77–90% yield). The materials were characterized by 1H and 13C nuclear magnetic resonance and elemental analysis (see Supplemental Information). All data agreed with the proposed molecular structures of the targeted compounds.    V. Gortz and J. W. Goodby, Chem. Commun., 2005, 3262–3264. D. H. Wang, Z. Shen, M. Guo, S. Z. D. Cheng, and F. W. Harris, Macromolecules 2007, 40, 889–900.Scheme 1. Synthesis of the target hockey-stick polycatenars An, A12F3 and A12F23. Reagents and conditions: i. N,N’-dicyclohexylcarbodiimide, dichloromethane (DCM) and 4-dimethylaminopyridine, stirring at room temperature (RT), 72 h; ii. DMF, stirring at RT, 48 h; iii. SOCl2, dry pyridine, reflux under argon atmosphere, 4 h; vi. SOCl2, dry pyridine, triethylamine, DCM, reflux under argon atmosphere, 4 h; v. Pd/C 10%, H2, dry tetrahydrofuran, stirring at RT, 24 h.  2.2. Characterization methodsPolarized optical micrographs (POMs) were acquired under the crossed Nicol’s condition. Differential scanning calorimetry (DSC) traces were recorded with DSC-7 and DSC-8000 Perkin Elmer setups, in Ar flow, at a heating or cooling rate of 10 K min−1. Small-angle X-ray scattering (SAXS) experiments were conducted at the BL16B1 beamline of the Shanghai Synchrotron Radiation Facility (SSRF). For powder SAXS, samples were filled into evacuated 1 mm diameter capillaries. Data calibration was conducted using silver borohydride and a series of n-alkanes. The conversion of 2D data to 1D plots was carried out using the Gauss equation implemented in the Irena and Nika macro packages within Igor64 software. After indexing and integrating SAXS peaks, electron density maps were reconstructed via Fourier transform as Comment by Iakoubovskii: I removed the 2D case because it follows from 3D case, and we might wish to delete 2D maps anyway ..For centrosymmetric structures with , the phase  is either 0 or , and hence all possible phase permutations can be tried. The best combination is decided by the physical merit of reconstructed electron density map and by other details, like the volume ratio of aromatic/aliphatic region and electron density distribution histogram. This method works well for liquid crystal systems exhibiting a small number of SAXS peaks.  Comment by Iakoubovskii: Unnecessary comment that may stimulate unneeded questions about the I23 LC phase. Samples for electro-optical measurements were prepared by filling commercial cells from EHC, Japan. The cells consisted of two 0.7-mm-thick glass plates coated with 20 nm of indium tin oxide. Epoxy spacers separated them by a distance of 4 µm (cells with a spacing of 2 and 9 were also tried, but the results were inferior, and hence were omitted). The overlapping area between the top and bottom electrodes was 4×4 mm. Cells were filled in vacuum, in darkness, at a temperature 5 °C above the isotropic point, within 1 hour. A small amount of powdered material was placed on the edge of a cell, where it melted, and was driven into the cell by the capillary force.   2.3. Thermal behavior: DSC and POMCombined POM and DSC results for An compounds are summarized in Table 1 and Figures S9-S10.  Table 1. Phase transitions of compounds An, A12F3 and A12F23.a Name X Y Phase transitions T/ºC [ΔH/kJ mol-1] A8 H H H: Cr 105 [51.9] Cub/I23[*] 114 [< 0.1] Cubbi/Iad 126 [1.3] IsoC: Iso 112 [-0.7] Colhex 111 [< 0.1] Cubbi/Iad 101 [1.1] Cubbi/I23[*] 80 [-62.3] Cr  A10 H H H: Cr 104 [73.2] Colhex 144 [1.3] IsoC: Iso 141 [-1.3] Colhex 76 [-71.5] Cr A12 H H H: Cr 100 [83.3] Colhex 135 [1.3] IsoC: Iso 131 [-1.3] Colhex 71 [-81.7] Cr A14 H H H: Cr 101 [55.5] Colhex 137 [1.5] IsoC: Iso 133 [-1.4] Colhex 76 [-58.8] Cr A16 H H H: Cr 84 [76.4] Colhex 139 [1.8] IsoC: Iso 135 [-1.5] Colhex 54 [-88.3] Cr A18 H H H: Cr 85 [136.8] Colhex 135 [2.0] IsoC: Iso 132 [-2.2] Colhex 56 [-90.0] Cr  A12F3 F H H: Cr 107 [58.1] Colhex 142 [1.6] IsoC: Iso 139 [-1.6] Colhex 65 [-66.8] Cr A12F23 F F H: Cr 115 [51.9] Colhex 121 [0.8] IsoC: Iso 118 [-0.9] Colhex 85 [-47.7] Cr[*]aPeak temperatures as determined by DSC upon 2nd heating (H:) and 2nd cooling (C:) with a rate of 10 K min–1. Abbreviations: Cr = crystalline solid phase; Iso = isotropic liquid phase; Colhex = hexagonal columnar phase; Cubbi/Iad = achiral bicontinuous cubic phase with Iad space group; Cubbi/I23[*] = chiral bicontinuous cubic phase with I23 space group.Initial phase characterization was performed using POM, both on heating and cooling. The phase behavior of the shortest homologue with n = 8 (A8) is different from other An members, and therefore it is discussed in detail. Upon cooling of A8 from the isotropic liquid state, a birefringent texture is observed under crossed polarizers (Figure S9a). This fan-shaped