Multifunctionalization of Nanostructured Metal OxidesView this Special Issue
Nanocrystallization of Vanadium Borophosphate Glass for Improving the Electrical and Catalytic Properties
75V2O5-10P2O5-15B2O3 ternary-system glasses were prepared and nanocrystallized to examine the catalytic effect and the variations in their structural and electrical properties. These glasses were annealed in a graphite mold above the glass transition temperature for 2 h and heat-treated at the crystallization temperature for 1, 3, and 5 h. Fourier transform infrared spectroscopy (FTIR) was used to analyze the structural changes in the B-O bonds after nanocrystallization, while X-ray photoelectron spectroscopy (XPS) analysis showed a decrease in V5+ and an increase in V4+. X-ray diffraction (XRD) analysis of the structure array (BO3 + V2O5 ↔ BO4 + 2VO2) verified these inferred changes. Structural changes induced by the heat treatment were confirmed by analyzing the molecular volume determined from the sample density. Conductivity and catalytic effects were discussed based on the migration of vanadate ions with different valence states due to the increase in VO2 nanocrystallinity at 275°C. Both conductance and the catalytic effect were higher after nanocrystallization at 275°C for 1 h compared to the annealed sample. Furthermore, compared to the sample heat-treated for 1 h, the conductance and catalytic effect were increased and decreased, respectively, for samples nanocrystallized at 275°C for 3 and 5 h.
Vanadate glasses containing large amounts of V2O5 have multivalent ions of various states mixed in their structure. These glasses have been developed for infrared transmission and as atomic-exchange electroconductive glasses using electron conduction. Studies focusing on improving the performance of these glasses involve changing the electrical conduction properties through low valency (V4+) to high valency (V5+) metal-ion electron hopping , changing the catalytic properties for the oxidation reaction by altering the valency change between V4+ and V5+ , and examining the correlation between the glass composition and its characteristics while focusing on the type and condition of V ions or the type and the amount of alkali or alkaline earth metals [3–7].
In 2012, we substituted B2O3 for P2O5 to improve the chemical resistance and electrical conductivity of vanadium phosphate-based glasses, and we were able to investigate the significance of the correlation between the resulting structure and the electrical, thermal, and chemical properties . However, the effect of the substitution was insignificant. Moreover, the analysis of the catalytic activity of the oxidation reaction, which is characteristic of V2O5, with respect to the vanadate glass was insufficient. On the other hand, if vanadate glass is crystallized and the structure becomes denser, electron tunneling in the glass would occur with a higher frequency . This behavior leads to frequent valency changes in the V cation. It is hypothesized that this change influences the change in the catalytic effect and electrical conductivity. In particular, if the nanocrystalline phases are generated, the specific surface area of the crystalline phases will be maximized. So, property changes due to crystallization appear to be more prominent.
In this study, P2O5 was used as a glass former, to which V2O5 or other metal oxides are added in order to achieve V2O5-like characteristics. We maximized the electrical conductivity and catalytic properties of V2O5-B2O3-P2O5 ternary glass by maximizing the amount of the substitute (B2O3 for P2O5) to form 75V2O5-15B2O3-10P2O5 ternary glass. It was confirmed that nanocrystallizing this ternary glass produces VO2(VO4), V2O5(VO5), and B2O3(BO4) crystalline phases, and the structural change in the nanocrystalline phase was caused by increasing the duration of heat treatment for crystallization. From the results, it was evident that this behavior affected the ratio of V5+ to V4+.
We presumed that the changes in the electrical conduction and catalytic properties were induced by the heat treatment. The causes behind the characteristic changes and the action mechanism were investigated through qualitative and quantitative analysis of X-ray diffraction (XRD) patterns, X-ray photoelectron spectroscopy (XPS), and Fourier transform infrared (FTIR) spectroscopy. We examined the quantitative physical properties such as density and molar volume to verify the predicted action mechanism.
2. Materials and Methods
2.1. Glass Preparation and Nanocrystallization
75V2O5-15B2O3-10P2O5 glass samples were prepared using reagent-grade NH4H2PO4, V2O5, and B2O3 (Junsei Chemical Co., Ltd., Japan). The constituents were mixed in an alumina mortar for 10 min to produce multiple batches. Well-mixed batches were first melted in an alumina crucible using an electrical furnace at 200°C for 1 h (first calcination), subsequently at 500°C for 2 h (second calcination), and finally at 800°C for 2 h. The resultant melt was quenched between two stainless steel plates.
