Photocatalytic Properties and Chemical Durability of CaO-B₂O₃-V₂O₅ Borovanadate Glasses

Abstract

The aim of this paper is to investigate the photocatalytic properties and chemical durability of the CaO-B₂O₃-V₂O₅ glasses system. The latter were synthesized by the melt-quenching technique. The amorphous nature of the prepared borovanadate glasses has been confirmed with X-ray diffraction. The chemical durability measured from their weights before and after immersion in deionized water and hydrochloric acid indicated that replacing V₂O₅ with B₂O₃ improved the chemical durability. The observed increases in chemical durability might be explained by the increase in the glass transition temperature, due to stronger bonding in the structural network. The photocatalytic performance was assessed by the degradation of methylene blue (MB) dye under irradiation, and the evolution of dye degradation was analyzed by UV-visible spectrometry. The vanadium content in the glass, the amount of catalyst, and the initial dye concentration showed a variable effect on the degradation of the MB dye. The photodegradation of methylene blue by the photocatalysts was found to follow pseudo-first-order rate kinetics. The photocatalytic activity for all the prepared photocatalysts showed a higher degradation performance, and the results indicated that 40 CaO-30 B₂O₃-30 V₂O₅ has the highest removal efficiency of about 99% in 180 min.

1. Introduction

Borate is one of the best materials for glass formation and the use of B₂O₃ as a network glass-forming oxide has many advantages: B₂O₃ forms two major structural units in borated glass networks, the triangle BO₃ and the tetrahedral BO₄ units, allowing the formation of anionic sites that accommodate various modified metal cations [1,2,3,4]. Additionally, vanadate (V₂O₅)-based glasses have attracted much interest in chemistry and materials science owing to their unique properties and applications in memory and switching devices [5,6]. Vanadate glasses are identified as n-type exceptional conductive properties due to the hopping between V⁴⁺ and V⁵⁺.

The addition of transition metals (TMs) to borate glasses has attracted more attention because of their wide applications in industry, science, and medicine [3], as they exhibit good semiconducting properties [7]. Semiconductor glasses could be made from different transition metal oxides (e.g., V₂O₅, MoO₃, etc.) and the semiconducting properties of V₂O₅ glasses containing results from the hopping of 3 dL electrons between the V⁴⁺ and V⁵⁺ states [7,8,9,10]. For instance, borovanadate glasses have important properties, such as a low melting temperature, high chemical durability, and thermal resistivity, making them attractive materials for various applications [11].

Our previous investigation indicates that glasses of the CaO-B₂O₃-V₂O₅ system with compositions of 40 CaO-(60−x) B₂O₃-x V₂O₅ (x = 0–40 mol%) have a low melting temperature at 900–950 °C, and a low glass transition temperature (381–650 °C) that is dependent on the composition [12]. Suyeon Choi et al. [13] reported that when P₂O₅ is substituted with B₂O₃, the dissolution rates of the different glass systems decrease, with an increase in the B₂O₃ concentration. However, the addition of B₂O₃ has been reported to improve the chemical durability [13]. Recently, semiconductor materials have received much attention and are at the forefront of the environmental sector for their application in photocatalysis [14]. In recent decades, the development of photocatalysis technology has received increasing attention and is one of the most interesting topics for scientists who want to use it for the removal and degradation of organic pollutants [15,16,17,18,19].

So far, the presence of organic pollutants in water is one of the main problems confronting modern society, causing serious and irreparable damage to the natural world [20]. Even low concentrations of dyes in water cause a change in its color, and thus the penetration of light, which has several consequences [20,21]. Therefore, these dyes will need to be removed from the wastewater before discharge to natural waters. New materials with photocatalytic and removal properties are investigated for their efficient use [22,23]. In general, these materials are oxides based on metal cations with d0 or d10 configurations. These metals have also been reported to use visible light more efficiently than traditional photocatalysts such as TiO₂ [24,25,26]. Transition metal vanadium group-5, block-d, is one of these metallates. V₂O₅ has a bandgap energy of 2.3 eV in the visible region, which shows high reactivity for the photocatalytic reaction [14,27]. Recent studies on bismuth vanadate (BiVO₄) [24,28], vanadium pentoxide (V₂O₅) [14], and zinc vanadate nanoparticles [29] revealed that these phases are semiconducting photocatalysts which exhibit good degradation efficiency. Studies using glasses and glass-ceramics as photocatalysts for wastewater treatment have been widely reported recently [30,31,32,33,34,35,36]. Herein, we have used the prepared materials as photocatalysts for the removal of methylene blue (MB) as a model pollutant in aqueous medium. MB is a thiazine dye that can cause eutrophication of surface waters [37], and it is characterized with a deep blue color and widely used in industry. This dye can have harmful effects on human health and our ecosystem. It can cause methemoglobinemia, vomiting, nausea, and tissue necrosis [37,38].

