Homogeneous alignment of liquid crystals using functionalized tungsten-oxide nanostructures: a route to improved electro-optical device performance Authors: Shikha Agarwal, Tarun Yadav, Bhupendra Pratap Singh, Santosh Y. Khatavi, C. V. Yelamaggad, Rajiv Manohar Published: 2026 ABSTRACT Alignment layers play a vital role in liquid crystal (LC) technologies as they not only control the initial orientation of LCs but also influence the key parameters like threshold voltage, response time etc., making them essential for high-quality optoelectronic applications. Conventional rubbed polyimide-based alignment layers exhibit poor ion-trapping, limited transparency and incompatibility with flexible substrates. Therefore, the quest for alternative options has become an important research field. In this study, we report the application of tungsten oxide nanostructures as an inorganic alignment layer for nematic LC devices, offering an effective alternative to conventional polyimide coatings. Precisely, chemically functionalized tungsten oxide monohydrate (WO3·H2O) nanodiscs are synthesized via the hydrothermal method. The WO3-based layers are fabricated on indium tin oxide (ITO)-coated glass substrates using a simple and versatile deposition technique. The LC molecules demonstrated uniform planar anchoring. Comprehensive dielectric and electro-optical characterization studies reveal that the LC cells with WO3 alignment layers exhibit significantly improved threshold voltage reduced by 41%, around 70% faster switching response, and enhanced dielectric anisotropy compared to their polyimide-based counterparts. The observed performance enhancement is attributed to the nanostructured surface morphology and favorable interfacial interactions between WO3 layers and LC molecules. These findings highlight the potential of tungsten oxide nanomaterials as functional alignment layers for high-performance and next-generation LC devices. Introduction Tungsten oxide (WO3) nanostructures, being highly stable 2D semiconductors, have emerged as versatile functional materials with wide applicability in optoelectronic,1,2 electrochromic,3 photocatalytic,4 and sensing domains,5 owing to their remarkable optical, electrical, and catalytic properties. Their excellent physicochemical attributes such as wide tuneable bandgap, high dielectric constant, and strong electro-chromic behaviour make them attractive candidates for applications ranging from smart windows and gas sensors to photonic and optoelectronic devices.6,7 WO3, being a transition metal oxide, exhibits these peculiar properties due to the partially-filled d-orbital in the tungsten atom.7 It exists in multiple polymorphic phases, including monoclinic, orthorhombic, and hexagonal with phase stability often governed by particle size and synthesis conditions.8 Nanostructured WO3—such as nanorods, nanoplates, nanowires, nanoparticles—can be synthesized through a variety of methods like the hydrothermal/solvothermal method,9,10 sol–gel processing,11 thermal evaporation,12 and acid precipitation,13 offering control over crystallinity and morphology that is critical for device performance. Smaller size nanostructures are preferred due to their higher surface area-to-volume ratio, offering more active sites and resulting in superior light absorption, enhanced charge transfer, increased epitaxial interaction and improved photochromic properties.8 In recent years, the incorporation of WO3 nanoparticles into soft matter systems has gained increasing attention, particularly for tuning and enhancing the physical properties of liquid crystals.14,15 Liquid crystals (LCs), particularly in their nematic phase, are perfect for tunable optical devices because they exhibit long-range orientational order while retaining fluidity. These optically anisotropic LC molecules are used in advanced electro-optical (EO) display technology. Traditionally, a unidirectionally rubbed polyimide (PI) layer is employed as a planar-anchoring agent in typical optically-transmissive liquid crystal displays (LCDs) to give the LC molecules planar anchoring.16–18 In such geometries, the long and flexible alkyl side chains in the LC molecules self-assemble along the rubbing direction on the PI surface, displaying a uniform director profile throughout the LC device.19,20 Even while this approach can provide high-quality LC alignment, it has certain substantial disadvantages that cannot be avoided.21 Usually a multi-step procedure, alignment layer coating calls for high temperatures and prolonged hard baking duration, followed by unidirectional rubbing that tends to generate electrostatic charges on the alignment layer surface.22 Additionally, the contaminants in the LC are increased by the fibre dust left over after the rubbing process. Furthermore, mechanical rubbing often results in an uneven pretilt angle distribution across the LC device's surface, which leads to non-uniform LC alignment and, ultimately, non-uniform LCD panel brightness. In addition, LC contamination is frequently caused by free ionic impurities stemming from the PI-based alignment layers.23,24 The sensitivity of these organic PIs to heat and light is another disadvantage of traditional PI-based alignment layers. As a result, the alignment characteristics of LC devices are altered over time by frequent exposure to heat and light.25,26 Over the past few decades, a number of unconventional contact-free methods for LC alignment have been continuously investigated. Some of them include photoalignment,27–29 dye-induced photoalignment,30 oblique evaporation method,31 controlled microgroove surfaces, Langmuir–Blodgett films,32 ion-beam irradiation,33 nanoparticles-mediated alignment.34–36 While these methods address some drawbacks of traditional approaches, they introduce new challenges. The materials and procedures used are more complicated than those used in traditional methods; the use of photodegradable polyimides reduces surface anchoring energy, which causes picture adherence and display flickering; and large-area homogeneity in LC alignment is still problematic.37 Therefore, it becomes crucial to develop alternative LC alignment techniques that use inorganic alignment agents in order to address these persistent issues with earlier technologies. One recent solution to these problems is the use of two-dimensional (2D) materials such as