High-Performance Electrochromic Devices Based on Size-Controlled 2D WO3 Nanosheets Prepared Using the Intercalation Method

It is difficult to obtain ultrathin two-dimensional (2D) tungsten trioxide (WO3) nanosheets through direct exfoliation from bulk WO3 in solution due to the strong bonding between interlayers. Herein, WO3 nanosheets with controllable sizes were synthesized via K+ intercalation and the exfoliation of WO3 powder using sonication and temperature. Because of the intercalation and expansion in the interlayer distance, the intercalated WO3 could be successfully exfoliated to produce a large quantity of individual 2D WO3 nanosheets in N-methyl-2-pyrrolidone under sonication. The exfoliated ultrathin WO3 nanosheets exhibited better electrochromic performance in an electrochromic device than WO3 powder and exfoliated WO3 without intercalation. In particular, the prepared small WO3 nanosheets exhibited excellent electrochromic properties with a large optical modulation of 41.78% at 700 nm and fast switching behavior times of 9.2 s for bleaching and 10.5 s for coloring. Furthermore, after 1000 cycles, the small WO3 nanosheets still maintained 86% of their initial performance.


Introduction
Electrochromic (EC) materials have attracted considerable interest in various applications, including transparent displays, smart windows, and automotive mirrors, owing to their ability to efficiently adjust optical properties, such as transmittance, reflectance, and absorbance with low power consumption [1][2][3][4].The optical properties of EC materials are reversibly altered through the injection and extraction of ions and electrons into the EC layer under a low applied voltage between the anode and the cathode.Among a number of EC materials, tungsten trioxide (WO 3 ) has been widely studied due to its good EC performance and electrochemical stability [5,6].To further enhance the EC performance of WO 3 , substantial efforts have been committed to producing nanostructured WO 3 [7][8][9].In particular, in contrast to bulk WO 3 , two-dimensional (2D) WO 3 nanosheets have been found to exhibit enhanced optical contrast and durability due to the expanded active surface area and facilitated ion penetration within EC materials [10,11].Various techniques are used to produce 2D WO 3 materials, achievable through two synthetic strategies: the top-down and the bottom-up approach [10,[12][13][14][15][16][17].The bottom-up methods aim to produce 2D WO 3 from smaller precursor molecules, whereas the top-down approach involves the direct exfoliation of bulk precursors.Exfoliation emerges as a highly efficient approach for producing a multitude of ultrathin 2D nanosheets from the bulk source.However, because of the strong covalent interaction between the layers of monoclinic WO 3 , the synthesis of Materials 2024, 17, 41 2 of 10 ultrathin 2D nanosheets from bulk WO 3 using a typical liquid-phase exfoliation method is still remarkably challenge.Only a few studies on the production of ultrathin 2D WO 3 nanosheets with liquid-phase exfoliation have been reported to date.This can be achieved through methods such as the direct exfoliation of bulk WO 3 , exfoliation and oxidation from tungsten disulfide (WS 2 ), and the exfoliation of hydrated WO 3 [11,15,16,18,19].Guan et al. achieved the direct exfoliation of WO 3 powder into thin nanosheets by coating the surface of WO 3 with bovine serum albumin (BSA) [18].In an acidic environment, the -NH 2 functional groups of BSA establish strong bonds with WO 3 via electrostatic interaction, thereby considerably improving the exfoliation process during sonication.However, due to strong binding in WO 3 , prolonged processing times (sonication for 48 h) are required, and the residual presence of BSA may restrict their subsequent applications.In addition, Azam et al. reported a solution-phase method for synthesizing 2D WO 3 nanosheets, involving the oxidation of 2D WS 2 nanosheets obtained through the exfoliation of bulk WS 2 powder [11].Typically, obtaining fully oxidized WO 3 nanosheeets requires distinct exfoliation and oxidation steps, making these processes more time-consuming compared to single-step methods.Subsequent investigations aimed to generate WO 3 intercalation compounds with the goal of producing WO 3 nanosheets without the need for additional oxidation processes.The layered WO 3 hydrates can serve as effective precursors for the preparation of ultrathin WO 3 nanosheets [15,16,20].Hydrated WO 3 is typically manufactured on a nanometer scale, resulting in exfoliated nanosheets with small lateral sizes (less than 500 nm).To our knowledge, no method has been reported for directly producing large 2D WO 3 nanosheets from bulk WO 3 without the hydration step.In order to prepare 2D nanosheets by exfoliating bulk WO 3 , an intercalation process was utilized to weaken the interlayer binding forces in the materials.
Intercalation occurs when molecules or ions are incorporated into the layers of bulk source materials, the interlayer distance is expanded, and the binding interactions between the adjacent layers are weakened.Depending on the bulk source materials, the intercalating species, and its concentration, a variety of different intercalation compounds with intriguing properties can be used for synthesis [21,22].The exfoliation method following intercalation isolates the atomic layers directly from the bulk materials.Importantly, an appropriate solvent plays a crucial role in the intercalation-based exfoliation and subsequent processing of layered materials [21,23,24].The solvent properties, such as polarity and surface energy (surface tension), significantly impact the quality of the final products.The solvent polarity influences intercalation efficiency by determining the dissolution or solvation of the intercalation compounds.Additionally, the surface energy (or surface tension) of the solvent is essential for effective exfoliation.
Here, we report a facile method to synthesize 2D WO 3 nanosheets with controllable sizes through the intercalation of K + into bulk WO 3 powder and exfoliation using N-methyl-2-pyrrolidone (NMP) with similar surface tension.In this method, lamellar intercalated WO 3 is initially prepared by intercalating K + into bulk WO 3 .Subsequently, WO 3 nanosheets with various sizes and thicknesses below 10 nm are exfoliated by carefully controlling the exfoliation conditions.The synthesized WO 3 nanosheets are then incorporated into an electrochromic device, and the impact of the WO 3 nanosheet size on electrochromic performance is examined.The electrochromic device using small nanosheets demonstrates a broad optical contrast, a rapid coloration/bleaching response, and robust cycling stability.

