Solvothermal synthesis and characterization of monoclinic WO3 nanoplatelets: investigation of their photocatalytic performance

A facile and original approach based on a one-pot non-aqueous sol–gel solvothermal process was developed to synthesize 2D Tungsten trioxide (WO3) nanoplatelets with an average size ranging from 30 to 50 nm, and a correspondingly high surface area. The structural, morphological, functional group, optical properties of the materials, and the properties of the adsorption surfaces were all investigated; the degree of surface hydroxyls (–OH groups) has been examined. Nuclear Magnetic Resonance Spectroscopy techniques indicated the formation of di-hexyl ether as a result of the solvothermal reaction. The optical absorption, measured using UV–Vis Diffuse Reflectance Spectroscopy, revealed a narrow bandgap (Eg = 2.18 and 2.48 eV for WO3-24 and WO3-48, respectively) compared to that of for bulk WO3 (2.7 eV), attributable to oxygen vacancies. The as-prepared WO3 nanoplatelets displayed excellent photocatalytic performance for degrading Rhodamine B under visible light-emitting diode light with up 99% degradation rate that was achieved in 120 min. Thus, the enhanced Rhodamine B photodegradation in the presence of WO3-24 along with H2O2 was assigned to the reactive oxygen species such as ·OH and RhB*+, involving in the strong synergistic effect between WO3 and H2O2, effectively separating of photocarriers and, as a consequence, boosting the photocatalytic efficiency.


Introduction
The quest for cost-effective, sustainable, and green energy sources to meet global energy needs is receiving much attention due to environmental pollution and the present energy crisis. As green chemistry process, semiconductor-based photocatalysis is an advanced oxidation technology operating at ambient temperature, employing solar energy to degrade organic contaminants completely. In general, titanium dioxide (TiO 2 ) is the most widely utilized photocatalyst because it is chemically stable, abundant, and has a highly oxidizing capacity for photoinduced holes [1,2]. On the other hand, Titanium dioxide has a bandgap of 3.0-3.2 eV and can only absorb light from the ultraviolet (UV) area; not only does this constitute a health hazard, but it also limits the use of solar energy to its full potential [3]. As a result, it is critical to shift to efficient semiconductors in photocatalysis employing solar energy, particularly in visible light. In the last decade, much attention has been focused on tungsten trioxide (WO 3 ) as an n-type semiconductor with a small, broadly tunable bandgap about Eg = 2.5-2.8 eV at room temperature capable of absorbing light in the visible region of the solar spectrum [4][5][6]. Moreover, due to its outstanding electrical, optical, and structural properties, and its excellent photocorrosion resistance in acid solutions [7]. It can oxidize water by photochemistry and photoelectrochemistry and degrade organic dyes [8]. In addition, recent work has shown that for the same catalyst composition, the use of nanoscale particles can improve photocatalytic reactions rates [9]. The increase in velocity can be attributed to the increase in the specific surface area of the material and the decrease in the electron-hole transport distance [8,10]. Consequently, much effort has been devoted to synthesizing tungsten oxide nanoparticles. Numerous approaches can be used to create various WO 3 morphologies, like Sol-Gel Synthesis [11] Electrochemical deposition [12], hydrothermal [13] and solvothermal processes [14]. In terms of tungsten oxide nanoparticles, we can say that different morphologies have been obtained, such as nanotubes [15][16][17], nano-sheet [18] nanoplatelets [8,19], quantum nanodot [5,20] and nano sphere [21]. But among various nanostructures, two-dimensional inorganic nanomaterials have attracted considerable attention because they can open up new possibilities in the field of applied surface materials. The 2D sheet structure generally has high surface energy, provides active sites for catalysts and, due to their large exposed surface area, has superior catalytic properties compared to their 1D and 3D counterparts in several uses. [22,23]. However, the synthesis strategy for a target material usually combines multiple techniques, requiring complex operating procedures and different surfactants for improved morphological control. [24], Furthermore, the utilization of tungsten precursors is frequently hazardous, expensive and requires a complicated synthesis procedure [8], As a result, a novel approach for producing tungsten trioxide nanoparticles with unique chemical and physical structural features that is reliable, easy, and economical is required. In the present study, we have synthesized 2D WO 3 NPs in the form of nanoplatelets 1 3 Solvothermal synthesis and characterization of monoclinic… by using two readily available reagents, including tungstic acid as the tungsten source and 1-hexanol as the reactif solvent, without the use of any other surfactant in a non-aqueous solvothermal route. The as-elaborated WO 3 NPs were characterized by X-Ray Powder Diffraction (XRPD), scanning electron microscope (SEM), High-resolution transmission electron microscopy (HR-TEM), Ultra-Violet-Visible-diffuse reflectance spectroscopy (UV-Vi-DRS), Fourier Transform Infrared Spectroscopy (FT-IR), Carbon-13 ( 13 C) nuclear magnetic resonance spectroscopy ( 13 C NMR) and Brunauer-Emmett-Teller techniques (BET) to investigate the structural, morphological, functional, optical and others spectroscopic properties of the synthesized WO 3 . The photocatalytic performance of the as-synthesized WO 3 catalyst was assessed by the white Vis-LED-light (15 Watt)-initiated degradation of Rhodamine B dye (RhB). The mechanism of photocatalytic degradation of RhB by WO 3 /H 2 O 2 system under white LEDs irradiation has been proposed.

