Application of the [WO₂(C₅H₇O₂)₂] Complex in Hydrothermal Synthesis of WO₃ Film and Study of Its Electrochromic Properties

Featured Application

Electrochromic devices, such as smart windows.

Abstract

The goal of this work was the synthesis study of the [WO₂(C₅H₇O₂)₂] complex and its application as a precursor for the growth of WO₃ films in hydrothermal conditions, as well as evaluating the microstructural features and electrochromic properties of the formed materials. Dioxotungsten acetylacetonate was synthesized in an aqueous medium and purified. It was found that during hydrothermal treatment of the alcohol solution of the complex, acetylacetonate ligands undergo partial destructive substitution by alkoxyl groups, intensifying at temperatures above 140 °C. Considering these data and using a [WO₂(C₅H₇O₂)₂] solution, WO₃ films were grown on glass and glass/ITO substrates. The resulting films had different microstructures according to scanning electron microscopy (SEM) and atomic force microscopy (AFM): the former consisted of submicron spheres (~500 nm), distinct nanoparticles (60–160 nm), and submicron- and micron-sized ridges, while the latter consisted of 1D structures (length 350 ± 25 nm, width 110 ± 25 nm). Using Kelvin probe force microscopy (KPFM), the electron work function was determined for the film on glass/ITO substrate (4.77 eV). It was found that the electrochemical coloration process of the obtained WO₃ film can proceed in two stages, and the optical contrast is about 17.5% (at the wavelengths of 600–1100 nm). The results obtained show the prospects of applying the proposed approach to obtaining WO₃ electrochromic films with a hierarchical microstructure with the hydrothermal method using the [WO₂(C₅H₇O₂)₂] complex as a precursor.

1. Introduction

Electrochromic materials are currently attracting a great deal of research interest due to their ability to controllably change optical absorption when an electrical voltage is applied [1]. This property makes them prospective functional components for a number of applications. For example, smart windows based on these materials can dramatically improve the energy efficiency of buildings by regulating the exchange of thermal energy between the room and the environment [2,3]. Reflective and transparent displays and e-paper based on electrochromic materials are also promising in terms of energy savings and are perfectly readable in outdoor lighting and safer for the eye than light-emitting displays [4,5,6]. The possibility of using such materials in devices for adaptive camouflage is also of great interest [7]. Furthermore, electrochromic materials can be used as auxiliary elements of non-optical devices—for example, as charge indicators in batteries—or they can perform electrochromic and energy functions simultaneously [8,9].

The most popular electrochromic materials are various conducting polymers (such as polyaniline) or transition metal oxides, such as NiO or WO₃ [1,2,10]. The latter have a number of advantages, such as high stability, spectral purity, and coloration efficiency. Among transition metal oxides, tungsten oxide WO₃ stands out due to its characteristic large optical contrast in the visible and near-infrared range, high ionic conductivity, and high coloration efficiency [11,12,13,14]. In addition to its use as an electrochromic material, tungsten oxide is also a promising photocatalyst for water splitting [15], photochromic material [16,17,18], photoreceptor [19], material for resistive memory with random access [19], gas sensors [16,20], lithium-ion batteries [21], etc.

Since electrochromic processes occur on the surface of a material, its dispersity and microstructure have a great influence on their course, so the use of nanomaterials with a developed surface, such as hierarchic materials, is promising in this field. Hierarchic materials may also have specific active centers, which may further improve the functional properties of such materials [22,23,24,25,26,27]. Hydrothermal synthesis is known as the most convenient approach to the preparation of hierarchic oxide nanomaterials [23,27,28,29,30]. Fine control over product morphology is possible by varying the duration and temperature of the synthesis and other parameters. Hydrothermal synthesis can also be used to grow oxide films, avoiding film deposition from powders [18,31,32,33,34]. The choice of precursor plays a key role in the production of material using the hydrothermal method. In the case of the hydrothermal synthesis of tungsten(VI) oxide, the following precursors are mostly used: peroxotungstic acid [29,35,36], ammonium or sodium tungstate [37,38,39], and freshly precipitated tungsten oxide [27,40,41,42]. In our opinion, expanding this list with complex compounds will allow a finer control over the product structure and properties by choosing a precursor that has the necessary geometry at the molecular level. In particular, obtaining metal oxides with the hydrothermal method from the corresponding acetylacetonates may be of interest. Tungsten acetylacetonate [WO₂(C₅H₇O₂)₂] use can result in special oxide structure due to presence of the WO₂²⁺ cation in its structure.