In this work, glass materials with different compositions of CaO-B₂O₃-V₂O₅ were successfully prepared. The chemical resistance of these glasses was examined in two solutions with different initial pH values, and the obtained results show how B₂O₃ influences the chemical durability in the glass system. Thus, the photocatalytic efficiency of the synthesized glasses was investigated by monitoring the degradation of methylene blue with various concentrations in an aqueous medium under illumination.

2. Experimental Procedures

2.1. Sample Preparation

Calcium borovanadate glasses with compositions of 40 CaO-(60−x) B₂O₃-x V₂O₅ (with x = 0, 10, 20, 30 mol%) were synthesized by a conventional melt-quenching method from stoichiometric mixtures of vanadium oxide (V₂O₅), boric acid (H₃BO₃), and calcium carbonate (CaCO₃). These starting products were precisely weighed and thoroughly ground in the agate mortar to homogenize the batch before being heated at 400 °C to remove volatile organic species. Then, the mixtures were gradually heat-treated at 900–950 °C for about 60 min. The melted liquid obtained was quenched in a steel mold preheated to around 200 °C.

Different methods were used to characterize the obtained glasses, namely density measurement and thermal analysis by differential scanning calorimetry (DSC), as reported in the previous work [12]. The compositions of the prepared glasses are provided in

Table 1. Composition (mol %), density ρ (g·cm⁻³), glass transition temperature Tg (°C), and dissolution rate (DR) (g/cm²/min) of the studied glasses.

Glass No.	CaO	B2O3	V2O5	ρ [12]	Tg ± 6° [12]	DR (pH = 7)	DR (pH = 4)
A1	40	60	0	2.603	650	(3.76 ± 0.20) × 10−7	(4.74 ± 0.20) × 10−7
A2	40	50	10	2.669	570	(1.58 ± 0.20) × 10−6	(4.18 ± 0.20) × 10−6
A3	40	40	20	2.749	480	(2.05 ± 0.20) × 10−6	(6.74 ± 0.20) × 10−6
A4	40	30	30	2.853	430	(5.48 ± 0.20) × 10−6	(8.95 ± 0.20) × 10−6

with the values of density and glass transition temperature. For confirmation of the vitreous nature of each glass sample, X-ray diffraction diagrams were recorded and reported in our earlier paper [12].

2.2. Durability Testing

To assess the chemical durability, the prepared samples were polished, cleaned, and weighed with an analytical digital balance. Then, the glass samples were attacked and completely immersed at room temperature for 28 days in deionized water (pH = 7) and for 5 days in an acidic HCl solution (pH = 4). The weight of each glass sample was measured after its immersion at regular intervals (144 h, 288 h, 432 h, 576 h, 672 h) in deionized water and (24 h, 48 h, 72 h, 96 h, 120 h) in an acidic HCl solution. The percentage of weight loss (WL) was determined from the following formula [39,40]:

$$\mathrm{WL}=\frac{\mathrm{WB}-\mathrm{WA}}{\mathrm{WA}}\times 100$$

(1)

where WA and WB are the weights of the glass after and before immersion in the solution.

Therefore, the dissolution rate was found from the weight loss with the expression [41,42]:

$$\mathrm{DR}=\frac{ \Delta \mathrm{W} }{\mathrm{A}.\mathrm{t}}$$

(2)

where ΔW is the weight loss (g), A is the area (cm²) before dissolution, and t is the immersion time (min).

