Abstract

Vanadium dioxide (VO2) particles were modified by grafting with poly(styrene sulfonate) (PSS) and poly(3,4-ethylenedioxythiophene) (PEDOT) via surface-initiated atom transfer radical polymerization (SI-ATRP). Critical transition temperature () of the modified VO2 ranging between 77 and 79°C was obtained. After mixing with acrylic-based emulsion, dispersion of the VO2 particles in the polymer matrix was significantly improved. The visible light transmittance through the composite films above 75% was maintained if a concentration of the modified VO2 particles loaded into acrylic polymer film was no greater than 1.0 wt%. The NIR transmittance through the acrylic/VO2@PSS : PEDOT also dropped by 9-10%, compared with that of the pure acrylic film (without any particles). Finally, glass substrates coated with the acrylic/VO2@PSS : PEDOT composite films could reduce the temperature inside a model house by 5-6°C, compared with that of the control system (pure acrylic coating film without VO2 particles). Overall, this work demonstrated that it was possible to improve the dispersion of VO2 particles in polymer films without sacrificing its NIR shielding ability by grafting the surface of VO2 particles with PSS : PEDOT chains, while providing the optimum grafting density and particle loading.

1. Introduction

Vanadium dioxide, particularly that in the form of monoclinic crystal structure (VO2(M)), is one of the most interesting types of thermochromic materials. This is because the critical phase transition temperature (, 68°C) [1] of VO2 is the lowest, compared to those of other thermochromic materials [2]. The transition temperature of VO2 can also be further reduced by doping it with chemicals such as molybdenum (Mo) [3], niobium (Nb) [4], fluorine [5], and tungsten [1]. At a temperature above , VO2(M) is transformed into a tetragonal/rutile phase, accompanied with a decrease in spectral transmittance in the near infrared (NIR) region [6, 7]. Considering this interesting behavior, the feasibility of utilizing VO2 as a coating layer on glass to enhance the energy efficiency in buildings has been explored. Practically, thin films of VO2 can be coated onto glass substrates by several techniques including sputtering [8], chemical vapor deposition (CVD) [9], pulsed-laser deposition [10], and ion implantation technique [11]. However, these techniques usually require complex and expensive equipment. Alternatively, if the VO2 material can be prepared in the form of particles, it is possible to develop polymer/VO2 composite [1, 1215], which can be further applied onto glass substrates by less complex techniques such as spray coating [16] and lamination [1, 12].

Despite the above advantages, some challenges have yet to be resolved. These include the fact that the NIR shielding efficacy of the V5O2-based film is usually achieved at the expense of its optical transparency [6, 7]. This is related to the poor dispersion of VO2 particles in many polymer matrices [17]. In this regard, surface modification of VO2 particles was undertaken to deal with the above problem.

Synthetic routes and structure-property relationships of polymer-grafted nanoparticles have been reviewed by Kumar and coworkers [18]. Efficacy of the grafted nanoparticles, for the enhancements of mechanical and optical properties of polymer nanocomposites, depends on many factors including the grafting density (), length of the grafting chains (), length of the matrix chains (), and the ratio (). Besides, the roles of size and shape of nanoparticles upon crossover of polymer chains and properties of the nanocomposites also deserve a consideration and have yet to be clarified.

Few studies concerning the surface modification of VO2 particles have been reported in the open literature; however, all those works focus on the improvement of oxidation stability and the enhancement of photocatalytic activity of the materials [1317, 19, 20]. For example, by coating the VO2 particles with a polyethylene (PE) layer, the stability of the VO2@PE particles against oxidation improved while the thermochromic behavior of the metal oxides was still maintained. Li et al. [19] prepared one type of core-shell VO2@TiO2 nanoparticles and reported that the chemical stability was enhanced by the TiO2 shell. After carrying out the surface modification, the transmittance values ( and ) of the VO2 particles increased from 7.35% and 6.93% to 27.46% and 17.63%, respectively.