The as-prepared sample was crushed in an automatic grinding mixer and sieved through a 325 mesh. This powder (particle size below 44 μm) was used for differential scanning calorimetry (DSC), XRD, XPS, FTIR, and thermogravimetric analysis (TGA) measurements.
DSC was performed on the samples using an SDT (SDT Q600, TA instrument, USA) to determine both the glass transition () and the crystallization () temperatures, as shown in Figure 1(a). The glasses were annealed in a graphite mold above for 4 h and heat-treated at for 1, 3, or 5 h. Detailed heating profiles are provided in Figure 1(b).
The glass samples used for the experiments (electrical conductivity, density, and molar volume) were rectangular parallelepipeds and their surfaces were polished using SiC paper (#220–1000).
XRD measurements were taken in a Rigaku Co. RINT 2000 using Cu-Kα radiation in the 10–80° range with a step size of 0.02°. The existence of the nanocrystal phase was confirmed using the resultant XRD patterns. The nanocrystal phases were observed by field emission scanning electron microscopy (FE-SEM, Carl Zeiss, Supra 25).
We investigated the structural changes using FTIR spectroscopy. Powdered samples mixed with KBr (mixing ratio 1 : 1200) were prepared and dried at 100°C in a desiccator to form pellets. A Spectrum GX (Perkin Elmer Co, USA) spectrometer was used to perform FTIR. Room temperature IR spectra were recorded from 400 to 1400 cm−1 (injection number of times: 20; resolving power: 2 cm−1).
XPS was used to analyze the change in the valency of vanadium cations in the glass. XPS measurements were taken using the ESCALAB250 XPS system and the Theta Probe XPS system using monochromatic Al-Kα ( eV) radiation. The analysis area was 400 μm and the data was compensated using C1s (284.6 eV) as the reference.
The catalytic properties and electrical conductivity are the main characteristics considered in this study. First, in order to measure the electrical conductivity of each specimen, the carrier density was measured at room temperature using a Hall-effect measurement system (HMS-3000 Hall Measurement System, ECOPIA). Furthermore, to examine the catalytic effect of the glass samples applied to the oxidation of fatty acids (linoleic acid and stearic acid) and carbon, we performed TGA on two samples obtained by mixing 20 mg of fatty acid with 20 mg of glass powder and gas chromatography with a mass spectrometer (GC-MSD) on a sample obtained by mixing 2 mg of carbon with 8 mg of glass powder. TGA measurements were obtained between room temperature and 600°C at a heating rate of 10°C/min in an atmosphere of 5 vol% using a simultaneous TGA/DSC (Q600, TA Instruments, Inc.). The fatty acids used in the experiments are reagent-grade stearic acid (95%) and technical grade linoleic acid (60–74%, GC), both from Sigma-Aldrich Co. GC-MSD measurements were obtained between room temperature and 450°C at a heating rate of 10°C·min−1 in an atmosphere of He and O2 using a 5975MSD-6890N GC (Agilent). He gas was used at a flow rate of 20 mL/min and O2 gas was used at a flow rate of 2 mL/min.
Changes in quantitative properties due to structural changes were confirmed using the molar volume and density measurements. The densities of the glass samples were measured using the Archimedes method using a GH-200 (AND Co, Korea), and the molar volume () was calculated from the measured density using , where is the average molar weight of glass.
3. Results and Discussion
3.1. Analysis of Nanocrystal Phases
Figure 2 shows the diffraction patterns of all the samples studied, which were annealed for 4 h and heat-treated at 275°C for different durations (1, 3, and 5 h). A V2O5 phase (V5+/JCPDS-PDF 85-0601) coexisting with the VO2 (V4+/JCPDS-PDF 73-0514) and B2O3 (BO4/JCPDS-PDF 760781) phases was detected in samples heat-treated at 275°C. As the holding time at 275°C increased, the relative amounts of VO2 and B2O3 increased while the relative amount of V2O5 decreased; that is, VO2 and BO4 structures are strengthened by nanocrystallizing the 75V2O5-15B2O3-10P2O5 ternary glass. Combining this fact with the results of the FTIR analysis (Section 3.3), which explains the structural change of BO3 to BO4, the nanocrystallization of the VO2 phase occurred and this nanocrystal was strengthened while the V2O5 phase was comparatively weakened as the heat treatment at 275°C progressed. This suggested that a structural change, indicated by the reaction BO3 + VO5(V2O5) BO4 + VO4(VO2), was induced by the heat treatment at 275°C. This structural change is a factor that affects the electrical conduction and catalytic properties of the vanadate glass.
