A PVA–Brookite Composite: The Effect of Plasma Pre-Treatment on the Thermal, Mechanical, and Photochromic Properties

Abstract

A composite material based on polyvinyl alcohol (PVA) and brookite-phase titanium dioxide (TiO₂) was synthesized using a straightforward method that involved combining the polymer with a sol as a filler. The composites were analyzed using Fourier transform infrared (FTIR) spectroscopy, scanning electron microscopy (SEM), atomic force microscopy (AFM), differential scanning calorimetry (DSC), and mechanical testing. The effects of treating the sol with underwater discharge plasma utilizing different electrode materials on the surface morphology, mechanical properties, thermal stability, and optical (photochromic) characteristics of the composites were investigated. FTIR spectral analysis indicated the presence of a chemical bond between the polymer matrix and the filler particles, as evidenced by the appearance of new peaks in the region of 700–500 cm⁻¹. Preliminary plasma treatment was shown to enhance the thermal stability, strength, and elasticity of the PVA-based composite. These improvements resulted from the modification of the filler (sol) due to plasma activity. The resulting composites exhibited a low photocolorization rate and a high bleaching rate. Such composites represent a promising material for use as inks in 3D printing.

1. Introduction

Composite materials that incorporate polymers and inorganic fillers demonstrate significant potential for a variety of applications, owing to the synergistic properties of both the polymers and the fillers involved [1]. The selection of fillers for the polymer matrix is influenced by the intended applications of the composites. Layered double hydroxides serve as effective fillers in the production of reinforced composite materials. The enhancement of mechanical properties is attributed to the intercalation of polymer molecules within the interlayer space, as well as the strong interphase adhesion with anions. An examination of published data indicates that, depending on the characteristics of Me²⁺, layered structures may function as either strengthening agents or plasticizers [2,3,4].

Polyvinyl alcohol is an attractive material due to its low cost, flexibility, transparency, and biodegradability [5]. However, the thermal properties of this polymer need improvement because the melting and decomposition temperatures are closely aligned. To address this issue, various composite materials based on PVA and different fillers—both organic and inorganic—have been developed, which influence not only the thermal properties but also the mechanical and optical characteristics [6]. This expansion of properties allows for a broader range of applications for such composites [7].

Titanium dioxide plays a crucial role as a heterogeneous photocatalyst. Early studies examined the photocatalytic properties of the three modifications of titanium dioxide: anatase, rutile, and brookite [8,9,10]. In addition to its photocatalytic capabilities, titanium dioxide also exhibits photochromic properties [11,12]. When combined with polymers, different oxides of Ti (Ti₂O₃, TiO₂ (anatase, rutile, brookite), Ti₄O₇) can serve as membranes for various applications or function as gamma ray shields [13,14,15,16,17,18,19,20,21,22,23,24,25]. Published results from the thermal measurements of composites based on PVA and TiO₂ reveal the effects of the phase composition and structure. The study referenced as [16] demonstrated that the incorporation of anatase increases the melting temperature. Furthermore, the addition of mixed-phase TiO₂ (comprising 80% anatase and 20% rutile) results in an elevation in the glass transition temperature while leaving the melting temperature (T_m) values unchanged [17]. The inclusion of amorphous titanium dioxide as a filler appears to have a minimal impact on the thermal properties of PVA [18].

An analysis of published data indicates that the mechanical properties of polymer composites can be either enhanced or diminished based on factors such as the filler concentration, phase composition, and the size of filler particles. Alfahed et al. determined that a polymer composite was only reinforced at a concentration of 0.04% TiO₂, within the range of 0.01–0.04 wt%. In a study conducted by Zamani et al., a reduction in mechanical properties was observed at filler concentrations of 1% and 2% (anatase, ~20 nm) when compared to the native polymer [10]. Conversely, the incorporation of anatase (10–12 nm) at concentrations ranging from 0.2% to 0.6% resulted in a significant improvement in the mechanical properties of polyvinyl alcohol (PVA) [12]. The authors demonstrated in a separate study that the introduction of just 1% TiO₂ (<10 nm) leads to a 1.5-fold increase in tensile stress [14]. Bozdogan et al. showed that the mechanical properties of composites containing 5% TiO₂ (rutile, 28 nm) declined when compared to both the original polymer and the composite with 50% filler [15].