However, enhancing the NIR shielding property of VO2-based film without sacrificing its optical transparency remains a challenge. In this study, it was proposed that if the VO2 particles were grafted with a kind of special material, which is intrinsically transparent in the visible light region but capable of blocking the transmission of NIR radiation, the above challenging problems should be resolved. One possible candidate material for such a demanding condition includes poly(3,4-ethylenedioxythiophene) (PEDOT) whose heat shielding efficacy is known [2124]. However, PEDOT is insoluble in water and many common organic solvents. In this regard, poly(styrene sulfonate) (PSS) was introduced into the PEDOT system to balance the polarity, dispersibility, and solubility of the polymer. Im et al. [22], for example, solved this problem by applying a copolymer, synthesized from styrene and sulfonated styrene (P(SS-co-St)) as a template for an oxidative polymerization of EDOT. By properly controlling the thickness of the P(SS-co-St) film, percentages of NIR shielding and visible light transmittance of 92 and 40 were achieved, respectively. Moreover, PEDOT:PSS-based materials in different configurations such as WO3/PEDOT:PSS composite for electrochromic glass [24] and PEDOT:PSS/SWCNT composite for smart window applications [23] have also been developed and reported. To the best of our knowledge, however, the hybrid materials based on a combination of VO2 particles and PEDOT:PSS have not been explored. In this study, it was hypothesized that by properly grafting the PEDOT:PSS chains onto the VO2 particles, a steric hindrance imposed by the presence of PEDOT:PSS chains could be introduced and that could lead to better dispersion of the modified VO2 (so called VO2@PEDOT:PSS herein) particles in polymer matrices. This means that the NIR shielding property of VO2-based film might be enhanced without sacrificing its optical transparency. Therefore, the aim of this work was to investigate the effects of composition and concentration of VO2@PEDOT:PSS particles on morphological, thermal, and optical properties of the acrylic polymer films, reinforced with the VO2-based particles.

2. Experiment

2.1. Materials

Vanadium pentoxide (V2O5, >98% pure), (3-aminopropyl) triethoxysilane (APTES, >99% purity), N,N,N,N-pentamethyldiethylenetriamine (PMDTA, 99%), copper (I) bromide (CuBr, >99% purity), α-bromoisobutyryl bromide (BIBB, C4H6Br2O, >99% purity), and sodium 4-vinylbenzenesulfonated (SSNa, C8H7NaO3S, ≥90% purity) were obtained from Sigma-Aldrich Co., Ltd. Citric acid (>99% purity) and methanol (99.8%) were obtained from Loba Chemie Pvt. Ltd. Toluene (analytical grade), chloroform (analytical grade), and ethanol (analytical grade) were obtained from Merck Co., Ltd. Triethylamine (TEA, >99% purity), 2,2-bipyridyl (Bpy, >99% purity), and 3,4-ethylenedioxythiophene (EDOT, >98% purity) were obtained from TCI group Co., Ltd. An acrylic-based emulsion (PLIOTEC HDT 12 from Synthomer), used as a binder for preparing polymer/VO2 composite films, was supplied by the RPSC Chemicals Co. Ltd. All chemicals were used without further purification.

2.2. Synthesis of VO2@PEDOT:PSS Particles

PEDOT and PSS chains were sequentially grafted onto VO2 particles through a “grafting from” approach via a surface-initiated atom transfer radical polymerization (SI-ATRP) mechanism. A schematic diagram, illustrating the chemical structures and synthetic route for preparation of PEDOT:PSS-grafted VO2 (designated as VO2@PEDOT:PSS), was proposed and is shown in Figure 1.

Basically, the synthesis is divided into 5 steps, as briefly described below. Firstly, a monoclinic phase VO2 particles were prepared via hydrothermal and calcination techniques using vanadium pentoxide (V2O5) and citric acid (C6H8O7) as raw materials and reducing agent, respectively. More details concerning the reaction conditions and procedures for these have been well described in our previous report [1]. After that, the VO2 particles were grafted with the silane coupling agent (APTES) to introduce amine functional groups onto the surface of VO2 particles. Again, more details concerning this reaction step have been reported elsewhere [25]. Next, α-bromoisobutyryl bromide (BIBB), which served as an initiating species for ATRP, was anchored onto the VO2@APETS particles. The reaction was proceeded via a condensation between the amine groups of APTES moieties and the acyl-bromide group of BIBB. After that, PSS chains were grafted from the surface of VO2 particles via a SI-ATRP of sodium styrene sulfonate (SS-Na). The polymerization was carried out under different reaction times, ranging from 12 to 24 h. Accordingly, the polymerized products (polystyrene sulfonate-grafted VO2 particles) were designated as VO2@PSS-12, VO2@PSS-16, VO2@PSS-20, and VO2@PSS-24, respectively (see Table 1).

