Introduction

With the acceleration of urbanization, the rapid development of industry and climate change of many rivers as well as lakes had been badly polluted, especially in developing countries, which directly affect the healthy drinking water of human beings (Mirzaeian et al. 2019; Gupta et al. 2015; Cheng and Jin 2022; Shi et al. 2022; Liu et al. 2021). Numerous organic pollutants have been identified in various sources of wastewater (Malinga and Jarvis 2020; Kamaludin et al. 2018; Masocha et al. 2022). For example, Ahmadi et al. reported the ZnO@ZIF-67 nanocomposite with high adsorption performance for the removal of pharmaceutical contaminant from aqueous media (Ahmadi et al. 2021, 2022). In recent years, the dyes can cause the decrease of water transmittance and hinder the growth of bacteria and other micro-organisms, inhibiting the aquatic photosynthesis and damage the ecosystem; at the same time, dyes are harmful to human beings (Robati et al. 2016; Rajabi et al. 2017). Cationic dyes methylene blue is widely used as colorants in industry and bacteriostatic agents. Researchers found that these azo dyes are carcinogenic, highly toxic, persistent, and mutagenic; it caused irretrievable damage to the environment if it discharged into water (Rajabi et al. 2019; Abdelmaksoud et al. 2021). It was reported that adsorption process is one of the effective methods to remove dyes and antibiotics from wastewaters (Moradi and Sharma 2021). The polymer, such as carboxylmethyl cellulose and guar gam, were composited with graphene oxide, Fe3O4 or TiO2, and the obtained composite adsorbents were used for removal of anionic and cationic dyes, such as methylene blue (MB), methyl orange (MO), and malachite green (MG), from aqueous solution (Naeini et al. 2022; Moradi et al. 2021). In past decades, some systems, including physical, chemical, and biological processes, were explored to treat contaminated water and improve effluent water quality (Rajendran et al. 2021; Moradi et al. 2009; Truong et al. 2022; Dhanalakshmi et al. 2020; Li et al. 2019), but there are still a lot of questions. Therefore, finding an efficient and economical approach to remove and degrade multiple pollutants from water is extremely critical (Babaei et al. 2022; Shi et al. 2020; Liu et al. 2019).

Since the discovery of semiconductor heterogeneous photocatalysis, it has become a potential solution for degrading toxic organic pollutants to achieve environmental sustainability (Mukhtar et al. 2021). And semiconductive photocatalyts have shown excellent development prospects in the field of industrial wastewater and domestic wastewater treatment (Ouyang et al. 2021; Chen et al. 2020). Tungsten trioxide (WO3) is an n-type semiconductor that has been vastly studied for applications in gas sensing, catalysis, solar energy conversion, and electrochromic displays (Lu et al. 2022). The low energy band gap of WO3 allows this material a very promising candidate for the green energy production to implement the world energy demand (Naeimi et al. 2022; Tahir et al. 2018). Layered WO3 nanocubes could inhibit the recombination of photoelectron-hole pairs (Parthibavarman et al. 2018), thereby enhancing the redox properties of nanoparticles, which has attracted much attention in the field of electrochemical and photocatalytic materials (Mehmood et al. 2018). However, in the semiconductor of nubbly WO3, the surface accumulation of photogenerated charge carriers was easy, which lead to increase their recombination rate and eventually tumbling the photocatalytic efficiency (Manikandan et al. 2022; Shah et al. 2020; Sun et al. 2020). To settle these problems, numerous effective efforts, such as construction of heterojunction (Sun et al. 2022), modification (Karami et al. 2022), metal doping (Raeisi et al. 2021; Al-Kuhailia and Drmoshb 2022), and quantum dot modification (Cui et al. 2020) which strengthen the separation of photocarrier and improve photocatalytic performance, have been researched. For improving the recyclability and performance of photocatalysts, natural non-metallic minerals were used as carriers to disperse the photocatalysts (Li et al. 2022; Shen et al. 2021). The loess is mainly composed of extremely small powdery particle. It has high hydrophilicity and pores of various sizes and shapes, so that the loess has certain adsorption properties (Babudurai et al. 2021; Xu et al. 2022; Gao et al. 2021). With the increasing shortage of non-renewable resources, the construction of WO3 nanocubes@Loess has good prospects for rapid photocatalytic degradation of organics in wastewater.

