Preparation and Orthogonal Analysis for Dual-Responsive Electrochromic Polymer Dispersed Liquid Crystal Devices

In this work, we provide a fabrication method for dual-responsive electrochromic (EC) polymer dispersed liquid crystal (PDLC) devices. The EC PDLC device was developed by combing the PDLC technique and a colored complex formed via a redox reaction without a specific EC molecule in a simple preparation method. The mesogen played dual roles in the device for scattering in the form of microdroplets and participating in the redox reactions. Orthogonal experiments were performed with the acrylate monomer concentration, the ionic salt concentration, and the cell thickness as factors to investigate the electro-optical performance for the achievement of optimized fabrication conditions. The optimized device presented four switchable states modulated by external electric fields. The light transmittance of the device was changed by an alternative current (AC) electric field while the color change was realized by a direct current (DC) electric field. Variations of mesogen and ionic salt species can modulate the color and hue of devices, which solves the disadvantage of a single color for traditional EC devices. This work lays the foundation for realizing patterned multi-colored patterned displays and anti-counterfeiting via screen printing and inkjet printing techniques.


Introduction
A liquid crystal (LC) is an intermediate state between the solid state and the liquid state, which demonstrates the anisotropy and fluidity similar to that of crystals and liquids, respectively [1]. LCs are able to respond to various external stimuli (heat, electricity, light, and magnet force) resulting in the rearrangement of the anisotropic LC molecules (mesogens) accompanied by variations in optical properties, which demonstrates substantial significance for diverse applications, such as displays [2][3][4], optics [5], photonics [6], biosensors [7,8], and biomedicines [9]. For example, cholesteric liquid crystals (CLCs) have been recently reported to switch within sub-microsecond by electrically modifying the orientational order of molecules and quenching their fluctuations [10]. In addition, a novel CLC-based device has been proposed for electrically active and thermally passive smart windows [11]. Its electrically active mode was due to the reversible transmission modulated by applied AC voltage while the passive mode was based on the controllable strength of voltage-induced electrohydrodynamic flow and temperature-dependent dynamic scattering for passive control. Additionally, LC devices with low-voltage driven scattering-controllability [12] and double-layer structures of dual-frequency [13] have been explored for smart windows as well. The introduction of polymers into liquid crystals endows the liquid crystal composite systems with expanded functionalities and applications, such as among which polymer dispersed liquid crystals (PDLCs) have attracted considerable attention on account of the excellent electro-optical properties and applications [14][15][16]. In a PDLC, LC microdroplets are dispersed in the polymer matrix, which is generated by the phase separation of a homogeneous mixture containing monomers and LCs via the polymerization with the treatment of heat or UV irradiation [17]. Sandwiching a PDLC Scheme 1. (a-d) Related chemical structures and (e) schematic illustration for the working pr of the EC PDLC.

Preparation Method
The electrochromic PDLC devices were prepared by photopolymerization-in phase separation under ultraviolet radiation. An acrylate monomer mixture compo

Preparation Method
The electrochromic PDLC devices were prepared by photopolymerization-induced phase separation under ultraviolet radiation. An acrylate monomer mixture composed of IBOMA (35 wt%), HPMA (30 wt%), CHMA (30 wt%), and HDDA (5 wt%), was applied throughout this work with TPO (0.6 wt%) as the photoinitiator. The acrylate monomer mixture, TPO, and 5CB were added into a clean container and magnetically stirred at room temperature followed by addition of DDAB to result in a homogeneous mixture. Then, the mixture was transferred to a glass cell through a capillary force and cured by continuous ultraviolet irradiation (wavelength: 365 nm, power density: 3 mW/cm 2 ) for 30 min at room temperature. DDAB-doped LC samples were prepared in similar method but without the addition of the acrylate monomer mixture and TPO. Empty glass cells were prepared by gluing two pieces of indium tin oxide (ITO) coated glass with the thickness controlled by spacers with different diameters (5 µm, 10 µm, 15 µm, and 20 µm).

Measurements
A white light source was applied to the sample and the transmitted optical signal was collected by a fiber optic spectrometer NOVA (Shanghai Fuxiang Optics Co., Ltd., Shanghai, China). Light transmittance of an empty cell was normalized as 100%. Test parameters including off-state transmittance (T off ), on-state transmittance (T on ), contrast ratio (CR), threshold voltage (V th ), saturation voltage (V sat ), and absorption were recorded to evaluate the device performance. The sample morphology was observed by a Nikon microscope LV100NPOL.

