Visible-Light-Driven Bio-Templated Magnetic Copper Oxide Composite for Heterogeneous Photo-Fenton Degradation of Tetracycline

: The development of a visible-light-driven, reusable, and long-lasting catalyst for the heterogeneous photo-Fenton process is critical for practical application in the treatment of contaminated water. This study focuses on synthesizing a visible-light-driven heterogenous bio-templated magnetic copper oxide composite (Fe 3 O 4 /CuO/C) by a two-step process of bio-templating and hydrothermal processes. The prepared composite was characterized by ﬁeld emission-scanning electron microscope (FE-SEM), Fourier-transform infrared spectroscopy (FTIR), thermogravimetric analysis (TGA), X-ray photoelectron spectroscopy (XPS), electrical impedance spectroscopy (EIS), and vibrating sample magnetometer (VSM). The results reveal that the prepared composite retains the template’s (corn stalk’s) original porous morphology, and a substantial amount of CuO and Fe 3 O 4 particles are loaded onto the surface of the template. The prepared Fe 3 O 4 /CuO/C composite was employed as a catalyst for heterogeneous photo-Fenton degradation of tetracycline (TC) irradiated by visible light. The prepared Fe 3 O 4 /CuO/C catalyst has high efﬁciency towards TC degradation within 60 min across a wide pH range irradiated by visible light, which is attributed to its readily available interfacial boundaries, which signiﬁcantly improves the movement of photoexcited electrons across various components of the prepared composite. The inﬂuence of other parameters such as initial H 2 O 2 concentration, initial concentration of TC, and catalyst dosages was also studied. In addition to high efﬁciency, the prepared catalyst’s performance was sustained after ﬁve cycles, and its recovery is aided by the use of an external magnetic ﬁeld. This research paper highlights the development of a heterogeneous catalyst for the elimination of refractory organic compounds in wastewater. runs. To evaluate the prepared catalyst’s chemical stability, we characterized its residue by XRD, XPS, and FTIR analysis. The resulting Figure 18b–d, showed no discernible change in intensity in the XRD, XPS, and FTIR analysis before and after the ﬁve cycles of reuse of Fe 3 O 4 /CuO/C composite. These results suggest that the prepared composite has good stability and can easily be adapted for water treatment and puriﬁcation.


Introduction
The shortage of clean water supplies is the most pressing challenge causing widespread concern in the world today [1]. The world has been placed under a lot of pressure as several organic chemicals have been recognized as possible emergent contaminants in the environment. Antibiotics are one of the most common evolving contaminants as they are commonly utilized for the treatment of ailments in Man and animals [2]. Tetracyclines (TCs) are among the most extensively utilized antibiotics in human medicine, animal disease management, and agricultural dietary supplements owing to their broad spectrum of action, high quality, and low cost [2][3][4]. Unfortunately, TC is weakly metabolized in the digestive tract, and about 70% of it is excreted by urine and feces leading to increasing TC contamination of water systems and posing a severe threat to ecosystems [5][6][7][8]. As pH value, catalyst dosage, and initial TC concentration were investigated. The prepared catalyst performed excellently in degrading TC between the pH range of 3-9 and exhibit remarkable stability after 5 cycles of consecutive reuse. A plausible reaction mechanism was proposed.

Synthesis of Bio-Templated CuO
Cornstalk was treated with 5% weak aqueous NH 3 solution at 80 • C for 3 h to remove lignin, hemicellulose, and other inherent impurities. The treated cornstalk was filtered and rinsed with distilled water several times before being dried at 60 • C for 24 h. A specific amount of the treated cornstalk was steeped in a solution of CuSO 4 •5H 2 O (80 mM) followed by the addition of NaOH (10 mM) and kept at 60 • C for 24 h. The cornstalk-loaded CuO was washed several times with distilled water after 24 h and then dried at 60 • C. The dried prepared cornstalk-loaded CuO was calcinated at 550 • C for 5 h in a tube furnace heating at a rate of 1 • C/min under a nitrogen atmosphere. The resultant product of this stage is a bio-templated copper oxide (CuO/C).