Sample preparation
All the reagents were analytically pure and used without further purification-First, tungstic acid was produced by reacting sodium tungstate with hydrochloric acid; the resulting yellow powder was then washed several times with distilled water and ethanol before being oven-dried at 60 °C for 24 h. In a typical procedure, (0.2498 g) or 1 mmol of the prepared tungstic acid was added to 25 ml of n-Hexanol, and the mix was transferred into a stainless-steel autoclave and sealed. In one hand, temperature-dependent experiments were carried out to determine the optimal temperature for the solvothermal process. For this purpose, two temperatures were selected: 160 and 180 °C, while maintaining the reaction time at 24 h. On the other hand, time-dependent experiments were carried out to understand and ensure the effective WO 3 nanoparticles formation. Consequently, the solvothermal synthesis was performed at various times of 6, 12, 24, and 48 h at a fixed temperature of 180 °C, the remaining synthesis conditions being unchanged. The resulting suspensions were centrifuged at 9000 rpm, and the precipitates were washed with dichloromethane and ethanol and then, oven-dried at 60 °C for 24 h.

Sample characterization
X-ray powder diffraction (XRPD) was used to determine the mineralogical and microstructural properties of the samples. The patterns were collected at room temperature on a "PANalytical X'Pert Pro, NL" automatic X-ray diffractometer, equipped with a fast RTMS detector (PIXcel 1D, PANalytical), operating with Cu-Kα radiation generated by a power of 45 kV and 40 mA. The XRD data were registered in a range of 20°-80° (2θ) with a step size of 0.02° (2θ) and 1 s per step. Diffuse reflectance (DR) spectra of the synthesized WO 3 nanoparticles were collected with a resolution of 0.2 nm on a Shimadzu UV-3100 spectrometer with an integrating sphere and a white Spectralon reference material, in the UV-Vis spectral region of (250-750 nm). The Brunauer-Emmett-Teller (BET) method on (Micromeritics Gemini 2380, US) was used to measure the specific surface area (SSA) of the samples. The morphology was investigated by scanning electron microscopy (SEM-Hitachi SU 70) coupled with energy-dispersive X-ray spectroscopy (Bruker EDS), Transmission electron microscopy (TEM) was performed using 120 kV JEOL 1210 (material institute Barcelona, Spain), samples for TEM/ observations were prepared by dispersing the WO 3 nanoparticles in ethanol, Fourier Transform Infrared Spectroscopy FT-IR spectra were recorded on (FT-IR, Shimadzu Affinity-1S spectrophotometer), and measurements were carried out over the wavenumber range 4000-350 cm −1 . The organic species produced in situ the reaction and persisting in the supernatant liquid after centrifugation were investigated using 13 C NMR spectroscopies; the spectra were measured in CDCl 3 solutions using a 300 MHz Bruker Ascend NMR spectrometer.