texture is fluid and shearable, which is typical for a columnar LC phase (Col). It exists in a narrow temperature range of ~1 K and is replaced by a highly viscous and completely dark phase (Fig. S9b), which persists until crystallization at ~80 °C. When one of the polarizers is slightly rotated either clockwise or anticlockwise, no dark and bright domains could be detected, while they are typically observed for a nanostructured double gyroid (Cubbi/Iad). On heating the birefringent crystalline state of A8 only the isotropic phase could be observed, indicating that the columnar phase is metastable. This observation agrees with the DSC heating scan (see Figure S10), where only one transition around 126 °C is observed besides the melting peak. In the DSC cooling scan, the isotropic-columnar transition could be detected at ~112 °C, followed by another transition at ~ 101 °C. For this transition, no structural change in the isotropic texture could be detected by POM, no matter the cell thickness and cooling rate. SAXS measurements revealed that the 101 °C transition leads to the second nanostructured Cubbi phase with I23 space group (Cubbi/I23), which usually do display chiral domains.[27,28,29,30,36] This represents the second case when macroscopic conglomerates were not detected in the I23 phase of a polycatenar material, following our recent report on linear polycatenars.[37] Owing to the spontaneous formation of two cubic phases (Cubbi/Iad and Cubbi/I23), the cubic domains are small and random. This leads to the absence of chiral conglomerates that could be detected by POM. Continuous heating and cooling scans were preformed to confirm the A8 phases. Two cubic phases were indexed, and the electron density map was reconstructed as shown in Figure 1. At 102 oC upon heating, a complex scattering pattern was observed and indexed as a mixture of two cubic LC and one crystal phases (Fig. S12a). The (211) Cubbi/Iad  peak becomes stronger, and both cubic phases coexist until 114 oC, when the Cubbi/Iad phase remains (Figure 1a). The isotropic state is observed at 126 oC. The reconstructed electron density map in Figure 1c suggests that the Cubbi/Iad phase is a meso-structure comprising two interwoven continuous networks with opposite handedness, which are conjugated by three-way junctions. The aromatic cores construct networks with high electron density. Flexible alkyl chains fill the rest of the lattice. Due to the asymmetric and twisted nature, the molecules first combine into molecular rafts that lie perpendicularly to the networks. Then the rafts twist along the networks, forming molecular helices of opposite handedness and cancelling the chirality (Figure 1e). Such molecular twists would also cancel the mismatch at the three-way junctions, and thus should be energetically favored. The lattice parameter of this cubic phase is 12.83 nm, and the number of molecules per unit cell is ~900. Considering the average spacing between molecules as 0.45 nm and the network period Lnet = 8.485aIa3d  ≈ 109 nm,[37] there are ~109/0.45 ≈240 molecular rafts in the lattice with about 900/240 ~ 3.7 molecules per raft as shown in Figure 1e. Given the fixed dihedral angle between neighboring junctions of 2×asin(1/√3) = 70.5o,[25] we can estimate the twist angle between rafts as 70.5°/(0.354acub/0.45nm) ≈ 7o.[25]Upon cooling from isotropic phase, a hexagonal columnar phase was first observed with a lattice parameter of 5.82 nm, leading to columns of about 5 molecules. This columnar phase swiftly converts into a I23 cubic phase (Figure 1b). Such a non-centrosymmetric space group provides the opportunity for supramolecular chirality while retaining the local chemical environment (three-way junctions).  In this phase, instead of a meso-structure, three networks were formed with molecular helices aligned as shown in Figure 1d, f.  In this arrangement, the chirality from molecular helices is not cancelled by the networks as in the Iad phase. The unit cell of this phase has a length of 20.24 nm and contains ~3525 molecules. The corresponding network has a length  Lnet ≈ 20.68aI23 ≈ 420 nm and contains ~420/0.45 ≈ 930 molecular rafts with ~3.8 molecules per raft, similar to the Iad phase. The twist angle between neighboring rafts is 90°/(0.290acub/0.45nm) ~ 6.9o.