Synthesis of Bio-Templated Magnetic Copper Oxide (Fe 3 O 4 /CuO/C)
The hydrothermal procedure was employed for the preparation of Fe 3 O 4 /CuO/C. In brief, a suitable amount of the prepared bio-templated copper oxide, FeSO4.7H 2 O (5.0 mM), and Na 2 S 2 O 3 ·•5H 2 O (5.0 mM) were dispersed in 30 mL of distilled water. The mixture was agitated for 10 min, before adding NaOH (10 mM) solution in drops. The mixture was then agitated for a further 5 min, before being placed into a 50 mL Teflon-lined autoclave. After sealing the Teflon-line autoclave, the temperature was raised to 140 • C for 12 h before being allowed to cool naturally to ambient temperature. After washing with ethanol and distilled water, the resultant residue was dried at 60 • C. The composition of the prepared bio-templated copper oxide (CuO/C) and FeSO 4 •7H 2 O were varied in the ratio of 1:1, 2:1, 1:2, and 1:3, respectively. The procedure was repeated for the synthesis of Fe 3 O 4 without the prepared bio-templated copper oxide.

Characterization
Analyzes were carried out using an XL-30 ESEM FEG scanning electron microscope and energy dispersive spectrometry (EDS). The crystal phase of all the composites was studied using X-ray diffraction (XRD) on a Rigaku D/Max 2550 diffract meter (Rigaku Corporation, Tokyo, Japan) with a Cu-k radiation source (k = 1.54056) at a 10 • C min −1 scan rate. The composite's elemental analysis was undertaken by X-ray photoelectron spectroscopy (XPS, ESCA LAB 220-XL, Al Kα radiation). Thermogravimetric analysis (TGA) of the samples was undertaken in air on an SDT Q600 thermal gravimetric analyzer (TA Instruments, USA) from room temperature to 800 • C at a 10 • C min −1 ramping rate. For the identification of functional groups in the region of 4000-400 cm −1 , Fourier transform infrared (FTIR) spectroscopy (Nexus 670, Nicolet, USA) was used. The prepared catalyst's electrochemical properties were determined on an electrical workstation with a threeelectrode cell system (CHI660E Shanghai Chenhua Instrument Co., Ltd., Shanghai, China). The Ag/AgCl served as the reference electrode, Pt wire served as the counter electrode, and Fe 3 O 4 and Fe 3 O 4 /CuO/C catalyst coated on flourine-tine oxide (FTO) conductive glass served as the working electrode. The electrochemical impedance spectroscopy was recorded at an open circuit potential of 0.1 m/l K 4 Fe(CN) 6 electrolyte solution at a frequency range of 1 × 105 to 102 Hz and an amplitude of 5 mA to analyze the conductivity of the prepared working electrode. A Lakeshore 735 vibrating sample magnetometer (VSM) was used to evaluate the magnetic properties of the prepared catalyst at ambient temperature.

Catalytic Performance of Prepared Catalyst on Tetracycline (TC)
The catalytic performance of the prepared Fe 3 O 4 /CuO/C catalyst was assessed by degrading an aqueous TC solution under visible light irradiation (λ = 420 nm). By putting 0.3 g of Fe 3 O 4 /CuO/C into 100 mL of TC aqueous solution (50 mg/L) in a glass beaker with continuous agitation, the reactions were carried out. To achieve a TC-catalyst adsorptiondesorption equilibrium, the aqueous suspension was agitated for 30 min in the dark. The catalytic heterogeneous photo-Fenton reaction was initiated by adding a specified amount of H 2 O 2 to the reaction vessel while simultaneously putting on a 100W Xe lamp (λ = 420 nm) for visible light irradiation. Approximately 5 mL aliquots were collected at a set interval and filtered immediately with 0.45 µm membrane filters. An ultraviolet-visible (UV-Vis) spectrophotometer was used to determine the concentration of residue TC by measuring its absorbance at 357 nm (SOPTOP 757). After the reaction, the catalyst was retrieved using an external magnetic field, rinsed with deionized water, and dried for 2 h at 60 • C before being recycled. Equation (1) was used to calculate the rate of TC degradation in each procedure using the synthesized Fe 3 O 4 /CuO/C catalyst. The experiments were carried out in triplicate, with the average results being presented. We retested the efficiency of the prepared Fe 3 O 4 /CuO/C catalyst for five consecutive cycles to assess its recyclability and stability.
where C 0 was the initial TC concentration at time t = 0, and C t was the concentration of TC at time t.  [48]. The diffraction peak of Cu in the Fe 3 O 4 /CuO/C composite spectrum is similarly stronger compared to that of Fe, implying that metallic copper possesses a high crystallinity [34,37].