In order to enhance the photocatalytic performance of WO3-based compounds, in this paper, the particles of LoP, a kind of extremely common and inexpensive natural silicate particles with metastable structure, were used as inorganic carrier; the WO3NCs was attached by in situ deposition, which afforded WO3NCs@LoP. After being characterized by SEM, XRD, FT-IR, XPS analysis, and UV–Vis DRS. In addition, its photocatalytic performance was investigated by photocatalytic degradation of dyes under visible light, and its synergistically photocatalytic performance was investigated.

Experimental

Materials

Loess was collected from the local hill near Lanzhou of China. Sodium tungstate dihydrate (Na2WO4.2H2O) used in the experiments was purchased by second branch of Shanghai Reagent General Plant. Citric acid monohydrate (C6H8O7.H2O, CA) was obtained by Tianjin Deen Chemical Reagent Co. Ltd. Anhydrous ethanol (C2H6O) was purchased by Sinopharm chemical Reagent Co. Ltd. Hydrochloric acid (HCl, Shanghai Zhongqin Chemical Reagent Co. Ltd.). Other reagents, such as isopropanol (IPA) (Tianjin Fuchen Chemical Reagent Factory), disodium ethylenediaminetetraacetate (EDTA-2Na) (Sinopharm Chemical Reagent Co. Ltd.), silver nitrate (AgNO3, Third Branch of Tianjin Chemical Reagent Sixth Factory), rhodamine B (RhB, Tianjin Tianxin Fine Chemical Development Center), methylene blue trihydrate (MB), malachite green (MG), crystal violet (CV), and Eriochrome Black T (EBT) were purchased from Sinopharm Chemical Reagent Co. Ltd.. All the above chemicals were of analytical grade and directly used without further purification. The redistilled water was used during the experiment procedures.

Separation of loess particles (LoP)

100 g of dried loess powder was dispersed in water and stirred for 12 h with mechanical at room temperature until loess particles being dispersed completely. Firstly, the supernatant suspension was quickly collected after 2 min standing of the loess powder turbid solution. Secondly, the obtained 300 mL of supernatant suspension was stirred for 20 min. Then, after standing for 10 min, the supernatant suspension was collected. Thirdly, the above 200 mL of supernatant suspension stood for 24 h, the clear water layer was sucked out. The small loess particles (LoP) were obtained by drying 12 h.

Preparation of WO3NCs@LoP

WO3NCs@LoP composite materials were prepared by hydrothermal method; the process was as follows: 1.0 g of Na2WO4.2H2O powder and 0.5 g of CA were dispersed into 50 mL of deionized water, and then 10 mL of 4 mol∙L−1 HCl was added dropwise. Then amounts of LoP were uniformly dispersed in the mixed solution, and the mixed solution was transferred to a 100-mL Teflon-lined stainless-steel autoclave and heated at 160 °C for 12 h. The light-yellow solid powder products were obtained by centrifugation and then washed with deionized water followed by ethanol for 3–5 times. Finally, a square WO3NCs@LoP photocatalyst was obtained at 60 °C by vacuum drying for 4 h. The WO3NCs@LoP-1, the WO3NCs@LoP-3, the WO3NCs@LoP-7, and the WO3NCs@LoP-9 samples were prepared via adjusting the amount of LoP, and the rest of steps was the same. The obtained photocatalysts were marked as WO3NCs@LoP-x (x = 1, 3, 5, 7, 9 denoted the mass of LoP). For comparison, pure WO3NCs was prepared without introducing LoP by using the same method.

Characterization

The mineralogical composition and crystalline phase structure were measured by X-ray diffraction (XRD) using Cu Ką radiation at 40 kV, 40 mA, and a step size of 0.02° in the range of 5 ~ 80°. FT-IR spectra were recorded between 4000 and 400 cm−1 through the KBr method with the FTS-3000 spectrophotometer. The morphology and particle size were observed by scanning electron microscope (SEM) (at 5 kV, ULTRA Plus microscope, Germany). UV–Vis diffuse reflectance spectra (UV–Vis DRS) were collected on a UV-2800A spectrophotometer in the range of 200–800 nm. The average particle size was measured by dynamic light scattering (DLS) (Nano Series, Malvern Instruments Ltd., UK). The elemental composition and chemical oxidation state of the sample were analyzed with X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha).