graphene, tungsten selenide (WSe2), transition metal dichalcogenides, hexagonal boron nitride (hBN), molybdenum disulphide (MoS2), etc., on conducting glass substrates, to provide the LC molecules the necessary planar anchoring.38–42 Substituting 2D materials for the PI alignment agent increases the polar anchoring strength, decreases free ion impurities, speeds up the dynamic electro-optic response, and improves the optical transmission of the LC cell.42–45 Over the recent years, diverse techniques have been established to achieve homogeneous orientations of 2D nanomaterials such as chemical vapor deposition (CVD), vacuum-assisted filtration, tape casting, wet spinning, centrifugal casting, direct bottom-up synthesis, hot pressing and self-assembly etc. Although these techniques provide fine free-standing layered structure of 2D materials on the substrate, however, they are generally complex, expensive and time-consuming processes exhibiting limited efficiency and restricted resistivity to high temperature.46–49 CVD is the most extensively utilized procedure, among others. It is a bottom-up technique in which the substrate is exposed to a volatile precursor of the desired 2D material at high temperatures and pressures. The precursor settles onto the substrate, forming a layer of 2D material. Ultimately, this layer is transferred to the ITO-coated glass substrates via a sacrificial layer, which is then removed. The resulting LC cells have regular alignment across small regions and fairly uniform birefringence. However, the technology is not suitable for large-scale manufacturing due to its limited domain size, the necessity for a sophisticated setup, and the demand for extremely high temperatures and pressure.50 Several studies reported that doping nanomaterials also tend to induce desired alignment for host LCs.51,52 However, NPs tend to form clusters, especially at higher concentrations, which compromises the stability of the mixture and leads to uneven properties. Also, increased concentration of NPs can lead to light scattering, resulting in reduced transparency and optical degradation of the LC display. As a solution, numerous reports indicate that different research groups have utilized a range of 2D nanomaterials as a stand-alone alignment layer and not as a doping agent for LC systems. For instance, M. Reznikov et al.53 proposed the use of dodecanethiol-capped gold NPs for a negative dielectric anisotropy LC named as MLC-6610, via ink-jet printing and achieved fair anchoring energy of around 6.8 × 10−4 J m−2 using 28 nm thick layer deposition. The technique was found to be able to provide uniform alignment from 30 µm to several square inches over the substrates. However, the requirement of vacuum and long hours of baking required for ensuring complete evaporation of solvent make the technique less feasible. Y. Liu et al. replaced the conventional PI coating on the ITO/glass substrates by aluminium-doped zinc oxide (AZO) coating through ion-beam (IB) irradiation.54 They used IB spurting of 0.6, 1.2, 1.8, 2.4 keV for 2 min over the substrates for AZO coating. In case of 1.8 keV energy, they achieved reduced pretilt angle, ∼30% increased polar anchoring energy, around 13% decrease in rise time and almost 38% faster fall time but increased threshold voltage. Though the idea of replacing conventional PI by AZO coating is beneficial, but it was observed that the pretilt angle is highly dependent upon ion-beam energy. Thus, ion-beam energy optimization makes it a complex process as a slight change in energy may lead to deterioration of alignment quality. The suitability of AZO based conducting layer has also been explored, by G. Pisharody et al.,55 together with coating hBN as an alignment layer via electrophoretic deposition method.41,55 The group observed that hBN/AZO based LC devices exhibit unidirectional alignment over large area (>3 cm2) for not only nematic LCs, but bent-core molecules as well. Besides this, it was observed that LC shows superior contrast ratio, reduced ionic concentration (∼70%), around 17% decrease in threshold voltage, but increased transit time on hBN/AZO based substrates as compared with the PI/ITO based ones. R. Basu also explored the use of hBN nanosheets in affecting the alignment properties of NLC by fabricating a hybrid device with one side PI coated and the other one hBN-coated and observed a significant enhancement in planar anchoring strength (×10) and around 45% faster fall time.43 Besides that, their group also studied the employability of different 2D nanomaterials such as graphene, carbon nanotubes (CNTs), WSe2etc. as potential alignment agent for LC systems.44,56,57 Though the materials exhibit fairly good performance, yet the CVD-grown nanomaterials are transferred onto the substrates using polymethyl-methacrylate (PMMA) aided wet-transfer method. In such scenario, the presence of PMMA residues in small domains over the substrates and the complexity and expenses of the technique still limit the practical applications. To address these issues, herein we report hydrothermal synthesis of tungsten oxide nanostructures which have been used here as an alignment layer for host NLC. The hydrothermal method has been adopted for nanomaterial synthesis due to its simplicity and high-quality product yield.58,59 Also, this method enables the growth of shape-specific nanostructures which enhances surface reactivity and charge transport.6 WO3-based nanostructures were selected for the study because, in contrast to other metal oxide nanomaterials, they show enhanced stability and durability, faster response times, and higher colouration efficiency, making them a good fit for electrochromic devices, smart windows, and displays.60 Incorporating nanoparticles into nematic liquid crystals (NLCs) is known to alter key electro-optic parameters such as threshold voltage, dielectric anisotropy, and response time. WO3 nanoparticles, due to their strong interaction with electric fields and surface anchoring capabilities, are especially promising for enhancing NLC performance. However, beyond bulk doping, the potential of WO3 nanoparticles as an alignment layer for liquid crystals remains largely unexplored. Here, in this work, WO3 is used not as a dopant but as a stand-alone alignment layer. A WO3-based alignment layer is fabricated using spray-coating technique to study its