Functional properties: photocatalytic experiments
The photocatalytic efficiencies of WO 3 nanoparticles in the liquid-solid phase were tested on the photocatalytic degradation of RhB. An aqueous dye solution (50 mL) with an initial concentration of 10 mg/L has been used in experiments, carried out at room temperature in a cylindrical photocatalytic reactor. Upon dissolution of the photocatalyst in the liquid, its concentration was 1 g/L. The mixture was magnetically stirred throughout the reaction to ensure a thoroughly mixed solution. Before turning on the light to begin the photocatalytic reaction, the suspension was stirred in the dark for 30 min, in order to evaluate the RhB adsorption/desorption capacity of the material. The reaction system was illuminated by mounting a 15-W led lamp on the top of the reactor; 0.1 mL of Hydrogen peroxide (33%) was added to the solution at the start of the photocatalytic test. At the end of the reaction, the catalyst particles were removed from the RhB solution by centrifugation at 9000 rpm.
The absorbance of RhB was measured at 554 nm using a Spectrum UV-200 s UV-VIS spectrophotometer.

XRPD and morphological analyses
The effect of the temperature of the solvothermal reaction on the WO 3 samples has been studied in carrying out experiments at two temperatures, of 160 °C, and 180 °C, during 24 h. Figure 1 shows typical XRPD patterns of WO 3 nanoparticles synthesized in such conditions. For WO 3 synthesized at 160 °C, the major diffraction peaks located at 2θ at 13.8°, 23 ). This result confirms the existence of an allotropic phase transition, from hexagonal to monoclinic WO 3 that occurred between 160 and 180 °C. Prior research has shown that the photocatalytic activity of monoclinic WO 3 is higher than that of hexagonal WO 3 [25], This led us to choose the synthesis temperature of 180 °C for the rest of our study, since it is suitable for obtaining the monoclinic form of WO 3 .
The XRPD patterns of WO 3 samples recorded after the solvothermal treatment at 180 °C for various times are shown in Fig. 2.
After 6 h of solvothermal treatment, the sample clearly displayed a crystalline structure that could be easily identified as the orthorhombic structure of the WO 3 .H 2 O phase (JCPDS-ICDD 84-0886, space group: Pnmb, a = 5.249 Å, b = 10.71 Å, c = 5.133 Å). However, two additional series of peaks were detected, the first being in the hexagonal WO 3 phase (very small proportions) characterized  by peaks at 2θ = 23.05°, 23.67°, 24.17° and 34.12°, respectively, and the second being in the monoclinic WO 3 phase (small proportions) characterized by peaks located at 2θ = 13.85°, 28°, and 36.59°, respectively. After 12 h of solvothermal treatment and beyond, the sample displayed a new crystalline structure that could be easily identified as the monoclinic structure of the WO 3 phase, with total disappearance of the orthorhombic phase. This confirms that an allotropic transition occurs from the orthorhombic WO 3 .H 2 O phase to the monoclinic WO 3 phase, beyond 6 h (i.e., 12, 24 and 48 h) of solvothermal reaction. As in the case of the sample obtained after 6 h of solvothermal reaction, the three samples obtained after 12, 24 and 48 h of reaction, respectively, also contain a small proportion of the hexagonal WO 3 phase, as evidenced by the presence of the four characteristic peaks located at 2θ = 23.05°, 23.67°, 24.17° and 34.12°, respectively. In addition, on the one hand the peaks of the hexagonal phase tend to fade with the increase in the reaction time, and on the other hand, after 48 h of solvothermal reaction, we can see a more orderly structure of monoclinic WO 3 -48, as evidenced by the improvement of the profiles of the characteristic peaks, reflecting a better crystallinity, and in particular, the (002) peak which regains its original intensity. The cell parameters of the WO 3 -6, WO 3 -12, WO 3 -24 and WO 3 -48 samples were refined using a least-squares fitting procedure with the "Unit cell" program [26], and the results are reported in Table 1.
At first sight, we see a marked decrease in the size of the unit cell in sample WO 3 -48 compared to the others. This can only be explained by the formation of oxygen vacancies, resulting in a non-stoichiometric oxygen deficiency of the WO 3-x type, as reported in the literature [27,28].
