Single step synthesis of WO3 nanoparticles by wire explosion process and its photocatalytic behaviour

Tungsten (W) wires are exploded in oxygen ambience to get tungsten oxide (WO3) nanoparticles (NPs). Energy stored in the capacitors (EC) is used to overcome the sublimation energy of wire. Energy ratio (K, ratio of EC and sublimation energy) and oxygen pressure (P) are two control parameters for the particle phase and morphology in the wire explosion process. X-ray diffraction (XRD) patterns confirmed the partial oxidation of W for low values of K. For K = 2, oxidation increases with increase in P. For K = 10, complete oxidation was achieved irrespective of P. Particles are spherical in shape as observed from scanning electron microscope (SEM) and transmission electron microscope (TEM) micrographs. Particle size follows a log-normal distribution with a least mean size of 24.1 nm. UV-vis diffuse reflectance spectroscopy (DRS) was used to measure the absorbance of NPs (complete WO3 with least mean size) for band gap measurement. The band gap was found to be 2.92 eV (visible region). NPs are used as photocatalyst to degrade aqueous solution of methylene blue (MB) under visible light irradiation. 500 mg l−1 of WO3 NPs were optimum to degrade 10 mg l−1 MB in 120 min.


Introduction
Functional materials and its nanostructures attract researchers due to its unique characteristics. Nano dimensions provide a wide control over the properties of a material. Nanoparticles (NPs; size range from 1 to 100 nm) are extensively studied for their unique traits for a particular application. Metal oxide NPs are an important constituent of many ceramic structures and smart devices. Nanostructure of tungsten oxide (WO 3 ), a transition metal oxide has versatile characteristics, a reason for being investigated in the fields of condensed matter physics to solid-state chemistry [1]. It finds its application in numerous fields like development of gas sensors [2], electrochromic devices [3], dye-sensitized solar cells [4], for field-emission applications [5], hightemperature superconductors [6] and as photocatalysts [7].
As a phtocatalyst, WO 3 is widely applied in the areas of air purification [8], CO 2 photoreduction [9], treatment of heavy metals [10], hydrogen evolution from splitting water [11], photodegradation of water pollutant [12], etc. Degradation of water pollutant like dye needs a lot of attention as the wastewater discharge from textile industries pollutes the surface water and in long run ground water too. Dye degradation using photocatalysis is widely researched using a number of metal oxide nanoparticles [13]. Starting from titanium dioxide (TiO 2 ), a classical photocatalyst for dye degradation [14] to zinc oxide (ZnO) NPs need ultra-violet (UV) light irradiation to overcome its band gap barrier [15]. But, sunlight consists of only 7% of the UV light. Visible light constituents for 40% of the sunlight we get on earth [16]. So, it is necessary to utilize NPs which can be used in the visible light.
WO 3 provides a viable solution as its band gap lies in the visible region [1]. Many researchers showed the applicability of WO 3 NPs for photodegaration of water pollutant. A number of methods were employed for synthesis of WO 3 NPs for this purpose, e.g. hydrothermal method [17], precipitation method [18], reverse microemulsion process [19], by annealing hexagonal ammonium tungsten bronze [20], flame aerosol route [21], ultrasound synthesis method assisted with CTAB [22], sol-gel method [23], spray-pyrolysis method with colloidal templating [24], arc discharge [25] etc. These synthesizing methods require many steps and/or various precursors. Wire explosion process (WEP) provides a single step method to produce NPs with least number of precursors.