Results
The electrochromic property of DDAB-doped LC samples was initially investigated. Homogeneous mixtures containing 5CB and DDAB with various DDAB concentrations (0 wt%, 5 wt%, 10 wt%, 15 wt%, 20 wt%, 25 wt%, and 30 wt%) were prepared and then sandwiched between two pieces of ITO glass with a thickness of 20 µm. The resultant cell samples were transparent and colorless (Figure 1a-g, top row) and displayed similar transmittance ( Figure 1h) without any electrical field (off state). With a DC voltage of 4.5 V applied (on state), all cells displayed green color in appearance with different transparencies (Figure 1a-g, bottom row) and their transmittance spectra demonstrated two absorption bands at 422 nm and 658 nm (Figure 1i). The absorption at 422 nm was more intense, which contributed to the physical green color of cell samples under the DC electric field. In the following study, light transmittance at 422 nm was utilized to evaluate the performance of samples. As shown in the bottom row from Figure 1a-g, the sample appearance color varied from light green to dark green as the concentration of DDAB raised, which was in accordance with the transmittance spectra that the transmittance decreased significantly from the initial T on value of 87% (pure 5CB) to final T on of 2% (30 wt% DDAB). With the DC electric field removed (off state), all samples returned to the transparent colorless state, which exhibited responsive abilities to reversibly switch color by a DC electric field. In the on state, ammonium cations (NH 4 + ) and bromide anions (Br − ) in the sample migrated to the cathode and anode, respectively [41]. At the cathode, NH 4 + reacted with the benzene ring from 5CB to form a green free radical oxide while at the anode bromine was generated. A higher NH 4 + concentration, namely DDAB, would accumulate more free radical oxides and a stronger transmittance in the green region. In the off state, the cell was bleached due to the recombination of the halide molecule and with the green free radical oxide. Since the sample with 5 wt% DDAB was inhomogeneous and the sample with 30 wt% DDAB was too sticky, the weight ration of DDAB to 5CB for the following study was controlled from 10 wt% to 25 wt%. EC PDLC devices based on 5CB, the acrylate monomer mixture, and DDAB were further investigated to explore the constitution with the best performance. The introduction of acrylate monomers can reduce the viscosity of samples. Additionally, the increased functionality of acrylates is beneficial to the formation of a cross-linked network with an improved cross-linking degree and an enhanced polymer strength, which, however, can raise the anchoring force of mesogens resulting in a larger driving voltage. Therefore, it is necessary to define an appropriate dosage of acrylate monomers. The thickness of LC cells also affects the electro-optical performance of PDLCs. In a thin LC cell, only a small number of meshes exist in the polymer network, which usually easily leads to the generation of large mesh size and an uneven distribution with respect to the polymer network. Accordingly, the resultant LC microdroplets exhibit a small number but large size and poor homogeneity. In the off state, such a PDLC device is generally suffering from light leakage due to the attenuated scattered light intensity by LC microdroplets, which can present a larger Toff value. Additionally, LC cells with too thick thickness are not welcomed since those PDLCs require large threshold voltage values to be working. As a result, an appropriate cell thickness is necessary for a high-performance and low-energy consumed PDLC. Moreover, the EC property is intimately linked to the concentration of DDAB. To achieve the optimal parameters for an EC PDLC device with a reduced quantity of work, an orthogonal experiment method was introduced in this study. We designed a factor-level table with three factors, A (the weight ratio of the acrylate monomer mixture to 5CB), B (the weight ratio of the ammonium salt to 5CB), and C (cell thickness), and four levels (Ⅰ, Ⅱ, Ⅲ, and Ⅳ) as shown in Table 1. To be specific, the weight ratio of the acrylate monomer EC PDLC devices based on 5CB, the acrylate monomer mixture, and DDAB were further investigated to explore the constitution with the best performance. The introduction of acrylate monomers can reduce the viscosity of samples. Additionally, the increased functionality of acrylates is beneficial to the formation of a cross-linked network with an improved cross-linking degree and an enhanced polymer strength, which, however, can raise the anchoring force of mesogens resulting in a larger driving voltage. Therefore, it is necessary to define an appropriate dosage of acrylate monomers. The thickness of LC cells also affects the electro-optical performance of PDLCs. In a thin LC cell, only a small number of meshes exist in the polymer network, which usually easily leads to the generation of large mesh size and an uneven distribution with respect to the polymer network. Accordingly, the resultant LC microdroplets exhibit a small number but large size and poor homogeneity. In the off state, such a PDLC device is generally suffering from light leakage due to the attenuated scattered light intensity by LC