Finally, EDOT monomer was polymerized on the surface of VO2@PSS particles via an oxidative polymerization mechanism, utilizing the PSS shell layer as a dopant and template for the reaction. Likewise, the final products (VO2 particles grafted with PSS : PEDOT) were designated as VO2@PSS : PEDOT-12, VO2@PSS : PEDOT-16, VO2@PSS : PEDOT-20, and VO2@PSS : PEDOT-24, respectively. More details regarding the synthesis of these particles are fully described in the supplementary information file (available here).

2.3. Preparation of Polymer/VO2 Composite Films

To prepare the polymer/VO2 composite films, different types of VO2-based particles, including the neat VO2, the VO2@PSS : PEDOT-12, and the VO2@PSS : PEDOT-24 were mixed with an acrylic-based emulsion. Depending on the concentration of particles to be loaded, a given amount of VO2 particles (0.03 g, 0.06 and 0.09 g) were dispersed in 15 g of the acrylic emulsion (PLIOTEC HDT 12) in a beaker and then placed in an ultrasonic bath and kept stirred for 150 min. The above mixtures were then coated onto a glass substrate () by using a bar coater. The final film thickness was controlled by selecting a suitable rod bar (with a specified gap between the rod and substrate of 80 nm). The coating was left overnight to allow complete drying.

2.4. Characterizations

Changes in chemical functional groups on the surface of the VO2 particles were monitored by using Fourier transform infrared spectroscopy (FTIR, Thermo, DSC-1) and the X-ray photoelectron spectroscopy (XPS, AXIS ultra DLD), equipped with Avantage Data System software. The XPS data were collected in the V2p, O1s C1s, N1s S2p, and Si2p binding energy regions.

Crystal structures of the modified VO2 particles were examined by using an X-ray diffractometer (XRD, Bruker AXS D8-Discover) in the range of 10–80°, using radiation (). The accelerating voltage and the current used were 40 kV and 40 mA, respectively. Thermogravimetric analysis (NETZSCH, TGA 209 model) was used to determine thermal stability and weight composition of the VO2-based particles. For TGA measurement, approximately, 8 mg of each sample was used and was scanned over temperatures ranging between 25°C and 800°C under nitrogen gas and a heating rate of 10°C/min. From the obtained TGA thermograms, grafting density () of the final product (VO2@PSS : PEDOT particles) was determined, using equation (1), which was adapted from the work of Babu and Raghavachari [26] where δ is the grafting density, is the percentage weight loss of polymer chains in the grafted VO2, is the surface area of VO2 particles determined by BET technique (15.9887 m2/g), is the percentage weight loss of VO2 particles before grafting, and is the molar mass of the immobilized molecules (the grafting chains).

In addition, thermal behaviors of VO2 particles and polymer/VO2 composites were investigated by differential scanning calorimetry, using DSC 204, NETZSCH instrument. Typically, about 8 mg of the sample was used and the specimens were scanned over temperatures ranging between 25°C and 800°C, under nitrogen gas atmosphere, at a heating rate of 10°C/min.

The morphology of VO2 particles and the acrylic/VO2 composite films were examined by using the scanning electron microscopy (SEM) and the transmission electron microscopy (TEM) techniques. The TEM experiment was carried out by using a JEOL JEM-2100 (JEOL, Peabody) microscope with an accelerating voltage of 200 kV. The TEM specimen was prepared by dispersing 1 mg of the VO2 particles in 10 ml of ethanol, followed by sonication for 30 min. The solution was then dropped onto a copper grid and dried at 60°C. The SEM experiment was carried out by using a JEOL (JSM 6610LV) machine, equipped with a secondary electron detector and energy-dispersive X-ray detector (EDX). The accelerating voltage used was 10-30 kV. The sample was coated with gold prior to carrying out the SEM experiment.

