Chemical Deposition Method for Preparing VO₂@AlF₃ Core–Shell-Structured Nanospheres for Smart Temperature-Control Coating

Abstract

Vanadium dioxide (VO₂) has become one of the most promising smart temperature-controlled thin-film materials due to its reversible phase transition between a metallic and an insulating state at approximately 68 °C, accompanied by negligible volume change and excellent optical modulation properties. However, the practical application of VO₂ is still limited by its relatively high phase transition temperature and susceptibility to oxidation. To address these two major shortcomings, this study employed a one-step hydrothermal method to prepare a VO₂ nanopowder, followed by a chemical precipitation method to form a VO₂@AlF₃ core–shell structure. The coated nanoparticles were then dispersed in a PVP ethanol solution, coated onto a glass substrate, and evaluated for performance. The experimental results indicate that when the molar ratio of VO₂ to AlF₃ reached 1:1, the phase transition temperature of VO₂@AlF₃ was effectively reduced to 50.3 °C, significantly lower than the original temperature of 68 °C. Additionally, the material exhibited favorable optical properties, with a solar modulation ability (ΔT_sol) of 17.2% and a luminous transmittance (T_lum) of 36.3%. After calcination in air at 300 °C for 3–6 h, the VO₂ core remained oxidation-resistant and maintained excellent phase-change thermal insulation properties.

1. Introduction

In the face of increasingly scarce natural resources and the trend of global warming, the development of new energy sources and the promotion of energy conservation have become critical global issues. Currently, building energy consumption accounts for approximately half of EU’s total social energy usage. Within this category, energy loss through glass doors and windows accounts for 25% to 35% of total building energy consumption. Therefore, improving the energy efficiency of windows is a key aspect of overall building energy conservation [1,2]. Solar radiation is primarily in the wavelength range of 200–2000 nm, with the following distribution: the ultraviolet (UV) wavelength band (290–380 nm) accounts for approximately 7% of the energy; the visible light wavelength band (380–760 nm) accounts for 46% of it; the near-infrared band (760–2500 nm) accounts for 44%; and the far-infrared band (≥2500 nm) accounts for only 3% [2]. Therefore, in summer, it is necessary to minimize the transmission of visible light and near-infrared radiation through windows to reduce the indoor temperatures. Conversely, in winter, maximizing the transmission of these same wavelengths can help increase indoor warmth and improve energy efficiency. Therefore, to reduce building energy consumption, researchers have explored many innovative window technologies. Among these, thermochromic smart windows have emerged as a particularly promising solution, attracting growing attention in global research efforts [3].

Since Morin’s [4] discovery in 1959 that VO₂ exhibits a semiconductor-to-metal phase transition, extensive research has been conducted on its various crystal structures. It has been discovered that VO₂ undergoes a reversible phase transition from the monoclinic phase (M) to the rutile phase (R) at temperatures exceeding 68 °C. During this transition, the material shifts into a metallic state, demonstrating high to infrared radiation (>70%), while maintaining high visible light transmittance (>50%) [5]. This unique dynamic optical modulation capability enables VO₂-based coatings to significantly reduce building energy consumption—simulation studies indicate that cooling energy consumption during summer can be reduced by 30%–50% [6]. In recent years, both domestic and international researchers have proposed intelligent radiative coatings based on VO₂ for comprehensive building applications, including roofs [7], walls [8], and windows [9]. However, the phase-change temperature (T_c) of VO₂ is higher than the optimal operational temperature range for practical applications; furthermore, VO₂ is easily oxidized under ambient conditions. These are critical challenges that hinder its widespread practical implementation [10].

