Thermochromic Behavior of VO₂/Polymer Nanocomposites for Energy Saving Coatings

Abstract

Vanadium dioxide (VO₂) is a well-known thermochromic material that can potentially be used as a smart coating on glazing systems in order to regulate the internal temperature of buildings. Most growth techniques for VO₂ demand high temperatures (>250 °C), making it impossible to comply with flexible (polymeric) substrates. To overcome this problem, hydrothermally synthesized VO₂ particles may be dispersed in an appropriate matrix, leading to a thermochromic coating that can be applied on a substrate at a low temperature (<100 °C). In this work, we reported on the thermochromic properties of a VO₂/Poly-Vinyl-Pyrrolidone (PVP) nanocomposite. More specifically, a fixed amount of VO₂ particles was dispersed in different PVP quantities forming hybrids of various VO₂/PVP molar ratios which were deposited as films on fused silica glass substrates by utilizing the drop-casting method. The crystallite size was calculated and found to be 35 nm, almost independent of the PVP concentration. As far as the thermochromic characteristics are concerned, the molar ratio of the VO₂/PVP nanocomposite producing VO₂ films with the optimum thermochromic properties was 0.8. These films exhibited integral solar transmittance modulation (overall wavelengths) ΔTr_sol = 0.35%–1.7%, infrared (IR) switching at 2000 nm ΔTr_IR = 10%, visible transmittance at 550 nm Tr_Vis = 38%, critical transition temperature T_C = 66.8 °C, and width of transmittance hysteresis loop ΔT_C = 6.8 °C. Moreover, the critical transition temperature was observed to slightly shift depending on the VO₂/PVP molar ratio.

1. Introduction

Versatile solutions to the expanding and rapid increase in worldwide energy demands as well as the restrictions concerning environmental pollution by utilizing renewable energy sources and energy-efficient materials have already attracted both scientific interest and commercial attention. The energy consumption in buildings alone is estimated to be approximately 40% of the world’s total energy consumption, and it is anticipated it will increase steadily [1,2,3]. Central heating, ventilation, and air conditioning are the main energy consumers, being responsible for about 30% of annual carbon dioxide emissions [3]. Moreover, heat transmittance and insulation inefficiency of windows are responsible for 15%–22% of a building’s energy loss [4,5,6,7].

In recent years, ‘smart’ windows have received widespread attention as one of the potential solutions to reducing energy consumption by air conditioning in modern architecture. ‘Smart’ windows are able to intelligently self-regulate the amount of transmitted heat, while keeping the visible transmission mainly unchanged. Vanadium dioxide (VO₂) is one of the most promising solid-sate materials for smart windows due to its unique optical properties related to its inherent and ultrafast reversible structural (phase-change) transition from a monoclinic VO₂(M) to tetragonal rutile VO₂(R) structure at a critical transition temperature of T_C = 68 °C, for pure monocrystalline material [8,9,10].

Conventionally, ‘smart’ thermochromic windows are fabricated by vapor phase deposition techniques such as sputtering [11,12,13,14], chemical vapor deposition (CVD) [15,16,17], and pulsed laser deposition (PLD) [18]. However, all these techniques are restricted by the cost and scale of vacuum systems. Moreover, most of these growth techniques, although they are of high fidelity and can produce good quality thermochromic VO₂ films, demand high deposition temperatures (>400 °C) [11,14,15,16,17,18], with very few employing sputtering techniques between 250 and 300 °C [12,13,19,20,21], making it impossible to utilize flexible (polymeric) substrates.

Another approach, called the ex-situ approach [22], to the fabrication of VO₂ thermochromic films is to first synthesize the desired material as a powder and then to deposit the material as a film onto the desired surface. Thermochromic VO₂ in the form of powders have been synthesized by various methods, such as thermolysis [23,24], rapid thermal annealing [25], pyrolysis [26], and, the most utilized method, hydrothermal (solvothermal) synthesis [27,28,29,30,31]. The latter is the most promising due to the high crystallinity of the resulting products, the precise phase control, the versatility on the synthetic parameters, and the possibility of large-scale synthesis [32].