Fourier Transform Infrared (FTIR) Spectroscopy Analysis
The FTIR spectrum of Fe 3 O 4 , CuO/C, and Fe 3 O 4 /CuO/C (2:1) are shown in Figure 2. Three essential bands of interest in the FTIR spectrum of Fe 3 O 4 nanoparticles are 585, 1637, and 3413.73 cm −1 , corresponding to Fe-O bonds in the crystal lattice of Fe 3 O 4 , C-H, and O-H bonds, respectively. The C-H and O-H bonds' appearance in this spectral is probably due to the use of ethanol and water during synthesis; this observation was corroborated and reported by previous researchers [49,50]. The FTIR spectrum of CuO/C has a sharp band at around 681 cm −1 , which is associated mostly with Cu-O stretching mode, 1619 cm −1 , related to the C-H band, confirming carbon's presence serving as the organic skeleton supporting the CuO nanoparticles [51,52]. The broadband at around 3313.31 cm −1 is due to stretching O-H molecules present in the CuO/C [53]. The spectrum of Fe 3 O 4 /CuO/C consists of the prominent bands of 581, 675, 1631, and 3500-3100 cm −1 corresponding to Fe-O of Fe 3 O 4 , Cu-O of CuO, C-H of bio-templated CuO, and O-H of inherent and adsorbed water molecules, respectively [47,53]. Thus, the formation of Fe 3 O 4 /CuO/C composite is also confirmed from the FTIR analysis.

Thermogravimetric (TG) Analysis
To determine the presence of carbon in the prepared Fe 3 O 4 /CuO/C catalyst, TG analysis was carried out in the air. The weight loss between room temperature and 280 • C is attributable to the elimination of absorbed water in Figure 3. The weight gain from around 280 and 335 • C is attributable to the oxidation of both CuO and Fe 3 O 4 [34,37]. After 335 • C, the weight loss may be attributable to carbon oxidation, confirming the existence of carbon in the prepared catalyst matrix [54].

Field Emission-Scanning Electron Microscopy (FE-SEM) Analysis
The morphology of the catalyst Fe 3 O 4 /CuO/C was examined by field emissionscanning electron microscopy (FE-SEM). Figure 4a shows that the CuO particles are dense and closely packed compare to CuO/C ( Figure 4b) which has been loosened because of the presence of carbon template; while Fe 3 O 4 (Figure 4c) are agglomerated particles with an irregular shape. Figure 4d, Fe 3 O 4 /CuO/C (2:1), shows that the prepared catalyst retains the cornstalk hierarchical porous organic structure after the template was removed.
The template creates a porous structure that favours particle interconnectivity, prevents photo-induced electron/hole recombination, and enhances TC molecules' adsorption.

Energy-Dispersive Spectrometry (EDS) Analysis
The EDS analysis confirms the elemental composition of the prepared Fe 3 O 4 /CuO/C catalyst. Figure 5 shows that Cu, Fe, C, and O are the main constituents of Fe 3 O 4 /CuO/C with different weight percentages of 11.68%, 19.51%, 42.24%, and 23.03%, respectively. This confirms the successful synthesis of the Fe 3 O 4 /CuO/C composite.  (Figure 6b). The high-resolution peaks appearing at 933.7 and 953.4 eV assigned to Cu 2p 3/2 and Cu 2p 1/2 reveal that the oxidation state of Cu is Cu 2+ in the prepared Fe 3 O 4 /CuO/C catalyst [35,55]. Also, two satellite peaks were observed at 942 and 962.2 eV, which correspond to the d9 configuration of Cu 2+ confirming the existence of CuO; this observation was similar to what has been reported previously [28,56]. Figure 6c shows a typical Fe 2p spectrum with two notable peaks at 711.2 and 724.5 eV, which correspond to the Fe 2p3/2 and Fe 2p1/2 spin-orbit split doublets, respectively; the results are consistent with earlier studies [34,57]. Moreover, when we compared the position of Cu 2p (Figure 6b) and Fe 2p (Figure 6c) in their pure and composite form, there was a redshift in binding energy, which really is advantageous for light excitation. Furthermore, the XPS deconvoluted spectrum of O1s with peaks at 533.3, 531.1, and 530.4 eV (Figure 6d) are the binding energy of lattice and adsorbed oxygen elements respectively [28,34,56]. The binding energies of the two peaks in the deconvoluted C1s XPS spectra of Fe 3 O 4 /CuO/C are 288.7 and 284.1 eV, respectively, corresponding to C-O, and C-C, with the conspicuous peak at 284.8 eV typically attributable to elemental carbon [58]. This means that some quantity of carbon from cornstalk has been doped into the CuO lattice, making it charge transfer sensitive when exposed to light [59,60]. Finally, the standard reference carbon may be assigned to the C 1s peak in the survey XPS spectra of Fe 3 O 4 [61].