WEP is also termed as pulsed wire discharge (PWD) or electric wire discharge (EWD), which needs wire and the reacting/cooling fluid as the starting materials [26][27][28][29][30][31]. In present work, tungsten (W) wires were used in oxygen ambience. High current was passed through the wire causing joule heating which leads to vaporisation of the wire. W vapour/plasma reacts with oxygen gas to form WO 3 NPs after simultaneous fast quenching. NPs synthesized by WEP were applied in a number of fields including photocatalysis under UV-light irradiation using WEP synthesized ZnO [30] and TiO 2 [31]. There is scarce report on usability of WEP synthesized NPs in visible light range. Previously, Arvinth et al attempted to synthesize WO 3 NPs by WEP but could not get complete oxidation of W vapour as there were traces of W were found in XRD patterns of their work [32].
Hence, in the present work, pure WO 3 NPs are synthesized without any traces of W. Size and oxidation control are discussed with variation in energy and oxygen pressure. Band gap of produced NPs is measured for its usability in visible light. Photoactivity of WEP synthesized NPs has been investigated under visible light irradiation taking methylene blue (MB) as a representative pollutant in water. Kinetics and mechanism of photodegradation of MB will be discussed.  2. Experimental studies 2.1. Synthesis of WO 3 NPs by wire explosion process Experimental setup used for the synthesis of NPs in this work is shown in figure 1. Wire of required dimension was fixed between the electrodes. The chamber was closed and vacuumed. Then, oxygen was purged in the exploding chamber till the required pressure (monitored by pressure gauge). Supply voltage of 230 V, 50 Hz AC was connected to a step-up transformer (230 V/25 kV, 25 kVA) through an autotransformer (230V/0-270V) to get the required voltage. The capacitor bank of total capacitance (C, 3 μF) was charged by the rectified voltage (V, diode was used for rectification). Energy (E C = 0.5CV 2 ) stored in C was discharged to the wire after triggering a trigatron gap. The energy injection/deposition to the wire leads to heating of the wire with subsequent explosion. Energy left after wire heating was deposited in arc discharge leading to the expansion of vapour/ plasma cloud. The vapour reacts and gets cool down simultaneously to yield NPs. Voltage and current during the explosion were measured using a voltage probe (Tektronix P6015A) and a current transformer (Pearson 101) respectively. Figure 2 shows the current, voltage and energy (integral sum of instantaneous power) deposited to the wire during explosion. Table 1 shows the experimental parameters used for the synthesis of NPs in present work. Based on the circuit parameters, it matches with the condition of an under damped RLC circuit [33] as per equation 1, The expression for the magnitude of current flow through the R, can be written as, The minimum energy per gram to vaporize the wire (sublimation energy), E S can be written as,  where, C S and C f are specific heat in the solid and liquid state respectively, T r , T m and T b are room, melting and boiling temperature of the metal respectively, and h m and h v are latent heat of fusion and vaporization of the material respectively. The data for the calculation were taken from CRC databook [34]. The energy ratio, K defined as the ratio of E C and E S is used in this work as the parameter to describe the effect of energy deposited to the wire. K and P are two parameters to control the morphology and/or phase of NPs produced by WEP.