Photocatalytic reaction

The photocatalytic activity of the composite was evaluated by measuring the degradation rates of MB in synthetic wastewater (Fig S1). In addition to the organic dyes MB, which were used as model pollutants, the synthetic wastewater mostly contains inorganic NaCl, Na2CO3, and NaH2PO4. And A 500 W Xe lamp with cut-off filter (λ > 420 nm) was served as light source to simulate sunlight. A set of photocatalytic degradation experiments were performed with the following procedure: 0.05 g WO3NCs@LoP was added to 50 mL of MB solution (20 mg∙L−1). The adsorption–desorption equilibrium of MB was established after stirring in the dark for 30 min and then degraded MB under light. At set intervals, the samples were withdrawn from the reactor and filtered by PES syringe filter (pore size: 0.22 μm) for removal of photocatalyst particles. Finally, the UV spectrophotometer was used to determine the absorbance of the solution at 664 nm. All photocatalytic experiments were repeated at least three times, and error bars were calculated. The degradation efficiency of the catalyst was determined by using Eq. 1.

$$\eta =\left[\left({C}_{\mathrm{O}}-{\mathrm{C}}_{\mathrm{t}}\right)/{C}_{0}\right]\times 100\%$$
(1)

where C0 is the initial concentration of the dye solution and Ct is the concentration MB after photocatalytic degradation for certain reaction time.

Capture of radical experiment

In the photocatalytic process, the active substances produced in the degradation process were captured by adding a trapping agent (1 mmol/L), where AgNO3 (1 mmol/L), benzoquinone (BQ, 1 mmol/L) isopropyl alcohol (IPA, 1 mmol/L), and EDTA-2Na (1 mmol/L) were viewed as the capture of e−, •O2−, •OH, and h+ species, respectively. Except the addition of trapping agent in the experimental process, other operations and conditions kept consistent with the photocatalytic process.

Results and discussion

Preparation and characterization of WO3NCs@LoP

The preparation process WO3NCs@LoP was shown in Scheme 1. Using LoP, a kind of extremely common and inexpensive natural silicate particles with metastable structure, as inorganic carrier, the WO3NCs loaded on loess particles in situ depositing, WO3NCs@LoP. The photocatalytic degradation process is the result of the synergistic effect of adsorption and degradation (Cao et al. 2022). Although WO3NCs has a certain extent photocatalytic activity, its adsorption performance is relatively poor. Dual-functional environmentally friendly composite photocatalyst was prepared by using the adsorption property of loose loess particles and synergistically with the photocatalytic properties of WO3NCs, which was favorable for enhancing the photocatalytic activity. It is of great guiding significance to construct environmentally friendly photocatalysts.

Scheme 1
scheme 1

Formation of WO3NCs@LoP photocatalyst

Full size image

The photocatalytic properties of the WO3NCs@LoP were studied by photocatalytic degradation of MB, which were exhibited in Table S1, pure hexagonal-WO3 nanocubes only 66.11% degradation rate of MB. Varying the ratio of the WO3NCs@LoP by adjusting the amount of LoP and photocatalytic performance for degradation of MB under visible light irradiation was compared. It could be observed that WO3NCs@LoP expressed the best photocatalytic performance (the amount of loess added was 5 g, 99.77%). Therefore, this WO3NCs@LoP was characterized and discussed in detail.

SEM analysis

The micromorphology of WO3NCs@LoP and its materials (LoP, WO3NCs) were observed by SEM, and the results were shown in Fig. 1. LoP were irregular particles with cracks with relatively loose structure (Fig. 1a), which was beneficial to exposing large surface area. The morphology of WO3NCs looked like nanocubes with smooth surface, and the average size was near 100 nm (Fig. 1b). But it was easy to agglomerate as reported results (Dhanalakshmi et al. 2020; Manikandan et al. 2022 and Palanisamy et al. 2021). In WO3NCs@LoP (Fig. 1c), the nanocubes of WO3NCs were tightly attached to LoP, which was well-dispersed on surface of the particles for avoiding agglomeration. The particle size of WO3NCs@LoP was measured by DLS. It showed that the average size of composites was near 780 nm (Fig. S2), which was similar to SEM results.