Intercalation and Exfoliation of WO 3 Nanosheets
A homogenously ground mixture of WO 3 powder (0.5 g; Sigma-Aldrich, Milwaukee, WI, USA), potassium metal (0.39 g; Sigma-Aldrich, Milwaukee, WI, USA), naphthalene (1.28 g; Sigma-Aldrich, Milwaukee, WI, USA), and tetrahydrofuran (10 mL; Sigma-Aldrich, Milwaukee, WI, USA) was reacted for 3 h.During the reaction, K + intercalated into layers of bulk WO 3 and expanded the bulk of WO 3 .The intercalated compounds of WO 3 were stirred and sonicated in NMP (30 mL; Sigma-Aldrich, Milwaukee, WI, USA) for 30 min at 10 • C and 40~50 • C for the exfoliation of large and small WO 3 nanosheets, respectively.In theory, increased energy has the potential to enhance the fragmentation of atomic layers, leading to a significant reduction in the lateral size of the nanosheets [25].Exfoliation achieved through gentle stirring at a low temperature can prevent in-plane damage, resulting in the formation of larger nanosheets in the solution [26].Following centrifugation at 1500 rpm for 30 min, a uniformly dispersed yellowish supernatant containing exfoliated WO 3 nanosheets was collected by discarding the precipitate of unexfoliated WO 3 nanosheets.For comparison, the exfoliated WO 3 powder in NMP without intercalation was also prepared.The corresponding preparation process is illustrated in Figure 1.
A homogenously ground mixture of WO3 powder (0.5 g; Sigma-Aldrich, Milwaukee, WI, USA), potassium metal (0.39 g; Sigma-Aldrich, Milwaukee, WI, USA), naphthalene (1.28 g; Sigma-Aldrich, Milwaukee, WI, USA), and tetrahydrofuran (10 mL; Sigma-Aldrich, Milwaukee, WI, USA) was reacted for 3 h.During the reaction, K + intercalated into layers of bulk WO3 and expanded the bulk of WO3.The intercalated compounds of WO3 were stirred and sonicated in NMP (30 mL; Sigma-Aldrich, Milwaukee, WI, USA) for 30 min at 10 °C and 40~50 °C for the exfoliation of large and small WO3 nanosheets, respectively.In theory, increased energy has the potential to enhance the fragmentation of atomic layers, leading to a significant reduction in the lateral size of the nanosheets [25].Exfoliation achieved through gentle stirring at a low temperature can prevent in-plane damage, resulting in the formation of larger nanosheets in the solution [26].Following centrifugation at 1500 rpm for 30 min, a uniformly dispersed yellowish supernatant containing exfoliated WO3 nanosheets was collected by discarding the precipitate of unexfoliated WO3 nanosheets.For comparison, the exfoliated WO3 powder in NMP without intercalation was also prepared.The corresponding preparation process is illustrated in