Particle size was calculated using the Debye-Scherrer formula [29]: D = K cos where: • D is the mean size of the crystalline domains of the particle (particles, supposed spherical) (nm). • K is the shape factor (dimensionless and close to unity), • λ is the X-ray wavelength (λ = 0.154178 nm), β is the line broadening at half the maximum intensity (FWHM) (°). • θ is the Bragg angle (°). For each sample, the mean particle diameter was calculated by selecting the peaks with profiles of the best possible quality, those who are least overlapped. These are as follows: The results are given in Table 2 and clearly show the influence of the solvothermal reaction time on the particle size of the samples. This decreased significantly from 278 Å (~ 28 nm) after 12 h of reaction to 218 Å (~ 22 nm) after 48 h of reaction, a decrease of approximately 22%.
Besides, this result shows unambiguously the positive effect of reaction time on particle size. It therefore seems clear that a longer reaction time would favor the formation of a oxygen-deficient monoclinic WO 3 phase, which would itself be very conducive to obtaining smaller particle sizes, as showed in a recent study where morphology and phase of a material can be improved by controlling the reaction time [30].
In general, a variety of variables, including antiphase domain borders, dislocations, and non-uniform lattice distortions, contribute to microstrain in nanocrystallites [31,32]. In this investigation, the microstrain (ε) was estimated by using the Williamson-Hall method [33,34] The data shown in Table 2 testify that increasing the synthesis time causes a greater micro-strain while also showing a reduction in the size of the nanoparticles.
In our case, the presence of organic ligands from the synthesis process decorating the growing NP surfaces detected by infrared spectroscopy may be correlated with the amplification of the microstrain observed. Indeed, with the increase in the synthesis time from 24 to 48 h, a greater amount of organic species is formed as evidenced by the increase in the peaks observed around 1600 cm −1 (Fig. 6), causing at the same time a decrease in the size of the particles because of the mechanical stress that may induce spontaneous compressive deformation of nanoparticles [35].
SEM pictures of the synthesized WO 3 samples are shown in Fig. 3 (a, b). The synthesis process produces NPs that are well crystallized, in the form of plate-like shaped particles, it can be seen that the synthesized sample tends to form agglomerated plate-like nanostructures. Additionally, it is clear that the W-48 sample depicted in Fig. 3-B appears to have a better crystallinity than W-24 sample. This is in perfect accordance with the results from the x-ray diffraction analysis. TEM images in Fig. 3.c and 3.d display that the WO 3 nanoplatelets are composed of quadrangular nanoplatelets with an average width of 35-50 nm. The majority of sheets stacking were principally formed by smaller particles. The SAED Fig. 3d in insert shows concentric circles, which are due to the polycrystallinity of the material.

Surface area analysis
Adsorption-desorption isotherms were used to determine the surface area and porosity of WO 3 nanoplatelets, as shown in Fig. 4 and shows nitrogen adsorption and pore-size distribution curve (inset) of the WO 3 -48 h nanoplatelets.
The isotherm exhibits a hysteresis loop in the range of 0.4-0.9 P/P 0 , The Barrett-Joyner-Halenda (BJH) method was used to determine the pore-size distribution of WO 3 nanoplatelets. The BET surface area of the WO 3 nanoplatelets was about 17.17 m 2 . g −1 . The porous structure of the materials is a significant component in addition to the SSA. Figure. 4. N2 sorption isotherms provide a description of this, which is a type-IV, typical of mesoporous materials. In addition, the trace shown in the insert of Fig. 4 has shown the occurrence of nanopores (pore size between 0.1 and 100 nm) [36], with a multimodal distribution [37]. Since the synthesized nanoparticles had identical and regular forms, the particle spacing may only have a little impact on the pore size distribution [38].

FTIR Spectroscopy
The FTIR spectroscopy is a powerful tool to identify the organics species on the surface of the materials. Figure 5 shows, a representative example, the FTIR spectra recorded for the both samples produced at two different dwell times of synthesis (WO 3 -24 and WO 3 -48). The band at 600 cm −1 is likened to the inter-bridge stretching O-W-O, and the band around 672 cm −1 is attributed to the stretching vibration mode W-O-W [39,40]. Meanwhile, the absorption band located at 941 cm −1 is assigned to (Metal = Oxygen) vibrations, which is characteristic for stretching vibrations of the W = O bond [39]. Increasing the processing temperature up to 48 h leads this shoulder to become gradually weak, which confirms the progressive destruction of tungsten-oxygen double bonds.