microdroplets, which can present a larger T off value. Additionally, LC cells with too thick thickness are not welcomed since those PDLCs require large threshold voltage values to be working. As a result, an appropriate cell thickness is necessary for a high-performance and low-energy consumed PDLC. Moreover, the EC property is intimately linked to the concentration of DDAB. To achieve the optimal parameters for an EC PDLC device with a reduced quantity of work, an orthogonal experiment method was introduced in this study. We designed a factor-level table with three factors, A (the weight ratio of the acrylate monomer mixture to 5CB), B (the weight ratio of the ammonium salt to 5CB), and C (cell thickness), and four levels (I, II, III, and IV) as shown in Table 1. To be specific, the weight ratio of the acrylate monomer mixture to 5CB (A) was from 33.4 wt% to 66.7 wt% while the weight ratio of the ammonium salt to 5CB (B) was from 11.2 wt% to 33.4 wt%. The cell thickness was controlled as 5 µm, 10 µm, 15 µm, and 20 µm, respectively. The following experiments were performed experiments based on the appropriate orthogonal table generated by Software Minitab 21 (Taguchi Design, Japan), which displayed the fabrication constitutions for sixteen EC PDLC samples with three variables controlled as shown in Table 2. Based on the experimental results, range analysis was applied to determine the factors' sensitivity. Samples were prepared according to the fabrication constitution shown in Table 2. Once fabricated, 1, 2, 5, 6, 7, 8, 9, 10, and 11 displayed an opaque color while samples 3, 4, 12, 13, 14, 15, and 16 were transparent colorless, which meant the absence of the PDLC feature. All samples were observed with a polarized optical microscope (POM) under crossed polarizers. The POM images of 1, 2, 5, 6, 7, 8, 9, 10, and 11 (Figure 2a-h) clearly presented the existence of LC microdroplets but with various diameters. Samples 1 and 2 exhibited similar POM images (Figure 2a,b), which displayed LC microdroplets with larger diameters than those of other samples. In addition, the LC microdroplets in 1 and 2 were of poor monodispersity and small density in the view. POM images of 5, 6, 7, 8, and 9 showed densely distributed LC microdroplets with relatively small diameters (Figure 2c-f). For samples without the PDLC feature, no LC microdroplet was found under POM. For instance, the POM image of 12 (Figure 2i) presented no existence of LC microdroplets but few typical Schlieren textures in the view. For 13-16, the disappearance of LC microdroplets was due to the high concentration of monomers and the low concentration of 5CB, which would result in the disability of light modulation. The absence of PDLC features for 4 and 12 was probably ascribed to the high concentration of the ammonium salt. The synergetic effect of monomer and ammonium salt concentration might decide the morphological and optical properties of 3. Therefore, 1, 2, 5, 6, 7, 8, 9, 10, and 11 would be discussed in the following study. The electro-dependent transmittance of 1, 2, 5, 6, 7, 8, 9, 10, and 11 was further characterized under an AC electric field ( Figure 3). All those samples were opaque in appearance in the beginning, which corresponded to the low transmittance values at the AC voltage of 0, namely the off state transmittance (Toff), in Figure 3. 1, 2, and 8 clearly showed higher Toff values than others, which was consistent with their POM images that showed larger LC microdroplets. As the AC voltage raised, their transmittance gradually increased differently. In essence, the randomly distributed mesogens in the microdroplets among the polymer network cause intensive light scattering leading to an opaque color without any electrical field. However, applying an AC electric field will orientationally rearrange mesogens according to the direction of the electric field resulting in a transparent appearance. At the same AC voltage value, those samples displayed different transmittance values due to the variation in the cell constitution. The orthogonal range analysis method was employed to add values at the same level under the same factor to generate the average value at each level, on which the range analysis diagram was based ( Figure  3b,c). According to the difference between the largest and smallest Toff values, the influence sequence regarding Toff was B > A > C (Figure 3b). The presence of ammonium salt affected the phase separation process. Based on the experimental results, once the DDAB concentration of the sample was above 33.4%, no electro-dependent optical switching could be observed. The monomer concentration affected the formation of the intermingled polymer network. As the monomer concentration raised, the polymer network became more compact, which added difficulty in the diffusion of mesogens and the formation of large LC microdroplets and accordingly attenuated Toff values. Values of Ton for those samples were basically in the range between 80-90% with inconspicuous changes ( Figure  3c). Thus, it was difficult to decide the impacting power for the three factors A, B, and C. The electro-dependent transmittance of 1, 2, 5, 6, 7, 8, 9, 10, and 11 was further characterized under an AC electric field ( Figure 3). All those samples were opaque in appearance in the beginning, which corresponded to the low transmittance values at the AC voltage of 0, namely the off state transmittance (T off ), in Figure 3. 