effectiveness in controlling LC orientation. To the best of our knowledge, this is for the first time when it is demonstrated that 2D WO3 film imposes planar-alignment in an EO device. For comprehensive study, the effect of functionalizing agent with different pH values is also studied. This integrated approach highlights the multifunctional potential of tungsten oxide nanoparticles in LC technologies and opens up new possibilities for the development of simplified, responsive, and high-performance liquid crystal devices. Experimental section Materials In the present investigation, a commercially available room-temperature (RT) nematic liquid crystal namely E204 (BDH Chemicals Ltd, UK) has been used. It is a low-viscous liquid crystalline mixture consisting of fluoro-biphenyl components. It exhibits dielectric anisotropy (Δε) of +6.4 at 1 kHz frequency and undergoes transition from its nematic mesophase to isotropic state at 89.8 °C.61 For the synthesis of chemically functionalized tungsten-oxide nanostructures, sodium tungstate dihydrate (Na2WO4·2H2O) (99%), sodium and ammonium sulphate (Na2SO4; (NH4)2SO4) (99%) and hydrochloric acid (HCl) were purchased from Sigma-Aldrich and were used without any further purification. Synthesis of WO3 nanostructures Hydrothermal method has been adopted for the synthesis of tungsten oxide nanoparticles. The reaction was carried out in an acidic medium to speed up the process. To initiate with, 1.73 g of sodium tungstate dihydrate (Na2WO4·2H2O), used as the precursor, was added to 40 mL distilled water (DI). The solution was kept on a stirrer at room temperature (RT) till the solution became transparent (>30 min). Two separate batches of this solution were prepared for functionalization purpose. In the first batch, 1.73 g of sodium sulfate (Na2SO4) and in the other, same quantity of ammonium sulphate ((NH4)2SO4) was added and the two mixtures were stirred for 20–30 min. 9 mL of HCl was added drop-by-drop in each of the mixtures while stirring them continuously. Soon, the color of the mixtures started changing and they were stirred for further 10 min. These mixtures were then transferred in autoclaves at 180 °C for duration of 1 and 2 h. The autoclaves were allowed to cool down naturally and the product so-collected was washed several times by centrifugation with DI water and ethanol. The nanostructures were obtained after drying the product in an oven at 60 °C overnight (Fig. 1). For the reaction time of 1 and 2 h and corresponding to two different reacting agents Na2SO4 and (NH4)2SO4, the nanostructures were named as W(Na)1, W(Na)2, W(NH4)1 and W(NH4)2 respectively, and the terminologies will be used in the subsequent sections accordingly. The precursor solution of sodium tungstate dihydrate is mildly alkaline with pH value of 8.0 to 10.5. The dropwise addition of HCl leads to the acidification of solution. Now, sodium sulphate (Na2SO4) being a normal neutral salt does not alter the pH of the material post functionalization. On the contrary, ammonium sulphate ((NH4)2SO4) is an acidic salt. The ammonium ion undergoes partial hydrolysis and pushes the localized surface environment towards a more acidic pH. Schematic illustration of procedure followed during tungsten oxide nanoparticles. Thin-film deposition and device fabrication A thin film of synthesized WO3 nanostructures was deposited on commercially available indium tin-oxide (ITO) coated glass substrates using spray-coating method with ethanol chosen as the dispersion solvent for its ability to produce uniform coatings due to its low contact angle,62 Therefore, the synthesized nanostructures were dispersed into ethanol (concentration: 1 mg mL−1) and sonicated for 3 h. The patterned ITO-coated substrates (refer SI for detailed procedure of device fabrication) were cleaned thoroughly using detergent solution, distilled water, acetone, and isopropyl alcohol (IPA). The substrates were kept on a hot-plate at 100 °C and the tungsten oxide–ethanol suspensions were uniformly sprayed on the glass plates while keeping the spray rate fixed at 1 mL min−1. The substrates were cooled down naturally. Once the uniform thin films were developed, spacers of 5 µm thickness were placed between two substrates and adhered with AB glue (with ITO-coated face towards each other) for the fabrication of required electro-optical LC device. Fig. S1 (SI) shows the schematic illustration of the procedure followed for the fabrication of EO LC device. Characterization techniques The synthesized WO3-based nanostructures underwent a series of characterizations for structural and phase confirmations. X-ray diffraction pattern has been recorded using RIGAKU Ultima IV X-ray diffractometer. Raman spectroscopy has been carried out using Horiba Scientific Xplora plus v1.2 Multiline instrument. A green laser of 532 nm wavelength with 2.5 mW power was used to record the Raman spectrum. The microscopic images of nanostructures were acquired using scanning electron microscope (JSM 6490 LV, JEOL, Japan) equipped with EDS 133 for energy dispersive X-ray spectroscopic (EDS) measurements. For dielectric and electro-optic measurements, pure NLC E204 has been filled in the fabricated LC device using conventional capillary action; as done in case of most of the LC devices. To examine the orientational properties of NLC on the surface of nanostructured WO3 film, polarized optical microscopy (POM) has been performed using microscope (Progress CT3 Radical Rx, USA) connected with a digital camera. Dielectric spectroscopy has been performed using an impedance spectroscopy measuring unit (HP-4194A) with an applied alternating field of 0.01 V µm−1 in the frequency regime of 100 Hz to 10 MHz to obtain the real and imaginary parts of the complex dielectric permittivity. Using the conventional optical switching technique, the electro-optical dynamic response time of LC was determined at RT.63 In this method, the LC sample holder is placed between a pair of crossed polarizers in such a way that the director forms an angle of 45° along the transmission axis of each polarizer. The sample holder is irradiated normally with light of wavelength 632.8 nm from He–Ne laser, after attenuating through the neutral-density (ND) filter. A square-wave signal having amplitude 10 V and frequency 10 Hz was applied at the LC cell using signal generator (Tektronix, AFG 3021B). A photodiode