The band positioned around 1600 cm −1 is due to the H 2 O bending mode (δH 2 O) [41]. The broad band near 3400 cm −1 represents the stretching vibrations of water molecules [42], and the presence of water molecules in the WO 3 -24 sample could be attributed to the OH − stretching mode ν(O-H), probably from the environment atmosphere. Thus, increasing the solvothermal time processes resulting in a reduction in the observed band [28,43]. Figure 6 displays the FT-IR spectra for WO 3 nanoplatelets enlarged in the region 1800-1000 cm −1 . A series of sharp bands, in the regions 1458, 1506 and 1546 cm -1 , are attributable to the C = O stretching vibration [44]. The bands observed at 1546 and 1458 cm -1 corresponded to asymmetric (ν as ) and symmetric (ν s ) stretching frequencies of carboxylate groups, respectively [45]. The frequency separation of these bands (Δν ≈ 88 cm −1 ) suggests a bidentate chelating interaction between the carboxylate groups and the metal ions located at the surface of WO 3. The fact that those peaks exist can be attributed to the presence of diverse solvent remnants in very small proportions. These features are more apparent in the sample synthesized at 180 °C for 48 h, suggesting that the organic species remain strongly bound to the external surfaces of the WO 3 nanoplatelets.

Liquid phase NMR spectroscopy
In order to study the reaction of 1-hexanol with H 2 WO 4 during the synthesis, the supernatant liquid was filtered to remove the WO 3 solid precipitate and analyzed using liquid 13 C NMR spectroscopy. The 13 C NMR spectrum is shown in Fig. 7.
Following the analysis of the collected spectra, we considered the signals according to their chemical shifts, which are listed in the table below (Table 3). Nonetheless, a good correlation with the di-hexyl ether with the C1 carbon at 72.85-ppm structure has been detected, indicating that an amount of 1-hexanol is converted into dihexyl ether during the synthesis. All other peaks from these hydrocarbons corresponding to hexane carbons C2 − C6 appear in the same region at 14 − 33 ppm.
Among the very large number of solvothermal and hydrothermal processes for the production of WO 3 from tungstic acid, thermal decomposition was the most reported by several authors [17,46]. In which the presence of the solvent is not necessary for the formation of the oxide but merely serves to create a homogeneous distribution of the metal oxide precursor, despite the fact that the exact method by which the oxide is produced is not always chemically obvious. However, it is also possible that condensation processes are responsible for the growth of early clusters or possibly the production of the oxide itself. In our case, the formation of dihexyl ether highlighted by the 13 C NMR spectroscopy can suggest an eventual reaction between the two regents leading to an ether elimination reaction [47].  Optical proprieties Figure 8 illustrates the UV-vis-DRS spectra of WO 3 nanoplatelets synthesized at 180 °C for 24 and 48 h. Using the Kubelka-Munk transform, the Reflectance spectra were converted into a pseudo-absorption function [48]. It has been observed that the nanoscale WO 3 exhibits a strong absorption band in the UV region around ≈ 330 nm. The WO 3 -24 sample presented a redshift which could be ascribed to the creation of newly localized in-gap-state defect band within the bandgap through the reducing solvothermal time process by the oxygen vacancies [28,49]. The majority of metal oxides have sub-stoichiometric oxygen concentrations, and the resulting oxygen vacancy states (VO), with two electrons remaining in the lattice for each oxygen atom missing, are thought to be the cause of these materials' n-type doping. The widely reported photo-and electrochromic behavior of WO 3 , which is known for its high intrinsic doping density, has been linked to intrinsic (VO) defects and, more particularly, the reduced W 5+ states that surround these lattice sites. UV-visible spectroscopy was used to observe approximately the chemical doping density (VO) of each sample. Figure 8 assumes that each oxygen vacancy results in the generation of two W 5+ species.
Indeed, W5 + exhibits broad absorption in the visible range due to the capture of photo-excited electrons at the trapping sites in WO 3 [50]. The polaronic transition between W 5+ and W 6+ can also be used to explain the optical absorption of substoichiometric WO 3 . The oxygen vacancies' localized electrons polarize the WO 3 lattice to produce tiny polarons. The polarons move from one tungsten site (A) to another by absorbing light photons (hv) switching between tungsten sites (A) (B). These transitions between intervals can be stated as follows [28,51].