In order to transform the powder into film, various deposition methods have been applied, most notably sol-gel methods. Sol-gel is a proper inexpensive method for large-scale deposition, utilizing fast processing techniques such as dip and spin coating [33]. In particular, the sol-gel method is suitable for the deposition of VO₂, either directly on rigid or on flexible substrates (membrane) attached to glass surfaces. In the sol-gel method, a mixture of a precursor-supporting material along with the particles of the functional material are used (e.g., usually a polymer or monomer along with VO₂ particles) in order to synthesize the films [33,34,35]. Hence the polymer has to be adequately selected to fulfill several demands, including: (i) Stabilization of the VO₂ particles dispersion, (ii) prevention of agglomerations, and (iii) protection of VO₂ particles, e.g., from oxidation/reduction by air or water vapor, which is especially important during the long storage period. Among the various prospective polymeric host matrices, such as polyoxymethylene, polyvinyl alcohol, polyacrylic acid, etc., the one fulfilling the above criteria to a fairly satisfying level has been identified to be Poly-Vinyl-Pyrrolidone (PVP) [36]. Although the sol-gel process is a promising technique for achieving a low deposition temperature for thermochromic VO₂ film fabrication, various concerns such as thickness control and repeatability have not yet been met equivalently in all the different deposition methods used (spin-coating, drop-casting, dip-coating, and spraying), thus sustaining the interest of the research community and industry [37,38].

In the present work, thermochromic VO₂/polymer nanocomposite films were deposited uniformly using a casting method on glass substrates. The method selected was polymer-assisted deposition employing the drop-casting technique. The polymer utilized was PVP. A parametric study was carried out in order to examine the effect of the VO₂/PVP molar ratio on the thermochromic properties. The quantities of both the dispersant (distilled water) and the VO₂ particles were kept constant, while the quantity of PVP was varied, in order to change the molar ratio of VO₂/PVP. It is the first time that such a study has been performed, since until now the majority of research [39,40,41,42,43,44] has mainly focused on the thermochromic behavior of VO₂ powder/composite, without showing how the thermochromic properties are affected by the presence of a host material.

2. Materials and Methods

2.1. Synthesis of VO₂ Particles

The reagents used were vanadium pentoxide, V₂O₅ (+98% pure), as the vanadium source and oxalic acid dihydrate, H₂C₂O₄·2H₂O (≥99.0% pure), as the reducing agent. All reagents were purchased from Sigma Aldrich (Merck, Germany) and were used without further purification.

VO₂ particles were synthesized via a typical hydrothermal procedure using a Parr Teflon-lined stainless steel autoclave. In a typical procedure, 2 mmol of V₂O₅ powder and 8 mmol of H₂C₂O₄·2H₂O were dissolved in 2.22 mol of deionized water, resulting in a 32 vol% filling of the Teflon vessel. After stirring, the original dark yellow mixture turned into a dark green-blue solution. For the hydrothermal treatment, the precursor mixture was transferred to the acid digestion vessel and into a furnace for treatment at 220 °C for 12 h. The obtained blue-black solid product was isolated via centrifugation, after drying at 80 °C for 4 h. Finally, to acquire the desired crystalline phase of thermochromic VO₂ the solid product was annealed at 700 °C for 2 h under a constant nitrogen gas flow.

2.2. VO₂/Polymer Nanocomposite Coatings

The as-prepared VO₂ particles were dispersed ultrasonically in deionized water at a concentration of 10 mg/mL for 30 min and stirred for 1 h to produce a suspension. Then, an appropriate amount of PVP K-30 was added in different molar ratios of VO₂/PVP, from 0.2 to 1.2, by decreasing the amount of PVP while keeping constant the amount of VO₂ particles. The suspension was stirred for 2 h to enhance its homogeneity. Prior to the deposition, fused silica glass substrates were treated with ethanol and propanol inside a sonication bath for 5 min, respectively. Finally, the mixture was uniformly casted on the substrates and dried for 1 h at 70 °C.