connected to a digital storage oscilloscope (Tektronix, TDS 2024C) was used to record the transmitted light intensity. The transmitted light intensity data was recorded as a function of time via software interfacing and corresponding data of rise and fall time was extracted. The transmission spectra have been recorded using Thermo Scientific Evolution 201 UV-visible absorption spectrophotometer. Results and discussion Characterization of WO3 nanostructures XRD analysis was performed on the synthesized WO3 nanostructures to get insights about their crystallographic phases and structures. The acquired XRD spectra are illustrated in Fig. 2(a). High degree of crystallinity has been determined via XRD analysis. Diffraction peaks with maximum relative intensities are observed at 2θ = 16.53°, 25.68°, 33.43°, 49.72°, 52.71°, 56.27° and 57.27° which can be allotted to (020), (111), (131), (202), (222), (311) and (113) crystal planes respectively. The XRD patterns closely match with the orthorhombic (Pmnb) phase of tungsten monohydrate (WO3·H2O) as per the JCPDS card no. 01-084-0886 (a = 5.249 Å, b = 10.711 Å, c = 5.133 Å). This orthorhombic phase of WO3·H2O is characterized by distorted octahedral coordination of tungsten atom; with 5 oxygen atoms and 1 water molecule. The configuration is shown in Fig. 2(b). Furthermore, the absence of any new diffraction peaks in the XRD spectra, with increased reaction time indicates that no structural modifications have been taken place with prolonged heat exposure. (a) XRD spectra of W(NH4)1, W(NH4)2, W(Na)1 and W(Na)2 nanostructures, (b) orthorhombic phase structure of WO3·H2O nanostructure, (c) FTIR and (d) Raman spectra of synthesized nanostructures. Quantitative analysis has been performed using Williamson–Hall (W–H) method for the determination of average crystallite size, crystallinity index and lattice strain of the nanostructures. The results are tabulated in Table 1. The W–H method used to determine these parameters is described briefly in SI and the corresponding plots are presented in Fig. S2. Table 1Lattice parameters derived using W–H plots and average diameter determined using SEM imagesMaterialParametersCrystallinity indexCrystallite sizeLattice strainAverage diameterW(Na)171.38%30.14 nm0.00114195.5 ± 17.3 nmW(Na)283.69%38.51 nm0.000686205.0 ± 18.5 nmW(NH4)177.19%42.01 nm0.00145357.1 ± 34.2 nmW(NH4)280.58%40.42 nm0.00104330.3 ± 34.2 nm Lattice parameters derived using W–H plots and average diameter determined using SEM images FTIR and Raman spectroscopy was performed to reveal information about the molecular interactions, surface modifications and functional groups in the synthesized nanostructures. The results are shown in Fig. 2(c) and (d) respectively. The information about in plane and out of plane segmental vibrations of atoms and/or molecules with the associated energies in IR region can be obtained using FTIR spectra. Using these measurements, the stretching, bending, bonding and wagging can be predicted. It is well-known that the crystal lattice of WO3 consists of packed corner-sharing WO6 octahedral, having 4 atoms and 6 fundamental normal modes of vibrations. The observed vibration bands, as seen from Fig. 2(c) are mainly the fundamental vibrations of W Created by potrace 1.16, written by Peter Selinger 2001-2019 O, W–O and W–O–W chromophores. The separation of normal modes as per the direction of their dynamic dipoles is facilitated by the local symmetry of WO chromophore and thus, IR-active vibrations are observed. The absorption peak in FTIR spectra positioned at 3405 cm−1 is assigned to symmetric stretching vibrations νsym in WO3 and intercalated water molecules. Similarly, a low-frequency in-plane bending (δ) is observed at 1624 cm−1 corresponding to H–O–H planes.64,65 A stretching ν(W–O–W) vibrational response is visible, as clear from the appearance of a distinct peak at 946 cm−1. A broad peak at 695 cm−1 represents out-of-plane wagging mode of vibrations γ(O–W–O). The absence of different functional groups shows that the synthesized nanomaterials are free of contaminants. Furthermore, the Raman spectra, shown in Fig. 2(d), reveals a characteristic terminal WO stretching at 946 cm−1 and the bridging O–W–O symmetric stretching at 638 cm−1 at a reaction time of 1 h. As the reaction time progresses to 2 h, another feeble peak at 806 cm−1 appears indicating the progressive condensation and ordering of the WO6 octahedral framework. The emergence of this mode reflects the formation of corner- and edge-sharing W–O–W linkages, signifying enhanced crystallinity and the development of orthorhombic hydrated tungsten oxide structure. For confirmation, deconvolution of Raman spectra has been performed (refer Fig. S3, SI). Upon deconvolution, it is seen that for reaction time of 1 h, only two characteristic peaks are observed located around 640 cm−1 and 943 cm−1 for all samples, as discussed above. Another peak appears around 806 cm−1 when reaction time is increased up to 2 h. This peak is the primary fingerprint of crystalline, anhydrous WO3. Literature suggests that when the reaction time is increased, WO3·H2O starts transforming into WO3 and the interlayer water gets absent. When reaction time is sufficiently high (up to 6 h), this peak gets prominent.62 Since in our case, the reaction time is not so high, therefore the intensity of this peak is quite low as complete transformation from WO3·H2O to WO3 does not occur. Therefore, this peak appears as a shoulder overlapping with 640 cm−1 peak. The surface morphology of synthesized nanostructures was determined using scanning electron microscopy and corresponding SEM images are shown in Fig. 3. SEM analysis shows that the synthesized WO3·H2O nanostructures have a uniform disc-like morphology with smooth surfaces and consistent dimensions, indicating controlled nucleation and growth. The synthesis process promotes discrete nanodiscs without significant agglomeration, which results in high surface-to-volume ratio, facilitating ordered packing and henceforth, providing improved interfacial interactions for device applications. SEM images of (a) W(NH4)1, (b) W(NH4)2, (c) W(Na)1 and (d) W(Na)2 nanostructures. The average diameter of the nanodiscs has been determined by processing these images through a public domain image analysis software (ImageJ) and the results are provided in Table 1. The size