Electrochemical Impedance Spectroscopy (EIS) Analysis
The electrochemical characteristics of the produced catalyst were measured using electrochemical impedance spectroscopy (EIS). A semicircle in the high-frequency region and a sloping line in the low-frequency region make up the Nyquist plot of the constructed electrodes (Figure 7). Compared to Fe 3 O 4 , the arc radius of the Fe 3 O 4 /CuO/C electrode has a low resistivity, implying that the prepared Fe 3 O 4 /CuO/C catalyst has a reduced electrochemical resistance, translating to a faster rate of electron transfer from Fe 3+ to Fe 2+ , thus contributing to a reduction in the associated energy consumption rate compared to Fe 3 O 4 , leading to an improved reaction process for the heterogeneous photo Fenton catalyst. value is approx. 48.7 emu/g, while the composite of Fe 3 O 4 /CuO/C is approx. 33.2 emu/g, suggesting that the prepared catalyst is superparamagnetic. The reduction in magnetic properties could be attributed to the presence of non-magnetic bio-templated CuO [20,25]. Following separation and manipulation by an external magnetic field, Fe 3 O 4 /CuO/C can also be easily re-dispersed for reuse, as seen in the inset of Figure 8. As a result, it can improve the Fe 3 O 4 /CuO/C catalyst's separation, recovery, and reusability.

The Catalytic Activities of Different Catalyst Systems
Evaluation of the synergistic effect of the synthesized catalysts on TC degradation was performed under visible light irradiation at pH 7 for 60 min and at a temperature of 25 • C. The result indicated that visible light does not have any significant impact on the degradation of the TC (Figure 9a). The combined effect of visible light-H 2 O 2 achieved 11.5% TC removal which is due to the slight photolysis of H 2 O 2 under the influence of solar light as the reaction vessel was insulated from the influence of solar light previous researchers have reported the instability of H 2 O 2 under the solar spectrum [62,63], while visible light-Fe 3 O 4 /H 2 O 2 (photo Fenton) achieved 76.3% degradation, and the combined impact of visible light-CuO (photocatalyst) could achieve 57% TC removal (Figure 9a). TC degradation with the visible light/Fe 3 O 4 /CuO/C/H 2 O 2 (heterogeneous photo Fenton) system achieved 96.1% TC removal. When the photo-Fenton activity of Fe 3 O 4 was compared to that of Fe 3 O 4 /CuO/C, the result showed that the performance of the latter was better than the former by 18.8% (Figure 9a). This result suggests that the interface binary complex of Fe 3 O 4 /CuO/C heterojunction contributed significantly to the heterogeneous photo-Fenton reaction. Therefore, we infer that the Fe 3 O 4 /CuO/C heterojunction interfaces significantly influence the photo-Fenton reaction using Fe 3 O 4 /CuO/C as the catalyst, which agrees with what has been previously reported [8,20]. From Figure 9b

Influence of Catalyst Dosage
The influence of the initial dosage of catalyst on TC's degradation was evaluated as shown in Figure 10. According to Figure 10, only around 17% of the TC molecules could be adsorbed onto the Fe 3 O 4 /CuO/C catalyst surface during the adsorption-desorption stage, implying that the adsorption influence was inadequate. At the degradation stage, there was an increase in TC degradation efficiency as the dosage of the Fe 3 O 4 /CuO/C catalyst was increased from 0.1 to 0.3 g, but a slight decrease was observed as the dosage was further increased to 0.5 g. The possible reason for this decrease is that as the initial dosage of the catalyst exceeds 0.3 g, the particles of catalyst in the resultant aqueous suspension impede the light penetration and decrease the usage of light, resulting in decreased TC degradation efficiency. Other possible reasons are the agglomeration of catalyst particles and the excessive initial dosages of Fe 3 O 4 /CuO/C may act as a scavenger for the free radicals generated [64,65].