Fig. 1
figure 1

The SEM images of LoP (a), WO3NCs (b), and WO3NCs@LoP (c)

Full size image

FT-IR analysis

The FT-IR spectra of LoP, WO3NCs, and WO3NCs@LoP were shown in Fig. 2. In WO3NCs, the characteristic peak near 800 ~ 645 cm−1 was attributed to the vibration of the W–O-W bond (Manikandan et al. 2022). The characteristic peaks of LoP appeared at 1084 cm−1 and 490 cm−1 (Deng et al. 2022), which were attributed to the stretching and bending vibration peaks of Si–O-Si in LoP. The wide peak at 3600 ~ 3300 cm−1 was related to the stretching vibration absorption peak of hydroxyl (-OH) in water. Comparing with pure LoP and WO3NCs, their characteristic peaks appeared in WO3NCs@LoP. It shows that WO3NCs is successfully loaded on the surface of loess, and the structure is extremely stable, which is consistent with the SEM results.

Fig. 2
figure 2

The FT-IR spectra of LoP, WO3NCs, and WO3NCs@LoP

Full size image

XRD analysis

The X-ray diffraction (XRD) was used to investigate the crystalline structure, and the results were shown in Fig. 3. For the WO3NCs XRD pattern, some peaks at 23.1°, 23.5°, 24.4°, 33.2°, 33.6°, 34.1° and 49.9° indexed to the (002), (020), (200), (022), (− 202), (202), and (140) planes, which was consistent with WO3NCs standard XRD patterns (JCPDS card no: 43–1035) (Zhang et al. 2021; Guo et al. 2019), proving that the material was successfully synthesized. The LoP XRD pattern could be seen; the diffraction peak at 28°, 26°, 50°, 60°, and 68° were the characteristic diffraction peak of amorphous silicate or aluminosilicate and quartz in LoP. The characteristic peaks of WO3NCs and LoP were appeared in WO3NCs@LoP, indicating that WO3NCs@LoP was successfully synthesized, and the crystal structure was not destroyed during the synthesis process, which was mutually confirmed with SEM and FT-IR.

Fig. 3
figure 3

The XRD diagram of LoP, WO3NCs, and WO3NCs@LoP

Full size image

XPS analysis

In order to illustrate the chemical composition and valence structure, WO3NCs@LoP and LoP were analyzed by XPS spectroscopy, and the results were shown in Fig. 4. In the XPS survey spectrum of WO3NCs@LoP (Fig. 4a), W was clearly appeared besides some typical elements of LoP, such as Si, Al, O, Fe, K, and C. The distribution of these elements was summarized in Table S2. In the XPS survey spectrum of LoP (Fig. 4b), W was not found. In the W4f high resolution spectrum (Fig. 4a1), it exhibited two typical peaks around 35.7 and 37.8 eV which were assigned to the W4f7/2 and W4f5/2 for W6+ oxidation states in WO3NCs@LoP composite, respectively (Dhanalakshmi et al. 2020), whereas the characteristic peaks of W4f element cannot be observed in the bare LoP (Fig. 4b1). It further proved the successful synthesis of composite materials. Besides, for Si2p as shown in Fig. 4a2, the peak value at 103.0 eV belonged to SiO2 (Ye et al. 2022), which was similar to LoP (Fig. 4b2), indicating the stability of the material. The result was consistent with FT-IR, SEM, and XRD, confirming that the WO3NCs@LoP was successfully synthesized.

Fig. 4
figure 4

XPS full spectrum, Si2p, and W4f high resolution spectra of WO3NCs@LoP (a, a1, a2) and LoP (b, b1, b2)

Full size image

In summary, based on SEM, XRD, FT-IR, XPS, and synthesis mechanism analysis, WO3NCs@LoP were successfully synthesized with regular morphology and microstructure. Meanwhile, comparing with WO3NCs, WO3NCs@LoP had a larger specific surface area, providing more photocatalytic active sites, which was beneficial to the photocatalytic degradation process.

Photochemical properties of WO3NCs@LoP

In order to explore the optical properties of the WO3NCs, LoP, and WO3NCs@LoP, UV–Vis diffuse reflectance spectra were conducted because it could verify the absorption of visible light and calculate the band gap energy of the material by Tauc equation. The results were shown in Fig. 5. It could be found that LoP exhibited better visible light absorption ability in the visible light band (400–800 nm), indicating that it utilized visible light energy. However, WO3NCs had a strong absorption ability in the ultraviolet region, but poor light absorption performance in the visible light region, indicating that it mainly absorbed ultraviolet light. After in situ synthesis of WO3NCs in LoP via a hydrothermal reaction, the obtained WO3NCs@LoP showed an enhanced absorption intensity in the visible light region, indicating an obvious red shift. The results manifested that the light absorption range of WO3NCs was significantly improved after the formation of WO3NCs@LoP. The results showed that the photochemical properties of WO3NCs could be enhanced by combining WO3NCs on LoP, separating photogenerated electron–hole pairs occur in WO3NCs@LoP under visible light.