distribution curves (histograms) for all the samples are provided in Fig. S4 (refer SI). From Table 1, it is clear that the growth of nanostructures is controlled significantly by the capping agent. The ammonium sulphate used for functionalization, tends to inhibit the growth of crystallites, and hence particles, with increased reaction time. Therefore, a decrement in crystallite size and particle diameter is observed when reaction time is increased from 1 h to 2 h in case of (NH4)2SO4. Furthermore, energy-dispersive X-ray spectroscopy (EDS) has been performed for elemental composition confirmations. EDS analysis confirms the presence of tungsten (65–75%) and oxygen, matching the expected stoichiometry and indicating high purity of the nanostructures (see Fig. S5). WO3 nanostructures—as alignment agent POM technique has been employed in order to assess the quality of LC alignment at different WO3 nanostructures-coated substrates. In the initial state, a uniform dark field of view is observed over the entire region. This condition changes when the cell is rotated by 45° in the azimuthal plane w.r.t. the plane of crossed polarizers; and a uniformly bright field of view is observed. This transition from dark to bright state shows that the director is unidirectionally oriented parallel to the polarizer/analyzer. For better comparison, the POM images for commercial PI-coated LC cell are also shown in Fig. 4(a). Absence of any aggregate formation or visible defects in Fig. 4(a), further reinforce the idea that nematic ordering is not disrupted on replacing the PI layer with WO3-nanostructured film. However, it must be acknowledged that the optical textures of WO3·H2O-aligned NLC appear visually somewhat more heterogeneous than the polyimide-aligned cell. The observed local intensity variations in the WO3·H2O micrograph should not be directly interpreted as spatial variations in LC alignment. The quantitative assessment of alignment (discussed in the subsequent sections) confirm that the LC molecules maintain a predominantly uniform planar orientation despite the optical intensity variations visible in the micrographs. The apparent discrepancy might be arisen due to the limited sensitivity of CCD camera setup or imaging artifacts. (a) Dark state and bright state POM images, respectively, of LC cells with different substrates: (a1 and a2) PI-coated substrates; (a3 and a4) W(NH4)1-coated substrates; (a5 and a6) W(NH4)2-coated substrates; (a7 and a8) W(Na)1-coated substrates and (a9 and a10) W(Na)2-coated substrates, (b and c) transmission spectra of LC cells as a function of azimuthal angle with the rectangular profile shown in (b) and corresponding polar graph shown in (c). The possible mechanism behind NLC alignment over the WO3-coated substrates can be understood as: The orthorhombic WO3·H2O exhibits a layered structure as shown in Fig. 2(b). Corner sharing WO6 octahedra form strong 2D networks in the ac-plane (a = 5.249Å and c = 5.133 Å) which drives rapid lateral growth. The layers are loosely stacked along the b-axis (b = 10.711 Å) via weak hydrogen bonding with interlayer water molecules, restricting growth along this direction. Previous studies66 reported that (020) plane forms the flat, extended face of disc-like WO3·H2O. The exposed (020) surface of nanodiscs provides an anisotropic interfacial environment for the NLC molecules. The augmentation of planar anchoring of LC molecules on the surface coated with WO3 nanodiscs is facilitated by the epitaxial interactions between the two entities. LC E204 is a liquid crystalline mixture containing fluorobiphenyl components. The core of its components consist of biphenyl group with a fluoro (–F) terminal group at one end, and some flexible alkyl chain necessarily at the other end to contribute to its mesogenic behaviour. The epitaxial interaction of NLC molecules on the surface of WO3-based nanostructures induces localized short-range ordering of LC molecules near the surface. At the interface, better planar alignment is observed owing to the presence of these ordered regions that coalign with the LC director (n̂). These interactions together with the anisotropic arrangement of surface oxygen and hydroxyl groups can produce direction-dependent anchoring of LC molecules over the surface. Consequently, the LC molecules preferentially adopt a planar orientation which corresponds to minimization of free-energy configuration, resulting in planar anchoring. To further quantify the unidirectional alignment quality of LC cells with WO3-coated substrates, the azimuthal angle (φ) dependence of transmitted light has been measured by placing the cell between crossed polarizers, irradiated with laser of wavelength 532.5 nm. The cell was rotated by 5° in the azimuthal plane and the transmitted intensity was recorded using photodetector. The corresponding graphs in rectangular and polar coordinate representations are illustrated in Fig. 4(b) and (c) respectively. From the graph, it is clear that successive maxima and minima are exactly 90° apart from each other, which further validates the true unidirectionality of the planar alignment for all LC cells. In fact, the LC cells coated with WO3 nanostructures exhibit better contrast ratio, as compared to the conventional PI-based cells; as evident from the variation in maximum and minimum intensity in Fig. 4(b) and (c). Moreover, pretilt angles have been measured using crystal rotation method67 and are found to be ∼2.2° for conventional PI and in the range of 3.6°–3.8° for WO3-based alignment layers. Though the pretilt angle is increased in case of WO3-based alignment layers slightly, but it is in standard range below 5° facilitating good alignment over the substrates. It is already known that the POM images of LCs exhibit color due to its birefringent nature. Therefore, birefringence analysis serves as another important tool for the quantification of uniform alignment of LCs over the substrates. The same has been done by performing gray-scale analysis of POM images using ImageJ software and the resultant scans are shown in Fig. 5(a–e). Ideally, if the alignment is perfectly uniform over the region, the gray-scale analysis results in a perfect straight line parallel to horizontal axis, indicating uniformity of birefringence over the scanned region. For a quantitative comparison, the determined standard deviation with respect