To see the effect of vanadium on the vibrational properties and according to our previous work [12], in infrared spectroscopy, there is a small band situated in the 1200–1600 region, and the maximum at 1230 cm⁻¹ starts to appear when the molar content of V₂O₅ exceeds 10%, which is explained by the presence of diborovanadate [B₂V₂O₉]²⁻ groups [43,44]. Furthermore, from the Raman spectra [12], the absence of a peak in the range 930–947 cm⁻¹ for the sample that is not containing V₂O₅ is normal as there is no V = O bond, while when V₂O₅ was added, this peak began to appear, which is attributed to the V = O vibrations of the tetrahedral units VO₄ or VO₅ [12,45].

The formation of diborovanadate groups is clarified by assuming that in glasses containing more than 10 mol% V₂O₅, a continuous random lattice is formed of either two BO₄ units and two VO₄ units or two BO₃ units and two VO₅ units, and the construction was proposed by Reddy and co-workers for the diborovanadate [B₂V₂O₉]²⁻ [43,44]. From the IR, Raman results, and the proposed structure construction, we infer that the V⁵⁺ content of the total vanadium in the structure has the highest proportion (see Supplementary Information).

2.3. Photocatalyic Activity

The photocatalytic degradation was performed in a circular quartz vessel under continuous UV exposure. The glass samples were irradiated with a 125 W Philips mercury vapor lamp (λ = 254 nm). Methylene blue (MB) was chosen as an organic pollutant for the photocatalytic degradation experiments. It is a representative strong cationic dye that has been extensively used as an organic model to examine the photocatalytic performance of photocatalysts [46,47,48]. For this, 150 mL of aqueous solution containing different concentrations of MB was suspended with each powder of the glass samples as photocatalysts. The distance between the lamp and the beaker is 10 cm. Before illumination, the solution was kept in the dark and under stirring for 45 min to establish adsorption/desorption, reaching the equilibrium. Afterward, the solution was irradiated to start the photocatalysis reaction. At the designated times, samples of the solution were taken and their UV-visible absorption spectra were measured to evaluate the photocatalytic degradation using a Jasco V-730 (Japan) UV-visible spectrophotometer (UV-2300). The 664 nm absorption band (Figure 1) represents the maximum absorbance of the methylene blue, and the change in MB concentration will be monitored at λ = 664 nm. The percent of MB dye degradation was obtained from the concentration difference, as shown in Equation (3) [49]:

$$\mathrm{Photocatalytic} \mathrm{efficiency} =\left(1-\frac{\mathrm{C}}{{\mathrm{C}}_0} \right)\times 100$$

(3)

C₀: the initial concentration of MB dye,

C: the concentrations of the solutions after irradiation.

Figure 1. Absorbance spectrum of MB.

Figure 1. Absorbance spectrum of MB.

3. Results and Discussion

3.1. Chemical Durability

3.1.1. Variation in Weight Loss and Dissolution Rate (DR)

Figure 2a,b illustrate the weight loss percent versus the leaching time in deionized water and in HCL solutions, respectively. As one can see, the mass loss increases proportionally with the immersion time. In deionized water (Figure 2a), the mass loss is dependent on the prepared sample, and different trends were recorded. Sample A4 containing less boron and more vanadium rapidly increased, reaching 43% in 720 h. On the other hand, sample A1, which has the highest percentage of boron, showed an increase in percentage weight loss of only 2% during the immersion time. However, for A2 and A3 samples, a similar weight loss behavior was observed, recording around 10% and 14% at 720 h, respectively.

In HCl solution, we observed a similar tendency: During 24 h, all samples showed a low weight loss which did not exceed 2%. After a long immersion time of more than 50 h, we noticed that:

• The weight loss of sample A4 rapidly increased to reach 11% in 120 h.
• The sample A1 showed a very low weight loss of nearly 1% in 120 h.
• A similar weight loss of 5% was recorded for A2 and A3 samples.

The increase in weight loss with the increasing immersion time is assigned to the fact that the glass components had adequate time to be liberated into the solution [40]. The increase of the weight loss of the sample A4 might be due to different reasons. For instance, the high content of B₂O₃ and/or V₂O₅ as well as the value of the glass transition temperature are the main factors that influence the weight loss. Below, we have found the dissolution rates of the four samples.