At the same time, the [WO₂(C₅H₇O₂)₂] complex is little described in the literature, and it was scarcely used for WO₃ preparation. As an example, we were able to find only work [43], in which tungsten(VI) oxide films were obtained by the decomposition of the complex under the influence of irradiation. This is due to the relatively complex method of synthesizing it compared to other precursors of tungsten oxide and acetylacetonates of other metals. It is usually prepared by the prolonged treatment of WO₂Cl₂ in benzene or toluene in the presence of acetylacetone [43,44,45]. However, we were unable to find in the literature any attempt to obtain this complex using a precipitation technique similar to that used to prepare β-diketonates of dioxomolybdenum(VI), which involves the precipitation of fresh MoO₃ from molybdate solution under the action of strong acid, which then reacts with β-diketone previously added to the system [46,47]. Perhaps the successful synthesis of [WO₂(C₅H₇O₂)₂] using a similar methodology will contribute to further wider application of this complex in the synthesis of various nanomaterials, including WO₃.

Thus, the aim of this work was to study the synthesis of the [WO₂(C₅H₇O₂)₂] complex and its application as a precursor in the hydrothermal synthesis of WO₃ film, as well as to evaluate its microstructural features and electrochromic properties.

2. Materials and Methods

2.1. Materials

The reagents used in this study were as follows: n-butanol (C₄H₉OH, 99%, Ekos–1, Moscow, Russia) for solution preparation, ammonium paratungstate ((NH₄)₁₀W₁₂O₄₁·4H₂O, >99%, Chemcraft, Kaliningrad, Russia) as starting reagent for tungsten complex preparation, HCl (35%, EKOS–1, Moscow, Russia) and acetylacetone (C₅H₈O₂, >98%, EKOS–1, Moscow, Russia) for complex preparation, polyethylene glycol (PEG or PEO) (M = 1500, Chimmed, Moscow, Russia), propylene carbonate (99.7%, Sigma Aldrich, St. Louis, MO, USA), and lithium perchlorate (LiClO₄, >99.9%, Sigma Aldrich, St. Louis, MO, USA) for electrolyte preparation.

2.2. [WO₂(C₅H₇O₂)₂] Synthesis

The procedure for the synthesis of the [WO₂(C₅H₇O₂)₂] complex was similar to that proposed in [46] for obtaining a similar molybdenum complex. An excess of acetylacetone was added to an aqueous solution of ammonium paratungstate, after which concentrated hydrochloric acid was added dropwise under stirring to the system until pH = 1.0. As acid was added, a pale-yellow precipitate formed in the system. After synthesis, the reaction system was kept for 12 h; as a result, the precipitate turned brown. The resulting precipitate was separated and dried at 40 °C for 12 h. After drying and characterization of the product by simultaneous TGA/DSC analysis and IR spectroscopy, the powder was placed in n-butanol medium followed by a heat treatment at 110 °C (2 h), which resulted in partial solubilization of the material. The obtained solution was separated from the remaining precipitate and further used to remove the solvent using a rotary evaporator. The brown powder thus isolated was further characterized using infrared spectroscopy and identified as [WO₂(C₅H₇O₂)₂].

2.3. Formation of WO₃ Films

WO₃ films were grown on pre-cleaned glass and glass substrates with a sublayer of transparent ITO electrode (resistance 6 Ohm/ϒ). For this purpose, 1.75 mL of [WO₂(C₅H₇O₂)₂] solution in n-butanol was placed in a 5 mL Teflon-lined stainless-steel autoclave in which the corresponding substrate was vertically fixed in a Teflon support. The autoclave was then placed in a muffle furnace, heated to 220 °C at the rate of 1.5 °C/min, and held at this temperature for 2 h. When growing the WO₃ film on a glass/ITO substrate, in order to increase the thickness of the material, this procedure was repeated three times. In order to select the temperature regime, studies on the behavior of the [WO₂(C₅H₇O₂)₂] complex during hydrothermal treatment of its solution in n-butanol were carried out beforehand. During these studies, the solution of the complex was successively heated to higher temperatures in the range of 120–220 °C in steps of 10 °C, followed by cooling for spectral measurements.