1, 2, and 8 clearly showed higher T off values than others, which was consistent with their POM images that showed larger LC microdroplets. As the AC voltage raised, their transmittance gradually increased differently. In essence, the randomly distributed mesogens in the microdroplets among the polymer network cause intensive light scattering leading to an opaque color without any electrical field. However, applying an AC electric field will orientationally rearrange mesogens according to the direction of the electric field resulting in a transparent appearance. At the same AC voltage value, those samples displayed different transmittance values due to the variation in the cell constitution. The orthogonal range analysis method was employed to add values at the same level under the same factor to generate the average value at each level, on which the range analysis diagram was based (Figure 3b,c). According to the difference between the largest and smallest T off values, the influence sequence regarding T off was B > A > C (Figure 3b). The presence of ammonium salt affected the phase separation process. Based on the experimental results, once the DDAB concentration of the sample was above 33.4%, no electro-dependent optical switching could be observed. The monomer concentration affected the formation of the intermingled polymer network. As the monomer concentration raised, the polymer network became more compact, which added difficulty in the diffusion of mesogens and the formation of large LC microdroplets and accordingly attenuated T off values. Values of T on for those samples were basically in the range between 80-90% with inconspicuous changes (Figure 3c). Thus, it was difficult to decide the impacting power for the three factors A, B, and C.   (Figure 4b,c). In general, the monomer concentration (A) and the cell thickness (C) had a positive correlation with both Vth and Vsat values of samples. It was clearly exhibited that the monomer concentration had the greatest effect on Vth and Vsat values. As the monomer concentration increased, the diameter of the LC microdroplet became smaller, in which case the interface of the polymer network displayed a stronger anchoring force towards LC microdroplets and accordingly higher energy (voltage), e.g., Vth and Vsat, was required to drive the mesogenic rearrangement. With the same voltage applied, the electric field strength in the device became weaker for an LC cell with a larger thickness, which therefore called for larger voltages including higher Vth and Vsat values, to rearrange the mesogens. On the contrary, Vth and Vsat decreased with the increase in the ammonium salt concentration (B). Since the ammonium salt played the role of the electrolyte, the increased ammonium salt concentration could reduce the device resistance and enhance the ability of the AC electric field to induce mesogens in the device, thus lowering the voltage for optical switching. Contrast ratio (CR), as another important factor for the electro-optical characteristics of PDLC devices, was defined as the ratio of Ton over Toff, which also reflected the size of LC microdroplets. The contrast ratio values for 1, 2, 5, 6, 7, 8, 9, 10, and 11 were graphed as shown in Figure 4d, which obviously exhibited the CR values of 5 (CR = 42.9) and 7 (CR = 36.7) were higher than the others.  (Figure 4b,c). In general, the monomer concentration (A) and the cell thickness (C) had a positive correlation with both V th and V sat values of samples. It was clearly exhibited that the monomer concentration had the greatest effect on V th and V sat values. As the monomer concentration increased, the diameter of the LC microdroplet became smaller, in which case the interface of the polymer network displayed a stronger anchoring force towards LC microdroplets and accordingly higher energy (voltage), e.g., V th and V sat , was required to drive the mesogenic rearrangement. With the same voltage applied, the electric field strength in the device became weaker for an LC cell with a larger thickness, which therefore called for larger voltages including higher V th and V sat values, to rearrange the mesogens. On the contrary, V th and V sat decreased with the increase in the ammonium salt concentration (B). Since the ammonium salt played the role of the electrolyte, the increased ammonium salt concentration could reduce the device resistance and enhance the ability of the AC electric field to induce mesogens in the device, thus lowering the voltage for optical switching. Contrast ratio (CR), as another important factor for the electro-optical characteristics of PDLC devices, was defined as the ratio of T on over T off , which also reflected the size of LC microdroplets. The contrast ratio values for 1, 2, 5, 6, 7, 8, 9, 10, and 11 were graphed as shown in Figure 4d, which obviously exhibited the CR values of 5 (CR = 42.9) and 7 (CR = 36.7) were higher than the others.  1, 2, 5, 6, 7, 8, 9, 10, and 11. The corresponding range analysis charts of (b) Vth, (c) Vsat, and (e) CR values with different levels of three factors A, B, and C.