Size of the VO2-based particles was analyzed by using SEM and dynamic light scattering (DLS) techniques, respectively. The DLS analysis was carried out by using a Zetasizer instrument (Nano-ZS90, Malvern, UK), at a measurement angle 173 backscatter (NIBS default). The samples were prepared by dilution and sonication of the sample in an aqueous solution before putting into disposable cuvettes. The average hydrodynamic diameter was determined by taking an average of 11 runs, and the duration for each run was 10 sec. Each analysis is the average of results from three consecutive measurements.

Optical properties of the acrylic binder films, containing different types and concentration of VO2-based particles, were tested by using a UV/visible/NIR spectrophotometer (Shimadzu, UV-3100). The percentage of light transmittance and reflectance of the film samples in the wavelength ranging between 200 and 2500 nm was investigated and calculated in accordance with the ISO 9050 standard method.

Heat shielding ability of polymer composite films containing VO2 particles was tested by using a model house, equipped with an infrared lamp, thermocouples, and glass windows. More details concerning the experimental setup of this equipment has already been described in our earlier report [1]. Changes of temperature inside the house chamber was then monitored and followed as a function of time and the types of polymer/VO2 window films.

3. Result and Discussion

3.1. Structural Characterizations of VO2 Particles

Figure 2 shows overlaid FT-IR spectra of the various types of VO2 particles, including the neat VO2, the VO2@APTES, VO2@Br, VO2@PSS, and VO2@PSS : PEDOT. A broad absorption band centered around wavenumbers ranging between 3200 cm-1 and 3600 cm-1 and a sharp absorption peak at 1630 cm-1 can be noted from all spectra. These bands are assigned to the stretching vibration of hydroxyl groups (OH) on the VO2 particles and the bending vibration of H-O bonds from the absorbed moisture, respectively [2729]. After grafting with APTES, an intensity of a broad absorption band ranging between 3350 and 3650 cm-1increased significantly. This can be ascribed to N-H stretching of NH2 group in the APTES [30]. New peaks at 1400cm-1, 1100 cm-1, and 984 cm-1 also emerged (sample (b) in Figure 2). These are attributed to the vibration of the Si-CH2 bond, the asymmetric stretching of the Si-O bond, and the stretching of the Si-OH bond, respectively.

To verify the existence of BIBB moieties on the surface of modified VO2 particles (VO2@Br), in this study, an XPS technique was used. From Figure 3, the presence of characteristic peaks in the XPS spectra at the binding energy of 68.25 eV and 182 eV was noted. These can be ascribed to the Br 3d and Br 3p orbitals, respectively. [31] The C 1s peak in the XPS spectrum was further analyzed by deconvolution. From Figure 4(a), the presence of a deconvoluted peak at 287.6 eV, representing the O=C-N group [5], can be noted. This indicates that the VO2@APTES particles were successfully grafted with the BIBB initiator via an amide linkage.

The BIBB moieties on the surface of these particles were then used as an initiator for the polymerization of the sodium styrene sulfonate (SSNa) monomer. From FTIR analysis, the absorption peaks at 1175 cm-1 and 1038 cm-1, which are attributed to asymmetric and symmetric stretching vibration of the SO3 groups in the monomer [24, 32], were observed, respectively (see the inset in Figure 2). Moreover, from the XPS spectrum of this product (sample (b) in Figure 3), peaks at 167.5 eV and 1070.5 eV, which represent the binding energy of the S 2p and Na 1s orbitals, can be noted. The intensity of the C 1s peak at the binding energy 284.5 eV also increased. This indicates that the surface of the modified VO2 particles was covered with more hydrocarbon component. In Figure 4(b), four deconvoluted peaks at the binding energy of 284.6 eV, 285.5 eV, 286.4 eV, and 287.6 eV were also noted. These can be assigned to C=C, C-C/C-H, C-O/C-S, and C=O/C=S bonds, respectively. In addition, direct evidence supporting the presence of a PSS amorphous organic layer on the surface of the modified VO2 particles can be seen from the TEM images of the products (Figure 5). The above results confirm that the monomer was successfully grafted on the VO2 particles.