Photoactivity test of NPs
Test pollutant, methylene blue (MB; Merck life sciences private limited, India) was dissolved in distilled water to make a stock solution of 1 g l −1 . For photodegradation, 10 mg l −1 MB solution was prepared from the stock solution. Figure 3 shows the photoreactor (30 cm height and volume of 130 ml, Heber scientific, India) used in this work. It is an immersion type reactor equipped with an outer cooling jacket (cools the reaction solution of MB and NPs by circulation of water with help of a pump), an oxygen port (flow was controlled by rotameter) and a port for taking out the sample. The cooling (water jacket) keeps the reaction solution at room temperature. A visible lamp (230 V, 150 W) was used for the irradiation of solution. Magnetic stirring keeps the nanoparticles in solution and prevents it from settling down.

Analytical methods
The phase of WEP-synthesized NPs was determined using an X-ray diffraction (XRD) technique. Data was obtained from Rigaku Smartlab X-Ray diffractometer with nickel-filtered Cu Kα radiation (λ=0.154 nm) at 40 kV and 100 mA, and Rietveld refinement was carried out using X'pert High Score software. Morphology was observed using a scanning electron microscope (SEM, FEI Quanta FEG 200) and a transmission electron microscope (TEM, Tecnai T20). TEM samples were prepared by dropping two droplets of mixture of a pinch of NPs and acetone on carbon coated copper grid. Particle sizes were determined using ImageJ software. The band gap was measured using absorbance spectrum given by UV-vis diffused reflectance spectroscopy (DRS, JASCO V-650 spectrophotometer ISV-722) with BaSO 4 as reference. 10 mg l −1 MB solution in distilled water was used for photoactivity test. Synthesized NPs of different quantities were mixed in 100 ml of MB solution and stirred for 30 min to reach the adsorption-desorption equilibrium. 50 ml minute −1 of oxygen flow was maintained in the reactor solution with continuous stirring and visible light irradiation. 3 ml of solution was collected through port of setup for analysis at regular intervals. Absorbance of solutions were measured using UV-vis spectrophotometer (Shimadzu, Japan). Ratio of C and C 0 was defined as the degradation ratio, D as follows, Where, C 0 is the absorbance value of untreated methylene blue solution and C is the absorbance value of treated solution for different interval of time. Absorbance was measured at 663 nm. Figure 4 shows the XRD patterns of NPs produced by WEP after exploding tungsten (W) wire at different K and in different oxygen pressure, P. The patterns match with standard pattern of WO 3 (JCPDS card no.− 20-1324) and for K other than 10, extra peaks of W (JCPDS card no.−47-1319, 100% peak at 40°) was detected. For K=10, complete oxidation of W vapour/plasma was achieved irrespective of the oxygen pressure used in the present work. For K=2, the intensity of peak corresponding to W decreases with increase in P, attributed due to the more oxidation of metal vapour in presence of relatively more molecular oxygen. For P=180 kPa, as we increase K from 2 to 6, decrease in intensity of peak corresponding to W indicates an increase in oxidation. This is due to decrease in density of metal vapour cloud produced after explosion for high energy deposition i.e. high K value [35]. For K =10, complete oxidation of W vapour was achieved irrespective of the value of P. The spread of vapour cloud was enough for the oxidation to WO 3 . In short, it is possible to get complete oxidation of W vapour in WEP for higher energy deposition leading to WO 3 nanoparticle formation without any traces of W metal.

SEM, TEM and particle size distribution studies
The morphology of NPs produced by WEP was checked through SEM and TEM micrographs. Figure 5 shows SEM images of NPs synthesized at different K and P values. Particles are spherical in shape except few. For low K value (K=2), bigger size nanoparticles were there with some having submicron size. This is due to the formation of NPs from molten metal instead of from vapour at low energy deposition to the wire [27]. Figure 6 shows the TEM images of NPs. Particles size (Heywood diameter, diameter of circle with equivalent area to the area of 2-dimensional projection of 3-dimensional particles) varies from few to 100 nm for all the cases presented in figure 6. Particle size of about 500 NPs were measured and it fits into Log-normal distribution [36] as per equations 5-7, where f(d) represents the log-normal distribution, d and D 1 are the particle and geometric mean diameter respectively, n i and d i are the number of particles and its diameter respectively, and σ g is the geometrical standard deviation. Figure 7 shows the log-normal distribution of NPs synthesized by WEP. The corresponding mean size and standard deviation is shown in table 2. Mean size of NPs reduces with increase in K for P=180 kPa. This is due to the fact that, more energy deposition leads to more expansion of metal vapour (keeping other parameters constant) due to increase in available energy in arc discharge after the explosion [27]. For K=10, size of NPs decreases with decrease in P due to less dense vapour cloud formation after explosion for low value of K. In short, small size oxide NPs is formed for high energy deposition and/or lower P.   Figure 8 (inset) shows the absorbance of WO 3 NPs produced by WEP at K=10 and P=20 kPa. The spectrum has a decrease in absorbance after 300 nm. Band gap of NPs was determined using equation 8 [37,38] as follows,

Band gap measurement by UV-vis DRS study
where, α is the absorption coefficient, h is Planck's constant, hν is photon energy, and m = 0.5 for semiconductors with direct allowed transition. α is replaced by absorbance 'a' as it is directly proportional to 'a'. Energy of photons are defined by hc/λ; where c and λ are the speed of light and the wavelength respectively. Figure 8 shows the variation of (ahν) 2 as a function of energy (eV). Band gap of NPs is the value of energy on X-axis at which the extended linear portion of the curve crosses. The band gap of NPs produced at K=10 and P = 20 kPa was found to be 2.92 eV which lies in visible region. It is comparable to the band gap of 2.88 eV for WO 3 NPs measured by Azbari et al [19]. In short, WO 3 NPs produced by WEP can be harnessed for photocatalysis in visible region.