to the mean gray scale value is presented in Table S1 (SI). The lower variation for the WO3·H2O nanodiscs-coated substrates when compared to the PI-coated ones indicates spatially uniform distribution of birefringence. Therefore, as evident from Fig. 5, the WO3·H2O nanodiscs prove to be a potential replacement of conventional PIs as an alignment agent. Gray-scale analysis for (a) PI-coated, (b) W(NH4)1, (c) W(NH4)2, (d) W(Na)1 and (e) W(Na)2 nanostructures-coated LC cells. Additionally, the wavelength-dependent optical transmission spectra of rubbed PI-ITO substrates and different WO3-ITO substrates are recorded in the visible region and presented in Fig. 6(a). It is clear from the figure that the WO3-ITO substrates exhibit good transparency, identical to that of commercially-available PI, over the entire visible region of EM radiation. Moreover, in the inset of Fig. 6(a), the images of LC cells with different alignment layers are shown; where it becomes evident that no color tinge is observed for WO3-ITO based substrates. Thus, it can be proven as a potential replacement for conventional PI-coated substrates without any compromise with the system's transparency or other performance parameters; as discussed below. The uniformity of thin film over the glass substrate can be observed using cross-sectional SEM image of thin film shown in Fig. 6(b). (a) Transmission spectra of LC cells as a function of wavelength (the inset shows the images of LC cells with different alignment layers, as seen by the unaided eye), (b) cross-sectional SEM image of thin film coated with W(NH4)2. Electrical and electro-optical characterizations were conducted to compare the EO performance of WO3-ITO based substrates with that of rubbed PI-based substrates. It is well-known that the LC molecules undergo an orientation from surface-dictated planar alignment condition to external field-driven homeotropic direction, when the applied voltage exceeds the threshold voltage (Vth). This director reorientation is a consequence of rotational torque and its dependence upon applied electric field (τ ∝ Δε × E2). The Fréedericksz transformation method allows the measurement of sample capacitance as a function of applied voltage to analyze the molecular reorientation and measure Vth. Therefore, the sample capacitance was measured for different cells while applying a voltage over the range 0.1–5 V at 1 kHz (a frequency well-below the relaxation frequency of any dielectric mode). The corresponding Fréedericksz profiles obtained using voltage-dependent capacitance, are illustrated in Fig. 7(a). On the application of external bias voltage, the orientation of nematic director changes; where the magnitude of applied force decides the extent of orientational change. Replacing the conventional rubbed PI-based substrates with different WO3 nanostructures-coated substrates causes a slight reduction in the threshold voltage of the system; as apparent from the shifting of CV curve towards a lower voltage end. The threshold voltage has been measured to be 2.2 V, 1.6 V, 1.5 V, 1.3 V, 1.3 V for PI-based, W(NH4)1, W(NH4)2, W(Na)1 and W(Na)2-coated substrates respectively. This decrement can be attributed to the inherent polarity of WO3 lattice and its resultant dipole moment. The presence of water molecules in WO3·H2O further enhances the polarity of molecules. This increased polarizability at the interface facilitates rapid switching dynamics with lower voltage requirements. The threshold voltage is observed to be dependent on reaction time (in turn, diameter of nanodiscs) and changes in surface chemistry of the alignment layer due to different functionalization agents. On increasing the reaction time by 1 h for (NH4)2SO4-functionalized nanodiscs (see W(NH4)1 and W(NH4)2), the threshold voltage undergoes a small reduction of ∼0.1 V from 1.6 V to 1.5 V. This can be attributed to the significant decrease (∼27 nm) in the diameter of these nanodiscs upon increasing reaction time. Smaller nanodiscs tend to offer higher surface are to volume ratio, and therefore, provide a larger functionalized interface accessible to NLC molecules. This can modify the local molecular organization of the LC molecules at the alignment layer. Thus, smaller nanodiscs offer slightly reduced threshold voltage. A maximum reduction of ∼32% is observed in case of (NH4)2SO4-capped nanodiscs. In contrast, such size-dependent modulation of threshold voltage is not observed in case of Na2SO4-capped nanodiscs based LC cells and ∼41% reduction is observed for these nanodiscs irrespective of reaction time. This slight deviation in case of two different capping agents can be attributed to the difference in pH levels of these two. (NH4)2SO4 being an acidic salt tends to slightly increase the energy barrier for director reorientation due to acidic environment, as compared to Na2SO4 being the normal salt. Therefore, a lower voltage is required by the system with Na2SO4-functionalized nanodiscs used as an alignment layer. The reduction in threshold voltage is further assisted by slightly increased pretilt angle with varying materials used as alignment layer as higher pretilt reduces the energy barrier for reorientation of LC molecules, thereby reducing threshold voltage. To mathematically clarify the observed reduction of threshold voltage (Vth) despite a reduction in bulk dielectric anisotropy (refer Fig. 8), we used the unified Taylor-expansion form of the Freedericksz transition. By combining the classic Rapini–Papoular surface boundary torque equation with the Nehring–Scheffer pretilt correction factor, the effective threshold voltage can be written as in eqn (1):68,691where, K11 denotes the splay elastic constant, ε0 is the permittivity of free space, Wp signifies the polar anchoring energy coefficient, θ and d denote the pretilt angle and cell gap respectively. Here, the contributions of separate parameters can be understood directly. While the drop in dielectric anisotropy tends to increase the threshold voltage for the system, the experimental parameters introduced by the functionalized tungsten oxide nanodiscs create two overriding surface terms: (a) CV curve and (b) variation of dynamic switching times of LC cells with different alignment layers. (a and c) Frequency dependence of experimental and corrected dielectric permittivity and (b and d) dielectric loss in planar configurations