2.3. Structural and Morphological Characterization/Thermochromic Properties

The structural characterization of VO₂ particles and nanocomposite films was performed by X-ray diffraction (XRD) using a Rigaku RINT-2000 diffractometer (Tokyo, Japan). The X-rays were produced by a 12 kW rotating anode generator with a Cu anode equipped with a secondary pyrolytic graphite monochromator. A Cu Kα radiation with wavelength λ = 0.154 nm was used. Measurements were performed with θ/2θ configuration, scanning from 20° to 80° with a step of 0.02°/min. The crystallite size of the VO₂ was calculated by using the Scherrer Equation;

$$d\left(\mathrm{nm}\right)=\frac{0.9·\mathsf{\lambda }\left(\mathrm{nm}\right)}{B·\cos \left({\mathsf{\theta }}_B\right)}$$

(1)

where λ = 0.154 nm, B the full width at half maximum (FWHM) at 2θ = 27.8° corresponding to the characteristic (011) direction of the VO₂ peak, and θ_B = θ.

The morphology of the VO₂ particles and nanocomposite films was determined by scanning electron microscopy (SEM, JEOL 7000, Tokyo, Japan) operating at 15 keV, while the microscopic nanostructures were observed by transmission electron microscopy (TEM) on a JEM-2100 instrument (JEOL, Tokyo, Japan) equipped with LaB₆ filament, operating at 200 kV.

The critical transition temperature of the VO₂ particles was determined by differential scanning calorimetry (DSC) using a PL-DSC system from Polymer Laboratories. Measurements were performed from 20 to 170 °C and back to 20 °C with a step of 10 °C/min, under a nitrogen flow of 20 cc/min. By identifying the critical transition temperature during heating (T₁) and cooling (T₂) procedures, the critical transition temperature (T_C) as well as the width of the hysteresis loop of the VO₂ powder were calculated as defined by the Equations below:

$$T_{\mathrm{C}}=\frac{T_1+T_2}{2}$$

(2)

$$\mathsf{\Delta }{T}_{\mathrm{C}}=T_1-T_2$$

(3)

The thermochromic properties of the nanocomposite films were examined by recording the transmittance spectra in the temperature range of 25 and 90 °C. For this, a Perkin Elmer Lambda 950 UV/Vis/NIR spectrophotometer, operating at λ = 250–2500 nm with a homemade heating sample stage attachment and a thermocouple on contact with the surface of the film to measure the temperature, was used. The temperature was controlled by an ACUSHNET FP-900 temperature controller (Taipei, Taiwan) at a heating rate of 1.5 °C/min. IR switching is defined as the difference of transmittance value at λ = 2000 nm, between 25 and 90 °C.

ΔTr_IR (%) =Tr_IR(25 °C) − Tr_IR(90 °C)

(4)

while integral luminous transmittance is defined as:

$$T{r}_{\mathrm{lum}}(\%)=\frac{{{\displaystyle \int }}_{350}^{750}{B}_{\mathrm{lum}}\left(\mathsf{\lambda }\right)\ast Tr\left(\mathsf{\lambda }\right)\mathrm{d}\mathsf{\lambda }}{{{\displaystyle \int }}_{350}^{750}{B}_{\mathrm{lum}}\left(\mathsf{\lambda }\right)\mathrm{d}\mathsf{\lambda }}$$

(5)

where Tr(λ) is the recorded transmittance spectrum at 25 or 90 °C and B_lum(λ) is the standard luminous efficiency function for photopic vision [45].

Moreover, solar transmittance modulation ΔTr_sol (%) is defined as the difference of integral solar transmission between 25 and 90 °C, obtained by the formula:

$$\Delta T{r}_{\mathrm{sol}}(\%)=\frac{{{\displaystyle \int }}_{250}^{2500}{B}_{\mathrm{sol}}\left(\mathsf{\lambda }\right)\ast Tr\left(T=25 {}^{\circ }\mathrm{C},\mathsf{\lambda }\right)\mathrm{d}\mathsf{\lambda }-{{\displaystyle \int }}_{250}^{2500}{B}_{\mathrm{sol}}\left(\mathsf{\lambda }\right)\ast Tr\left(T=90 {}^{\circ }\mathrm{C},\mathsf{\lambda }\right)\mathrm{d}\mathsf{\lambda }}{{{\displaystyle \int }}_{250}^{2500}{B}_{\mathrm{sol}}\left(\mathsf{\lambda }\right)\mathrm{d}\mathsf{\lambda }}$$

(6)

where Tr(λ) is the recorded transmittance spectrum at 25 or 90 °C and B_sol(λ) is the solar irradiance spectrum for an air mass (AM) of 1.5 (corresponding to a solar zenith angle of 48.2° [46].