The EC property of 1-12 was studied as well. The transmittance spectra of 1-12 were collected with a DC current of 4.5 V applied (Figure 5a). Considering that the opaque initial state could scatter light and affect the EC-induced absorption measurements, samples were kept in the isotropic state with transmittance normalized to 100%. Samples were divided into three groups according to the monomer concentration, namely, 1-4 in the first group with the monomer concentration of 33.4 wt%, 5-8 in the second group with the monomer concentration of 42.9 wt%, and 9-12 in the third group with the monomer concentration of 53.9 wt%. For samples in the same group, the absorption at 422 nm (A422) increased as the ammonium salt concentration raised and saturated when the ammonium salt concentration exceeded 25.0 wt% (Figure 5b). The maximized value of A422 could reach 94.2% for samples 3 and 4. The effect of the ammonium salt on the absorption of EC PDLC was similar to that of 5CB. In addition, the absorption at 422 nm (A422) was not affected by the monomer concentration and the cell thickness.   ratio of 1, 2, 5, 6, 7, 8, 9, 10, and 11. The corresponding range analysis charts of (b) V th , (c) V sat , and (e) CR values with different levels of three factors A, B, and C.
The EC property of 1-12 was studied as well. The transmittance spectra of 1-12 were collected with a DC current of 4.5 V applied (Figure 5a). Considering that the opaque initial state could scatter light and affect the EC-induced absorption measurements, samples were kept in the isotropic state with transmittance normalized to 100%. Samples were divided into three groups according to the monomer concentration, namely, 1-4 in the first group with the monomer concentration of 33.4 wt%, 5-8 in the second group with the monomer concentration of 42.9 wt%, and 9-12 in the third group with the monomer concentration of 53.9 wt%. For samples in the same group, the absorption at 422 nm (A 422 ) increased as the ammonium salt concentration raised and saturated when the ammonium salt concentration exceeded 25.0 wt% (Figure 5b). The maximized value of A 422 could reach 94.2% for samples 3 and 4. The effect of the ammonium salt on the absorption of EC PDLC was similar to that of 5CB. In addition, the absorption at 422 nm (A 422 ) was not affected by the monomer concentration and the cell thickness. The EC property of 1-12 was studied as well. The transmittance spectra of 1-12 were collected with a DC current of 4.5 V applied (Figure 5a). Considering that the opaque initial state could scatter light and affect the EC-induced absorption measurements, samples were kept in the isotropic state with transmittance normalized to 100%. Samples were divided into three groups according to the monomer concentration, namely, 1-4 in the first group with the monomer concentration of 33.4 wt%, 5-8 in the second group with the monomer concentration of 42.9 wt%, and 9-12 in the third group with the monomer concentration of 53.9 wt%. For samples in the same group, the absorption at 422 nm (A422) increased as the ammonium salt concentration raised and saturated when the ammonium salt concentration exceeded 25.0 wt% (Figure 5b). The maximized value of A422 could reach 94.2% for samples 3 and 4. The effect of the ammonium salt on the absorption of EC PDLC was similar to that of 5CB. In addition, the absorption at 422 nm (A422) was not affected by the monomer concentration and the cell thickness.  From the perspective of T off , T on , V th , V sat , CR, and A 422 , the electro-optical property of the EC PDLC device was comprehensively investigated by the orthogonal range difference method. When A (the weight ratio of the acrylate monomer mixture to 5CB) was in level II, the corresponding device presented better T off and T on , the highest CR, and the smallest V th and V sat . When B (the weight ratio of the ammonium salt to 5CB) belonged to level III, the corresponding device exhibited the lowest T off and higher T on , lower V th and V sat , and the highest CR, which also had saturated absorption and obvious coloration. When C (cell thickness) was in level II, the corresponding device demonstrated better T off and T on , the smallest V th and V sat , and higher CR. Consequently, we choose the weight ratio of the acrylate monomer mixture to 5CB of 42.9 wt%, the weight ratio of DDAB to 5CB of 25.0 wt%, and the cell thickness of 10 µm as the optimal constitution parameter for the fabrication of EC PDLC devices. In the end, a device with the optimal parameter was prepared and tested. It was able to display four different states by varying the external electrical field (Figure 6a-d). The device was initially in an opaque colorless state (Figure 6a). With an AC electric field of 20 V applied, the device turned into a transparent colorless state and reversibly returned to the initial state once the electric field was removed. With a DC electric field of 4.5 V applied, the device changed into an opaque green state and restore to its initial state after removing the electric field. Additionally, with AC (20 V) and DC (4.5 V) electric fields simultaneously applied, the device altered into a transparent green state and recovered after removing the electric field. The transmittance spectra associated with the status in Figure 6a-d were collected as shown in Figure 6e, which agreed with the corresponding visual appearance. Variations of mesogen species could produce similar results but different colors. For example, replacing 5CB with E7 resulted in a series of yellow-colored states instead of green-colored ones. To overcome the limitations of a single color, the specific work regarding multicolor EC PDLC devices with fabrication accessibilities to printing techniques, such as screen printing and ink-jet printing, is still under study.