and homeotropic configurations (respectively) [inset of (d) shows the zoomed-in version of peak dielectric loss] for different alignment layers. (i) The slightly increased pretilt angle θ reduces the threshold voltage. (ii) The polar surface dynamics of the nanostructured film tends to weaken the polar anchoring energy slightly due to induction of localized dipolar screening. Therefore, the WO3-based alignment layer transitions the boundaries into a finite anchoring regime, contracting the reduction multiplier term . These combined modifications strongly overpower the dielectric anisotropy effect and therefore, contributes in overall lowering the threshold voltage. Switching dynamics of LC devices are often controlled by the anchoring conditions at the substrates; therefore, the quantification of electro-optic dynamic response time becomes pivotal for the study. The rise time defines the time required by any LC layer to switch from surface-mediated planar condition to voltage-driven homeotropic configuration resulting in a change of transmission from 10% to 90% of its maximum value when the device is turned on from Vth to 10 V. On the other hand, fall time describes the time taken in the reduction of transmitted intensity from 90% to 10% of its maximum value on turning off the device from 10 V to lower than Vth. The expressions of rise time and fall time are given by the eqn (2) and (3):2and3where, γ is the rotational viscosity, K11 is the splay elastic constant and V(>Vth) is the driving voltage. The measurement of EO switching time has been performed using the standard EO setup63 and the variation of rise time and fall time on varying alignment layers is shown in Fig. 7(b). On changing alignment layers, the rise time reduces from 5.27 ms for conventional PI-coated substrates to 2.98 ms, 3.66 ms, 2.44 ms, 2.9 ms for W(NH4)1-, W(NH4)2-, W(Na)1-, and W(Na)2-coated substrates respectively. Similarly, the fall time is also observed to be reduced from 14.26 ms for PI-coated substrates to 4.63 ms, 5.02 ms, 3.1 ms and 4.74 ms respectively for W(NH4)1-, W(NH4)2-, W(Na)1-, and W(Na)2-coated substrates. The corresponding transmittance vs. time profiles (obtained experimentally) are shown in Fig. S6 (refer SI). Both rise time and fall time are observed to be reduced on replacing PI-based alignment layer by WO3-based thin film, but the rate of decrement of fall time is significantly large as compared to that of rise time. The reduction of response time can be accredited to the increased pretilt angle and reduced threshold voltage. Increase in initial pretilt reduces the torque needed for molecular reorientation during turning-on process and also reduces the deformation torque required for the LCs to return to their initial state, resulting in faster turning-off process and consequently smaller fall time. A maximum of 78% reduction of fall time and 54% reduction in rise time is achieved in case of W(Na)1 used as alignment layer whereas in case of W(NH4)1, the maximum reduction of fall and rise time is achieved as 68% and 77% respectively. Here also, the electro-optical response for LC cells with different alignment layers is observed to be dependent on both nanodisc size and surface functionalization. The Na2SO4-functionalized nanodiscs offer less threshold voltage as compared to (NH4)2SO4 due to lower energy barrier because of the chemical neutrality of the functionalizing agent, and as a consequence, molecular reorientation becomes faster. The ionic radius of sodium ion (∼1.02 Å) is much smaller than that of ammonium ion (∼1.48 Å). Smaller ionic radius allows Na+ allows denser chemical packing of sulphate groups on the nanodiscs, resulting in more uniformity in surface energy. This minimizes local steric drag and surface distortions at the interface, thereby enabling faster switching dynamics. Dielectric spectroscopy was performed to better understand the molecular dynamics of various alignment layer coated substrates. The static operational domain, i.e. standard operating frequency range of most of the LC devices and LCDs falls between 100 Hz–10 kHz. At frequencies below this range, director reorientation is too rapid in the LC device and flickering is observed at the output. On the other hand, higher frequencies (f > 5 MHz) are also generally not preferred for device operation because of performance degradation due to increased resistivity of connecting wires and ITO electrodes, along with the molecular dipolar relaxation. Therefore, we have studied the dielectric spectra of materials at different interfaces over a wide frequency range of 100 Hz–10 MHz. The experimentally-determined frequency-dependent trend of dielectric permittivity and loss for planar aligned cells (at 0 V) and in homeotropic geometry (at 24 V) are provided in Fig. 8(a–d). The figure also includes the dielectric spectra corrected after the removal of parasitic effects. It is well-known that parasitic effects often appear in the experimentally-recorded dielectric spectrum; arisen due to several factors including surface resistance of electrodes, resistance and inductance of the connecting leads, electrode polarization, ionic conductivity of the material, etc. For their removal and required corrections, the dielectric spectra of NLC E204 at interfaces with different alignment layers have been fitted using Cole–Cole model (for details of model and corresponding fitting parameters, please refer SI). The fitting has been done using Origin 2016 software and it was characterized by high R-square values, consistently close to 1. It is observed that dielectric modifications are influenced by stable anchoring forces at the interface of WO3-nanostructures coated substrates. This is evident from the significant alteration in dielectric properties of the NLC E204 on replacement of conventional rubbed PI-based conducting glass substrates by chemically functionalized WO3-coated substrates. A change in permittivity is observed at both HA and VA conditions with considerable shift in relaxation frequency, on varying the functionalizing agent from (NH4)2SO4 to Na2SO4 and with different reaction time. The orientation of LC molecules on the surface of alignment layer is strongly affected by the pretilt angle and anchoring energy offered by the layer. The decrement of permittivity is influenced by the epitaxial interaction between LC molecules and