Finally, after reacting the VO2@PSS particles with the EDOT monomer, the FTIR spectrum of the final product (VO2@PSS : PEDOT) (sample (e) in Figure 2) shows additional absorption peaks at 1541 cm-1, corresponding to C=C stretching of the thiophene rings in the PEDOT repeating units. Absorption peaks at 1358 cm-1 (C-C stretching of SO3H), 1217 cm-1, 1093 cm-1 (C-O-C stretching), and 982 cm-1 (C-S stretching) representing the PSS chains were also noted [4]. Consideration of the XPS spectra (Figure 4(c)) shows that the intensity of the deconvoluted XPS peaks (C-O/C-S and C=O/C=S bonds) increased after grafting with the EDOT. This suggests that the monomer was polymerized on the surface of the VO2 particles via the PSS shell, which served as both a template and a dopant.

Figure 6 shows SEM images of VO2-based particles. The particles are in a spherical shape, and sizes of the individual particles are in the range of 3000 and 5000 nm. Morphological changes of the VO2 particle surface after grafting with polystyrene sulfonate (PSS) chains were noticed. This can be used as the evidence indicating the presence of polymer coating on VO2 particles. In addition, results from dynamic light scattering analysis (DLS curves) of the various samples are presented in Figure 7. Some parameters derived from the DLS analysis, including size distribution and size at the maximum intensity of the major peaks () for those particles, are also summarized in Table 2. In this study, since the VO2-based particles were agglomerated and exhibited multimodal characteristics, no further attempts were made to determine the average size of the particles from the above techniques.

3.2. Thermal Properties of the Modified VO2 Particles

Figure 8 shows DSC thermograms of the various types of VO2 particles. An endothermic peak at 73°C can be observed from the thermogram of the neat VO2. This refers to the critical transition temperature () of the monoclinic VO2. The critical transition temperature () of grafted products (VO2@PSS) shifted slightly to a higher temperature ranging between 76.9°C and 79.6°C (Table 3). This was most likely due to the presence of grafting polymer chains on the surface of the VO2 particles, which affected the heat transfer and the onset of the phase transition temperature of the VO2. After polymerization with the EDOT monomer, the critical transition temperature of the products (VO2@PSS : PEDOT) still existed and those ranged between 77.8°C and 79.6°C.

Notably, after grafting of the VO2 particles with PPS chains, an additional peak, centered at around 164-174°C, was observed. This peak also existed in a thermogram of PSS homopolymer (see also the inset in Figure 8). The transition temperature of this peak is much lower than the melting peak of PSS, which is around 450°C [33]. Interestingly, after running a second heating DSC scan, the above peaks disappeared. The similar effect was observed by Zhou et al. in a study on temperature-dependent microstructure of PEDOT/PSS films [34]. Accordingly, the above peak was described to the release of water during heating. Finally, after polymerization of EDOT monomer, the position of this peak shifted to the lower temperatures (ranging between 145.1 and 147.2°C, see also Table 3). A similar trend was noticed in the case of PSS homopolymer and PSS/PEDOT blend systems (which were polymerized without the presence of VO2 particles). In our opinion, the above changes were related to changes in polarity and hydrophilicity on the surface of VO2@PSS particles after coupling with the polymerized PEDOT chains.

Figure 9 shows TGA thermograms of the synthesized PSS : PEDOT and various types of VO2 particles. The thermogram of the PSS : PEDOT shows three major transitions including an initial weight loss, which occurred at a temperature below 110°C, owing to the loss of some absorbed moisture. The second transition occurred over temperatures ranging between 250°C and 400°C, and that was attributed to the elimination of the sulfonate groups (-SO3) attached to the styrene rings in the PSS molecules [35, 36]. The third transition occurred over temperatures ranging between 300°C and 550°C, and that was attributed to the degradation of the polymers (PEDOT [28] and PSS [32, 37] chains).

By contrast, the neat VO2 particles were thermally stable up to 900°C. TGA thermograms of the intermediates, which included VO2 particles functionalized with silane (VO2@APTES) and VO2@APETS treated with the BIBB initiator (the so called VO2@Br), show a gradual weight loss transition over temperatures between 120 and 600°C. This can be ascribed to the loss of functional groups on the surface of the chemically modified VO2 particles. Finally, after grafting with PSS followed by coupling with PEDOT molecules, all these transitions can be observed from the TGA thermograms of the final product (VO2@PSS : PEDOT).

