N-Doped HNT/TiO2 Nanocomposite by Electrospinning for Acetaminophen Degradation

In this study, we combined electrospinning of a large amount of halloysite (HNT, 95%) with nitriding to produce N-HNT-TiO2 composite nanofibers (N-H95T5 hereafter) to be used for acetaminophen (ACT) photodegradation. Investigation of the morphological and structural properties of the obtained materials did not highlight any significant difference in their morphological features and confirmed that nitrogen was evenly distributed in the samples. Photocatalytic tests under visible light showed that acetaminophen photodegraded faster in the presence of samples with nitrogen (N-H95T5) than without (H95T5 nanofibers). Moreover, the N-H95T5 nanocomposite photocatalytic activity did not change after repeated utilization (five cycles). The addition of scavengers during photocatalytic tests showed the key implication of OH•−, O2•− and h+ radicals in acetaminophen degradation. These results indicated that N–H95T5 composite nanofibers could be considered a cheap multifunctional material for photodegradation and could open new prospects for preparing tunable photocatalysts.


Introduction
Water pollution is a major problem leading to health and environmental issues [1]. With proper water treatment, such as by heterogeneous photocatalysis, it is possible to reduce water pollution by degrading pollutants [2][3][4]. TiO 2 is the most used material in water purification systems because of its low cost, low toxicity, relatively high efficiency, and high chemical and thermal stability [5,6]. However, two of its major drawbacks are the limited recovery after water treatment and the rapid recombination of the photogenerated charges [7]. Therefore, substantial research has been conducted to optimize TiO 2 photocatalytic activity [8,9]. For instance, doping and surface modification have been studied to improve the utilization of sunlight. TiO 2 has been doped using metal ions (e.g., Pd [10,11], Ag [12,13], Pt [10]), non-metal ions (e.g., B, N, Cu, Ni), and semi-conductors (e.g., BN, ZnO, CuO) [14][15][16][17] to enhance its photocatalytic activity.
According to the atomic orbital theory, nitrogen doping increases the amount of nitrogen in the first energy level N 1s and in the O 2p orbital of TiO 2 . This allows for increasing the number of electrons in the conduction band and the number of holes in the valence band and limiting the in situ recombination of electron-hole pairs after excitation with visible light [18]. In nitrogen-doped TiO 2 , the semiconductor bandgap is reduced (thus facilitating the photocatalyst activation), the light response range is broadened, and the number of photogenerated carriers is increased [19,20]. Recently, Asahi et al. showed that TiO 2 optical absorbance into the visible light region is enhanced after nitrogen doping [21]. Nitrogen-doped TiO 2 (N-TiO 2 ) can be prepared using different methods, such as hydrothermal synthesis and calcination in the presence of NH 3 [22] and by postsynthesis nitriding at 500 • C. In the last study, the authors highlighted the improved activity of N-TiO 2 for acetone oxidation under visible light [23]. Moreover, nitrogen doping by treatment with ammonium nitrate in water at pH 5.87 strongly increased N-TiO 2 -mediated degradation of 2,4-dichlorophenol following exposure to visible light [24]. These findings indicate that nitrogen doping improves TiO 2 photocatalytic activity by promoting hydroxyl and superoxide radical production.
Recently, Abid et al. prepared TiO 2 -based composite nanofibers with 95% halloysite (H95T5) and demonstrated that these nanocomposites degraded 91% of acetaminophen under visible light irradiation in 360 min [25]. Here, we explored the preparation and photocatalytic performance of nitrogen-doped H95T5 (N-H95T5) composite nanofibers. We found that these new nitrogen-doped nanocomposites eliminated more than 95% of acetaminophen in the presence of visible light in only 270 min. We then used highperformance liquid chromatography (HPLC) and toxicity assays to determine the intermediates and reaction products generated during the photocatalytic process and their toxicity and found out that the percentage of bacterial luminescence inhibition by ACT decreased to 6% after 24 h using N-H95T5 compared to 52% with H95T5 [25]. We identified the main reactive species explaining acetaminophen degradation using scavenger tests.

Materials
All chemicals listed in Table 1 were used without further purification.