2.4. Instrumentation

An InfraLUM FT-08 infrared spectrometer (Lumex, St. Petersburg, Russia) was used to record vibrational spectra of the complex solution. The spectra were recorded in the range of 350–2000 cm⁻¹. For analysis of the powders, their suspensions in nujol mull were prepared and then placed as films between KBr lenses.

A UV-Vis spectrophotometer SF-56 (OKB Spectr, St. Petersburg, Russia) was used to record spectra in the UV, visible, and near-IR ranges using a quartz cuvette (1 mm thick). The precursor solution was not additionally diluted during the study of its hydrothermal treatment behavior and was returned to the autoclave after recording spectra for further thermal treatment.

Thermal behavior of the prepared mixture of the [WO₂(C₅H₇O₂)₂] complex and tungsten oxide in the range of 25–600 °C was studied using the synchronous thermal analyzer SDT Q600 (TA Instruments, New Castle, DE, USA) in air flow (250 mL/min), heating speed 10°/min.

X-ray diffraction analysis (XRD) of the WO₃ film grown on glass was performed using a D8-Advance diffractometer (Bruker, Billerica, MA, USA; CuKα = 1.5418 Å, Ni filter, E = 40 keV, I = 40 mA, range 2θ 5–80°, signal accumulation time 0.3 s/point, resolution 0.02°).

Raman spectroscopy of the WO₃/glass film was performed in the range of 200–1200 cm⁻¹ on a R532 spectrometer (EnSpectr, Chernogolovka, Russia). The scan accumulation time was 500 ms, the number of scans was 150, and the laser wavelength was 532 nm.

The microstructure of the films was studied using scanning electron microscopy (SEM, NVision 40 dual-beam workstation, Carl Zeiss, Inc., Jena, Germany). In the course of SEM measurements, we also used the Oxford Instruments X MAX 80 microprobe analyzer, with the help of which we clarified the chemical composition of the formed oxide films.

The grown WO₃ films were also studied using atomic force microscopy (AFM) on the NT-MDT Solver PRO-M (NT-MDT, Zelenograd, Russia). Both surface morphology and local electrophysical properties of the material were characterized. For the latter purpose, Kelvin Probe Force Microscopy (KPFM) technique was used. Scanning was performed in the semicontact mode using ETALON HA_HR probes (ScanSens, Bremen, Germany) with a conductive W₂C coating (probe tip curvature radius ≤ 35 nm). Within the framework of the KPFM, we also measured the value of the electron work function φ_oxide for the oxide films.

To study the electrochromic properties of the WO₃ film on a glass/ITO substrate, a specialized two-electrode cell was assembled. The glass/ITO substrate was used as the auxiliary electrode. A polymer electrolyte composed of a mixture of polyethylene glycol, propylene carbonate, and lithium perchlorate was placed between the WO₃ film and the auxiliary electrode (the ratio of polyethylene glycol monomers to the amount of lithium perchlorate was 8:1). After the cell was sealed, it was placed in a chamber of a UV-Vis-spectrophotometer SF-56 to measure its optical properties. The difference in potentials was supplied by potentiostat-galvanostat P-45X with an attachment for measurement of electrochemical impedance FRA-24M (Electrochemical Instruments, Chernogolovka, Russia). The same device was used to record cyclic voltammetry curves (speed of voltage change 50 mV/s).

3. Results

3.1. Synthesis and Characterization of [WO₂(C₅H₇O₂)₂]

After drying, the synthesis product was subjected to synchronous thermal analysis in an air current up to 600 °C (Figure 1a). In the initial stage of heating, a degree of mass loss was observed up to 240 °C, associated with the removal of sorbed atmospheric gases and residual solvent (Δm value reached 7.7%). A further increase in temperature led to a sharper decrease in mass, as can be seen from the TGA curve, with Δm ~ 7.6% at temperatures of 240–280 °C. This process was accompanied by intense energy release (with a maximum of thermal effect at 280.4 °C), which is probably associated with the oxidation of the organic fragments of the [WO₂(C₅H₇O₂)₂] complex. The third mass loss step was at 280–430 °C. As can be seen, it is wider and less intense (Δm = 3.5%) compared to the previous one. In this case, the process is also associated with the release of energy (the maximum of the corresponding exo-effect is around 393 °C), which can be attributed to the oxidation of residual carbon formed during the thermal destruction of the complex. Following further temperature increase, the mass of the material stabilized, i.e., the main processes of material transformation completed at ~430 °C. The total mass loss throughout the temperature range under study was ~18.8%. The obtained results confirmed the presence of organometallic tungsten compound in the studied material composition and allowed for the establishment of the peculiarities of its thermal behavior in the air atmosphere.