The brookite phase was long considered thermodynamically unstable until Arnal et al. proposed a low-temperature sol–gel method to obtain this phase [26]. Recently published data on the PVA-TiO₂ composite (brookite) are limited to the general properties of the obtained material (X-ray, IR, AFM/TEM, spectroscopy). In this work, we present for the first time a method for the preparation of a composite based on a polymer matrix and a sol. Polyvinyl alcohol was used as the matrix and a titanium dioxide sol (brookite) as the filler without and after underwater plasma treatment at a low temperature. Pre-treatment with underwater plasma facilitates the creation of doped TiO₂ structures [27,28,29]. Utilizing underwater plasma to modify both organic and inorganic materials is regarded as an environmentally friendly method, as the primary modifier is the plasma itself, along with the electrode materials [30,31,32]. The thermal, mechanical, and photochromic properties of the composites were evaluated and compared to those of the untreated sol composite.

2. Materials and Methods

2.1. Preparation of Samples

Titanium oxide was synthesized using sol–gel technology. The detailed procedure is provided in the Supporting Information (SI) and has been described in greater detail elsewhere [27]. The prepared sols were subjected to an underwater diaphragm discharge using an alternating current. Two metal electrodes were immersed in a volume of liquid sol, with one electrode placed in a quartz ampoule featuring a small hole (diaphragm) with a diameter of 2 mm. When a voltage was applied to the electrodes in the diaphragm zone, a vapor–gas bubble was formed, within which an electric discharge occurred (plasma zone). The formation of the vapor–gas bubble interrupted the electric circuit, resulting in an electrical discharge inside the bubble. The subsequent collapse of the bubble generated sound and shock waves that propagated along the diameter of the membrane. This collapse also induced the convective flows of plasmolysis products (chemically active particles) into the main volume of the solution due to temperature and concentration gradients. Consequently, this resulted in the localized heating of the polymer, which facilitated subsequent processes involving chemically active particles, such as polymerization, depolymerization, and degradation. Conversely, the collapse of the bubble led to the spraying of the electrodes, with the speed of this process being determined by the material used. Plasma treatment durations varied from 1 to 5 min. In the experiments involving titanium oxide sols, molybdenum (Mo), tungsten (W), and niobium (Nb) wires were employed as electrodes. The dopant content was determined by the rate of sputtering of the electrode material under specific parameters of the underwater diaphragm discharge (I = 45 mA; U = 0.6 kV). X-ray phase analysis indicated that titanium dioxide had been successfully synthesized in the brookite phase (see Figure S1 in the Supplementary Information). The plasma treatment of the sols did not result in the formation of new phases but did cause changes in the intensities of the main peaks.

Poly(vinyl alcohol) (PVA) with a molecular weight of 2 × 10⁵, grade 16/1 (Khimreactive, Nizhny Novgorod, Russia), was dissolved in distilled water heated to 80 °C. Sols, both with and without plasma treatment, were added to the polymer solution while maintaining constant stirring. The resulting solutions were then poured into polystyrene Petri dishes (PERINT). The obtained films were dried at 40 °C for 48 h. The filler concentration was set at 1 wt%. A total of five samples were produced, including the original polymer. The designations of the samples are presented in

Table 1. Experimental conditions.

Sample	Sol	Plasma Treatment of Sol (Material of Electrode)/Dopant Content, %	Film Thickness, mm
1	-	-/0	0.10
2	TiO2	-	0.09
3	TiO2	+(Mo)/3.8	0.05
4	TiO2	+(Nb)/3.0	0.06
5	TiO2	+(W)/1.3	0.07

.

2.2. Physical Measurements

Fourier transform infrared (FTIR) spectra were registered using the Fourier spectrometer FT-801 (Simeks, Novosibirsk, Russia) in the range of 4000–400 cm⁻¹ with a resolution of 0.5 cm⁻¹. The thermal properties of the samples were studied through differential scanning calorimetry using the DSC 204 F1 t-sensor (NETZSCH, Selb, Germany) in the temperature range of 25–250 °C at a heating rate of 10 K min⁻¹ in an argon atmosphere. The degree of crystallinity of the samples was calculated using [33,34]

$$x={\displaystyle \frac{∆H_m}{∆H_m^0(1-n)}}\times 100\mathrm{\%}$$

(1)

Here, ΔH_m is the measured enthalpy of melting from the DSC analysis, $∆H_m^0$ is the enthalpy of 100% crystalline PVA melting (138.6 J g⁻¹) [35], and n is the mass fraction of the filler’s structures.