Fabrication of Electrochromic Devices (ECDs)
Bare indium tin oxide (ITO)-coated glass, used as an electrode, was cleaned via sonication in ethanol, acetone, and isopropanol for 15 min, respectively.Thin films composed of WO3 ultrathin nanosheets were then spin-coated on the cleaned ITO/glass at 4000 rpm for 60 s and dried at 120 °C for 10 min.Subsequently, 1 M lithium perchlorate (LiClO4) in propylene carbonate (PC), used as the electrolyte, was injected into an adhesive spacer between two ITO electrodes.Finally, another ITO electrode was placed on top of the spacer, and then the device was sealed (Figure 2a).The prepared ECDs with the structure of ITO/WO3/LiClO4 + PC/ITO were used to evaluate the performance of the ultrathin WO3 nanosheets.

Fabrication of Electrochromic Devices (ECDs)
Bare indium tin oxide (ITO)-coated glass, used as an electrode, was cleaned via sonication in ethanol, acetone, and isopropanol for 15 min, respectively.Thin films composed of WO 3 ultrathin nanosheets were then spin-coated on the cleaned ITO/glass at 4000 rpm for 60 s and dried at 120 • C for 10 min.Subsequently, 1 M lithium perchlorate (LiClO 4 ) in propylene carbonate (PC), used as the electrolyte, was injected into an adhesive spacer between two ITO electrodes.Finally, another ITO electrode was placed on top of the spacer, and then the device was sealed (Figure 2a).The prepared ECDs with the structure of ITO/WO 3 /LiClO 4 + PC/ITO were used to evaluate the performance of the ultrathin WO 3 nanosheets.

Characterization of WO 3 Nanosheets
The morphologies and structures were examined via scanning electron microscopy (SEM, MiRA3-LMH, Tescan, Brno, Czech Republic), transmission electron microscopy (TEM, JEM-F200, JEOL, Tokyo, Japan), and atomic force microscopy (AFM) in tapping mode under ambient conditions using the X2-70 system from Park Systems Corp, Suwon, Republic of Korea.The crystal structure of the prepared WO 3 films was investigated by utilizing X-ray diffraction (XRD) with the MiniFlex 600 instrument from Rigaku, Tokyo, Japan and employing a Cu radiation source at a scan rate of 5 • min −1 .Raman spectra were measured using a Raman spectrometer (LabRAM HR, HORIBA, Longjumeau, France) from 200 to 900 cm −1 with 325 nm laser excitation.The samples' chemical compositions were examined using X-ray photoelectron spectroscopy (XPS) with a K-Alpha X-ray photoelectron spectrometer from Thermo Scientific, Loughborough, UK.

Characterization of WO3 Nanosheets
The morphologies and structures were examined via scanning electron microscopy (SEM, MiRA3-LMH, Tescan, Brno, Czech Republic), transmission electron microscopy (TEM, JEM-F200, JEOL, Tokyo, Japan), and atomic force microscopy (AFM) in tapping mode under ambient conditions using the X2-70 system from Park Systems Corp, Suwon, South Korea.The crystal structure of the prepared WO3 films was investigated by utilizing X-ray diffraction (XRD) with the MiniFlex 600 instrument from Rigaku, Tokyo, Japan and employing a Cu radiation source at a scan rate of 5° min −1 .Raman spectra were measured using a Raman spectrometer (LabRAM HR, HORIBA, Longjumeau, France) from 200 to 900 cm −1 with 325 nm laser excitation.The samples' chemical compositions were examined using X-ray photoelectron spectroscopy (XPS) with a K-Alpha X-ray photoelectron spectrometer from Thermo Scientific, Loughborough, UK.

ECD Performance Test
Electrochemical measurements for the EC device were conducted using a BioLogic SP-150 Potentiostat (BioLogic Science Instruments, Seyssinet-Pariset, France) within a three-electrode system, where the working electrode was WO3 films on the ITO, the reference electrode was Ag/AgCl (1 M KCl), and the counter electrode was graphite.The cyclic voltammetry (CV) curves were obtained by applying a potential range from +1.0 V to −1.0 V at a scan rate of 20 mV s −1 .Additionally, a chronoamperometry test was performed, as shown in Figure 2b, using an operating voltage window between +3.0 V and −3.0 V with a 30 s interval.Furthermore, UV-Vis spectroscopy (UV-3600, Shimadzu, Kyoto, Japan) was used for measuring the optical color contrast and switching response time of the ECD, with an operating voltage ranging from +3.0 V to −3.0 V.