Figure 3a,b show the dissolution rate (DR) of glasses as a function of B₂O₃ content, after 720 h and 144 h of the leaching test in different solutions: deionized water (pH = 7) and HCl (pH = 4), respectively.

In an acidic medium (Figure 3b), the dissolution rate slightly decreased between 30 and 50 molar% of B₂O₃. When more than 50% B₂O₃ was added, the dissolution rate rapidly decreased. In our case, we supposed that the lower DR for more than 50% B₂O₃ corresponds to the absence of diborovanadate groups in the solution [12]. According to our previous work [12], diborovanadate groups are present in glasses containing more than 10% V₂O₅ and less than 50% B₂O₃. Thus, the addition of less than 50 mol% B₂O₃ indicates that B₂O₃ has a lesser effect on the glass structure. A similar effect was found for minor additions of MnO in (50−x) K₂O-x MnO-50 P₂O₅ glasses [50]. Furthermore, when the samples were immersed in deionized water (Figure 3a), the DR was also reduced at the immersion time of 720 h. In general, the dissolution rate depends on the chemical compositions, bond forces, bond lengths, and the number of the coordination of atoms in the glass matrix [39,40]. The increase in the glass transition temperature with increasing B₂O₃ content (Table 1) led to a higher chemical durability due to increased cross-linking between the borate chains, and therefore a stronger bond in the structural network [51,52,53]. The results of the DR show that the increase in B₂O₃ content of borovanadate glasses led to a more effective chemical durability improvement. Addition of B₂O₃ has been reported to improve the chemical durability in several works [13,53,54]. As can be seen from Table 1, A1 glass has the lowest DR value (3.76 × 10⁻⁷ and 4.74 × 10⁻⁷ g/cm²/min in deionized water and HCL, respectively), which is quite close to that of window glasses (10^−7.2–10^−8.4 g/cm²/min) [41,55,56], and to those (1.1 × 10⁻⁷–6.6 × 10⁻⁹ g/cm²/min) of iron phosphate waste-forms [41,55].

According to Suyeon Choi et al. [13], the greater DR of glass systems containing high concentrations of V₂O₅ (70 V₂O₅-x B₂O₃-(70−x) P₂O₅) confirms that V₂O₅ does not enhance the chemical durability, which is in good agreement with our results.

3.1.2. Variation of the pH

Figure 4a,b show the evolution of the pH value as a function of immersion time in water and HCl, respectively, for all prepared glasses. In acidic solution (Figure 4b), we found that the pH curves consist of two regions: The first region occurs before 24 h, with a fast increase in the pH of the solution resulting from the sudden ion exchange between the glass and the solution [50,52]. The second region is between 24 h and 120 h, exhibiting a plateau that can be assigned to the saturation of the surrounding solution [57]. An identical pH behavior was observed in water (Figure 4a). However, small pH fluctuations around 8–9 were noted in the studied timespan.

The pH increases of the leaching solutions (HCl pH = 4 and deionized water pH = 7) could be ascribed to the extraction via ion exchange processes between the solution’s proton and the lattice-modifying elements. This exchange process is behind the obtained pH values because of the increase of OH⁻ ions remaining in the solution [50,52].

3.2. Photocatalytic Activity

3.2.1. Effect of V₂O₅ and Modeling of Photodegradation Kinetics

The photocatalytic reaction was performed by choosing methylene blue as the model organic pollutant. The prepared materials of CaO-B₂O₃-V₂O₅ glasses with different percentages of vanadium have been used as photocatalysts under UV light conditions. Generally, under illumination, the catalyst leads to the creation of a hole (h⁺) in the valence band and an electron (e⁻) in the conduction band (CB). The electron (e⁻), before being recombined with the hole, reacts with the water molecule on the surface to create OH^- radicals. These radicals lead to the degradation of organic molecules [30,58,59].