Photocatalytic behaviour of WEP synthesized WO 3 nanoparticles
NPs produced at K=10 and P=20 kPa gives complete oxidised yield with least mean size of 24.1 nm (table 2) and a band gap of 2.92 eV (section 3.3). For photoactivity tests, NPs synthesized at these conditions were utilized in this work. Tests to demonstrate the effect of NPs size on the photoactivity was not carried out as the completely oxidised NPs has close size range (24-30 nm) as depicted from table 2. The influence of size on the photoactivity of NPs will be carried out in the next work. Figure 9 As we increase the concentration of NPs from 200 mg l −1 in steps of 100 mg l −1 till 500 mg l −1 , C/C 0 has decreased with increasing concentration indicating more degradation. This is because of the availability of more number of reaction sites to generate electron-hole pairs with increase in number (concentration) of NPs [30,31]. But, if the concentration of NPs was increased further to 600 mg l −1 , the C/C 0 value increases indicating a decrease in degradation. This is attributed to the fact that at higher concentration of NPs in solution, (i) NPs will tend to agglomerate leading to decrease in active sites, (ii) high turbidity of solution leads to inhibition and scattering of light which prevents all photon energy to reach NPs to produce electron-hole pairs for degradation reactions [15]. For 500 mg l −1 of WO 3 , we achieved 86% degradation of MB solution (10 mg l −1 ) for 150 min of treatment time. Martinez et al [18] observed about 90% degradation of rhodamine B dye (5 mg l −1 ) in 150 min which is comparable as present work used 10 mg l −1 dye solution.

Kinetics and mechanism of photodegradation
Kinetics of photodegradation was studied using Langmuir-Hinshelwood model. As the concentration of MB is low (10 mg l −1 ), we approximated the degradation kinetics using pseudo first order equation [37] given as where k psuedo is the pseudo-first-order rate constant (minute −1 ) which is equal to the slope of the line indicating the variation of ln(C 0 /C) as a function of time, t. Figure 10 shows the kinetics curve of MB degradation using WO 3 NPs as catalyst with a concentration of 500 mg l −1 . Linear fit for data has a slope of 0.0114 per minute. The degradation of dye/pollutant in water comprises of many steps. The reactions involved in photodegradation is given as follows [18,39]:   where, equation 10 depicts the generation of holes (h vb + ) and electrons (e cb − ) from valance and conduction band of WO 3 NPs respectively. A semiconductor yields electron-hole pair if it is irradiated with photon energy more than its band gap energy. The reaction solution contains dissolved oxygen which reacts with electrons to give superoxide radical ( -•O 2 ) as shown in equation 11. Water molecules react with holes to yield hydroxyl radical (•OH) and hydrogen ion (equation 12). More hydroxyl ions are generated through reactions represented by equations 13-17 with formation of hydrogen peroxide (H 2 O 2 ) and its radical. Superoxide and hydroxyl radicals are oxidising in nature which lead to degradation of pollutant to carbon dioxide (CO 2 ) and small degraded compounds as shown in equation 18.

Conclusions
Completely oxidized WO 3 NPs were successfully synthesized using wire explosion process. High energy and high pressure aids in oxidation of W vapour/plasma. Size of NPs decrease with increase in K and/or decrease in P. Particle size data follows log-normal distribution. Least mean size of NPs was found to be 24.1 nm. Band gap of NPs was found to be 2.92 eV by DRS measurement. Optimum quantity of NPs was found to be 500 mg/l for the degradation of 10 mg l −1 MB solution under visible light irradiation.

Data availability statement
The data that support the findings of this study are available upon reasonable request from the authors.