Results and Discussion
In the experimental process, WO3 nanosheets were synthesized from bulk WO3 using a two-step procedure.This involved initially intercalating WO3 with K + ions, followed by the subsequent exfoliation of layered intercalated compounds into nanosheets, as illustrated in Figure 1.The morphology of exfoliated WO3 nanosheets with controllable sizes was investigated via SEM, TEM, and AFM.SEM images showed the successful production of WO3 nanosheets with a lamellar structure (Figure 3a,d).However, there were more wrinkles between the structures in the large WO3 nanosheets compared with the small nanosheets.The size of the small WO3 nanosheets was sub-1 µm, whereas the size of the large WO3 nanosheets was greater than 1 µm (Figure S1).During the sonication process, higher temperatures in the sonication bath promote the exfoliation and reduce the average size of the nanosheets, due to the increased sonic pressure at elevated temperatures [27,28].Figure 3b,e show high-resolution transmission electron microscopy (HRTEM) images of individual small and large WO3 nanosheets, respectively.HRTEM images of the

ECD Performance Test
Electrochemical measurements for the EC device were conducted using a BioLogic SP-150 Potentiostat (BioLogic Science Instruments, Seyssinet-Pariset, France) within a threeelectrode system, where the working electrode was WO 3 films on the ITO, the reference electrode was Ag/AgCl (1 M KCl), and the counter electrode was graphite.The cyclic voltammetry (CV) curves were obtained by applying a potential range from +1.0 V to −1.0 V at a scan rate of 20 mV s −1 .Additionally, a chronoamperometry test was performed, as shown in Figure 2b, using an operating voltage window between +3.0 V and −3.0 V with a 30 s interval.Furthermore, UV-Vis spectroscopy (UV-3600, Shimadzu, Kyoto, Japan) was used for measuring the optical color contrast and switching response time of the ECD, with an operating voltage ranging from +3.0 V to −3.0 V.