Furthermore, in order to determine the critical transition temperature of VO₂/polymer nanocomposite films through optical measurements, the transmittance as a function of temperature (25–90 °C) by a step of 1.5 °C/min at λ = 2000 nm was recorded for both the heating and cooling procedure. The derivative of transmittance (dTr/dT) versus temperature was plotted and fitted by a Gaussian curve for both procedures. The minimum of the fitting curves determined the critical transition temperature T₁ and T₂ for the heating and cooling procedure, respectively. Thus, the critical transition temperature T_C and the width of transmittance hysteresis loop were obtained by Equations (2) and (3), respectively.

3. Results and Discussion

3.1. Structural and Morphological Characterization of VO₂ Particles

Figure 1a shows the XRD pattern of VO₂ powder particles (for 2θ = 20–80°), from which the monoclinic phase of VO₂ (M) is confirmed (JCPDS card No. 43-1051). The crystallite size of the VO₂ powder was calculated from the XRD pattern using Scherer’s Equation (Equation (1)) and was found to be around 35 nm. TEM imaging revealed the material to be composed of nanoflakes, with particles varying from less than 25 to 400 nm, as presented in Figure 1e,f. The thermochromic behavior of the VO₂ particles was confirmed by the DSC diagram of Figure 1b, from which the critical transition temperature and the hysteresis width were calculated and found to be T_C = 66.6 °C and ΔT_C = 12.1 °C. The morphology of the VO₂ particles was investigated by SEM, presented in Figure 1c,d, in which nanoflake agglomerates with sizes up to about 500 nm can be observed.

3.2. Structural and Morphological Characterization of VO₂/Polymer Nanocomposite Films

The XRD pattern of VO₂/polymer nanocomposite films at the optimized ratio of VO₂/PVP = 0.8 deposited on fused silica glass substrate is shown in Figure 2a, along with a picture of the film in Figure 2b. The monoclinic phase of VO₂ was confirmed by the two characteristic peaks at 2θ = 27.8° and 37.0° corresponding to the VO₂ (011) and (200) crystallographic directions, respectively, according to JCPDS card No. 44-0252. The small peak at around 25° is attributed to the glass substrate. Thus, the polymer PVP does not affect the crystallinity of the VO₂. In addition, the crystallite size of VO₂ in the composite was calculated and found to be the same (35 nm) as that of VO₂ in powder form. However, the rest of the peaks corresponding to the VO₂ powder (Figure 1a) could not be detected in the composite, probably due to the presence of the host material (PVP).

In order to investigate the effect of the amount of PVP in the morphology of the surface of VO₂/polymer nanocomposite films, SEM at ×1500 magnification was performed. VO₂ particles, as shown in Figure 3, were found to have almost entirely been submerged, particularly at the optimized ratio of VO₂/PVP = 0.8, within the polymer matrices exhibiting a smooth composite finish surface without cracks. Moreover, it was found that the presence of polymer for all ratios from 0.2 to 1 prevented the agglomeration of particles with an average grain size of more than 500 nm.

3.3. Effect of VO₂/PVP Molar Ratio on Thermochromic Properties

The transmittance spectra from 250 to 2500 nm of VO₂/polymer nanocomposite films with a different VO₂/PVP molar ratio at 25 and 90 °C are presented in Figure 4a–d. It can be seen that all films exhibit a thermochromic behavior. Utilizing these spectra along with Equations (4)–(6) IR switching at 2000 nm, Tr_lum and Tr_sol, were calculated and are presented in

Table 1. Optical characteristics of VO₂/polymer nanocomposite films.

VO2/PVP *	Trlum(%)	ΔTrIR(%)	ΔTrsol(%)
0.2	24.5	7	1.72
0.4	27.2	5.7	0.36
0.6	26.9	7.8	1.00
0.8	37.1	10.2	1.22
1	36.8	7.8	1.35
1.2	41.2	4.8	0.35

* molar ratio.

.