The morphology of composite films was investigated using atomic force microscopy (AFM) (Solver P47-PRO, Zelenograd, Russia) and scanning electron microscopy (SEM) (Quattro S, Thermo Fisher Scientific, Praha, Czech Republic). AFM analysis was carried out in the contact mode with a resolution of 3 μm and scanning area of 30 × 30 μm².

Mechanical properties were examined at room temperature using a Criterion C42 universal tensile testing machine (MTS System, Eden Prairie, MN, USA). The film size was 10 × 55 mm. The cross-head speed was 20 mm min⁻¹. Measurements were carried out according to a standard test (ASTM D882) [36].

Polymer composite films were irradiated with light (wavelengths of 254 and 365 nm). The samples were placed on the substrate at a distance of 10 cm from the radiation source (15W, NDTRADE, Saint Petersburg, Russia). The reversibility of the photochromism effect was studied at room temperature. The absorption spectra of irradiated composite films were studied using the spectrophotometer SF 2000 in the range of 400–1100 nm, which was equipped with a thin film sample holder for six samples (SDB Spectr, Saint Petersburg, Russia).

The morphology of the polymer composites was studied using atomic force microscopy (AFM) (Solver P47-PRO) in the contact mode with a resolution of 3 μm (area of 30 × 30 μm).

3. Results

3.1. FTIR Measurements

The investigation into the chemical functional composition of the composites was conducted using Fourier transform infrared (FTIR) spectroscopy measurements (Figure S2,

Table 2. FTIR characteristic peaks (cm⁻¹) of samples.

Peak Designation	Sample 1	Sample 2	Sample 3	Sample 4	Sample 5
O-H stretching	3265	3270	3267	3270	3266
C-H stretching	2938	2915	2916	2915	2918
C=O stretching	1704	1729	1729	1728	1729
δ(O-H)	1652	1639	1644	1643	1647
CH2 bending	1423	1417	1416	1422	1417
C-O stretching	1092	1092	1092	1091	1091
C-C	917	917	918	917	915
C-H out-of-plane	844	845	845	845	844
Me-O-Me	–	652	650	654	652

). The incorporation of fillers led to alterations in the shapes, intensities, and positions of the principal peaks. The emergence of new peaks within the 800–400 cm⁻¹ range indicates the formation of oxide structures. Shifts in the peaks associated with -CH₂ bending vibrations suggest a chemical interaction between the polymer matrix and the oxide structures, indicative of a cross-linking process. Variations in the peak intensity related to C-O stretching vibrations (within the C-O-H group) signify changes in the polymer’s degree of crystallinity. A decrease in the intensity of the O-H bands (3100–3600 cm⁻¹) was observed, indicating a reduction in the quantity of hydroxyl (OH) groups within the polymer matrix (PVA). This reduction may be attributed to the cross-linking process. Similar findings were reported in a series of studies on the physicochemical properties of composites based on PVA and TiO₂ (anatase + rutile) [37,38], which contrast with the results presented in another study [39], where the authors could not confirm the chemical interaction between the PVA polymer matrix and titanium dioxide.

3.2. Thermal Analysis

The thermal study data are presented in Figure 1 and

Table 3. DSC data of composites.

Sample	Tg, °C	Tc, °C	Tm, °C	Tm − Tc, °C	Tc/Tm	ΔHm, J g−1	x, %
1	34.67	186.64	226.99	40.35	0.82	45.05	32.51
2	31.11	188.24	242.2	53.96	0.78	18.46	13.18
3	43.75	189.61	241.90	52.29	0.78	67.06	47.89
4	44.88	194.33	246.60	52.27	0.79	97.66	69.75
5	44.09	192.49	245.90	53.41	0.78	88.64	63.32

. The introduction of a titanium dioxide sol without plasma treatment led to a reduction in the glass transition temperature and degree of crystallinity, a phenomenon that may be attributed to the plasticization effect. The incorporation of fillers following plasma treatment facilitated the production of composites with improved heat resistance. Conducting a comparative analysis with previously published data poses challenges due to the variability in the reported effects among different authors. In reference [19], it was observed that composites based on TiO₂ (anatase) and PVA exhibit a melting point of approximately 250 °C, with cross-linking resulting in the formation of an amorphous structure within the composite. Ahmad et al. noted that the melting temperature values remained constant with the addition of mixed-phase TiO₂ (80% anatase and 20% rutile), while the degree of crystallinity of the composites increased [20]. As this is the first time that thermal data have been obtained for composites made of PVA and brookite, our results are qualitatively consistent with the results of studies [19,20,21].