Figure 3 shows the SEM images of all samples studied, which were annealed for 4 h and heat-treated at 275°C for different durations (1, 3, and 5 h). Nanocrystallines of VO2 phase were detected in samples heat-treated at 275°C. As the holding time at 275°C increased, the amounts of nanocrystalline increased.
3.2. XPS Spectroscopy
To quantitatively determine the change in the vanadium ion valency, we performed an XPS analysis on each glass sample. The V 2p spin orbit doublet spectra of 75V2O5-15B2O3-10P2O5 glasses are shown in Figure 4. The XPS analysis was performed while maintaining the same X-ray energy, X-ray-to-sample positioning, and sampling time for all the four samples in accordance with the nanocrystallization conditions. V 2p peaks appeared asymmetrically in each sample, indicating that the vanadium ions are present in more than one oxidation state . According to previous studies on vanadium phosphate glasses, core level spectra of V5+ appear at a binding energy (BE) of 517.3 eV, while that of V4+ appears at 516.0 eV [11–14]. Therefore, when the V 2p3/2 core level spectrum is fitted to the BE of 517.3 eV of the V5+ ions and 516.1 eV of the V4+ ions, quantitative analysis for the concentration of V5+ and V4+ became possible . The V5+ and V4+ peaks of the 75V2O5-15B2O3-10P2O5 glass, obtained from the vanadium Gaussian-Lorentzian peaks, fitted using the least squares method, are shown in Figure 5. The relative content ratios of and obtained using this fit are shown in Table 1.
It was found that the content of the V5+ in glass was reduced and V4+ was increased by nanocrystallizing 75V2O5-15B2O3-10P2O5 glass and increasing the heat treatment time for crystallization. This means that the V5+ to V4+ reduction reaction occurs during nanocrystallization at 275°C and increments in the heat treatment duration. This behavior is in agreement with the trends in the catalytic properties (Section 3.5) and electrical conductivity (Section 3.4). We have confirmed that the changes in concentration and state of vanadium ions due to nanocrystallization contribute directly to the characteristic change in the 75V2O5-15B2O3-10P2O5 glass.
3.3. FTIR Spectroscopy
XPS analysis was performed to verify the changes in the concentration and state of the vanadium ions in the glass due to nanocrystallization. The structural changes were expected to induce a change in the concentrations of V5+ and V4+. Results of FTIR spectroscopy provided a direct evidence for the BO4 restructuring observed in the XRD analysis. The infrared spectra (at 400–2000 cm−1) of the 75V2O5-15B2O3-10P2O5 glass annealed for 4 h and heat-treated at 275°C for different durations (1, 3, and 5 h) are shown in Figure 6. The vibration modes of the borate network were found to be active mainly in three infrared regions, as previously reported [16, 17]. The group of bands located at 1200–1600 cm−1 is attributed to the B-O bonds in the trigonal BO3 units. The second group of bands, located at 800–1200 cm−1, could be ascribed to the B-O bond stretching in the tetrahedral BO4 units. A third group of absorption bands was observed at around 600 cm−1, which can be attributed to the bending of O-B-O linkages in the borate network [18, 19].
The absence of a peak at 809 cm−1 indicates that no boroxol ring was formed , which suggests that the glass system under investigation consists of randomly connected BO3 and BO4 groups. The absorption peaks at 1195 and 1465 cm−1 are related to the fundamental asymmetrical stretching vibration of the B-O bond in the trigonal BO3 units [21, 22]. Mitigation of the slope of the absorption band indicated that B-O bond stretching in the trigonal BO3 units occurred during heat treatment at 275°C. In addition, the absorption peak intensity at 1468 cm−1 (BO3 structure) shifted to lower wave numbers (1468–1430 cm−1) as the heat treatment was significantly reduced. The stretching vibration of the B-O bond in the tetrahedral BO4 units appeared at 1050–900 cm−1. The absorption peak at 1019 cm−1, attributed to the B-O bond stretching vibration in the tetrahedral BO4 units, shifted to higher wave numbers (1019–1020 cm−1) upon heat treatment at 275°C [17, 23].