Jiang [11] successfully synthesized VO₂ thin films with different oxygen flow ratios on quartz glass substrates through reactive magnetron sputtering. Without doping elements, the phase transition temperature can be precisely controlled between 46 °C and 72 °C by adjusting the oxygen flow ratio. Compared to conventional magnetron sputtering techniques, this method facilitates the deposition of high-quality thin films. The VO₂ thin films deposited using this method exhibit excellent performance characteristics. Zou [12] and his collaborators revealed an efficient one-step rapid hydrothermal synthesis strategy, employing ammonium metavanadate as the precursor and hydrazine as the reducing agent to successfully prepare monoclinic VO₂ nanoparticles. This study demonstrated that, under optimized precursor concentration and reaction temperatures exceeding 340 °C, pure VO₂(M) nanoparticles can be effectively synthesized. However, several challenges remain in the practical application of VO₂ in real-world scenarios [13,14]. Furthermore, VO₂ exhibits a relatively high refractive index, which may result in reduced visible light transmittance [15]. Additionally, VO₂ has poor oxidation resistance and weatherability [16]. To mitigate these issues, researchers commonly employ multilayer structures incorporating low-refractive-index membranes or apply a protective outer shell with a low refractive index to the VO₂ layer. LONG Shiwei [17] used magnetron sputtering to make a three-layer structure consisting of WO₃/VO₂/WO₃, which not only effectively improved the optical properties of VO₂ but also provided protection for the VO₂ layer. ZHANG Jing [18] constructed a VO₂/SiO₂ bilayer film, significantly improving visible light transmittance. ZHU Jingting [19] prepared V_xW_(1−x)O₂@SiO₂ core–shell nanoparticles that showed greatly enhanced oxidation resistance, with moderately enhanced visible light transmittance. GAO Yanfeng [20] prepared VO₂@SiO₂ core–shell-structured nanoparticles, mixed them with polyurethane (PU), and coated them onto polyethylene terephthalate (PET) to produce excellent thermochromic films with good transmittance and high stability under acidic conditions. Tong [21] used an alumina shell to protect VO₂ nanoparticles from oxidation at high temperature and corrosion in H₂O₂ solution. Chen et al. [22] reported that VO₂@ZnO core–shell nanoparticles significantly enhanced the environmental durability of VO₂-based membranes.

Some scholars have applied an inert layer to VO₂ to form a core–shell structure. AlF₃ nanoparticles, known for being chemically stable and corrosion-resistant, are commonly utilized in the preparation of electrode protective materials. VO₂@AlF₃ core–shell-structured nanopowders with an AlF₃ coating on VO₂ were synthesized by a solvothermal method and showed excellent stability during high-temperature air annealing (350 °C) and exposure to acid solutions. Furthermore, an appropriately thick AlF₃ shell layer has almost no effect on the solar modulation ΔT_sol of VO₂ nanoparticles.

This study used a one-step hydrothermal method to synthesize VO₂ powder with high solar light regulation capacity (T_sol). Then an amorphous AlF₃ layer was prepared via chemical precipitation and uniformly deposited onto the surface of VO₂ nanoparticles, forming a core–shell structure VO₂@AlF₃ composite powder. The influence of the VO₂ to Al³⁺ ratio on the thickness of the shell was systematically investigated. Based on the findings, a novel core–shell-structured powder was developed and combined with the polymers PVP and ethanol to produce a temperature-responsive film. Finally, the film’s phase transition temperature, optical properties, and oxidation resistance were evaluated.

2. Experimental Section

2.1. Materials

Ammonium metavanadate (NH₄VO₃, 99%), hydrazine monohydrochloride (N₂H₄·HCl, 99.7+%), aluminum nitrate nonahydrate (Al(NO₃)₃·9H₂O, 99%), ammonium fluoride (NH₄F, 99.5%), and polyvinylpyrrolidone (PVP, K-30) were purchased from Aladdin Reagent Company (Shanghai, China). Hydrochloric acid (HCl, 99.7+%), ammonia solution (NH₃·H₂O, 99.7+%), ethylene glycol (EG, 99.7%), acetone (C₃H₆O, 99.7+%), and absolute ethanol (C₂H₆O, 99.7+%) were purchased from Sinopharm, Shanghai, China. All chemical regents were analytic-grade and used without any further purification.

2.2. One-Step Hydrothermal Method to Synthesize VO₂ Nanopowder

A solution of 1.0 g of hydrazine monohydrochloride in 10 mL of deionized water was slowly added, under continuous stirring, to a solution of 4.5 g of ammonium metavanadate dissolved in 60 mL of deionized water, which had been stirred for 30 min. Stirring was continued for an additional 40 min. Subsequently, 15 mL of concentrated hydrochloric acid (37%) was added dropwise to the resulting mixture. The mixture was stirred for 60 min until it became clear. Then, ammonia water (25%) was added to adjust the pH to 8.5 ± 0.2, followed by stirring for 30 min. The resulting precipitate was separated by centrifugation and thoroughly washed with deionized water. The precipitate was then transferred into a reaction kettle, and deionized water was added to fill approximately 80% of the kettle’s volume. The hydrothermal reaction was maintained at 275 °C for 24 h. After cooling to room temperature, the product was washed sequentially with deionized water, anhydrous ethanol, and isopropanol by centrifugation. The final product was dried in an oven at 80 °C for 6 h and then ground into a fine powder using an agate mortar to obtain VO₂ (M) nanopowder.