the molecules of alignment layer.70 The alterations in dielectric loss are associated with reduction of dielectric permittivity and additional surface polarization of WO3·H2O under an applied a.c. field of 100 mV. Also, it is important to mention that the observed changes in dielectric and electro-optical response of NLC can be affected by the potential leaching of ions into the host NLC matrix by different alignment layers. Direct experimental quantification of ionic impurities for such systems is required for better comprehension. In this study, the use of inorganic alignment layer is proposed as a replacement of conventional rubbing-enabled polyimide-based alignment layers for single-pixel electrodes. Further evaluation over in-plane switching (IPS) and fringe-field switching (FFS)-type interdigitated electrode-based systems is required to examine wider applicability of these systems. This examination was beyond the scope of present study. Nevertheless, this study presents a potential replacement of conventional polyimide-based alignment layer by inorganic nanostructured film offering comparably uniform planar alignment with better electro-optical performance. A comparative summary of similar other studies with the present study is provided in Table 2. Table 2Comparative summary of reported inorganic alignment-layer based studies with the present study against the benchmark of PI-based alignment layeraStudyLC usedNanomaterial dimensionsAlignment layer materialDeposition techniques V th θ τ on and τoff% TOther key metricsD. P. Singh et al.715CB25–30 nmAnatase TiO2 nanoparticlesSpin-coating of NPs-ethanol suspension———↓ γ C: ↓; ε′: ↓; TN–I: ↓G. Pisharody et al.41E7124 nmh-BN nanosheetsElectrophoretic depositionUnchanged—Improved↓ ε′: ↑; CR: ×2R. Basu et al.72E7—Hexagonal boron nitride (h-BN) nanosheetsCVD-grown monolayer transferred on ITO-coated glass substrates using wet-transfer method↑∼14%——↑—R. Basu et al.56E7—Tungsten diselenide (WSe2) nanosheetsCVD-grown WSe2 monolayer transferred on ITO-coated glass substrates using wet-transfer method↓——↑—Present studyE204Nanodisc diameter: W(Na)1: 195 nm, W(Na)2: 205 nm, W(NH4)1: 357 nm, W(NH4)2: 330 nmWO3·H2O nanodiscs functionalized with Na2SO4 and (NH4)2SO4Spray deposition technique↓∼41%3.6°–3.8° τ on: ↓∼54%; τoff: ↓∼78% [maximum reduction for W(Na)1]↑CR: ↑aAbbreviations, Vth: threshold voltage, θ: pretilt angle, τon: rise time, τoff: fall time, γC: surface energy, Δε: dielectric anisotropy, ε′: dielectric permittivity, TN–I: nematic-to-isotropic transition temperature, CR: contrast ratio. The arrows ↓ and ↑ denote the decrement and increment of corresponding parameter respectively. Comparative summary of reported inorganic alignment-layer based studies with the present study against the benchmark of PI-based alignment layera Abbreviations, Vth: threshold voltage, θ: pretilt angle, τon: rise time, τoff: fall time, γC: surface energy, Δε: dielectric anisotropy, ε′: dielectric permittivity, TN–I: nematic-to-isotropic transition temperature, CR: contrast ratio. The arrows ↓ and ↑ denote the decrement and increment of corresponding parameter respectively. Conclusion To conclude, we have demonstrated the potential of chemically functionalized hydrated WO3 nanostructures as a novel alignment layer in LC devices as an alternative to the conventionally used rubbed polyimide-based layer. The WO3·H2O nanodiscs were synthesized via hydrothermal route. The thin films of these nanodiscs, prepared using spray-coating, provide effective planar alignment to the nematic LC without requiring any mechanical rubbing process and thereby, overcoming the limitations of traditional rubbing-based alignment methodology to a great extent. The study systematically explores the effects of reaction time and capping agent on the alignment properties of these nanostructures. The new alignment layer demonstrated several performance enhancements over traditional systems, including uniform planar alignment, subtly enhanced dielectric anisotropy, up to 41% reduction in threshold voltage, high transparency over visible region, improved contrast ratio and around 70% faster response time. These are attributed to the enhanced epitaxial interaction between the 2D material and NLC molecules, tailored surface morphology, and potential ionic interactions of WO3·H2O nanostructures with LC, which collectively foster strong LC anchoring and reduced ionic contamination. The uniform planar anchoring of NLC molecules over WO3-based alignment layer was confirmed basically through independent techniques: (1) intensity variation as a function of azimuthal angle demonstrates a perfectly periodic variation, with consecutive maxima and minima positioned exactly 45° apart. This angular precision provides a definitive mathematical validation of planar alignment; (2) the grayscale pixel-intensity distribution analysis across the real-space polarized optical microscopy (POM) images exhibited negligible variance in grayscale intensity and narrow standard deviation mathematically which proves that the induced birefringence is uniform across the entire cell substrate, ruling out the possibility of local micro-domain defects or alignment fluctuations; (3) the measured pretilt angles remain stable and below 5° proving that the inorganic nanodiscs successfully anchor the long molecular axes parallel to the substrate. Although it is acknowledged that a systematic spatial mapping of pretilt angle and uniform distribution of nanodiscs over larger substrates would be required to demonstrate quantitatively the large-area alignment uniformity, the spray-coating-based deposition employed here provides a potentially scalable route for extending the WO3·H2O alignment layer to larger areas, which will be investigated in future work. The promising results of this study open new avenues for the integration of inorganic nanostructured materials as functional alignment layers in next-generation LC devices. Author contributions S. Agarwal: writing – original draft, writing – review and editing; conceptualization, synthesis and data curation; T. Yadav: SEM characterizations; B. P. Singh: pretilt-angle measurements; S. Khatavi: Raman and FTIR measurements; C. V. Yelamaggad: resources, methodology; R. Manohar: supervision, visualization and validation. Conflicts of interest There is no conflict of interest. Supplementary Material