From the FTIR analysis, shift of the vibrational band indicated that nanocrystallization at 275°C with increasing durations of heat treatment strengthened the BO4 unit structure and weakened the BO3 unit structure. This fact, coupled with the XRD analysis results, allowed us to examine the BO3 + VO5(V2O5) BO4 + VO4(VO2) structural change that occurred in the glass. This structural change affects the electrical conduction and catalytic properties of the vanadate glass.
3.4. Electrical Conductivity
The changes in the electrical conductivity of the heat-treated specimens are shown in Table 2. Electrical conductivity was increased by heat-treating the samples at 275°C and increasing the duration of the heat treatment.
The resulting conductivity value of 1.42003 × 10−7 Ω−1 obtained by heat treatment at 275°C for 1 h was approximately 5.1 times greater than the conductivity of the annealed glass. Furthermore, the resulting value of 7.51445 × 10−7 Ω−1 obtained by heat treatment at 275°C for 5 h was approximately 26.9 times greater than the conductivity of annealed glass.
Combining result of the electrical conductivity with those of the XRD and XPS analyses, it was hypothesized that the change in the electrical properties was because of the changes in the VO5 and VO4 structures caused by the change in the coordination number, which was due to the nanocrystallization. Electrical conductivity of this glass system is related to the metal ion value ratio C, which for vanadium-containing oxide glasses is defined as . According to Morinaga and Fujino , the electrical conductivity of vanadate glasses increases as C approaches 0.5. Heat-treating the 75V2O5-15B2O3-10P2O5 glass at 275°C nanocrystallizes and strengthens the VO2 phase while reducing the V2O5 phase. This indicates that the V5+ ions changed to V4+ as values of C approached 0.5. From these inferences, it can be predicted that this behavior leads to an increase in the electrical conductivity and that increasing the conduction path length, number of electronic transfer ports, and the VO2 nanocrystalline phase contribute to this increased conductivity.
3.5. Catalytic Property
TGA results of two samples containing 20 mg of glass powder nanocrystallized at each heat treatment condition mixed with 20 mg of stearic acid or linoleic acid are shown in Figure 7 and Table 3. Linoleic acid (CH3(CH2)4CH=CHCH2CH=CH(CH2)7CO2H) is a typical unsaturated fatty acid with two double bonds, and stearic acid (CH3(CH2)16COOH) is a typical saturated higher fatty acid with a carbon number of 18. Weight loss occurs when each fatty acid was partially converted to CO2 gas by an oxidation reaction.
The finish temperature of weight loss, which is the temperature at which the oxidation reaction of linoleic acid and stearic acid is complete, associated with the conversion of the acid to CO2 gas caused by crystallization due to 1 h heat treatment at 275°C, shifted to lower temperatures (356.05 to 324.15°C for linoleic acid and 378.67 to 341.75°C for stearic acid). Subsequently, the finish temperature of weight loss increases gradually by increasing the heat treatment duration. The finish temperature of samples heat-treated for 5 h was 340.96°C and 362.16°C for linoleic and stearic acids, respectively.
GC-MSD results of a sample containing 2 mg glass powder nanocrystallized at each heat treatment condition mixed with 8 mg of carbon are shown in Figure 8. Carbon converted into CO2 gas by an oxidation reaction, and hence, the desorption intensity of CO2 gas increases.
The sudden increase point of the desorption intensity, which is the temperature at which the oxidation reaction of carbon is complete, associated with the conversion of the carbon to CO2 gas because of nanocrystallization after 1 h heat treatment at 275°C, shifted from 395.13°C to 281.25°C (Table 4). Thereafter, the sudden increase point of the desorption intensity gradually decreased by increasing the heat treatment duration. The sudden increase point after 5 h of heat treatment was 387.98°C.