Preparation of N-H95T5 Composite Nanofibers
N-H95T5 samples were prepared by electrospinning and nitriding following previously described protocols [25,26]. Polyvinylpyrrolidone solution was prepared by dissolving PVP powder into ethanol solution. Then, halloysite was added to TiO 2 prepared by hydrolyzing titanium tetraisopropoxide into a mixture of acetic acid and ethanol. The precursor mixture was stirred for 1 h at room temperature and electrospun. The obtained nanofibers were left exposed to air overnight to hydrolyze followed by heating at 400 • C for 4 h in air and sintering in a tubular furnace at 500 • C for 1 h (under nitrogen atmosphere).

Structural Characterizations
The samples' structures, phase, and crystallinity were determined by scanning electron microscopy (SEM; Hitachi S-4800) and transmission electron microscopy (TEM; JEOL ARM 200F).
Surface and micropore areas were described using the Brunauer-Emmett-Teller (BET) method, with different data points and relative pressures (P/Po) from 0 to 1. Their structural and crystallinity properties, elemental composition, and oxidation states were analyzed by X-ray diffraction (XRD; PANAlytical Xpert-PRO diffractometer equipped with an X'celerator detector using Ni-filtered Cu-radiation), Fourier-transform infrared spectroscopy (FT-IR) was recorded with the NEXUS instrument, equipped with an attenuated total reflection accessory in the frequency range of 400-4000 cm −1 , Raman (Horiba Xplora, 532 nm), and X-ray photoelectron spectroscopy (XPS) with an ESCALAB 250 spectrometer (Thermo Electron; excitation source: Al Kα monochromatic source, 1486.6 eV), respectively.

Photocatalytic Activity
For photocatalytic activity testing, acetaminophen degradation in the presence of different samples and of visible light was carried out as in our previous work [25] and quantified with Equation (1) [28].
where C 0 and C are the pollutant concentrations before and after irradiation. Acetaminophen was chosen for these tests because it is consumed in all countries of the world, has been detected in water samples from different origins, and is very stable in water [14,17,25,[29][30][31][32]].

Kinetics
Acetaminophen photocatalytic degradation kinetic data were fitted using a pseudofirst-order kinetic model [33]: ln (C 0 /C) = K app t, where C 0 is the initial concentration, C is the concentration at time t, and K app is the apparent rate constant.

Micro-Toxicity Tests
To assess the toxicity of the solution during acetaminophen photodegradation, microtoxicity tests were carried out using the bioluminescent marine bacterium Vibrio fischeri as described in [25].