The results of IR spectroscopy of the synthesis product (Figure 1b) confirm that it contains tungsten acetylacetonate. The spectrum shows bands characteristic of C=O and C=C group vibrations in the coordinated chelate ligand with maxima at 1586 and 1537 cm⁻¹; according to the literature [43,48], it is at these wave numbers that vibrations of C₅H₇O₂- ligands in the complex [WO₂(C₅H₇O₂)₂] are observed. However, these bands are low-intensity, and in the region of 520–900 cm⁻¹, there is a broadened band of complex shape formed by a combination of different absorption bands from the vibration of W–O bonds. The shape of this band is characteristic of tungsten oxide. When the synthesis product is dissolved in n-butanol, some amount of a grayish-yellow precipitate does remain, presumably related to tungsten oxide. The IR spectrum of the material obtained by recrystallization from alcohol solution exhibits more intense absorption bands at 1586 and 1537 cm⁻¹ and significant changes in absorption in the 520–900 cm⁻¹ range. The obtained result thus indicates that the [WO₂(C₅H₇O₂)₂] complex purified from WO₃ was obtained.

3.2. Transformations of Tungsten Complex in Hydrothermal Conditions

After the isolation and characterization of the complex, its behavior during the stepwise hydrothermal treatment of its solution at different temperatures was studied. The temperatures were varied from 120 to 220 °C in steps of 10 °C. Changes in the composition of the complex were monitored using UV spectrophotometry (Figure 2a). It can be seen that the intensity of the absorption band with a maximum at 272 nm, corresponding to the coordinated acetylacetonate ligands, decreases with hydrothermal treatment, with the process intensifying after transition to temperatures above 140 °C. The observed changes in the spectral properties can be explained by the gradual destruction of the chelate ligand and its replacement by butoxyl groups, as occurs with similar complexes during the thermal treatment of their alcohol solutions [49,50]. Using the data obtained, the values of the degree of substitution of chelate ligands at each stage of solution processing were calculated using the following formula:

$$\mathsf{\alpha }=100\cdot \frac{{(\mathrm{D}}_{272 \mathrm{nm}}^0{-\mathrm{D}}_{272 \mathrm{nm}}^{\mathrm{T}})}{{\mathrm{D}}_{272 \mathrm{nm}}^0},$$

(1)

where α is the degree of C₅H₇O₂–ligand substitution for alkoxyl groups (%), ${\mathrm{D}}_{272 \mathrm{nm}}^0$ is the optical density of the solution at 272 nm before heat treatment, and ${\mathrm{D}}_{272 \mathrm{nm}}^{\mathrm{T}}$ is the optical density of the solution at 272 nm after heat treatment at the temperature T. The dependence of the resulting α values on the hydrothermal treatment step can be represented as a curve (Figure 2b). This curve shows that the process intensifies at temperatures above 140 °C but gradually stabilizes at temperatures above 170 °C. As a result, after the heat treatment of the solution at 220 °C, the α value was 83%. Thus, the process of partial destructive substitution of acetylacetonate ligands for C₄H₉O groups with the formation of a heteroligand complex of the composition [WO₂(C₅H₇O₂)_2−x(C₄H₉O)_x] was studied using UV spectrophotometry.

Taking into account the spectral data obtained, the following regime was chosen for the growth of WO₃ films under hydrothermal conditions: heating to 220 °C at a rate of 1.5 °C/min and holding at this temperature for 2 h.