The data presented in Table 3 indicate that TiO₂ fillers facilitate the nucleation of the polymer matrix at early stages, specifically at high temperatures during the cooling cycle (T_c), when compared to native PVA. Among the fillers, undoped TiO₂ exhibited a relatively low Tc value. The initial phase of crystallization is the nucleation stage. Once the critical size of the nuclei is attained, they serve as centers for the crystallization of the polymer. The overall crystallization rate is the sum of the nucleation and growth rates, with the degree of cooling (T_m − T_c) being a crucial factor. A lower (T_m − T_c) corresponds to a higher overall crystallization rate [40]. Consequently, the high degree of cooling observed in sample 2 correlates with a low overall crystallization rate, which is reflected in a reduced degree of crystallinity. This suggests that a titanium dioxide sol is not an effective nucleating agent for PVA in the absence of plasma treatment. This inefficacy may be attributed to the morphology of the filler particles (refer to Section 3.3). Currently, no consistent relationship between the crystallization rate and the degree of crystallization has been identified. Nevertheless, similar trends have been noted in the variations in the melting point values and degrees of crystallinity [41,42].

3.3. Surface Morphology

AFM measurements indicated that the incorporation of any filler altered the surface roughness (

Table 4. Roughness of composites (nm).

Sample 1	Sample 2	Sample 3	Sample 4	Sample 5
3.80	17.94	5.90	16.59	46.77

) without inducing local inhomogeneities (Figure 2). The surface of pure PVA had the lowest roughness (Figure 2a). The increase in the surface roughness observed in the composite (sample 2) was likely related to the shape of the particles (see Figure 2b and Figure 3a). The variations in the roughness values of the films from samples 3–5 (sols after plasma treatment used as fillers) (Figure 2c–e) may be attributed to changes in the molecular weight of the stabilizer and the cross-linking process between the polymer and the filler [29]. As demonstrated in reference [43], the surface roughness increases with a higher cross-linking density. The use of sols promotes a uniform distribution of filler particles within the polymer matrix. Taking into account the scale of 2D AFM images, they are consistent with the SEM analysis data (Figure 3).

Figure 3 presents scanning electron microscopy (SEM) images of films containing fillers. The filler was uniformly distributed in samples 2, 3, and 5, although its shape varied. In the absence of plasma treatment, the filler particles in the sol exhibited a disk-like morphology (sample 2, Figure 3a). Following plasma treatment, the sol could either form a nearly continuous layer on the polymer surface (sample 3, Figure 3b) or aggregate into small and large rod-shaped structures within the polymer matrix (sample 5, Figure 3d). The sol demonstrated uneven aggregation both on the surface and within the volume of the polymer after plasma treatment with niobium electrodes (sample 4, Figure 3c). A change in the shape of the titanium dioxide particles was observed post-plasma treatment, likely due to alterations in the molecular weight of the stabilizer under the influence of the plasma [29]. This hypothesis is supported by previously published data indicating that the molecular weight of the stabilizer affects the shape of the nanoparticles [44,45,46].

3.4. Mechanical Properties

The mechanical test results are presented in Figure 4,

Table 5. Mechanical properties of composites.

Sample	Stress, MPa	Young’s Modulus, MPa	Elongation at Break, %
1	3.17	2.78	11.2
2	0.11	0.19	3.21
3	0.42	2.00	10.9
4	0.25	1.21	7.71
5	0.29	0.98	11.2