Fig. 5
figure 5

a Diffuse reflectance UV–Vis, b band gaps curves of LoP, WO3NCs, and WO3NCs@LoP

Full size image

Furthermore, the band energy (Eg) is calculated by the reported Tauc function in the literature (Xiang et al. 2020): (αhv)n = A (hv − Eg) where α, hν, A, Eg, and n indicate the absorption coefficient, Planck constant, photon energy, proportionality constant, energy gap, and the transition properties of semiconductors (indirect semiconductors: n = 1/2, direct semiconductors: n = 2), respectively (Zhao et al. 2021). Herein, n = 2 due to the as-prepared solid samples have direct semiconductor properties. According to the calculation, the band gap of WO3NCs was 2.90 eV; for the WO3NCs@LoP, its Eg was 2.49 eV, which were exhibited in Fig. 5b. Of note, WO3NCs@LoP had narrow band width in contrast with WO3NCs. As a consequence, WO3NCs@LoP photocatalyst used visible light to promote more efficient separation of photogenerated electron–hole pairs, which further improved the photocatalytic performance. Moreover, the conduction band potential (CB) and valence band potential (VB) can be calculated via the following formula (Eqs. (3), (4)), where Xis the electronegativity of the semiconductor, EVB is the VB potential, ECB is the CB potential, and Ee is the energy of the e− in the Hscale (~ 4.5 eV) (Roy et al., 2022). According to the above equation, the ECB and EVB values of WO3NCs and WO3NCs@LoP were calculated as follows: for WO3, ECB = 0.63 eV and EVB = 3.53 eV; for WO3NCs@LoP, ECB = 0.845 eV and EVB = 3.335 eV. The results indicated that WO3NCs@LoP composite materials could greatly enhance optical absorption in visible light due to synergistic effect among WO3NCs and LoP, which was beneficial to enhance the photocatalytic activity.

$${\left(\mathrm{\alpha hv}\right)}^{\mathrm{n}}=\mathrm{A}\left(\mathrm{hv}-\mathrm{Eg}\right)$$
(2)
$${\mathrm{E}}_{\mathrm{CB}}=\mathrm{X}-{\mathrm{E}}_{\mathrm{C}}-0.5\mathrm{Eg}$$
(3)
$${\mathrm{E}}_{\mathrm{VB}}={\mathrm{E}}_{\mathrm{CB}}+\mathrm{Eg}$$
(4)

Photocatalytic activity

To explore the photocatalytic activity of the prepared photocatalysts, a common ionic dye MB in the textile industry was selected as the degradation substrate. A series of degradation experiments performed were shown in Fig. 6. The effects of photocatalyst dosage, MB concentration, irradiation time, and pH value were explored. Discussion and validation of the effects of experimental conditions on MB degradation experiments were in the Supporting Information (Fig S3–S6). The results denoted that the optimal conditions were photocatalyst 0.05 g, MB 20 mg/L, irradiation time 20 min, and no pH adjustment. Figure 6a shows the degradation curves of MB by different catalysts, and it was found that MB was a relatively stable dye with hardly self-degradation without any catalyst. The removal rate of LoP with the favorable adsorption capacity reached 54.4% after the equilibrium of adsorption and desorption, and the final removal rate reached 62.1%, indicating its poor photodegradation performance. After the adsorption–desorption equilibrium of WO3NCs, the removal rate was only 29.0%, while the final MB removal rate reached 66.1% under visible light irradiation, indicating that WO3NCs presented photocatalytic degradation of MB. WO3NCs@LoP as a photocatalyst photodegrades MB under the same conditions, and it could be observed that the removal rate of MB was higher than that of pure WO3NCs and LoP, with a final removal rate of 99.8% and an adsorption capacity of 47.9%. It was worth mentioning that the degradation rate of WO3NCs@LoP reached 99.5% under visible light irradiation for only 20 min and proved that the prepared photocatalyst was an efficient photocatalyst with excellent photocatalytic performance.