Results and Discussion
In the experimental process, WO 3 nanosheets were synthesized from bulk WO 3 using a two-step procedure.This involved initially intercalating WO 3 with K + ions, followed by the subsequent exfoliation of layered intercalated compounds into nanosheets, as illustrated in Figure 1.The morphology of exfoliated WO 3 nanosheets with controllable sizes was investigated via SEM, TEM, and AFM.SEM images showed the successful production of WO 3 nanosheets with a lamellar structure (Figure 3a,d).However, there were more wrinkles between the structures in the large WO 3 nanosheets compared with the small nanosheets.The size of the small WO 3 nanosheets was sub-1 µm, whereas the size of the large WO 3 nanosheets was greater than 1 µm (Figure S1).During the sonication process, higher temperatures in the sonication bath promote the exfoliation and reduce the average size of the nanosheets, due to the increased sonic pressure at elevated temperatures [27,28].Figure 3b,e show high-resolution transmission electron microscopy (HRTEM) images of individual small and large WO 3 nanosheets, respectively.HRTEM images of the nanosheets display distinct lattice fringes, indicating their remarkable crystal structure.All the clear lattice spacings were 0.37 nm, corresponding to the (002) plane of monoclinic WO 3 [29,30].In addition, the selected area electron diffraction patterns of the small and large nanosheets illustrate their single crystal nature, respectively (inset of Figure 3b,e).AFM was used to further analyze the thickness of the WO 3 nanosheets.As can be seen from Figures 3c  and S2, the thickness of the small WO 3 nanosheets was about 8 nm, corresponding to the sub-layer structure of the WO 3 .The thickness of the large WO 3 nanosheets was about 10 nm, and there were more wrinkles between the structures (Figures 3f and S3), which was consistent with the SEM analysis.Using the intercalation and exfoliation of bulk WO 3 , the majority of nanosheets were found to exhibit a fundamental thickness below 10 nm, approximately 10 layers of monoclinic WO 3 unit cells [16].
(200) planes of the WO3 monoclinic structure (JCPDS 43-1035), respectively [11,31].All the peaks of the exfoliated WO3 nanosheets were quite similar to the monoclinic structure of WO3 powder, indicating that the exfoliation had no effect on the crystalline structure of WO3 in the nanosheets.In contrast, the peak positions of intercalated WO3 shifted toward smaller angle values compared to the WO3 powder due to the intercalation and expansion of the interlayer distance.Therefore, intercalated WO3 can be successfully exfoliated to produce a large quantity of individual 2D WO3 nanosheets via sonication in NMP.Raman spectroscopy is additionally utilized to examine the crystalline structure of the exfoliated products.The Raman spectra of (a) large WO3 nanosheets, (b) small WO3 nanosheets, (c) NMP exfoliation, and (d) WO3 powder are shown in Figures 4b and S4.In the Raman spectrum of WO3 (curve d), the peaks at 277 and 302 cm −1 were assigned to the bending modes of O-W-O, while the peaks at 718 and 815 cm −1 were identified as the stretching modes of W-O-W of monoclinic phase [11,31,32].The Raman spectra of the The XRD pattern analysis of nanosheets also confirmed the crystalline structure.As presented in Figure 4a, the crystalline structures of the prepared large WO 3 nanosheets, small WO 3 nanosheets, NMP exfoliation, and WO 3 powder were determined through XRD analysis.For comparison, the XRD pattern of intercalated compound of WO 3 (K x W 1-x O 3 ) used for exfoliation was also studied.In the XRD pattern of WO 3 powder, there were well-defined peaks located at 23.1 • , 23.7 • , and 24.2 • , corresponding to the (002), (020), and (200) planes of the WO 3 monoclinic structure (JCPDS 43-1035), respectively [11,31].All the peaks of the exfoliated WO 3 nanosheets were quite similar to the monoclinic structure of WO 3 powder, indicating that the exfoliation had no effect on the crystalline structure of WO 3 in the nanosheets.In contrast, the peak positions of intercalated WO 3 shifted toward smaller angle values compared to the WO 3 powder due to the intercalation and expansion of the interlayer distance.Therefore, intercalated WO 3 can be successfully exfoliated to produce a large quantity of individual 2D WO 3 nanosheets via sonication in NMP.
Raman spectroscopy is additionally utilized to examine the crystalline structure of the exfoliated products.The Raman spectra of (a) large WO 3 nanosheets, (b) small WO 3 nanosheets, (c) NMP exfoliation, and (d) WO 3 powder are shown in Figures 4b and S4.In the Raman spectrum of WO 3 (curve d), the peaks at 277 and 302 cm −1 were assigned to the bending modes of O-W-O, while the peaks at 718 and 815 cm −1 were identified as the stretching modes of W-O-W of monoclinic phase [11,31,32].The Raman spectra of the nanosheets showed all the characteristic peaks of WO 3 in curves a and b, demonstrating that the nanosheets were successfully exfoliated from the WO 3 powder.It should be noted that the band intensity of nanosheets at 277 cm −1 decreased compared with the WO 3 powder.This indicates a substantial reduction in the layer number of exfoliated nanosheets, which was consistent with the remarkable decrease in the XRD intensity.
In order to further study the components of the nanosheets, XPS characterizations were carried out.Figure 4c,d shows the XPS survey and W 4f of the prepared samples.In the survey scan spectra of the prepared WO 3 samples, the presence of W and O was evident.The peaks and the atomic ratios of O to W in the exfoliated WO 3 nanosheets closely resemble those of the WO 3 powder, suggesting that the exfoliation process had no impact on the WO 3 components in the nanosheets.The W 4f core level spin split into two energy states, namely W 4f 7/2 and W 4f 5/2 , and the peak positions of these two energy levels thus were located at 35.2 and 37.3 eV, respectively.These peak positions were aligned with those reported in the literature, confirming that the nanosheets of WO 3 existed in the highest oxidation state (W 6+ ) [11,31,33].XPS analysis, combined with XRD and Raman analyses, verified that the nanosheets were successfully exfoliated with controllable sizes.Hence, it was anticipated that the electrochromic performance of the exfoliated WO 3 nanosheets would surpass that of bulk WO 3 .