In this regard, the grafting density of the final products was determined using equation (1), and the results are summarized in Table 3. Notably, grafting density values ranging from 5.35 to 7.16 chains/mm2 were obtained, and the values changed with the polymerization time in a nonlinear fashion. In our opinion, the above effect can be ascribed to the nature of the SI-ATRP mechanism. The longer the reaction time, the greater the monomer conversion [38]. This led to the higher degree of particles covered by grafting chains [39]. Moreover, the mechanism of SI-ATRP is basically a kind of controlled free radical polymerization, which means that the length of the polymerized chains also gradually increased with the reaction time [40]. In this regard, it was possible that both parameters contributed to the number of chains per area (the grafting density values) of the VO2@PSS : PEDOT reported in this study. It is also worth mentioning that the grafting density values of the modified VO2 particles in this study are relatively high compared to those of the Fe3O4 nanoparticles grafted with poly(methyl methacrylate) chains, reported by Babu and Raghavachari [26]. In our opinion, the discrepancy was attributed to the different sizes and surface areas of the particles used. Specifically, the surface area of VO2 (15.9887 m2/g) is relatively low compared to that of the Fe3O4 nanoparticle (115 m2/g).

3.3. Morphological, Optical, and Thermochromic Properties of Polymer/VO2 Films

Figure 10 shows SEM images of acrylic polymer composite films loaded with 1 wt% of different types of VO2 particles. It can be clearly seen that voids or gaps in the interface between particles and the polymer matrix disappeared after the neat VO2 particles were replaced by the VO2@PSS : PEDOT particles. This indicates that the state of compatibility between the acrylic polymer and the grafted VO2 improved. In addition, from the EDX dot maps (Figure 11), greater dispersion of the particles in the acrylic polymer matrix can also be noticed.

The optical properties of acrylic film containing different types and concentrations of VO2 particles are presented in Figures 12 and 13. The percentage visible light transmittance () through the acrylic polymer film dropped from 87% to 81-82%, after loading with 0.5% of the neat VO2 particles. The transmittance values tended to decrease with increasing concentration of VO2 particles. A similar trend is also observed for the polymer composite films containing the modified VO2 (VO2@PSS : PEDOT) particles. The percentage visible light transmittance (, %) of each composite film decreased to the lowest values, ranging between 68% and 71%, when the polymer was loaded with 1.5 wt% of VO2-based particles. These values became lower than those of some existing commercial products such as heat control window films [41]. Therefore, in this study, no further attempts were made to increase the concentration of VO2 particles beyond 1.5 wt%.

Figure 14 shows the appearance of polymer composite films containing different types of VO2 particles. The optical transparency of these films can hardly be differentiated by the naked eye, probably because the percentage loading of the particles is only 1.0 wt%. However, consideration of the percentage transmittance (Figure 12) reveals that a value above 75% of the composite films could be reached when the neat VO2 particles were replaced with VO2@PSS : PEDOT particles. Again, the improvement could be attributed to better dispersion of the particles after surface modification. However, as the concentration of VO2@PSS : PEDOT was further increased to 1.5 wt%, the values of the composite film decreased again and became closer to the value of the acrylic/neat VO2 film. This was due to more agglomeration of the modified VO2 particles at higher loading (Figure 11). Interestingly, by using PSS and PEDOT (polymerized without the presence of VO2 particles) as fillers, the transparency of the acrylic film also decreased. This was probably due to poor miscibility between the acrylic polymer and PSS/PEDOT. Moreover, the actual content of PSS/PEDOT chains loaded into the acrylic film was 1.0 wt%, which is much higher (about 10 times) than the actual weight of the PSS : PEDOT grafted onto the modified VO2 particles. These factors contributed to the decrease in transparency. Nevertheless, the values of the acrylic polymer/VO2@PSS : PEDOT composite films in this study are much greater than those of the EVA/VO2@PE composites in our earlier report [12], provided that the same volume of VO2 particles were loaded (1.0 wt%) (see Table 4). In our opinion, the discrepancy could be attributed to several factors, including the fact that different polymers and mixing processes were used. Hereafter, acrylic-based emulsion was used, and the mixing was carried out via a solution process. By contrast, EVA pellets were compounded with additives and VO2 particles via an extrusion process under shear force and heat. In this regard, a different state of dispersion of the particles in the polymer matrices can be obtained. Besides this, the PSS and PEDOT grafting chains used in this study are intrinsically transparent. This feature contributed significantly to the greater of the acrylic film (Table 4). The above effect was not the case for the VO2 coated with polyethylene, which is opaque.