2.3. Chemical Deposition to Synthesize the VO₂@AlF₃ Core–Shell Structure

The masses/volumes of the various samples/elements added are listed in

Table 1. Preparation of the samples required for the preparation of VO₂@AlF₃ core–shell powders with different molar ratios.

Sample	A1	A2	A3
VO2/g	0.6	0.6	0.6
Al(NO3)3·9H2O/g	1.3569	2.7137	5.4275
NH4F/g	0.4020	0.8062	1.6077
EG/mL	60	60	60

. To prepare VO₂@AlF₃-coated structures with molar ratios of n(AlF₃): n(VO₂) of 0.5, 1.0, and 2.0, three VO₂ aliquots of 0.6 g were each dispersed in 60 mL of ethylene glycol. To each group, 0.1 g of polyvinylpyrrolidone (PVP) was added, followed by continuous stirring for 24 h. Subsequently, 1.3569 g, 2.7137 g, and 5.4275 g of Al(NO₃)₃·9H₂O were weighed for the respective above-mentioned preparations, and each amount was dissolved in 10 mL of deionized water under stirring for 1 h. These solutions were then added to the corresponding VO₂ suspensions that had been stirred for 24 h. The resulting mixtures were sonicated for 1 h. Separately, 0.4020 g, 0.8062 g, and 1.6077 g of NH₄F were weighed, dissolved in 10 mL of deionized water, and stirred for 1 h. The NH₄F solutions were then added dropwise at a rate of 4 drops per minute to the ultrasonicated VO₂-Al(NO₃)₃ mixtures under continuous stirring in a 55 °C oil bath. After 12 h, the resulting products were centrifuged at 10,000 rpm for 5 min, washed four times with deionized water and anhydrous ethanol, and dried in a vacuum oven at 80 °C for 6 h. The dried samples were then ground into powder and calcined in a tube furnace at 400 °C under a nitrogen atmosphere for 5 h to obtain the VO₂@AlF₃ core–shell-structured powder. The core-shell structure may be shown in Figure 1.

Figure 1. Schematic diagram of the formation mechanism of VO₂@AlF₃.

Figure 1. Schematic diagram of the formation mechanism of VO₂@AlF₃.

The obtained core–shell-structured nanopowder was mixed with PVP and ethanol at a ratio of 0.1 g: 0.1 g: 3.5 mL in a 10 mL reagent vial. The mixture was stirred on a magnetic stirrer at room temperature for 24 h, followed by 12 h of settling. The supernatant was then spin-coated in two layers to obtain VO₂@AlF₃ smart temperature-controlled films. The specific parameters for the spin-coating operation were set as follows: in the initial phase we employed a dual-mode operation, with rotation at 800 rpm for 10 s followed by acceleration to 300 rpm; and subsequently, the speed was increased to 1500 rpm for 20 s, with acceleration adjusted to 500 rpm during this process. Following the spin-coating process, the glass substrate was placed on a heating plate and heated at 80 °C for 5 min to achieve film curing. Repeating this operation enabled the preparation of thin-film materials with varying thicknesses.

2.4. Characterization

X-ray diffraction (XRD) was used to determine the crystal structure of the samples. The experiment was conducted using a copper target under an operating voltage of 40 kV and an operating current of 40 mA. The scanning range was set from 20° to 80°, with a scanning speed of 0.1°/s. The X-ray photoelectron spectroscopy (XPS) model used was Escalab 250Xi, purchased from Thermo Fisher Scientific in the Waltham, MA, USA. XPS was employed for qualitative and quantitative analyses of elemental composition and oxidation states. This experiment utilized Mg K_α radiation to activate photoelectrons. A Zeiss Sigma500 scanning electron microscope (SEM) (Oberkochen, Germany) was utilized to examine the surface morphology with high precision. Transmission electron microscopy (TEM) was used to investigate the morphological features and core–shell structure of the samples. EDS was used to analyze the types and concentrations of elements. The American company TA’s Q2000 calorimeter (Boston, MA, USA) was used for differential scanning calorimetry (DSC) analysis conducted under a N2 atmosphere at heating and cooling rates of 5 °C/min. The Shimadzu Corporation’s UV-Vis-NIR spectrometer model UV-3600 (Kyoto, Japan) was employed to obtain detailed transmittance data of the film under varying environmental conditions, including high temperatures (90 °C) and low temperatures (20 °C).