the EC PDLC device was comprehensively investigated by the orthogonal range difference method. When A (the weight ratio of the acrylate monomer mixture to 5CB) was in level II, the corresponding device presented better Toff and Ton, the highest CR, and the smallest Vth and Vsat. When B (the weight ratio of the ammonium salt to 5CB) belonged to level III, the corresponding device exhibited the lowest Toff and higher Ton, lower Vth and Vsat, and the highest CR, which also had saturated absorption and obvious coloration. When C (cell thickness) was in level II, the corresponding device demonstrated better Toff and Ton, the smallest Vth and Vsat, and higher CR. Consequently, we choose the weight ratio of the acrylate monomer mixture to 5CB of 42.9 wt%, the weight ratio of DDAB to 5CB of 25.0 wt%, and the cell thickness of 10 μm as the optimal constitution parameter for the fabrication of EC PDLC devices. In the end, a device with the optimal parameter was prepared and tested. It was able to display four different states by varying the external electrical field (Figure 6a-d). The device was initially in an opaque colorless state ( Figure  6a). With an AC electric field of 20 V applied, the device turned into a transparent colorless state and reversibly returned to the initial state once the electric field was removed. With a DC electric field of 4.5 V applied, the device changed into an opaque green state and restore to its initial state after removing the electric field. Additionally, with AC (20 V) and DC (4.5 V) electric fields simultaneously applied, the device altered into a transparent green state and recovered after removing the electric field. The transmittance spectra associated with the status in Figure 6a-d were collected as shown in Figure 6e, which agreed with the corresponding visual appearance. Variations of mesogen species could produce similar results but different colors. For example, replacing 5CB with E7 resulted in a series of yellow-colored states instead of green-colored ones. To overcome the limitations of a single color, the specific work regarding multicolor EC PDLC devices with fabrication accessibilities to printing techniques, such as screen printing and ink-jet printing, is still under study.

Conclusions
In summary, a dual-responsive EC PDLC device was successfully fabricated by combing the PDLC technique with the reversible redox reaction between the benzene ring of mesogens and the cation of an ammonium salt under DC to form a colored complex. In the device, the mesogen 5CB not only played the role of scattering light in the form of microdroplets but also worked as a reactant for color development. Orthogonal experiments and range analysis methods were applied to evaluate the electro-optical character- Figure 6. The physical appearance of the EC PDLC sample: (a) initial opaque colorless without no external electrical field, (b) transparent colorless with the application of an AC electric field (20 V), (c) opaque green with the application of a DC electric field (4.5 V), (d) transparent green with the application of both AC and DC electric fields, and (e) the corresponding transmittance spectra.

Conclusions
In summary, a dual-responsive EC PDLC device was successfully fabricated by combing the PDLC technique with the reversible redox reaction between the benzene ring of mesogens and the cation of an ammonium salt under DC to form a colored complex. In the device, the mesogen 5CB not only played the role of scattering light in the form of microdroplets but also worked as a reactant for color development. Orthogonal experiments and range analysis methods were applied to evaluate the electro-optical characteristics and discoloration performance of those devices, which exhibited the fabrication parameter for the optimized device. The device presented four switchable states with the application of AC and DC electric fields. The AC electric field could change the light transmittance of the device while the DC electric field altered the device color. This work lays the foundation for realizing patterned multi-colored patterned displays and anti-counterfeiting via screen printing and inkjet printing techniques.