[37] It is slightly smaller compared to the Iad phase because of the lower temperature range for the I23 phase. The coherence length of the I23 phase was estimated as 190 nm from the width of SAXS peaks. Such a small value explains the absence of chiral conglomerates in POM images.Figure 1. SAXS data of A8: a) Cubbi/Iad phase observed at 114 °C upon heating and b) Cubbi/I23 phase at 100 °C upon cooling. Electron density maps of the c) Cubbi/Iad and d) Cubbi/I23 phases; high electron density is in purple and low electron density (red) is omitted for clarity. Geometric model of the e) Cubbi/Iad and f) Cubbi/I23 phases; molecular rafts are represented by green rods.2.3.2. Hexagonal columnar phasesAn compounds with n ≥ 10 exhibited only columnar phases, and their temperature ranges (~35–55 K depending on n, see Figures S9c,d for examples) are much wider compared to A8. As can be seen from Table 1, in the DSC heating and cooling traces of compounds A10–A18 only one sharp transition (Col-Iso) is observed. Its enthalpy increases with n from 1.3 to 2.0 kJ mol-1. The melting temperatures are almost the same for A10-A14; they decrease for A16 and A18, resulting in wider ranges of columnar phase in the last two cases.As illustrated in Figure S13, SAXS patterns for all A10–A18 compounds exhibited four major peaks whose positions scaled as 1:. Such a pattern strongly implies a presence of a single hexagonal p6mm phase.  2.4. Polarization by electric fieldWe have introduced oxadiazole to the core of the studied molecules with the aim to polarize them by applied electric field, relying on the relatively large dipole moment of oxadiazole (3.04 D [[endnoteRef:64]]). Such polarization was achieved, as demonstrated for the columnar phase of A12 in Figure 2. In particular, application of a modest DC electric field of 5 V µm‑1, normal to the sandwich-type sample, darkened the area between the electrodes (Fig. 2a), while the areas  outside the electrodes retained their birefringence (Fig. 2b).  b)a) K. Iakoubovskii, M. Yoshio, Chem. Commun., 2023, 59,7443–7446.  Figure 2. Columnar phase of A12 viewed between crossed polarizers at 118 °C, after cooling from the temperature above the isotropic point under 5 V µm-1 electric field: a) inside, b) outside the electrode area. Scale bar is 200 µm.  To achieve or alter this alignment, it was necessary to heat the sample above the isotropic transition temperature. The alignment was not observed at lower temperatures, presumably due to the high rigidity of the oxadiazole unit in the molecular core. To overcome this problem, we have created a polar and easily rotatable fluorobenzene group in the LC core. The resulting compounds A12F3 and A12F23 (Table 1) have the same chain length as A12 and different F substitution patterns. In A12F3, a fluorine atom is introduced at an ortho position with respect to the terminal alkoxy chain, while in A12F23, F substitutes for hydrogens both in the ortho and meta positions. Comparing these two materials with their nonfluorinated analogue A12, it could be seen from Table 1 that in both cases the melting temperature is increased, presumably due to enhanced core-core interactions. Therefore, the melting temperature of A12F23 (115 °C) is higher than that of the monofluorinated derivative A12F3 (107 °C). Both derivatives show the same Colhex phase observed in A12, as confirmed by POM and SAXS  (see Table 1 and Supplemental Information). Fluorination decreases the lattice parameter from 5.70 nm in A12, to 5.50 nm in A12F3, and 5.38 nm in A12F23. This decrease can be explained by the strong inductive effect of the fluorine, which reduces the electron density of the core and increases the core-core interaction.   d)c)b)a)Figure 3. Columnar phase of A12F23  viewed between crossed polarizers at 118 °C: a) at zero field, b) 1 minute after applying an electric field of 5 V μm-1 (scale bar 200 µm), c)  relaxation of dark current after switching off the field; d) "switching time" required to reduce the birefringence signal intensity twice, plotted vs. applied field.  After fluorination of A12, we have achieved switching of birefringence not only in the isotropic, but also in the columnar LC phase, as demonstrated on example of A12F23 in Figure 3: Application of the same DC electric field of  5 V µm‑1 to the LC phase completely erased the birefringence within one minute (Figure 3a,b). A current of ~5 µA was measured under the field; it quickly dropped by ~1000 times, and yet remained at the 1 nA level for almost one hour after switching off the field, revealing residual ferroelectric polarization (Fig. 3c). The "switching time",  which we arbitrarily defined  as the time to reduce the birefringence intensity twice, was relatively slow at 5 V µm‑1, but it could be reduced to under one second by doubling the field (see Fig. 3d).  