In order to further study the components of the nanosheets, XPS characterizations were carried out.Figure 4c,d shows the XPS survey and W 4f of the prepared samples.In the survey scan spectra of the prepared WO3 samples, the presence of W and O was evident.The peaks and the atomic ratios of O to W in the exfoliated WO3 nanosheets closely resemble those of the WO3 powder, suggesting that the exfoliation process had no impact on the WO3 components in the nanosheets.The W 4f core level spin split into two energy states, namely W 4f7/2 and W 4f5/2, and the peak positions of these two energy levels thus were located at 35.2 and 37.3 eV, respectively.These peak positions were aligned with those reported in the literature, confirming that the nanosheets of WO3 existed in the highest oxidation state (W 6+ ) [11,31,33].XPS analysis, combined with XRD and Raman analyses, verified that the nanosheets were successfully exfoliated with controllable sizes.Hence, it was anticipated that the electrochromic performance of the exfoliated WO3 nanosheets would surpass that of bulk WO3.The electrochemical properties of the WO 3 were analyzed using the CV method.Figure S5 presents the CV curves of these deposited films carried out in a potential range of −1.0 to 1.0 V at a scan rate of 20 mV s −1 .The CV curves show cathodic and anodic currents due to the intercalation and extraction processes as follows [5][6][7]11,32,34]: where Li + , in this study, denotes the ions in the polymer lithium perchlorate electrolyte.By applying a negative potential, charges and Li ions were inserted into the WO 3 films, which rapidly changed the color of the film to dark blue (colored state).In contrast, with the application of a positive potential, the charges and Li ions were extracted from the WO 3 films and showed a milky white color (bleached state).The shapes of the CV curves of the prepared WO 3 samples were very similar and the potentials of the redox peaks did not exhibit significant differences (Figure S5).Nevertheless, the anodic and cathodic peak currents of the WO 3 nanosheets were much higher than those of the WO 3 powder and exfoliation using NMP, confirming the higher electrochemical activity of Li + in the nanosheets.The utilization of WO 3 nanosheets in the devices enabled greater charge insertion in each coloring-bleaching cycle and was expected to necessitate shorter coloration times.The increase in the cathodic and anodic peak currents was an indication of an increasing amount of active mass deposited on the substrate, due to the larger surface area of the ultrathin nanosheets.
In addition to evaluating the electrochemical performance, the electrochromic properties of the WO 3 nanosheets were further studied.The transmittance spectra of WO 3 powder, NMP exfoliation, small WO 3 nanosheets, and large WO 3 nanosheets in the bleached and colored states are presented in Figure 5a-d following the application of ±3.0 V across the two ITO electrodes.The WO 3 -based ECD demonstrated reversible color changes across the entire surface, changing between the bleached state at 3.0 V and the colored state at −3.0 V (Figure 2b).The ECDs exhibited notable variations in transmittance within the visible light spectrum range of 350 to 800 nm.The maximum optical contrast of the small WO 3 nanosheets was found to be 41.78% at 700 nm (Figure 5c), which was relatively high compared to the WO 3 powder (14.15%),NMP exfoliation (24.31%), and large WO 3 nanosheets (39.38%).Moreover, the distinct observation of the blue color was evident in the ECD by utilizing small WO 3 nanosheets, as shown in the inset of Figure 5c.This suggests that the electrochromic properties of the small WO 3 nanosheet are reasonably good within the visible light spectrum range.Ultimately, the use of smaller WO 3 nanosheets can enhance the modulation between the bleached and colored state, where a large optical modulation is highly desired in ECDs.In addition, the transmittance at 700 nm under the alternating potential between +3.0 and −3.0 V for 30 s is displayed, and the corresponding times for bleaching and coloring, defined as the duration to achieve 90% of the final change in transmittance, are recorded in Figures 5e-h and S6.Fast switching speeds of 9.2 s for bleaching and 10.5 s for coloring were obtained using small WO 3 nanosheets; these times were faster compared to those of the WO 3 powder (18.0 s for bleaching and 18.9 s for coloring), NMP exfoliation (15.9 s for bleaching and 16.8 s for coloring), and large WO 3 nanosheets (10.5 s for bleaching and 12.1 s for coloring).The ECD using small WO 3 nanosheets exhibited a significant optical contrast and reduced switching time, indicating the facile intercalation of Li + ions into the smaller WO 3 nanosheets (electrochromic layer) from the electrolyte.The coloration time of the small nanosheets was comparable to the times of the previously reported WO 3 nanosheets [11,20,35,36] (Table S1).Unfortunately, the bleaching time of the small nanosheets was a little higher than the times of the previously reported WO 3 nanosheets [11,20,35,36] (Table S1).
Another important aspect of electrochromic performance is reversibility.Coloringbleaching cycling performance was also evaluated via transmittance at 700 nm under an alternating potential of ±3.0 V (Figure 5i-l).As shown in Figure 5k, there was no obvious degradation in the transmittance modulation during the cycling process of 5000 s and a reversibility rate of 86% was found for 30,000 s (1000 cycles), demonstrating the fine electrochromic stability of the small WO 3 nanosheets.However, reversibility rates of 16%, 35%, and 74% were observed using WO 3 powder, exfoliated WO 3 without intercalation, and large WO 3 nanosheets in ECDs, respectively.The excellent cycling durability of the small nanosheets is comparable to the times of the previously reported WO 3 nanosheets [11,18,20,35] (Table S1).Another important aspect of electrochromic performance is reversibility.Coloringbleaching cycling performance was also evaluated via transmittance at 700 nm under an alternating potential of ±3.0 V (Figure 5i-l).As shown in Figure 5k, there was no obvious degradation in the transmittance modulation during the cycling process of 5000 s and a reversibility rate of 86% was found for 30,000 s (1000 cycles), demonstrating the fine electrochromic stability of the small WO3 nanosheets.However, reversibility rates of 16%, 35%, and 74% were observed using WO3 powder, exfoliated WO3 without intercalation, and large WO3 nanosheets in ECDs, respectively.The excellent cycling durability of the small nanosheets is comparable to the times of the previously reported WO3 nanosheets [11,18,20,35] (Table S1).