3.3. Characterization of the Grown WO₃ Films

XRD analysis and Raman spectroscopy (Figure 3) were employed to study the structure and phase composition of the obtained WO₃ film on glass. The diffractogram shown in Figure 3a reveals no discernible crystalline phases—the only feature is a wide halo in 13–38°, which is characteristic of the glass substrate. This indicates that the grown film is X-ray amorphous, due to being either too highly dispersed or amorphous in nature. Further study with Raman spectroscopy (Figure 3b) revealed that all visible bands are also characteristic of glass. Namely, the bands with maxima at 1183, 1097, and 954 cm⁻¹ can be attributed to the Q⁴, Q³, and Q² modes of Si-O stretching vibrations, the wide and low-intensity band around 794 cm⁻¹ is related to Si vibration in its tetrahedral oxygen cell, and bands at lower wave numbers are caused by Si-O-Si bending vibration modes, including from the rings of three Si around 570–600 cm⁻¹ [51,52,53,54,55]. Many of the band’s characteristics of W-O-W and O-W-O in various modifications of WO₃, such as those with maxima around 800–820 cm⁻¹, 650 cm⁻¹, and 320–330 cm⁻¹, overlap with glass bands. However, a rather intense band around 690–715 cm⁻¹ is usually observed for WO₃, and an overlap with glass bands would be minimal in this region [56,57]. In our case, no band can be observed with certainty in this region. Thus, the grown oxide film is most likely too highly dispersed and thin to be observed by Raman spectroscopy. Combined with the XRD analysis results, it follows that the WO₃ film seems to be very highly dispersed, which should result in a very developed surface with a large specific area. This, in turn, should positively impact the electrochromic properties of the material since associated processes occur at the oxide’s surface.

To further characterize the obtained films, their transmittance spectra were recorded in the visible range (Figure 4). It can be seen that the WO₃ film on the glass substrate has a very low transparency—only 53% transmittance at a wavelength of 600 nm. Subtracting the transmittance spectrum of the original glass substrate shows that the contribution to the absorption directly from the oxide film is about 51% (at 430 nm). The WO₃ film on the glass/ITO substrate, despite the fact that it was deposited in three cycles, contributes significantly less to the absorption of the corresponding sample—only up to 21.5% (at 460 nm). Such a difference in transparency between the two films formed on the surfaces of different substrates can partially be explained by the antireflection effect in the case of the glass/ITO substrate.

Then, the WO₃ coating grown on a glass substrate was examined using SEM. The microphotographs (Figure 5) show that the oxide film is formed from several microstructural units uniformly occurring on the surface of the sample. Firstly, spherical formations with a diameter of 400 to 550 nm consisting of smaller nanoparticles 15–55 nm in size can be distinguished (Figure 5). Second, elongated ridges of varying lengths, mostly less than 1 μm, and widths from 100 to 400 nm can be distinguished. Some of these ridges are formed from two or more units which were originally submicron spheres. Finally, nanoparticles of rounded or slightly elongated shape can be distinguished—the rounded ones are generally 50–55 nm in diameter, while the more elongated ones are 80–140 nm in length. It can be assumed that such particles act as nuclei for the formation of submicron spheres and ridges during hydrothermal film growth. Thus, the SEM results clearly demonstrate the complex nature of the microstructure of the grown film formed from both anisotropic structures (submicron ridges, elongated nanoparticles) and hierarchically organized submicron spheres. The absence of a clear texture and uniform microstructure may indicate that during the chosen hydrothermal treatment regime, mostly nucleation occurs (resulting in nanoparticles), with processes of further nuclei aggregation and Ostwald ripening (resulting in ridges and hierarchical submicron spheres) only in the initial stages. Such nonhomogeneous microstructure is also facilitated by amorphous substrate since, as can be seen below from the AFM results, the microstructure for film grown on the ITO layer (which is crystalline) differs significantly and is more uniform. The microstructure of the WO₃ film on glass is also consistent with its XRD results and Raman spectrum, which clearly indicate its high dispersity and X-ray amorphous nature. The surface of the formed film was also studied using X-ray elemental microanalysis, the results of which confirmed the formation of the WO₃ film and the absence of impurities not related to the substrate material.

The microstructural features and local electrophysical properties of the formed oxide films were additionally studied using different AFM techniques.