Figure 5a illustrates the photocatalytic tests for MB (10 mg/L) degradation in the presence of the three borovanadate glass catalysts. As the figure shows, the concentration of MB is reduced with the irradiation time in the presence of different catalysts. This decrease is dependent on the vanadium percentage. Photocatalytic performance was enhanced as the nominal vanadium oxide content increased, reaching the highest MB decolorization for the 40Ca0-30B₂O₃-30V₂O₅ photocatalyst. The decrease in MB concentration in the dark was a result of adsorption, whereas that under illumination occurred as a result of both adsorption and decomposition of MB [60]. After the reaction time, there was almost complete removal of the MB molecule. Over 99% of MB was photo-decomposed, confirming the excellent photocatalytic performance of the 40Ca0-30B₂O₃-30V₂O₅ catalyst. Furthermore, the percentages of degradation for the other catalysts, 40 CaO-(60−x) B₂O₃-xV₂O₅ with x = 20 and x = 10, were 96% and 84%, respectively (Figure 5b). It can be observed from previous studies [61,62,63] that the addition of vanadium with an optimal amount improves the photocatalytic activity.

The kinetic rate is an importance parameter for degradation studies as it could predict the rate at which a pollutant is eliminated from the water solution. For this purpose, the kinetic characteristics of the different glasses used as photocatalysts for the photodegradation of methylene blue were carried out. Figure 6a shows the kinetic fitting curves of ln (C₀/C) versus irradiation time, t, for the different photocatalysts, which indicate that the degradation follows a pseudo-first-order kinetics obtained from the Langmuir–Hinshelwood equation [64,65,66]:

$$\mathrm{Ln}\left(\frac{{\mathrm{C}}_0}{\mathrm{C}} \right)=\mathrm{k}.\mathrm{t}$$

(4)

where C₀ is the initial MB concentration, C is a concentration of MB at any time, t, k is the rate constant for the first order, and t is time.

The rate constants are presented in Figure 6b, allowing a quantitative comparison to represent the photocatalytic activity. According to the figure, it can be clearly observed that the constant rate was significantly enhanced from 0.0113 min⁻¹ to 0.0274 min⁻¹ by raising the V₂O₅ content from 10% to 30%, which accounts for an almost 2.5-fold increase in the catalytic efficiency. In addition, it can also be noticed that the 40 CaO-30 B₂O₃-30 V₂O₅ photocatalyst presented the greatest constant k value. From the kinetic data, it was concluded that as the content of vanadate increased, the constant rate was improved, indicating that vanadium is a key parameter to enhance the photodegradation process.

The high rate constant observed for the A4 sample can be principally due to the change in morphologies, and voids in the glass matrix. This enhanced the MB adsorption and yielded more active sites for the photocatalytic degradation of MB molecules [67]. The existence of vanadium oxide promoted the migration of photogenerated electrons to vanadium, thereby enhancing the electron–hole separation [62].

3.2.2. Effect of Catalyst Loading and of Initial MB Concentration

The effect of catalyst mass on the MB photodegradation was examined. In this regard, we have selected the best-performing photocatalyst (40 CaO-30 B₂O₃-30 V₂O₅) at different loadings ranging from 0.5 to 3 g/L. Figure 7 displays the obtained results. In the dark, the highest adsorption amount was around 44%, recorded by a catalyst amount of 3 g/L. Other loadings have recorded around 30% of MB adsorption. Once the irradiation was turned on, the 3 g/L catalyst loading showed the fastest degradation rate, explained by its higher MB uptake in the dark through the adsorption process. After 90 min of irradiation, a 1 g/L loading recorded the lower degradation efficiency compared to 0.5, 2, and 3 g/L. Interestingly, the 1 g/L loading has also showed the lowest adsorption uptake in the dark, and this explains its low active sites availability and hence low photocatalytic activity compared to the other loadings. The small decrease of the photocatalytic efficiency of the 1g/L loading compared to the 0.5 g/L loading can be explained by the fact that the light transmission field decreased as the opacity of the solution increased [15,33].

Keeping the catalyst loading of 40 CaO-30 B₂O₃-30 V₂O₅ constant at 0.5 g/L, the effect of various initial MB concentrations on the photocatalytic activity is presented in Figure 8. The initial dye concentration was varied in the range of 5 to 15 mg L⁻¹. During the adsorption process, when the MB concentration increased, the removal efficiency of MB decreased. After illumination, we observed that MB degradation with the 15 mg/L concentration was the slowest. This was expected due to the higher amount of the dye and the low-light transmission field. This can also be explained by the large quantity of simulated illumination absorbed by MB molecules, more than by the catalyst at high MB concentrations [15,68]. The catalyst with different concentrations of MB (5, 10, and 15 mg/L) reached almost the same degradation value around 99% during 180 min of irradiation.