, Figure S3, and Table S1 (refer to the SI file), indicating that the incorporation of titanium-containing fillers significantly influenced the mechanical properties of the composites compared to the original polyvinyl alcohol (PVA). The changes in the mechanical properties of the composites relative to the native PVA cannot be attributed to local filler aggregation phenomena (refer to Section 3.3). Furthermore, the plasma pre-treatment of the sols affected the mechanical properties. A notable reduction in the mechanical properties of sample 2 compared to the original PVA (sample 1) may have been due to insufficient mechanical adhesion between TiO₂ and the polymer matrix (PVA), which can be attributed to the surface and shape characteristics of the filler. Another contributing factor could have been the lack of chemical bonding between PVA molecules and titanium dioxide. The introduction of sols after plasma treatment significantly enhanced the mechanical strength. The preliminary plasma treatment of the sols led to modifications in both the stabilizer (polyvinylpyrrolidone, PVP) and the oxide structures. Plasma treatment resulted in the formation of additional defective oxide structures, thereby improving chemical bonding with the polymer matrix. Conversely, plasma treatment induced the depolymerization of PVP [29]. The interaction of the stabilizer with PVA during cross-linking via the radical mechanism (Scheme 1) [47] resulted in a modification of the syndiotacticity of the polymer matrix. Data from the thermal analysis (melting points, the degree of crystallization) [48] and mechanical tests (samples 3–5 compared to sample 2) support this conclusion. A similar assumption was established previously [49]. The findings are consistent with previously published data [13,14,18]. Increasing the filler concentration further enhanced the elasticity of the resulting composites (Figure S3 and Table S1). The increase in the elasticity of the composite with a TiO2 filler without plasma treatment may have resulted from the formation of oval micropores near the disk-shaped filler particles, which elongated upon stretching. An analysis of the data from the thermal and mechanical tests indicates that the obtained composites can be considered non-toxic analogs of ink for 3D printers (Figure S4).

3.5. Photochromic Properties

In the absorption spectra of the irradiated composite films (Figure 5a), a poorly resolved band in the range of 500–530 nm could be identified. In comparison to the absorption spectra of the liquid sols after irradiation, which exhibited a maximum at 530–550 nm (with a shoulder at 650 nm) [29], the absorption maximum in the composite films was shifted toward the shorter wavelength region. This shift is attributed to the presence of monomeric and dimeric photochromic titanium dioxide particles [50,51]. The photochromism observed in the polymer composites suggests the existence of intermolecular interactions between the polymer matrix and the oxide structures.

Studies on the dynamics of photocoloring and bleaching have demonstrated that composites containing tungsten and molybdenum oxide structures exhibit a slow coloring process under light irradiation, yet they possess a significantly high rate of discoloration (see Figure 5b). It is important to note that composites with untreated fillers display a lower coloring rate, taking approximately 7 to 8 h. Conversely, plasma treatment does not influence the rate of discoloration. Previous research involving colloidal solutions of defective anatase (characterized by oxygen vacancies and the presence of Ti³⁺) and rutile indicated that noticeable coloration occurred after 45 and 15 h, respectively [49,51].

In previous studies examining the photochromic properties of composites composed of polymer matrices and transition metal oxides, the coloring process was regarded as a redox reaction involving the oxidation of polymer matrix units. However, the presence of oxidation products was not documented [52]. Further research into the mechanisms of photochromism suggested that photocoloration could result from two processes: (1) excitation by a photon leading to the formation of an excited state of the metal oxide, and (2) the abstraction of one or two electrons from the organic matrix without decomposition, followed by the subsequent addition of an electron, resulting in the formation of the Me⁽ⁿ⁻¹⁾⁺ state.

Discoloration in the absence of light is a process of oxidation influenced by atmospheric oxygen. Polyvinyl alcohol (PVA) exhibits a low oxygen permeability [53], which would typically result in low bleaching rates. However, the unexpectedly high rates of discoloration observed in composites containing TiO₂ fillers are noteworthy. This phenomenon may be attributed to the cross-linking effect, as it is known that the chemical cross-linking of PVA enhances its permeability to oxygen [53].

Currently, filaments made of polylactide (PLA) or acrylonitrile butadiene styrene (ABS) are used in 3D printers. When using PLA, the substrate must be heated, and in the case of ABS, the room must be ventilated. In this respect, PVA–brookite-based composites are promising materials for 3D printing, as their temperature properties are on average between those of ABS and PLA, they are non-toxic, and the substrate does not need to be heated.

4. Conclusions

The thermal, mechanical, and optical properties of the first synthesized composites based on PVA and a brookite sol were studied. The composite material properties were compared depending on the preliminary plasma modification of the filler. It was found that the brookite sol was a plasticizer of PVA. The plasma modification of the filler improved the thermal and mechanical properties of the composite. The photochromic properties of PVA–brookite composites were studied for the first time. It was shown that the modification of the brookite sol had a positive effect on the optical properties. The obtained results of the thermal and mechanical tests showed that such composites can be considered as non-toxic analogs of ink for 3D printers.