It is clear that the absorption at longer wavelengths varies significantly between samples, despite the band-edges of the samples being just slightly different. Since W 5+ centers connected to oxygen vacancies cause this absorption, a higher proportion of absorption corresponds to a higher (VO). The outstanding performance of the "high" (VO) sample in our case, WO 3 -24, shows that electron transport is more boosting than hole transport in this nanostructured sample and is especially high in the "high" (VO) sample, indicating of electron conductivity mobility as a result of high donor density [52]. The bandgaps Eg of the sample was calculated from the adsorption spectra using the Tauc plot method (αhν) 1/n = A(hν-Eg), where h is Planck's constant, ν is the frequency of the light, E g is the optical energy bandgap, and A is a material constant [53]. The n factor is dependent on the electron transition's type, in our case; the value of n is 2, assuming the WO 3 to be an indirect semiconductor [54]. The plot of the modified Kubelka-Munk function versus the energy of incident light for the WO 3 (Fig. 9) results in bandgap energy (Eg) of 2.18 and 2.48 eV for WO 3 -24 and WO 3 -48, respectively, which are much lower values than the reported bandgap for bulk WO 3 (Eg: 2.62 eV) [55]. Whereas an obvious blue shift toward higher bandgap energy (Eg = 2.48 eV) was observed in the absorption threshold of WO 3 -48 sample due to the different crystalline structure and quantum confinement effect [56,57]. Theoretical and experimental results suggested that the distortion of WO 6 octahedra in the crystal structure of WO 3 can increase the bandgap [58]. In addition to the effect of the crystal phase, the size of the nanostructures also plays a crucial role in the optical properties. As shown earlier, nanostructures smaller than 100 nm exhibit a weak quantum confinement effect (W-QC) and give a broaden bandgap. The QC effect gets stronger once the size of the nanomaterials approaches the Bohr radius which is 3 nm for WO 3 [59]. In sight of this, the apparent blue shift in the Eg for the nanoplatelets WO 3 -48 phase can be further explained by the substantial size difference compared to the bigger nanoplates of WO 3 -24 sample.
The valence band (VB) and conduction band (CB) potentials of a semiconductor at the point of zero charges (PZC) can be measured using the Mulliken electronegativity theory [60] Eqs. (3-2): where, E VB and E VB are the VB and CB potential, respectively, χ is the absolute electronegativity of the semiconductor, E C is the energy of the free electron on normal hydrogen electrode (NHE) scale (approximately 4.5 eV) [61] and Eg is the bandgap energy of the semiconductor. The absolute electronegativity of the semiconductor, which is the geometric mean of the electronegativity of the constituent atoms, is assessed by Eq. (4): where, a, b, and c are the number of atoms in the compounds A, B and C, respectively [62] x (A), x (B) and x (C) are the arithmetic mean of the atomic electron affinity and the first ionization energy, correspondently. Herein, the χ value of WO 3 is about 6.57 eV, similar to that reported in literature [63,64]. The calculated E VB and E CB values in the Normal Hydrogen Electrode (NHE) scale for the WO 3 -24 are found to be 0.98 eV and 3.16 eV, whereas for the WO 3 -48, the E VB and E CB are 3.31 eV and 0.83 eV, respectively.

Photocatalytic Experiments
The comparison of photocatalytic performance of WO 3 -24 nanoplatelets for RhB degradation from simulated wastewaters assisted by of H 2 O 2 under different conditions is shown in Fig. 10.