Fig. 6
figure 6

The degradation curves (a) and the pseudo first-order kinetic (b) of MB degradation with different photocatalysts, UV–Vis spectra MB solution during photodegradation (c), and comparison of photocatalytic degradation of WO3NCs@LoP by different dyes (d)

Full size image

To study the catalytic activity of the photocatalysts, a pseudo-first-order kinetic model was used for fitting to compare the degradation rates of MB under different photocatalysts, and the reaction rate constants were shown in Fig. 6b (Dai et al. 2021; Valério et al. 2020). It could be observed that the size of the apparent reaction rate was WO3NCs@LoP (0.0495 min−1) > WO3NCs (0.0074 min−1) > LoP (0.0047 min−1). WO3NCs@LoP showed the highest apparent reaction rate, which was about 10.53 and 6.69 times higher than pure LoP and WO3NCs, respectively. In WO3NCs@LoP, the photocatalytic activity of WO3NCs was significantly enhanced after WO3NCs loaded on the LoP surface. The reason was that WO3NCs uniformly attached on the surface of LoP, which not only improved the agglomeration of WO3NCs, but also increased the specific surface area of the catalyst, so that the catalyst provided more active sites during the photocatalytic degradation process. At the same time, the synergistic effect of LoP and WO3NCs enhanced the photocatalytic performance of the catalyst. After being photodegraded by WO3NCs@LoP, the change of MB concentration under different illumination times was measured by UV–Vis spectroscopy as shown in Fig. 6c, with the maximum absorption peak at 664 nm. With the increase of photocatalytic degradation time, the intensity of the absorption peak obviously gradually weakened, which indicated that the residual MB concentration in the solution gradually decreased.

The lowest value was almost reached at 20 min, which was not much different from the maximum degradation value at 90 min, indicating that MB in the solution was almost completely degraded within 20 min of photodegradation. In another side, near 50 % of dyes were absorbed in darkness from -30 to 0 min. It proved that adsorption played an important role in the degradation process, and it was demonstrated that this process synergistic effect of photocatalytic degradation and adsorption. In general, most photocatalysts were not widely used because they do not possess degradation universality. In this study, the degradation performance of WO3NCs@LoP for other ionic dyes was investigated, and the results were shown in Fig. 6d. It could be clearly found that WO3NCs@LoP showed a good degradation effect on EBT, MG, CV, and RhB, and the lowest degradation rate also reached more than 75%, indicating that the material denoted a wide range of dye degradation properties.

It is well known that WO3NCs is a typical photocatalyst (Shandilya et al. 2022), and several studies denote reported its photocatalytic degradation performance. The degradation performance of the prepared photocatalysts for MB was compared with the reported literatures (Table 1). The results were shown that WO3NCs@LoP expressed excellent performance in terms of dosage, irradiation time, and final removal rate, indicating that the photocatalyst prepared in this experiment expressed a good degradation effect.

Table 1 Comparison with other WO3NCs@LoP-based photocatalysts on MB degradation
Full size table

Cycling stability of photocatalytic degradation

The cycling stability of photocatalysts is an important indicator to investigate (Gu et al. 2022). In view of this factor, the cycle stability of the catalyst was investigated, and the results were shown in Fig. 7. The photocatalytic degradation of MB was consistent with the above-mentioned process. The used photocatalyst was collected, washed with redistilled water and absolute ethanol by centrifugation, and dried, and the recovered catalyst was recycled. The MB degradation rate of WO3NCs@LoP still reached above 85% after 9 cycles (Fig. 7a). It was proved that WO3NCs@LoP had excellent photocatalytic activity, which might be related to the tight interface formed by the combination of WO3NCs and LoP due to the hydrothermal reaction. At the same time, WO3NCs@LoP also possessed excellent photocatalytic stability. The structure of the photocatalyst (WO3NCs@LoP) before and after the degradation reaction was characterized by FT-IR (Fig. 7b). It indicated that characteristic peaks of W–O-W bond near 800 ~ 645 cm−1 were weakened after cycling, and the characteristic peaks of Si–O-Si bonds still existed, indicating some WO3NCs were run away after 9 cycles of degradation process. In conclusion, WO3NCs@LoP is a photocatalyst with excellent photocatalytic activity and stability.