Influence of Initial Concentration of H 2 O 2
The influence of the initial concentration of H 2 O 2 was examined by using different concentrations of H 2 O 2 in the Fe 3 O 4 /CuO/C-H 2 O 2 -visible light system. Figure 11 shows that increasing the H 2 O 2 concentration from 15 to 30 mM improves the degradation efficiency somewhat. The efficiency of TC degradation decreases when the initial H 2 O 2 concentration rises from 30 to 45 mM. The decrease in TC degradation efficiency with a rise in the concentration of H 2 O 2 is attributable to the fact that H 2 O 2 acts as a scavenger at a higher concentration, thereby decreasing the number of active radicals responsible for oxidizing TC molecules. Also at a higher concentration, H 2 O 2 could generate HOO • , which has a lower oxidation potential than hydroxyl radical ( • OH), thus inhibiting the efficiency of the process for TC degradation [66]. However, these results revealed that the initial concentration of H 2 O 2 had minimal influence on the overall degradation rate because of the H 2 O 2 low radical scavenging effect attributable to the low initial concentration of the H 2 O 2 that was used and the quick reaction between the TC molecule and the OH radicals generated. Therefore, an optimal initial concentration of H 2 O 2 adequate in this work H 2 O 2 is 30 mM.

Influence of Initial pH
In TC degradation, the influence of the initial pH value cannot be overlooked [2,67]. Figure 12 depicts the effect of the initial pH value on the efficiency of TC degradation in this study. The maximum degradation efficiency of 96.1% was achieved at pH 3 and 7 while at pH 9 the efficiency of degradation is 91.4% which is also high. The results show that Fe 3 O 4 /CuO/C composites can function over a wide range of pH. This effectiveness may be attributed to the development of acidic intermediates, even if the starting pH is 7.0 or 9.0, the final pH is retained at an acidic medium under pH of 4.0 after 60 min [20,68]. As a result of this research, Fe 3 O 4 /CuO/C composites catalyzing photo-Fenton TC degradation have outstanding photodegradation efficiency over a wide initial pH value range.

Influence of Initial TC Concentration
At a pH of 7 and a catalyst dosage of 0.3 g/L, the effects of initial TC concentration on photo-Fenton degradation were examined. The proportion of TC eliminated decreased from 96.1% to 83.3% when the starting concentration of TC was increased from 50 to 100 mg/L. (Figure 13). As the TC initial concentration increases from 50 to 75 and 100 mg/L, the number of TC molecules available for absorption on the Fe 3 O 4 /CuO/C surface increases too, thus reducing the production of oxidant radicals and hence the photodegradation reaction. The decline in TC removal percentage could be due to the non-availability of active sites on the catalyst at higher TC concentrations, increasing the number of un-adsorbed TC molecules, because there was no commensurate rise in catalyst mass as TC concentration increased. An increased initial TC concentration reduces light irradiation penetration (screen effect), reduces • OH production; therefore, its photo-Fenton degradation efficiency is ultimately reduced.

Kinetic Studies
Kinetic studies of TC degradation reactions exhibited a pseudo-first-order kinetic model concerning the irradiation time ( Figure 14 and Table 1), with the rate of degradation given by Equation (2): where C 0 (mg/L) is the initial TC concentration at time t = 0, C t (mg/L) is the TC concentration at time t, and k is the rate constant (min −1 ).  When the molar ratios of Fe 3 O 4 and CuO/C were varied with irradiation time, a linear relationship for TC degradation was obtained ( Figure 14). The slope and intercepts of these plots were used to calculate the rate constants, as indicated in Table 1. Table 2 shows an assessment of the catalytic activity of various catalysts utilized by various researchers for tetracycline degradation.