As depicted in Fig. 10, in the direct photolysis, RhB demonstrated a strong chemical stability, implying that the photosensitization process under LED irradiation is neglected. However, in the homogeneous oxidation process, H 2 O 2 is a key parameter that significantly impacts the photodegradation of RhB. Thus, only 18.92% of RhB degradation rate was achieved in 120 min in the presence of H 2 O 2 under LED light irradiation [65,66], indicating that small amounts of hydroxyl radicals ( · OH) are generated, which can oxidize RhB molecules through direct oxidation process. In the adsorption process employing tungsten-based catalyst only without H 2 O 2 , up to 22.36% RhB was decomposed at the same reaction time, implying that, under our experimental conditions, the photoexcited electrons in WO 3 photocatalyst are Solvothermal synthesis and characterization of monoclinic… inactive for O 2 reduction to form · OH radicals [67]. In the other hand, the WO 3 nanoplatelets showed an excellent photocatalytic performance in the presence of H 2 O 2 as an electron-trapping agent, as a result of 99% of RhB degradation was obtained in 120 min under LED light irradiation. The highest photocatalytic efficiency of the tungsten-based catalyst nanoplatelets can be attributed to the strong synergetic effect between WO 3 -24 and H 2 O 2 , resulting in the increased production of · OH radicals and the remarkably promotion of photocatalytic efficiency. Moreover, the large active surface areas of WO 3 facilitate contact with the organic contaminants, the high number of reactive sites for catalysis, the low bandgap energy, the availability of the oxygen vacancies, and modest thicknesses, that allow electron-hole pairs to migrate easily to the surface. Additionally, the nanostructures having corners and edges (such nanoplatelets) are more adsorbent than other nanostructures due to their low coordination numbers, corners and edges. [68,69]. Figure 11a exhibits the evolution of time-dependent absorption spectra of RhB during the photodegradation process under LED light irradiation using WO 3 nanoplatelets. We can observe from Fig. 11b that the characteristic absorption band of RhB positioned at 554 nm blue-shifted and its intensity reduced gradually, as the stimulation time rises, almost disappeared 120 min, implying that the photosensitization process is considered since photosensitization involves the continual N-deethylation of RhB B [70]. In this process, the λ max = 554 nm of RhB solution blue-shifted to 540.4 nm only in 30 min, attributing to the N, N, N'-triethylation. The N, N'-diethylation of RhB demonstrated by the blue-shift of λ max to 522 nm in 40 min, then N-ethylation of RhB occurred at 510 nm after 120 min, that is attributed to the photoisomerization of dye chromophores [71].
As illustrated in Fig. 12, the linear plots of (ln C 0/ C) against irradiation time (t) result in a straight line with a slope equal to K app , according to the relationship C 0 /C = exp (K app t). The dye degradation efficiency of WO 3 catalytic systems under LED light irradiation appears to follow a pseudo-first-order kinetics model, as evidenced by the regression coefficients values (R 2 = 0.98-0.97). The rate constants of the first order for the WO 3 -24 and WO 3 -48 during the first 60 min were 0.535 and 0.0478 min −1 , respectively. The presence of oxygen vacancy defects in the WO 3 -24 matrix (as confirmed by XRPD and UV-vis-DR measurements), which act as electron traps, could be able to contribute to a reduction in the recombination rate of the hole-electron couples and, consequently, an increase in the photocatalytic activity.

Mechanism of photocatalytic degradation
The photocatalytic efficiency of WO 3 nanoplatelets photocatalysts is assigned to the light absorption with energy greater than or equal to its bandgap, transfer of the lightinduced electron-hole pair and its charge separation in WO 3 NPs interfaces. The possible mechanism scheme of photocatalytic activity of WO 3 nanoplatelets on the degradation of RHB along with H 2 O 2 under white light-emitting diode (LED 15 W) lamp is shown in Fig. 13. WO 3 is a n-type semiconductor, which always stimulated by visible light illumination, absorbing at 500-568.81 nm (Eg = 2.48-2.18 eV/NHE). The relative positions of the redox potentials of superoxide radical O 2 /O 2 ·− (−0.28 eV/NHE), hydroxyl radicals O 2 / · OH (+ 0.7 V/NHE), OH − / · OH (1.9 eV/NHE), H 2 O/ · OH (2.8 eV/ NHE)) and (H 2 O 2 / · OH (1.14 eV/NHE)) as well as water (H 2 O 2 /H 2 O (1.76 eV/NHE)) along with the predicted band edge potentials of WO 3 catalysts are presented in Fig. 13.