Figure 4. Ferroelectric polarization at 118 °C under triangular voltage pulses with an amplitude of ±5 V μm‑1 for A12F23 and  ±19 V μm‑1 for A12 and A12F3.We have characterized the ferroelectric polarization in A12, A12F3 and A12F23 compounds not only under DC, but also under AC bias. Figure 4 summarizes a standard poling experiment, in which a triangular voltage was applied at a frequency of 5 Hz, and the switching current was recorded as a voltage drop over a 10 kΩ resistor. A clear polarization peak was observed for A12F23 at a field of 2-5 V µm‑1, but not for other materials. The polarization peak appeared in A12F3, but only at a bias exceeding 18 V µm‑1, and it was not observed in A12 even near the breakdown field of ~20 V µm‑1. These results confirm that the fluorobenzene fragments are responsible for the ferroelectric switching, and the lower threshold field in A12F23 can be naturally explained by the higher dipole moment of the difluorobenzene compared to fluorobenzene (2.46 vs. 1.60 D [[endnoteRef:65]]).  R. D. Nelson, D. R. Lide, A. A. Maryott, “Selected Values of Electric Dipole Moments for Molecules in the Gas Phase”. National Standard Reference Data Series 10, 1967. We have performed a standard impedance analysis for the AC switching behavior in A12F23 and summarized the results in the Supplemental Information. They reveal that application of a 0.175 V µm‑1 bias (which is the limit of our impedance analyzer) increased the dielectric constant of the columnar phase from 7 to 10 at 0.1 MHz, and that this change originated from the bulk rather than interfacial layers.2.5. Photochromism   Figure 5. Columnar phase of A12 viewed at 90 °C upon cooling between crossed polarizers (left) and parallel polarizers (right), before (top row) and after irradiation with 350 nm light (bottom row). Scale bar is 200 µm. a)b)Figure 6. a) Optical absorption spectra from A12: initial; right after the sample was irradiated with 350 nm light at 90 °C and cooled under UV light to 30 °C; and 12 hours after keeping the sample at 30 °C in the dark. b) Normalized kinetics of 450 nm absorption: increase under 350 nm light at 90 °C, and relaxation under visible illumination at 30 °C. The core of all the studied molecules contains azobenzene, which is known to undergo cis-to-trans transition under UV illumination. This transition has been widely studied in azobenzene-based LCs in a solution, where the intermolecular interactions are weak and do not hinder the photoisomerization. Meanwhile, photoisomerization studies in the solids are rare. Here we characterize photochromism in the LC state.When An LCs were irradiated by 350 nm light of 40 mW cm-2 intensity, their domain texture and birefringence disappeared within one second, as demonstrated in Figure 5 on example of A12. The birefringence would recover within one minute after switching off the light. The corresponding optical absorption spectra (Figure 6a) showed a reduction in the 370 nm band and an increase in the 450 nm absorption after UV illumination. The absorption completely recovered after storing the sample at ambient conditions in the dark overnight, or after illuminating the sample for one minute with bandpass-filtered visible light (490-760 nm, 100 mW cm-2). Although the optical absorption could be recovered at room temperature, the restoration of domain texture and birefringence required remelting the sample.The disappearance of LC domains under UV illumination reveals amorphization, which was confirmed by SAXS and DSC measurements (see Figure S17).  These results can be rationalized as follows: UV illumination changed the molecular shape via isomerization of the azobenzene groups. This modification changed the optical absorption spectra and resulted in semi-permanent amorphization. The optical spectra could be restored by one-minute irradiation with visible light. This phenomenon can be utilized in optical information-storage devices.3. Summary and conclusionsAcknowledgementsM. Alaasar acknowledges the German Research Foundation (DFG) for the financial support (AL2378/1-2). A. F. Darweesh acknowledges the support by the Alexander von Humboldt Foundation for the research fellowship at Martin Luther University Halle-Wittenberg, Germany.M. Yoshio gratefully acknowledges financial support from the Japan Society for the Promotion of Science (no. 21H02021), Japan Science and Technology Agency (no. JPMJPR23QB) and Iketani Science and Technology Foundation (no. 0351202-A).References1image2.emfOC16H33OC16H33OC16H33OONNOOH2n+1CnONNOXYimage3.pngimage4.jpegimage5.jpegimage6.jpegimage7.jpegimage8.pngimage9.pngimage10.pngimage11.jpegimage12.jpegimage13.jpegimage14.jpegimage15.pngimage16.pngimage1.emfBnOOHO+FFFFFHOi.BnOOOFFFFF123NHONH2HOBnONHOHNOOH45NNOOHBnO6OC16H33OC16H33OC16H33OC16H33OC16H33OC16H33OONNOBnO789H2n+1CnONNOC16H33OC16H33OC16H33OON NOOH2n+1CnONNOXY10nXYAn,X=Y= H; n = 8,10,12, 14, 16, 18A12F3,X=F;Y=H;n=12  A12F23, X=Y= F; n = 12iii.ii.OHOvi.OOHvi.OC16H33OC16H33OC16H33OONNOHOv.