The results of the AFM study of the WO₃ film formed on the surface of the glass substrate are shown in Figure 6. The topographic scans show three types of formations constituting the film microstructure: ridges, individual 350–450 nm formations resembling submicron spheres of slightly distorted shape, and rounded nanoparticles between them. The latter evenly cover the area between the ridges and individual submicron formations, which is especially well seen in Figure 6c,e,f, and are mostly about 60 ± 15 nm in size, although larger particles as big as 160 nm also occur. The ridges detected using AFM are 2–5 µm long and widen toward the middle, reaching a maximum width of more than 1 µm, and consist of elongated rods adjacent to each other (Figure 6c,e). The map of the surface potential distribution shows that there is some concentration of potential on the ridges themselves and nanoparticles between them, and on the contrary, at the edges of the ridges the potential is somewhat reduced. This indicates a shift in the density of charge carriers from the boundaries between the surface areas with different morphologies and the presence of homojunction at the ridge boundaries in contact with individual nanoparticles. At the same time, the difference in the potentials between the different surface areas mostly does not exceed 100 mV and indicates the presence of sufficient (although not high) conductivity of the material. Thus, the AFM data—both the topography and the surface potential distribution map—reliably indicate the continuous nature of the grown WO₃ film.

The surface of the WO₃ film grown on the glass/ITO substrate was also studied using AFM. It can be seen that the film significantly differs in its microstructure from both the pure section of the substrate with an ITO layer and the WO₃ film on a glass substrate and consists of one-dimensional structures 350 ± 25 nm long and 110 ± 25 nm wide. At the same time, under these rods, as can be seen from the topographic scans and the image in the phase contrast mode (Figure 7b,c,e,f), are either similar but partially hidden rods or rounded nanoparticles with a diameter of 80–90 nm, which clearly do not belong to the ITO sublayer, whose nanoparticles are several times smaller (Figure 7a). The significant difference in the microstructure of this film from that deposited on the glass substrate can be explained by the influence of the ITO sublayer, whose particles serve as seeds in the formation of the WO₃ film. The surface potential is very uniformly distributed over the film surface—it is difficult to distinguish any areas with clearly different potential values in the studied area, and the variation in values is only about 25 mV (Figure 7d). This indicates higher conductivity of this film compared to the film on the glass substrate and the absence of any significant barriers for charge transfer at the boundaries between the particles.

Using the recorded surface potential distribution maps, we also calculated the values of the electron work function for both films. Thus, the work function for the WO₃ film on the glass substrate was 4.22 eV, and on the glass/ITO substrate it was 4.77 eV. In general, the work function for WO₃ can, according to the literature, range from 4.6 to 6.4 eV [58,59]. Accordingly, the value of the work function for the film on the glass/ITO substrate is in good agreement with the data reported in the literature. If we consider the WO₃ film on the glass/ITO substrate, its work function is closer to the lower boundary of the marked range, which may indicate a relatively high content of oxygen vacancies and positively influence the conductivity of the tungsten oxide since it is an n-type semiconductor.

Thus, the AFM results indicate that a hierarchically organized WO₃ film is formed on the surface of a glass/ITO substrate using the [WO₂(C₅H₇O₂)₂] complex as a precursor under hydrothermal conditions.

3.4. Electrochromic Properties of the Grown WO₃ Film

WO₃ electrochromism is explained by reversible reduction and oxidation reactions accompanied by the intercalation/deintercalation of lithium ions, which can be expressed by the following equation:

$${\mathrm{WO}}_3\left(\mathrm{colorless}\right)+\mathrm{x}\left({\mathrm{Li}}^{+}{+\mathrm{e}}^{-}\right){=\mathrm{Li}}_{\mathrm{x}}{\mathrm{W}}_{\left(1-\mathrm{x}\right)}^{+\mathrm{VI}}{\mathrm{W}}_{\mathrm{x}}^{+\mathrm{V}}{\mathrm{O}}_3(\mathrm{dark} \mathrm{blue})$$

(2)