H95T5 and N-H95T5 Morphology and Structure
H95T5 and N-H95T5 morphological features were characterized by SEM. In both samples, nanofibers were uniform, continuous, and randomly oriented ( Figure S1a,b). This confirmed that the introduction of nitrogen did not change TiO 2 shape, in good agreement with the literature [22,25,34].
Crystallinity analysis of H95T5 and N-H95T5 by XRD showed ( Figure S2) the characteristic 001 reflection of HNT (7.18 [25]. XRD patterns indicated that the main TiO 2 reflection was shifted from the initial position after nitriding. Likewise, no significant difference was found between H95T5 and N-H95T5 by Raman and Fourier-transform infrared spectroscopy ( Figure S3).
High-resolution TEM (Figure 1a-c) indicated that N-H95T5 nanofibers had a rough and large surface area with a lattice spacing of 0.350 nm, fully consistent with the distance of the (101) crystalline plane of the TiO 2 anatase [25,35,36]. Doping with nitrogen changed neither the morphology of N-H95T5 compared with H95T5 [25] nor the crystal lattice values (selected area electron diffraction images). In fact, elemental mapping can be used for qualitative (the type of elements) as well as quantitative (the percentage of the concentration of each element of the sample) analysis. In N-H95T5, there are more than four elements. Then, it is recommended to proceed with the quantitative analysis by indicating the concentration of elements by a change in color: from blue (low concentration) to red (high concentration) to green in the middle. Figure 1d-i confirmed the homogenous distribution of nitrogen with T, O, Al, and Si elements. and large surface area with a lattice spacing of 0.350 nm, fully consiste of the (101) crystalline plane of the TiO2 anatase [25,35,36]. Doping wit neither the morphology of N-H95T5 compared with H95T5 [25] nor th ues (selected area electron diffraction images). In fact, elemental mapp qualitative (the type of elements) as well as quantitative (the percenta tion of each element of the sample) analysis. In N-H95T5, there are ments. Then, it is recommended to proceed with the quantitative an the concentration of elements by a change in color: from blue (low co (high concentration) to green in the middle. Figure 1d-i confirmed th tribution of nitrogen with T, O, Al, and Si elements.  Figure S4 shows the samples before and after nitriding. For further analysis of the chemica H95T5 samples, high-resolution spectrum of XPS was used to identif sent in the N-H95T5 nanofibers in three areas. The Ti 2p region near the N 1s region near 400 eV (Figure 2c), and the O 1s region near 530 This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.
N-H95T5 surface structure and chemical state were investigated by XPS. The survey spectrum of H95T5 and N-H95T5 (Figure 2a) contained the dominant signals of Ti 2p and O 1s and the weak signals of C 1s, Al 2p, and Si 2p. In comparison, nitrogen content can be detected in the nitrogen-treated H95T5 sample. Figure S4 shows the N 1s peak for these samples before and after nitriding. For further analysis of the chemical structure of the N-H95T5 samples, high-resolution spectrum of XPS was used to identify the elements present in the N-H95T5 nanofibers in three areas. The Ti 2p region near 460 eV (Figure 2b (Figure 2c). The N 1s binding energy peak was broad, ranging from 396.32 eV to 403.91 eV and centered at 399.82 eV, which was greater than the typical binding energy of 397.2 eV in Ti-N; therefore, Ti-N bonds were excluded and this peak could be assigned to the presence of O-Ti-N linkage [37,38]. In the O 1s spectrum (Figure 2d), the peaks at 529.84 and 531.50 eV were assigned to O-Ti and O-H. respectively. The XPS spectra displayed a third peak that shifted to a higher binding energy (+0.7 eV) after nitriding ( Table 2). This bond could be attributed to the Ti-O-N or O-Ti-N bonds mainly on the surface [39,40]. The XPS spectra confirmed nitrogen distribution into H95T5 nanofibers.
Nitrogen is commonly used for BET surface analysis because of its high purity and strong interaction with most solids. Since the interaction between gas and solid phases is generally weak, the surface is cooled with liquid N 2 to obtain detectable levels of adsorption. A known amount of nitrogen gas is then gradually released into the sample cell. Relative pressures less than atmospheric pressure are achieved by creating conditions of partial vacuum. After the adsorption layer is formed, the sample is removed from the nitrogen atmosphere and heated at room temperature to release the adsorbed nitrogen from the material and quantify it. Heating from −200 • C to 25 • C has no great effect on the surface morphology and architecture of N-H95T5 nanofibers. Figure 3 demonstrates that the N-H95T5 exhibited a type IV isotherm and a type H2 hysteresis loop at a lower relative pressure region [41]. The BET method gave specific surface area values of 36.6 m 2 /g for H95T5 and 67.85 m 2 /g for N-H95T5. This increase was due to nitrogen incorporation in TiO 2 nanofibers and should be beneficial for photocatalytic activity by creating more active adsorption sites [42]. The Ti 2p spectrum (Figure 2b) presented two peaks (458.73 and and Ti 2p1/2). After deconvolution of the N 1s peak, only one peak was ing energy at 399.82 eV (Figure 2c). The N 1s binding energy peak w from 396.32 eV to 403.91 eV and centered at 399.82 eV, which was grea binding energy of 397.2 eV in Ti-N; therefore, Ti-N bonds were excluded be assigned to the presence of O-Ti-N linkage [37,38]. In the O 1s spectr peaks at 529.84 and 531.50 eV were assigned to O-Ti and O-H. respective displayed a third peak that shifted to a higher binding energy (+0.7 eV) a 2). This bond could be attributed to the Ti-O-N or O-Ti-N bonds ma [39,40]. The XPS spectra confirmed nitrogen distribution into H95T5 na

Electrochemical Activity
Electrochemical impedance spectroscopy with the previously describe