To quantitatively evaluate the thermochromic performance of the composite film within the visible spectrum (380 nm to 780 nm) and its overall transmittance characteristics for sunlight (wavelength range from 300 nm to 2500 nm), this study incorporated both visible light transmittance and total solar transmittance, which were calculated using the following equations:

$$T_{lum}={\displaystyle \frac{\int {\phi }_{lum}\left(\lambda \right)T\left(\lambda \right)d\lambda }{\int {\phi }_{lum}\left(\lambda \right)d\lambda }}$$

(1)

$$T_{sol}={\displaystyle \frac{\int {\phi }_{sol}\left(\lambda \right)T\left(\lambda \right)d\lambda }{\int {\phi }_{sol}\left(\lambda \right)d\lambda }}$$

(2)

T(λ) denotes the transmittance at a specific wavelength λ, φ_lum denotes the standard luminous efficiency function under human myopic vision [23], T_lum denotes the transmittance of visible light in the 380–780 nm wavelength range, while φ_sol denotes the solar radiation spectrum under air quality condition 1.5 (corresponding to a solar elevation angle of 37° above the horizon) [24]. The solar light control efficiency formula is as follows:

$$\mathsf{\Delta }{T}_{sol}=T_{sol,l}-T_{sol,h}$$

(3)

T_sol,l represents the solar transmittance at 20 °C, while T_sol,h corresponds to the solar transmittance at 90 °C [25].

3. Results and Discussion

3.1. Phase and Composition Analysis of VO₂@AlF₃ Core–Shell-Structured Powders

The XRD patterns of the samples are presented in Figure 2. The XRD profiles of the uncoated group (A0) and the coated group (A1, A2, A3) were compared with the standard PDF card for metallic VO₂ (M) (PDF#82-0611). The results showed that the major diffraction peaks were in excellent agreement with those of the reference pattern. The interplanar spacings d of the crystal planes (110), (211), and (111) matched those of the standard values. Characteristic diffraction peaks corresponding to the (011), (220), and (022) planes were observed at 27.8°, at 37.1°, and at 55.3°, respectively, with negligible angular deviation (Δ2θ < 0.15°). These findings confirmed the successfully synthesis of monoclinic VO₂ (M) and demonstrated that the AlF₃ coating process did not induce crystal lattice distortion or phase transformation. It is worth noting that even under high Al/F feed ratios (as in the A3 sample), no characteristic peaks of α-AlF₃ (PDF#44-0231) or β-AlF₃ (PDF#33-0019), such as the α-AlF₃ (012) peak at 23.5°, were detected. This observation, together with the XPS depth profiling results shown in Figure 3, revealed a stable Al-F bonding state, confirming that the AlF₃ layer was present in an amorphous form.

The formation of amorphous aluminum fluoride could be attributed to the extremely rapid kinetic migration of fluoride ions in the solution, leading to the formation of an irregular network. Additionally, the strong coordination interaction between the hydroxyl (-OH) groups on VO₂ surface and Al³⁺ ions resulted in a high surface concentration of Al³⁺ and a decrease in interfacial energy. This is supported by the observed reduction in Al³⁺ 2p binding energy from 77 eV to 74 eV, as shown in Figure 3c.

Qualitative analysis of the elemental composition and oxidation states of the samples was performed using XPS. Data processing was conducted using Advantage software, yielding the full spectrum of VO₂@AlF₃, along with high-resolution spectra for V 2p_3/2, Al 2p, and F 1s, as shown in Figure 3. In the spectrum in Figure 3a, clearly identifiable features corresponding to the V 2p_3/2 Al 2p, F 1s, and O 1s peaks are clearly observed. The presence of these characteristic peaks strongly confirmed the composition integrity of the VO₂@AlF₃ composite material system. In the detailed analysis of Figure 3b, the binding energies of the V 2p_3/2 peak and V 2p_1/2 peak were observed to be 516 eV and 528.8 eV, respectively. These values exhibited slight deviations from those typically reported for V⁴⁺ ions in the literature, with a minor shift toward lower binding energies. However, this result is still generally consistent with the standard range associated with V⁴⁺ species, demonstrating a reasonable degree of alignment with the theoretical expectations [26]. As shown in Figure 3c,d, the peaks centered at 74.8 eV and 686 eV were attributed to F 1s and Al 2p, respectively, indicating the presence of AlF₃. Based on the XPS results, AlF₃ was clearly deposited on the surface of the VO₂ particles, which is indicative of the formation of a VO₂@AlF₃ core–shell structure [27].