Up on illumination by white LED (15 W) lamp, the electrons (e − ) in the VB of WO 3 were excited to its CB, resulting in the simultaneous generation of the same number of holes (h + ) in the VB (Eq. (5)). According to the band energy levels, the photo-induced CB electrons (e − ) cannot directly reduce neither O 2 into O 2 ·− nor O 2 to · OH, because of the redox energies of O 2 /O 2 −· (−0.28 V/NHE) and O 2 / · OH (+ 0.7 V/ NHE) are more negatives than the CB edge potential of WO 3 (+ 0.98 eV/NHE). The photogenerated VB holes (h + ), by contrast, can effectively oxidize both H 2 O and OH − to produce · OH (Eqs. (6-7)) due to their high-lying oxidation potentials. In the other hand, the addition of H 2 O 2 in the photo-oxidation process could act as an alternative electron acceptor to oxygen, thereby producing · OH (Eq. (8)) and H 2 O (Eq. (9))) and inhibiting the e _ and h + recombination. Furthermore, in aqueous solution, UV-light absorption (λ < 400 nm) splits H 2 O 2 into · OH radicals (Eq. (10)). The · OH radicals undertake reaction with RhB molecules to yield the mineralized products (Eq. (11)).
The possible pathways involved in the photo-oxidation of RhB under LED light irradiation using WO 3 as photocatalyst could be described as follow. As a probe organic polluante, RhB is often used for exploring the efficiencies of photocatalysts. Three approaches on the photocatalytic degradation mechanism of RhB have been considered: (i) the chemisorption of functional groups, (ii) the photosensitized degradation and (iii) the selective photocatalytic oxidation of RhB. Dye sensitization process is further used in this study to extend the absorption of the catalyst into LED-light illumination. It consists of the light absorption by a molecule called sensitizer (S) that, during the lifetime of its excited state, can transfer its excess energy to a substrate, obtaining the excited state of the latter. Sensitizer absorbs a photon in the visible light range and an electron passes from its HOMO (highest occupied molecular orbital) to its LOMO (lowest unoccupied molecular orbital); then, this electron is transferred into the CB of the semiconductor.
To achieve a thermodynamically possible and efficient electron transfer, the excited state of the photosensitizer should be higher in energy than the CB of catalyst. Rhodamine B (N, N, N', N'-tetraethylrodamine) belongs to the oxygen-containing heterocyclic xanthene dyes family. It could be seen from Fig. 11, that there is an absorption band at positioned at 554 nm, which originates from the π → π* transitions from the binding HOMO to the anti-binding LUMO along the longest dimension of the conjugated system. However, the shoulder at 521 nm is ascribed to the dimmer [72].
All generated reactive oxygen species (ROS) such as · OH; O 2 ·− , RhB ·+ and RhBO 2 during the photocatalytic reactions via the simultaneous (RhB/WO 3 -24/LED 15 W) and dye photosensibilization processes (RhB/LED 15 W), facilitating charge (7 carrier localization, reducing their recombination and hence, prolonged separation by trapping at energy levels within the bandgap and effectively decompose RhB into less toxic molecules or degraded or mineralized products Eq. (17).
Furthermore, as demonstrated by XRPD, FTIR, and UV-vis DRS measurements, the presence of oxygen vacancies may operate as an electron acceptor (and/or hole donor) to enhance charge carrier localization, as previously mentioned, minimizing their recombination and, as a result, allowing for a longer separation time [74,75]. The photocatalytic mechanism of WO 3 nanoplatelets in the presence of H 2 O 2 is illustrated in Fig. 13.

Conclusion
Using a unique low-temperature (180 °C) solvothermal method, we produced crystalline WO 3 nanoplatelets at low temperatures. In this study, we explored a commonly available short-chain alcohol 1-hexanol as both a solvent and as one of the reagents, resulting in a straightforward procedure that produces WO 3 nanoplatelets with small diameters. The reaction between H 2 WO 4 and n-Hexanol avoids expensive and hazardous precursors. By analyzing the UV-vis DRS data, our results showed that E g for the elaborated WO 3 nanoplatelets were found to be 2.18 and 2.48 eV for WO 3 -24 and WO 3 -48, much lower than that for bulk WO 3 (2.7 eV). The superior photocatalytic efficiency observed with the WO 3 nanoplatelets may be attributed to the strong synergetic effect between heterogeneous photocatalysis (RhB/WO 3 -24/ LED 15 W) and dye photosensibilization processes (RhB/LED 15 W), facilitate charge carrier localization, suppresses the recombination of photocarriers and, as a consequence, enhances the photocatalytic efficiency and effectively decompose RhB into less toxic degraded molecules or mineralized products.