Accordingly, to study the electrochromic properties of WO₃, it is necessary to monitor both the changes in the optical parameters of the film and the electrochemical processes taking place. Therefore, first of all, to characterize the electrochromic properties of the WO₃ film grown on the glass/ITO substrate surface, cyclic voltammetry (CV) was recorded in parallel with monitoring changes in the sample transmittance at λ = 600 nm during this experiment (Figure 8a,b). During the first two cycles of recording, the film and its electrochemical behavior were stabilizing; Figure 8 shows CV and data on the change in transmission during its recording for the stabilized sample. Several effects can be distinguished. First, a broad and intense peak during oxidation of the material starting at about 0.2 V, peaking at 1.31 V (I = 1.24 mA), and ending at about 2 V. This effect clearly refers to the oxidation of the material with the deintercalation of lithium ions and bleaching of the cell—Figure 8b shows that at around 0.2 V, the bleaching begins, and the curve of change in transmittance begins to reach a plateau at around 1.6 V, when the intensity of the electrochemical process had already passed through the maximum and began to decrease noticeably. It is interesting that the coloration after the stabilization of the sample seems to possess a two-step character: at first, upon the decrease in potential from 2.2 V to ~0.9 V, the film is still slightly bleached; upon reaching ~0.9 V, it begins to color gradually, and then, upon reaching −2.5 V, the rate of the coloration process sharply increases. Indeed, some increase in the current started at around 0.9 V on the cathode branch, reached its maximum at 0.295 V (I_max = −0.023 mA), and gradually subsided. We could not distinguish any separate peak at −2.5 V—the process is obviously low-intensive in terms of current. The presence of two stages can probably be explained by the formation of two phases after the stabilization of the material—one begins to oxidize earlier but makes a smaller contribution to coloration, and the second begins to oxidize later and makes a larger contribution to coloration. The second phase most likely intercalates fewer lithium ions than the first, which is why it has a lower current intensity on CV. It could then be assumed that the first phase is closer in its crystal structure to the intermediate phase between WO₃ and LiWO₃ formed during lithium intercalation upon reduction—the formed intermediate phase is apparently not completely oxidized back to WO₃, so the structure is preserved and ready for the intensive intercalation of lithium ions upon earlier reduction.

During CV recording, the maximum optical contrast at 600 nm reaches 7.5%, but it can be seen from Figure 8c that when a constant potential is applied, a much greater effect is achieved—the contrast is approximately 17.5% and remains at this level (Figure 8c). From the transmittance spectra of the film in the different states (Figure 8d), it can be seen that the sample does not return to its original state as a result of cycling, darkening by an average of 5%, and the contrast between the colored and bleached states in the 550–1100 nm range varies between 11 and 18%, increasing with wavelength. In the region up to 550 nm, i.e., in the violet and blue parts of the visible spectrum, the transmittance of the film drops significantly less when colored, due to which the colored film shows a blue coloration, which is typical for WO₃-based electrochromic materials. Figure 8c also shows that the processes of coloration and bleaching are asymmetrical: the coloration is much slower to plateau than the bleaching process, so the response time (during which 90% of the maximum optical contrast during the corresponding half cycle is reached) for coloration is 25.5 s, while for bleaching it is only 1.5 s. Such asymmetry may indicate the predominance of an oxide phase with a highly stable structure in the sample, which makes the processes of tungsten(VI) reduction and lithium ion intercalation difficult.

From the obtained data on optical contrast, coloration efficiency (CE) at 600 nm was calculated according to the following formula:

$$\mathsf{\eta }=\frac{\log \left(\frac{{\mathrm{T}}_{\mathrm{bleached}}}{{\mathrm{T}}_{\mathrm{colored}}}\right)}{\mathrm{Q}},$$

(3)