Conclusions
Ultrathin WO3 nanosheets with varying sizes were synthesized using the intercalation method.The small WO3 nanosheets were smaller than 1 µm and the large WO3 ranged from 2 to 5 µm.The effects of nanosheet size on the EC properties of WO3 were elucidated.The obtained high optical contrast was attributed to the high optical modulation between the colored and bleached states of the small WO3 nanosheets.In addition, the smaller nanosheets showed a decrease in coloring and bleaching times compared with the WO3 powder, NMP exfoliation, and large WO3 nanosheets, respectively.Moreover, the small WO3 nanosheets exhibited reversibility of 86%.

Supplementary Materials:
The following supporting information can be downloaded at: www.mdpi.com/xxx/s1.

Conclusions
Ultrathin WO 3 nanosheets with varying sizes were synthesized using the intercalation method.The small WO 3 nanosheets were smaller than 1 µm and the large WO 3 ranged from 2 to 5 µm.The effects of nanosheet size on the EC properties of WO 3 were elucidated.The obtained high optical contrast was attributed to the high optical modulation between the colored and bleached states of the small WO 3 nanosheets.In addition, the smaller nanosheets showed a decrease in coloring and bleaching times compared with the WO 3 powder, NMP exfoliation, and large WO 3 nanosheets, respectively.Moreover, the small WO 3 nanosheets exhibited reversibility of 86%.

Figure 1 .
Figure 1.Schematic illustration of the intercalation and exfoliation process used for the synthesis of 2D WO3 nanosheets from bulk powder.

Figure 1 .
Figure 1.Schematic illustration of the intercalation and exfoliation process used for the synthesis of 2D WO 3 nanosheets from bulk powder.

Figure 2 .
Figure 2. Schematic of (a) the fabrication of ECD and (b) the ECD performance test.

Figure 2 .
Figure 2. Schematic of (a) the fabrication of ECD and (b) the ECD performance test.
Figure S1: Size distribution of (a) small synthesized WO 3 nanosheets and (b) large WO 3 nanosheets; Figure S2: AFM image of small WO 3 nanosheets; Figure S3: AFM image of large WO 3 nanosheets; Figure S4: Magnified Raman spectra of (a) large WO 3 nanosheets, (b) small WO 3 nanosheets, (c) NMP exfoliation, and (d) WO 3 powder; Figure S5: CV curves of the prepared WO 3 samples at 20 mV s −1 in 1 M LiClO 4 containing PC; Figure S6: Magnified transmittance modulation of (a) WO 3 powder, (b) NMP exfoliation, (c) small WO 3 nanosheets, and (d) large WO 3 nanosheets during the cycling process between 14,000 and 15,000 s; Table S1: Performance comparison of EC devices made up of WO 3 nanosheets.Author Contributions: The manuscript was written with contributions from all authors.The experiments were conceptualized by D.-J.L. and S.-H.S., and the conduction of experiments and organization of data were performed by C.-A.L., K.-H.P., B.K., J.-G.A. and T.P.All authors contributed to the data analysis and discussion of the results.All authors have read and agreed to the published version of the manuscript.