3.2. Surface Morphology and Structural Analysis of the VO₂@AlF₃ Core–Shell-Structured Powder

To investigate the encapsulation of the VO₂@AlF₃ core–shell-structured powder, SEM analysis was performed on sample A2 (n(VO₂):n(AlF₃) = 1), which showed the most effective encapsulation. As shown in Figure 4, compared with uncoated VO₂, the VO₂@AlF₃ core–shell-structured powder demonstrated a significantly larger particle size and improved dispersion. In contrast, the uncoated VO₂ nanopowder exhibited obvious agglomeration. Statistics analysis showed that the average particle size was 45 nm before encapsulation and increased to 58 nm after encapsulation. This increase indicates that AlF₃ successfully encapsulated VO₂ and effectively prevented interparticle agglomeration.

Further TEM and EDS testing of the samples was conducted, as shown in Figure 5. In Figure 5a–d, since the atomic number of V is higher than those of Al and F, V appears as a dark region, whereas Al and F appear as bright regions in the TEM image. This observation indicates that AlF₃ was uniformly coated on the surface of VO₂. As shown in Figure 5b, the shell of sample A1 consisted of nanoscale amorphous particles with an average size of approximately 2 nm. In Figure 5c, the shell of sample A2 appeared to consist of nanoscale amorphous crystals approximately 4 nm in size. Similarly, as shown in Figure 5c, the shell of sample A3 consisted of nanoscale amorphous crystals approximately 6 nm in diameter. It can be observed that as the molar ratio of the shell layer in the VO₂@AlF₃ composite powder increased, the thickness of the shell also increased accordingly. As shown in Figure 6, the EDS results were consistent with the previous XPS analysis, indicating that the VO₂@AlF₃ core–shell structure powder synthesis was successful.

3.3. VO₂@AlF₃ Core–Shell Powder Phase Transition Temperature

The effect of AlF₃ shell thickness on the phase transition behavior of VO₂ was studied using a differential scanning calorimetry (DSC) system. Figure 7a–c show the DSC spectra of samples A1 to A3, respectively. The results show that during the heating and cooling processes, all samples exhibited significant endothermic and exothermic peaks. To further clarify the regulatory effect of AlF3 shell layers with different thicknesses on the phase transition temperature of the VO₂ nanoparticles in the core, the key data from Figure 7 were summarized and integrated into

Table 2. Phase transition parameters of VO₂@AlF₃ core–shell powders with different shell thicknesses.

Sample	Heating Cycle Tc,h (°C)	Cooling Cycle Tc,c (°C)	Phase Transition Temperature Tc (°C)	Hysteresis Width ∆Tc (°C)
A0	58.73	31.62	45.17	27.11
A1	62.96	35.52	49.24	27.44
A2	65.08	35.42	50.25	29.66
A3	65.25	44.64	54.94	20.61

, which lists the detailed phase transition parameters for samples A0 to A3. The phase transition temperature of untreated pure VO₂ is 45.17 °C. During the thermal cycling process, the difference between the endothermic and the exothermic peak, referred to as thermal hysteresis width, was measured at 27.11 °C. In VO₂ materials containing an AlF₃ coating, an overall increase in phase transition temperature was observed, with the temperature rising as the shell layer thickness increased. Specifically, sample A1 exhibited a phase transition temperature of 49.24 °C. Although AlF₃ coating led to a higher phase transition temperature compared to that of the untreated VO₂ nanoparticles, the temperature remained significantly lower than the 68 °C reported in the literature. This phenomenon indicates that although the coating layer could moderately modulate the phase transition temperature, the final phase transition characteristics were primarily influenced by intrinsic factors such as nanoparticle size, morphology, and dispersion state, which ultimately contributed to the relatively low phase transition temperature.