Prior research results have shown that the V5+ to V4+ reduction reaction causes a catalytic effect in the oxidation reaction of reactants (linoleic and stearic acids and carbon), and the XRD, XPS, and FTIR analyses of the structural changes due to the glass nanocrystallization lead to the following conclusion. We hypothesize that, by nanocrystallizing a given composition of glass at 275°C, nanocrystal phases are produced and the glass structure is changed to a relatively more closed structure (see Section 3.6). It was determined that the tunneling effect, which is the electron transfer mechanism in the vanadate glass structure, becomes more active. However, as the heat treatment time became progressively longer, the catalytic effect diminished in strength. By increasing the heat treatment time, the relative amount of the VO2 phase increased while that of VO5 decreased, changing the vanadium ion valency from V5+ to V4+. Thus, the V5+ activity and the reduction reaction of V4+ decrease gradually, weakening the catalytic property. Therefore, glass nanocrystallization due to heat treatment creates a denser structure than annealed glass. This causes the V5+ to V4+ reduction reaction and catalytic action of the oxidation of fatty acid to become active. On the other hand, as the heat treatment time increased, the reduction reaction decreased, indicating that the catalytic effects declined gradually.
3.6. Other Properties
We examined the density and molar volume to obtain indirect evidence of the structural changes and the associated catalytic properties described above. Looking at the changes in density and molar volume of 75V2O5-15B2O3-10P2O5 glasses shown in Figure 9, we observe that the density greatly increased and the molar volume greatly decreased after heat treatment at 275°C. On the other hand, when the heat treatment duration was increased, the density slightly decreased and the molar volume slightly increased.
The results from Figure 9 provide indirect evidence that describes the changes in the catalytic properties. When the glass is nanocrystallized by heat treatment, its structural density is much higher compared to the annealed glass, which leads to the enhanced catalytic activity. Increasing the heat treatment duration then leads to weaker catalytic activity because the structure gradually becomes less dense.
The purpose of this study was to examine and verify the correlation between the changes in the catalytic and electrical properties and the structure due to the nanocrystallization of 75V2O5-15B2O3-10P2O5 glass containing a mix of V5+, V4+, BO4, and BO3. Using various equipment of structural analysis, we conformed the BO3 + VO5(V2O5) BO4 + VO4(VO2) change in structure by nanocrystallizing and increasing the duration of heat treatment.
The changes in the catalytic properties and electrical conductivity were due to the nanocrystal structure and the state of the vanadium ions. The heat treatment used for nanocrystallization caused the electrical conductivity to increase continually; this is because the ion ratio, C value, approaches 0.5 owing to the transition of V5+ to V4+. On the other hand, after being improved for the purpose of nanocrystallization, the catalytic effect decreased as the duration of heat treatment increased. This is because, during nanocrystallization, the glass structure is dense and the tunneling effect occurs more actively. This behavior accelerates the reaction responsible for the catalytic activity. However, by increasing the duration of heat treatment, the vanadium ion valency changes from V5+ to V4+, and the activity of V5+ decreases; that is, the reduction reaction site of V4+ decreases gradually, causing the catalytic property to weaken. In addition, we indirectly and quantitatively confirmed the structural change through the increase in density and the decrease in molar volume due to nanocrystallization and the decrease in density and the increase in molar volume due to increasing the duration of heat treatment.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Y. Kawamoto, M. Fukuzuka, Y. Ohta, and M. Imai, “Electronic conduction and glass structure in V2O5-BaO-K2O-ZnO glasses,” Physics and Chemistry of Glasses, vol. 20, p. 54, 1979.View at: Google Scholar
A. M. Nassar, P. A. Moustaffa, and M. A. Salem, “Electrical properties of some vanadate semiconducting glasses,” Indian Journal of Pure and Applied Physics, vol. 20, no. 5, pp. 337–340, 1982.View at: Google Scholar
A. Mekkia, G. D. Khattaka, and L. E. Wengerb, “XPS and magnetic studies of vanadium tellurite glasses,” Journal of Electron Spectroscopy and Related Phenomena, vol. 175, no. 1–3, pp. 21–26, 2009.View at: Google Scholar
Y. D. Yiannopoulos, G. D. Chryssikos, and E. I. Kamitsos, “Structure and properties of alkaline earth borate glasses,” Physics and Chemistry of Glasses, vol. 42, no. 3, pp. 164–172, 2001.View at: Google Scholar
E. E. Horopanitis, G. Perentzis, A. Beck, L. Guczi, G. Peto, and L. Papadimitriou, “Correlation between structural and electrical properties of heavily lithiated boron oxide solid electrolytes,” Journal of Non-Crystalline Solids, vol. 354, no. 2–9, pp. 374–379, 2008.View at: Publisher Site | Google Scholar