Electrochemical Activity
Electrochemical impedance spectroscopy with the previously described model was used to investigate N-H95T5 electrochemical activity [25]. Figure 4a,b show that the impedance arc radius of electrodes in the dark was much bigger than that under visible light irradiation, which indicated that there were few electrons across the electrolyte interfaces in the dark. While under visible light, the arc radius of the N-doped H95T5 electrode was smaller than that of the undoped electrode. This demonstrated that the N-H95T5 displayed greater separation efficiency of photogenerated electron-hole pairs and faster charge transfer than H95T5. It can be seen that the resistance of N-H95t5 decreased by 89% after its exposure to visible light irradiation compared with 85% for H95T5. Charge transfer rate was 10% faster and photogenerated electron-hole separation improved, as indicated by the lower R 2 value (1148 Ω for N-H95T5 versus 1280 Ω for H95T5). This indicates that heat treatment with an inert nitrogen atmosphere is a promising way to improve the efficiency of photocatalyst.
TiO2 nanofibers and should be beneficial for photocatalytic activity by creating more active adsorption sites [42].

Electrochemical Activity
Electrochemical impedance spectroscopy with the previously described model was used to investigate N-H95T5 electrochemical activity [25]. Figure 4a,b show that the impedance arc radius of electrodes in the dark was much bigger than that under visible light irradiation, which indicated that there were few electrons across the electrolyte interfaces in the dark. While under visible light, the arc radius of the N-doped H95T5 electrode was smaller than that of the undoped electrode. This demonstrated that the N-H95T5 displayed greater separation efficiency of photogenerated electron-hole pairs and faster charge transfer than H95T5. It can be seen that the resistance of N-H95t5 decreased by 89% after its exposure to visible light irradiation compared with 85% for H95T5. Charge transfer rate was 10% faster and photogenerated electron-hole separation improved, as indicated by the lower R2 value (1148 Ω for N-H95T5 versus 1280 Ω for H95T5). This indicates that heat treatment with an inert nitrogen atmosphere is a promising way to improve the efficiency of photocatalyst. Then, acetaminophen degradation quantification in the presence of N-H95T5 or H95T5 and visible light for 4.5 h (Figure 5a) showed an increase in photocatalytic efficiency with N-H95T5 at the end of the experiment (83% and 95% of acetaminophen degraded with H95T5 and with N-H95T5, respectively). These data suggest that nitrogen doping increases oxygen vacancies that promote the trapping of photoinduced electrons and that act as a reactive center for photocatalysis [23,43]. Then, acetaminophen degradation quantification in the presence of N-H95T5 or H95T5 and visible light for 4.5 h (Figure 5a) showed an increase in photocatalytic efficiency with N-H95T5 at the end of the experiment (83% and 95% of acetaminophen degraded with H95T5 and with N-H95T5, respectively). These data suggest that nitrogen doping increases oxygen vacancies that promote the trapping of photoinduced electrons and that act as a reactive center for photocatalysis [23,43]. Table 3 summarizes the degradation activity of previously studied photocatalysts for organic pollutants in water, highlighting the comparable performance of N-H95T5.
Acetaminophen degradation kinetics was explained by a pseudo first-order reaction (curve linearity and linear coefficient R 2~1 ) (Figure 5b). The lower electron-hole recombination rate explained the enhanced acetaminophen degradation with N-H95T5.  Table 3 summarizes the degradation activity of previously studied photocatalysts for organic pollutants in water, highlighting the comparable performance of N-H95T5.   After the photocatalysis experiment, the photocatalyst was removed by filtration, washed with deionized water several times, and dried at 80 • C for 12 h, then treated for 15 min at 500 • C to remove all impurities and water molecules. Then, the catalyst was analyzed by BET. The BET method gave specific surface area values of 678,512 m 2 /g for N-H95T5 and 60.8532 m 2 /g for regenerated N-H95T5. A loss of 11% was found but the obtained value remained higher than raw H95T5. N-H95T5 reusability was confirmed by monitoring acetaminophen (10 mg/L, pH 7) degradation over five cycles (same conditions as before). After each experiment, N-H95T5 was filtered, washed in water, and dried (100 • C for 12 h). Acetaminophen degradation reached 95.45% in the first run and then was 93.54%, 92.11%, 86.95%, and 83.04% (Figure 6a). The decrease of 17% after five consecutive cycles could be related to nitrogen loss after each run or to the accumulation of degradation by-products on the catalyst surface that decrease the number of available active sites [48]. Despite this loss of activity after five cycles, N-H95T5 can be considered a promising stable catalyst for industrial applications.
N-H95T5 and 60.8532 m 2 /g for regenerated N-H95T5. A loss of 11% was found but the obtained value remained higher than raw H95T5. N-H95T5 reusability was confirmed by monitoring acetaminophen (10 mg/L, pH 7 degradation over five cycles (same conditions as before). After each experiment, N-H95T5 was filtered, washed in water, and dried (100 °C for 12 h). Acetaminophen degradation reached 95.45% in the first run and then was 93.54%, 92.11%, 86.95%, and 83.04% ( Figure  6a). The decrease of 17% after five consecutive cycles could be related to nitrogen loss afte each run or to the accumulation of degradation by-products on the catalyst surface tha decrease the number of available active sites [48]. Despite this loss of activity after five cycles, N-H95T5 can be considered a promising stable catalyst for industrial applications Then, acetaminophen degradation assays were performed using the same experimental conditions, but in the presence of N-H95T5 and different scavengers (0.06M [49]) to determine what reactive radicals are implicated in this photocatalytic process [49 ( Figure 6b). The addition of isopropanol ( • OH scavenger) strongly decreased the acetaminophen degradation rate. The strongest inhibitory effects were obtained with benzoquinone (O2 •− scavenger) and with EDTA (h + scavenger). Therefore, all three radica types are implicated in acetaminophen photodegradation [17,25].
Lastly, toxicity assays were performed to monitor the formation of harmful by products during acetaminophen degradation. When Vibrio fischeri was incubated with N H95T5 and acetaminophen, its natural fluorescence was inhibited by 27% after 15 min and up to 86% after 2 h of exposure to visible light to induce acetaminophen degradation Then, acetaminophen degradation assays were performed using the same experimental conditions, but in the presence of N-H95T5 and different scavengers (0.06M [49]) to determine what reactive radicals are implicated in this photocatalytic process [49] (Figure 6b). The addition of isopropanol ( • OH scavenger) strongly decreased the acetaminophen degradation rate. The strongest inhibitory effects were obtained with benzoquinone (O 2 •− scavenger) and with EDTA (h + scavenger). Therefore, all three radical types are implicated in acetaminophen photodegradation [17,25].
Lastly, toxicity assays were performed to monitor the formation of harmful byproducts during acetaminophen degradation. When Vibrio fischeri was incubated with N-H95T5 and acetaminophen, its natural fluorescence was inhibited by 27% after 15 min and up to 86% after 2 h of exposure to visible light to induce acetaminophen degradation (Figure 6c). Fluorescence inhibition progressively decreased: 83% at 6 h, 65% at 8 h, 42% at 12 h, 12% at 20 h, and 6% at 24 h. This indicates that after 24 h, the toxic aromatic by-products generated during acetaminophen photodegradation were transformed into nontoxic compounds [33,50,51].