The elevation in the phase transition temperature may be attributed to the significantly enhanced cohesive effects among nanoparticles during the deposition of the amorphous aluminum fluoride (AlF₃) shell on VO₂ surface. This process leads to aggregation and increased particle diameter. Consequently, the energy required for the material to transition from one phase to another increases, thereby raising the phase transition temperature. Further analysis revealed that after AlF₃ shell encapsulation, the temperature difference between the endothermic and the exothermic peaks remained constant for the VO₂@AlF₃ sample A1, while it significantly increased for the VO₂@AlF₃ sample A2 and decreased for the VO₂@AlF₃ sample A3.

This observation revealed that VO₂@AlF₃ in sample A3 exhibited a lower thermal hysteresis effect. This phenomenon may be attributed to the shell structure enhancing the elastic strain energy of the VO₂ nanoparticles. A smaller thermal hysteresis width can improve the sensitivity of VO₂@AlF₃ nanoparticles to temperature-induced phase transition [26]. The increase in phase transition temperature and the reduction in thermal hysteresis can be partly attributed to the surface deposition of AlF₃ on VO₂ nanoparticles.

3.4. Analysis of the Optical Properties of VO₂@AlF₃ Thin Films

UV-Vis-NIR transmittance tests were conducted on the obtained smart temperature-controlled glass and compared with those on films prepared using untreated VO₂ nanoparticles. The results are shown in Figure 8 and

Table 3. Standard parameters for the optical properties of the smart temperature-controlled A0~A3 VO₂@AlF₃ thin-film samples.

Sample	Tlum,l (%)	Tlum,h (%)	Tsol,l (%)	Tsol,h (%)	T550 nm (%)	ΔTsol (%)
A0	46.9	44.2	52.5	34.5	45.4	18.0
A1	40.2	40.2	51.4	34.1	39.5	17.3
A2	36.3	33.5	59.2	42.0	49.7	17.2
A3	46.44	59.8	64.6	52.6	43.7	12.0

. The VO₂-based film demonstrated excellent solar control performance (∆T_sol = 18.0%) with a visible light transmittance of 45.4% at 550 nm (T_{550 nm}). The A1 sample, which had a thinner core–shell structure, showed a slight decrease in solar light modulation capability (ΔT_sol = 17.3%) and a corresponding decrease in visible light transmittance (T_{550 nm} = 39.5%). Compared with the A2 sample, which had a slightly thicker core–shell structure, the solar light regulation performance decreased by 4.4% (ΔT_sol = 17.2%). However, visible light transmittance increased by 9.5% (T_{550 nm} = 49.7%) compared to that of pure VO₂, possibly due to the interference effects between reflected light waves caused by the thick core–shell structure, which in turn enhanced visible light transmittance. The A3 sample, with the thickest core–shell thickness, exhibited a 3.6% decrease in visible light transmittance (T_{550 nm} = 43.7%) and a significant 33.3% reduction in solar light regulation performance (ΔT_sol = 12.0%). This decline can be explained by the decreasing proportion of VO₂ nanoparticles within the entire core–shell structure, leading to a reduction in the sample’s solar light regulation capability.

We also compared our best sample A2 with uncoated VO₂, VO₂@SiO₂, and VO₂@ZnO [21,28,29]. The results are presented in

Table 4. Standard parameters for sample A2, uncoated VO₂, VO₂@SiO₂, and VO₂@ZnO smart thermoregulation film samples.

Sample	Tc (°C)	Tlum,l (%)	Tlum,h (%)	Tsol,l (%)	Tsol,h (%)	ΔTsol (%)
VO2	68.0	38.9	36.2	47	29.8	17.2
VO2@ZnO	63.6	47.7	43.6	53.9	35.0	18.9
VO2@SiO2	68.0	70.0	68.8	72.0	66.9	5.0
VO2@AlF3	50.3	36.3	33.5	59.2	42.0	17.2

. It can be observed that VO₂@AlF₃ exhibited improved performance and phase transition temperature compared to the others.