Figure 9 shows the evolution of the UV-vis absorption spectrum as a function of the photodegradation time of the MB dye for the best-performing catalyst, 40 CaO-30 B₂O₃-30 V₂O₅ glass (0.5 g/L), at the optimal conditions of catalyst amount and dye concentration. As can be seen from Figure 9, the color of the 10 ppm MB solution completely disappeared after 180 min of illumination.

In order to further investigate the photocatalytic mechanism, the main active species generated during the process of MB degradation were inspected by the scavenging experiments.

The scavenging tests for hydroxyl radicals, superoxide ions, and holes were performed using isopropanol, ascorbic acid, and EDTA, respectively [69,70,71,72]. Figure 10 shows the effect of these scavengers on the degradation of MB. We found that in the presence of EDTA and ascorbic acid, the degradation efficiency was reduced to 43% and 39%, respectively. However, a low inhibition was observed for MB degradation with the addition of isopropanol (70%). Thus, it could be inferred that h⁺ holes and O₂•⁻ ions should be the dominant active species in the photocatalysis of MB, while hydroxyl radicals play a minor role in the photocatalytic reaction. Since the reduction potential of photogenerated electrons from the photocatalyst was higher than that of O₂•⁻/O₂ (0.28 eV), the released electrons can be easily transferred to oxygen (O₂) molecules adsorbed on the surface of the borovanadate photocatalyst, producing the superoxide O₂•⁻, which is a strong oxidizing species to degrade organic molecules.

Table 2. Comparison with the state-of-the-art photocatalysts for MB degradation.

Catalyst	Concentration of Catalyst	Concentration of Dye	Light Source	Photocatalytique Degradation	Ref
Zn3(VO4)2/BiVO4	0.7 g/L	-	Visible light	98% at 90 min	[73]
CuO-BiVO4	0.6 g/L	10 mg/L	Visible light	92% at 240 min	[74]
SrBi3VO8	0.17 g/L	10 mg/L	Visible light	80% at 180 min	[75]
2Bi2O3–B2O3 (BBO) glass		50 mg/L	UV light	25% at 180 min	[36]
Fe doped TiO2	1 g/L	10 mg/L	UV light	38% at 120min	[64]
CdS-Borosilicate glass (SiO2-B2O3-Na2O-ZnO)	1 g/L	10 mg/L	sunlight	68% at 270 min	[34]
titanium-doped hydroxyapatite	0.5 g/L	10 mg/L	UV light	99% at 240 min	[76]
SnO2/CdSe/Bi2S3 composite	0.5 g/L	30 mg/L	Visible light	99% at 60 min	[77]
Tin Oxide (SnO2)	0.5 g/L	30 mg/L	Visible light	53% at 60 min	[77]
CeO2–Fe2O3–NiO nanocomposite	0.6 g/L	5 mg/L	sunlight	73% at 50 min	[78]
Fe2O3 nanoparticles	0.4 g/L	10 mg/L	sunlight	53% at 50 min	[79]
40CaO–30 B2O3-30 V2O5	0.5 g/L	10 mg/L	UV light	99% at 180 min	Present work

reports the comparison with state-of-the-art photocatalysts for MB photocatalytic degradation. As one can see, the efficiency presented in this paper is one of the most promising catalysts for MB removal.

4. Conclusions

This paper reported a facile synthesis of borovanadate glasses, and the study of the chemical durability and photocatalytic activity of CaO-B₂O₃-V₂O₅ glasses. The substitution by increased B₂O₃ has led to the increase of Tg temperature, leading to a decrease in the dissolution rate, therefore enhancing the chemical durability of the glass. The photocatalytic performance of the prepared glasses was examined under UV irradiation by the degradation of methylene blue (MB) dye. The prepared 40CaO-30B₂O₃-30V₂O₅ photocatalyst with a catalyst loading of 0.5 g/L had the highest photodegradation efficiency of MB (10 mg/L), whereby it reached about 99% after 180 min of irradiation. This study shed light on the importance of glass materials as a sustainable method for the depollution of water through introducing more transition metals (i.e., V₂O₅).