Conclusions
This study describes the fabrication of H95T5 and N-H95T5 composite nanofibers by combining electrospinning and nitriding. After doping with nitrogen, no significant difference was found between H95T5 and N-H95T5 by Raman, FT-IR, XRD, and SEM. In addition, nitrogen was homogeneously distributed with T, O, Al, and Si as proved by elemental mapping. In the XPS survey spectrum, Ti 2p and O 1s were the dominant signals, while C 1s, Al 2p, Si 2p, and N 1s were weaker. The charge transfer rate was faster and Membranes 2023, 13, 204 9 of 11 electron-hole separation was improved by nitrogen doping. Nitrogen doping also enhanced the catalytic properties of the prepared sample, with a degradation rate of 0.0089 min −1 . Recyclability tests were promising, as indicated by the loss of only 17% of activity after five cycles. The scavenging experiments revealed that • OH, h + and O 2 •− were strongly implicated in acetaminophen photodegradation. Toxicity (i.e., V. fischeri fluorescence inhibition) was high in the first 4h of acetaminophen photodegradation in the presence of N-H95T5, due to the generation of aromatic by-products (1,4-benzoquinone, benzoic acid, and benzaldehyde) that were later transformed into nontoxic compounds. This study has confirmed that a cheap photocatalyst with low TiO 2 and nitrogen concentrations can be used for the degradation of organic molecules and has opened prospects for mass production and practical applications.