where η is CE, T_bleached is transmittance in the bleached state, T_colored is transmittance in the colored state, and Q is charge density (charge inserted into the film per area unit). The resultant value for the grown WO₃ film was 15.5 cm²/C. In the literature, CE values for WO₃ vary in a large range, reaching up to 141 cm²/C [11] and even 193 cm²/C [12]. Thus, our value seems to be relatively low. However, it should be noted that in most works, CE and other electrochromic properties are studied in three-electrode cells with liquid electrolyte (usually LiClO₄ in propylene carbonate), which is not the case in our experiment, where a polymeric gel electrolyte was used in a two-electrode cell. Polymeric and liquid electrolytes can differ significantly in their ionic conductivity, which might lead to lower CE in the case of electrolytes with less mobile ions. Other parameters, such as the voltage at which coloration and bleaching is observed, can also differ between two electrolytes. Consequently, it would be more correct to compare the electrochromic characteristics of our material with those reported in other works in which polymeric electrolytes were used (preferably, also based on polyethylene glycol and LiClO₄). In work [60], WO₃-PEO composite was used as electrochromic material in a cell with gel electrolyte consisting of propylene carbonate, PEO, LiClO₄, and ferrocene (which also exhibits electrochromic properties and can act as an anodic species [61]). The CEs for their composites were 24.3–61.4 cm²/C at 700 nm, depending on the WO₃:PEO ratio. Considering that we have calculated the CE for 600 nm, and contrast at 700 nm is higher in our case, our CE for 700 nm would most likely be close to the lower values from [60]. The fact that ferrocene also contributed to coloration and thus to CE values in [60] should also be taken into account. It should also be noted that the PEO:LiClO₄ ratio was different in [60] compared to ours. In work [62], the optical contrast for a WO₃ electrochromic reflective device based on polymeric electrolyte was 5–6% in the potential range of −3–3 V, which is a smaller contrast than ours and a larger potential window. No values for CE were provided by the authors. However, that was a reflective and not a transparent device, and a different polymer was used by the authors of [62]. In [63], the researchers stated that the WO₃ film obtained by sol-gel could not be colored in LiClO₄-PEG electrolyte, and even the addition of TiO₂ nanoparticles to the electrolyte resulted only in contrast about two times lower than that achieved by us in this study. In [64], a device comprising electrodeposited WO₃ film and Prussian blue film as electrochromic materials and a mixture of PEO, propylene carbonate, and LiCLO₄ as gel electrolyte showed a CE of 82 cm²/C; however, coloration and bleaching proceeded much slower than in our case—60 and 30 s (at 40 °C) vs. 25.5 and 1.5 s. Lastly, the stability of the electrochromic properties was studied for the grown WO₃ film. As can be seen from Figure 9, the sample showed good stability, retaining its contrast value after 100 cycles at −3.2 and 2.0 V (the time of each half-cycle was 5 s).

Overall, the results of the conducted measurements of the electrochromic properties of the material show good prospects for the application of the proposed approach to the growth of hierarchically organized WO₃ films in hydrothermal conditions using the [WO₂(C₅H₇O₂)₂] complex as a precursor.

4. Conclusions

In this study, the synthesis of the [WO₂(C₅H₇O₂)₂] complex was investigated. Its efficiency as a precursor in the hydrothermal synthesis of WO₃ film was shown, and microstructural features and electrochromic properties of such a film were evaluated. Thus, the behavior of the synthesized complex in n-butanol medium under hydrothermal conditions in the temperature range of 120–220 °C was studied. It was found that the chelate ligand undergoes partial destructive substitution for the alkoxyl fragment upon an increase in temperature, and the process is significantly intensified at a processing temperature above 140 °C. Using these data, the formation of WO₃ films on glass substrate and glass/ITO substrate under hydrothermal conditions using an alcohol solution [WO₂(C₅H₇O₂)₂] was studied. It was found that the oxide films on different substrates exhibit different microstructures: the WO₃ film on a glass/ITO substrate consists of nanorods 350 ± 25 nm long and 110 ± 25 nm wide, while the film on a glass substrate is formed by a mixture of submicron spheres (~500 nm diameter), ridges of submicron and micron sizes, and individual nanoparticles 60–160 nm in size. According to KPFM data, the WO₃ film on a glass/ITO substrate is characterized by an electron work function of 4.77 eV. This film also showed sufficient transparency for use as a transparent component of optical devices, in particular, electrochromic ones. The optical contrast for the studied material was about 17.5% at 600 nm, as well as in the near-infrared range, which indicates the prospects for using the proposed approach to the growth of electrochromic hierarchically organized WO₃ films in hydrothermal conditions using the [WO₂(C₅H₇O₂)₂] complex as a precursor. It was also found in the course of CV recording that the coloration process for the obtained WO₃ film at varying potentials can proceed in two steps after the stabilization of the electrochemical behavior of the system, which provides additional information on the mechanism of the processes taking place. The obtained WO₃ film showed good stability of electrochromic properties after 100 cycles of coloration and bleaching, practically not losing any contrast, and a coloration efficiency of 15.5 cm²/C in the tested two-electrode cell with polymeric electrolyte.