Apart from the optical transparency, the percentage NIR transmittance () through the polymer films also decreased when the VO2 particles were loaded. The higher the concentration of modified VO2 particles, the lower the values. However, the percentage NIR reflectance () of these films did not increase, regardless of the type of concentration of VO2-based particles. It is important to remember that the UV/Vis/NIR spectroscopy test was carried out at ambient temperature, which is below the of the VO2 particles. Therefore, the metal oxide particles remained in a monoclinic phase, VO2(M), which is incapable of reflecting the transmission of NIR radiation effectively. In this regard, the decrease in percentage NIR transmittance through the acrylic/VO2@PSS : PEDOT composite film could be attributed uniquely to other factors including a light-scattering effect and the presence of PSS and PEDOT chains on the surface of the modified VO2 particles. In addition, a significant amount of the NIR radiation was absorbed by metal oxide particles in the polymer composite films. A similar effect was also observed in the PVC/VO2 composites [42], and that influenced the thermal stability and durability of the composite films.

The heat shielding properties of the composite films containing VO2 particles, however, began to appear and became remarkable when these samples were tested in a model house, installed with thermocouples and an infrared lamp. It is worth mentioning that when the lamp was turned on, the actual temperature in front of the model house was equal to 85°C, which is above the of the VO2 particles. In this regard, it was possible that the monoclinic crystal structure of the VO2 particles changed to the rutile phase, which is thermochromic and capable of exhibiting NIR shielding properties. From Figure 15, the temperature of the normal acrylic polymer film (without VO2 particles) increased steadily with time and reached a plateau at 85°C, after 30 min. However, when VO2 particles were loaded into the acrylic polymer film, the temperature, after reaching equilibrium, decreased by 4.3°C.

Finally, when the neat VO2 particles were replaced by the modified VO2 particles, similar trends still existed. The final temperature after equilibrium of the system equipped with these films was also slightly lower than that installed with the neat VO2/acrylic film. However, the differences were not obvious. Since the critical transition temperature of VO2 increased slightly after the surface modifications (see Table 3), it was possible that the phase transition of the modified VO2 particles did not occur completely and so the NIR shielding efficacy of the materials was not fully exhibited. In addition, further work has yet to be carried out to optimize the grafting density and percentage loading of modified VO2 particles in this system.

4. Conclusions

Vanadium dioxide particles grafted with polystyrene sulfonate/poly(3,4-ethylenedioxythiophene) (VO2@PSS : PEDOT) were successfully prepared via surface functionalization and surface-initiated atom transfer radical polymerization. The thermochromic behaviors of the modified VO2 particles were observed with phase transition temperatures ranging between 78 and 79°C. Window films made from the acrylic/VO2@PSS : PEDOT (1.0 wt%) were capable of reducing the temperature inside a model house, of up to 5-6°C, compared with the control system (the acrylic coating without any particles). Overall, this study has demonstrated that it is possible to improve the dispersion of VO2-based particles in acrylic polymer by grafting with PSS : PEDOT chains.

Data Availability

The data concerning the detailed synthesis procedure for preparation of various VO2-based particles are described and reported within the supplementary information file. Additional data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

Ms. Srirodpai is sincerely grateful to the scholarship provided by King Mongkut’s University of Technology Thonburi through the “KMUTT 55th Anniversary Commemorative Fund (Petchra Pra Jom Klao Doctoral Scholarship).” Special thanks go to Mr. Alongkorn Nukulpakdi, from the RPSC Chemicals Co. Ltd., for providing the acrylic emulsion used as a binder for making polymer films in this work.

Supplementary Materials

Detailed synthesis procedures of the various vanadium dioxide-based particles. (Supplementary Materials)