Influence of Different Radical Scavengers and Mechanism of Degradation
To figure out the main reactive species involved in this heterogeneous photo-Fenton process, a series of radical scavengers were introduced into the reaction vessel. The scavengers employed were TBA (5 mM) for • OH, BQ (5 mM) for superoxide radical O 2 •− (5 mM) [29], AgNO3 (5 mM) for photo-generated electrons (e − ), and formic acid for photogenerated holes (h + ) [25]. As shown in Figure 15, the photo-Fenton degradation of TC is partially suppressed after the addition of the radical scavengers. The photo-degradation reaction's suppression is in the following order BQ > TBA > AgNO 3 >Formic acid. Thus, it was deduced that • OH, O 2 •− , e − , and h + oxidizing species were photo-generated onto the surface of Fe 3 O 4 /CuO/C. Therefore, in this heterogeneous photo-Fenton system of TC degradation, the O 2 •− , • OH, e − , and h + play an essential role in the catalytic photo-Fenton reaction. The plausible mechanism of photo-Fenton activity of the Fe 3 O 4 /CuO/C in this combined system is based on the above results of radical scavenging as illustrated in Figure 15.  (4)). The photogenerated electrons at the conduction band combine with oxygen (O 2 ) to produce • O 2 − (Equation (5)) and promote regeneration of Fe 2+ from Fe 3+ (Equation (7)). Fe 2+ combines with H 2 O 2 to generate • OH (Equation (6)) while the adsorbed water is also photo catalytically oxidized into • OH by photogenerated holes in the valence band (Equation (8)). Therefore, we conclude that as it can be seen in Figure 16 the degradation of TC into CO 2 and H 2 O (Equation (9)) is a result of the synergistic effect between • OH generated in Equations (4), (6) and (9) with O 2 •− in Equation (5). Compared with the homogeneous Fenton process, photo-induced electrons from the CuO surface enhances the Fe 3+ /Fe 2+ cycle efficiency (Equation (6)). The Fe 3+ release from the heterogeneous photo-Fenton process performs the role of an electron acceptor which hinders photo-generated electron-hole recombination, taking advantage of the well-suited interfacial interaction between Fe 3 O 4 /CuO/C heterostructure. The carbon content in the prepared composite affords an enhanced microenvironment for trapping TC molecules from the aqueous solution. Thus, the heterojunction of Fe 3 O 4 /CuO/C com-posite improves the heterogeneous photo-Fenton performance process by enhancing the separation and use of photo-induced carriers.

Mineralization Study for TC Degradation
After each treatment, the residual total organic carbon (TOC) content was measured to conduct the mineralization investigation for each of the catalyst systems. According to the results presented in Figure 17 Figure 17.

Reusability and Stability Test
The recoverability, reusability, and stability of a catalyst is an essential consideration in adopting the catalyst for practical application. For reusability, the prepared Fe 3 O 4 /CuO/C (2:1) composite was recycled five consecutive times, and there was not much loss of photocatalytic performance, as shown in Figure 18a, as 93.5% TC degradation was achieved after five runs. To evaluate the prepared catalyst's chemical stability, we characterized its residue by XRD, XPS, and FTIR analysis. The resulting Figure 18b-d, showed no discernible change in intensity in the XRD, XPS, and FTIR analysis before and after the five cycles of reuse of Fe 3 O 4 /CuO/C composite. These results suggest that the prepared composite has good stability and can easily be adapted for water treatment and purification.

Conclusions
In conclusion, a visible-light-driven heterogeneous bio-templated magnetic copper oxide composite was created by a two-step of bio-templating and hydrothermal method. The produced Fe 3 O 4 /CuO/C catalyst composition was verified by XRD, SEM, EDS, FTIR, TGA, EIS, and VSM. The prepared catalyst performed excellently in photo-Fenton catalytic degradation of tetracycline (TC) over a wide range of pH using visible light as an irradiation source. A microenvironment was generated by the presence of cornstalk in the prepared catalyst composite to adsorb TC molecules, and the Fe 3+ /Fe 2+ cycle efficiency was improved by interfacial interactions between Fe 3 O 4 and CuO. Quite significantly, even after 5 cycles, the photo-Fenton activity remained high, indicating that the catalyst is highly stable and reusable. Using an external magnetic field, the catalyst was also easily recovered from the reaction medium. Because of its excellent performance in the visible region of the solar spectrum, Fe 3 O 4 /CuO/C is a promising catalyst for heterogeneous photo-Fenton TC removal under visible light irradiation, maximizing the potential of solar energy and being more environmentally benign.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy reasons.