3.5. VO₂@AlF₃ Core–Shell Structure Powder Oxidation Stability Test

VO₂ is easily oxidized in non-acidic environments, particularly in air, where it readily oxidizes from a blue-black to a yellow-brown appearance due to the formation of V₂O₅. To investigate the oxidation resistance of the VO₂@AlF₃ core–shell structure prepared in this experiment, samples A1, A2, and A3 were subjected to calcination in air at 300°C for 180 min. As shown in Figure 9a–c, no visible color change was observed in the VO₂@AlF₃ nanoparticles. This suggests that the prepared VO₂@AlF₃ core–shell-structured powder exhibits oxidation resistance at certain high temperatures. Since Sample A2 exhibited superior overall performance, it was selected for further comparison with uncoated VO₂ under identical calcination conditions at 300 °C, as shown in Figure 9c,d. The results indicated that the VO₂@AlF₃ nanopowder retained its original color even after extended calcination, while the uncoated VO₂ nanopowder acquired a yellowish-brown hue, indicating significant oxidation.

To further investigate the oxidative stability of the VO₂@AlF₃ core–shell-structured powder, sample A2, which was annealed at 300 °C for 480 min, was further processed into a slurry, spin-coated into a thin film, and subsequently coated in one to three layers (denoted as S1–S3). The UV-Vis-NIR transmittance results are shown in Figure 10. Further data processing yielded the results presented in

Table 5. Optical properties of the films prepared from sample A2 after annealing in air at 300 °C for 480 min.

Sample	Tlum,l (%)	Tlum,h (%)	Tsol,l (%)	Tsol,h (%)	T550nm (%)	ΔTsol (%)
S1	46.1	47.3	35.2	22.9	46.2	12.3
S2	41.8	43.7	43.8	30.0	42.7	13.8
S3	27.3	34.2	31.5	20.3	30.7	11.2

. As shown in the table, sample A2 exhibited a solar radiation modulation capability (ΔT_sol) of 12.3% with a single-layer coating, representing a 28% reduction compared to the value obtained for the smart temperature-controlled thin film prepared before annealing (ΔT_sol = 17.2%), thereby retaining approximately 71% of its original solar radiation regulation capability. With a two-layer coating, ΔT_sol increased to 13.8%, indicating a 19% decrease compared to the pre-annealing value recorded for the smart film and the retention of approximately 80% of its solar light modulation capability. When sample A2 was coated with three layers, the solar light regulation capability ΔT_sol was 11.2%, showing a 32% decrease compared to that of the smart temperature-controlled film prepared before annealing (ΔT_sol = 17.2%) and a retention of approximately 65% of the original solar light regulation capability. Compared to conventional VO₂ powder, which underwent completely discoloration and lost its solar light regulation capability, the sample displayed significantly improved antioxidant performance.

4. Conclusions

In this study, a smart temperature-control coating was developed using VO₂@AlF₃ core–shell-structured nanospheres synthesized via one-step hydrothermal and chemical deposition methods. The objective was to reduce the transition temperature (T_c), while enhancing oxidation stability and dispersion performance. The experimental results showed that the T_c of uncoated VO₂ was 45.17 °C, but the VO₂@AlF₃ core–shell-structured nanospheres exhibited a higher T_c of 50.25 °C. Furthermore, the VO₂@AlF₃ nanospheres exhibited excellent optical properties with the optical molar ratio of 1.0 (n(VO2):n(AlF3) = 1.0), including a highly luminous transmittance (T_lum,h) of 33.5%, T_lum,l of 36.3%, T_{550 nm} of 49.7%, and ΔT_sol of 17.2%. After the oxidation test, the material still had approximately 80% of its original regulatory capacity and retained comparable optical performance with T_lum,h of 43.7%, T_lum,l of 41.8%, T_{550 nm} of 42.7%, and ΔT_sol of 13.8%, which demonstrated its outstanding oxidation resistance. Meanwhile, as regards the morphology, compared to uncoated VO₂, VO₂@AlF₃ showed better dispersion. The one-step hydrothermal and chemical deposition methods provide convenient and novel designs to prepare core–shell structures for VO₂, effectively reducing their transition temperature and enhancing their oxidation stability and dispersion. These finding suggest that this strategy holds promise for the preparation and modification of smart temperature-control coating. However, despite the improvements, the phase transition temperature of 50.25 °C remains relatively high compared to the ambient temperature of 25 °C, indicating a need for further optimization. Additionally, VO₂@AlF₃ visible transmittance at 550 nm is only 49.7%, with still low visible light transmittance at low temperatures and a noticeable yellowish tint in the coated glass.