Fig. 7
figure 7

a The catalytic degradation performance after repeated experiments, b FT-IR spectra of WO3NCs@LoP photocatalyst before and after nine experiments

Full size image

Capture of radical experiment

It is well known that active species play an extremely crucial role in the photocatalytic degradation process (Cai et al. 2022). With a view to identifying the active species in the process of degrading MB and providing a strong proof for the degradation mechanism, the capture experiment of active species in the photodegradation process of MB by WO3NCs@LoP was produced under visible light. The results were shown in Fig. 8; it could be found that with the addition of AgNO3, IPA and EDTA significantly inhibited the MB degradation, while BQ had little effect on the degradation process. Therefore, e−, h+, and •OH were the active substances that played a role in the photocatalytic degradation of MB.

Fig. 8
figure 8

Active substance capture experiment

Full size image

Photocatalytic mechanism of WO3NCs@LoP

On the basis of aforementioned characterization and experimental result analysis, the possible mechanism for high degradation efficiency of WO3NCs@LoP composites is proposed under visible light, and showed in Scheme 2. When visible light irradiated on the surface of the catalyst, it produced the electron–hole pairs for photocatalytic degradation of MB (Eqs., 5, 6 and 7). Obviously, the higher VB potential of WO3NCs@LoP (3.33 V vs NHE) than H2O/•OH (2.13 eV vs NHE) was capable of oxidizing and adsorbing H2O to produce •OH. Meanwhile, parts of the hole could act as an active substance to degradate MB, whereas the CB accumulated electrons of WO3NCs@LoP could not reduce O2 to form •O2− owing to the higher CB potential of WO3NCs@LoP (0.845 V vs NHE) than O2/•O2− (− 0.33 eV vs NHE). As a consequence, it was demonstrated that decomposition of hydroxyl radicals (•OH), e−, and photogenerated holes (h+) was the more effective active substance in the process of photodegradation, which could directly react with MB to further oxidize it (Eq. (8)). This was consistent with the results of the active substance capture experiment above. Thus, the significantly improved catalytic activity of WO3NCs@LoP photocatalyst under visible light could depend on the effective separation mechanism of photogenerated electron–hole pairs.

Scheme 2
scheme 2

Photocatalytic mechanism of WO3NCs@LoP

Full size image
$${\mathrm{WO}}_{3}{\mathrm{NC}}_{\mathrm{s}}\to {\mathrm{WO}}_{3}{\mathrm{NC}}_{\mathrm{s}}\left({\mathrm{e}}^{-}+{\mathrm{h}}^{+}\right)$$
(5)
$${\mathrm{WO}}_{3}{\mathrm{NC}}_{\mathrm{s}}@\mathrm{LoP}+\mathrm{hv}\to {\mathrm{WO}}_{3}{\mathrm{NC}}_{\mathrm{s}}\left({\mathrm{e}}^{-}+{\mathrm{h}}^{+}\right)$$
(6)
$${\mathrm{H}}_{2}\mathrm{O}+{\mathrm{h}}^{+}\to \bullet \mathrm{OH}$$
(7)
$$\bullet \mathrm{OH}+{\mathrm{h}}^{+}+{\mathrm{e}}^{-}+\mathrm{MB}\to \mathrm{Products}$$
(8)

Conclusions

In summary, a novel WO3NCs@LoP composite photocatalyst has been triumphantly synthesized by a facile one-pot hydrothermal with the loading of WO3NCs on LoP, which not only improves the agglomeration of WO3NCs, but also enhances the photocatalytic activity. The morphologies, functional groups, crystal phase, band gap, composition, and chemical structure were analyzed by SEM, FT-IR, XRD, UV–Vis DRS, and XPS, which demonstrated the successful synthesis of composite material and deep look into the possible mechanism. Under visible light irradiation, compared to the pure WO3NCs and LoP, WO3NCs @ LoP composite material exhibited an excellent degradation effect on MB, and the removal rate in 20 min reached more than 99%, with the apparent reaction rate of 0.0495 min−1, which was about 10.53 and 6.69 times higher than pure LoP and WO3NCs, respectively. At the same times, composite photocatalyst WO3NCs@LoP demonstrated favorable stability, and the degradation rate of MB remained reach more than 85% after recycling 9 times. A photocatalyst based on the loess substrate presents good performance and low cost; there will have broad application prospects and also respond to environmentally friendly materials for organic pollutions in wastewater treatment.