According to Table 1, it is clear that the luminous transmittance is decreasing as the concentration of PVP increases (i.e., the ratio of VO₂/ PVP is decreases) due to the anticipated increase of film thickness, varying between 3 and 5 μm, as a result of the casting method. Concerning the effect of PVP on IR switching, there is an optimal ratio of VO₂/PVP = 0.8, for which IR switching becomes maximum and equal to 10.2%, as presented in the graphical representation of data in Figure 5a. Furthermore, the calculated solar transmittance modulation varied from 0.35% to 1.7% without any evident indication on whether there is a correlation with the VO₂/PVP molar ratio, rather than a moderately noticeable non-linear increase of ΔTr_sol with the increase of PVP in solution concentration (i.e., a decrease of the VO₂/PVP molar ratio). Although solar transmittance modulation is low enough when compared to thermochromic films prepared by other techniques at higher deposition temperatures [1,4,16], a systematic investigation on the dispersion of VO₂ powder in the solvent should lead to an improvement of thermochromic characteristics. Additionally, a different casting method such as dip coating or spin coating will be employed in order to increase the homogeneity of the films, resulting in an increase of luminous transmittance.

The transmittance hysteresis loop of VO₂/PVP nanocomposites with different amounts of PVP, at λ = 2000 nm is presented in Figure 6. From these heating–cooling loops, it is evident that the transition is a fully reversible process for all VO₂/PVP nanocomposite samples. By using Equations (2) and (3), the critical transition temperature and the hysteresis width, respectively, were calculated. The results are presented in

Table 2. Critical transition temperature T_C and width of transmittance hysteresis loop (ΔT_C) of VO₂/PVP nanocomposite films for different amounts of PVP.

VO2/PVP *	ΔTC (°C)	TC (°C)
0.2	3.5	62.2
0.4	5.0	65.6
0.6	6.0	67.6
0.8	6.8	66.8
1.0	4.9	67.6
1.2	8.1	67.8

* molar ratio.

, while in Figure 5b the variation of T_C is graphically presented as a function of the VO₂/PVP molar ratio. It is observed that T_C is increasing in regards to the molar ratio of VO₂/PVP from 62 °C (VO₂/PVP = 0.2) to 68 °C (VO₂/PVP = 1.2). This behavior can be attributed to the increasing concentration of polymer in the films (from the 1.2 to the 0.2 VO₂: PVP molar ratio), since a higher polymer concentration seems to inhibit the agglomeration of VO₂ nanoparticles. Thus, the transition is enhanced by reducing the energy needed for VO₂ particles in the film to transition from the monoclinic to the full tetragonal rutile phase (and vice versa), leading to lower T_C values. A similar phenomenon is observed for the width of transmittance hysteresis loop, since it is increased from 3.5 °C for VO₂/PVP = 0.2 to 8.1 °C for VO₂/PVP = 1.2, indicating that the increase of PVP facilitates the transition, thus lowering the hysteresis width.

4. Conclusions

In this work, hydrothermal synthesis was employed to formulate pure monoclinic VO₂ particles. Subsequently, VO₂/PVP nanocomposite films were fabricated by polymer-assisted deposition on fused silica commercial glass at low treatment temperatures (<70 °C). Both thermochromic and optical properties were examined as a function of the VO₂/PVP molar ratio varying from 0.2 to 1.2, by only changing the amount of PVP in the solution. The IR switching at 2000 nm was found to depend on the molar ratio of VO₂/PVP and the optimal ratio was determined to be 0.8, for which ΔTr_IR = 10.2%. The integrated luminous transmittance varied from 24.5% to 41.2% as the ratio of VO₂/PVP increased from 0.2 to 1.2 (i.e., the amount of PVP was decreased), probably due to the increase on thickness of the produced films. Additionally, the solar transmittance modulation of the VO₂/PVP nanocomposites non-linearly increased from 0.35% to 1.7% as the amount of PVP increased (i.e., a lower VO₂/PVP molar ratio). Finally, it was observed that the critical transition temperature non-linearly increased from 62.2 to 67.8 °C as the ratio of VO₂/PVP increased. This was attributed to the presence of the polymer which inhibits the agglomeration of the VO₂ particles, resulting in lower T_C values for the transition.