Novel Sodium Alginate/Polyvinylpyrrolidone/TiO 2 Nanocomposite for Efﬁcient Removal of Cationic Dye from Aqueous Solution

Featured Application: Novel SA/PVP/TiO 2 nanocomposite beads can be used as an effective, eco-friendly adsorbent-photocatalyst for treating cationic dye-contaminated wastewaters. Abstract: The combination of adsorption and photodegradation processes is an effective technique for the removal of dye contaminants from water, which is motivating the development of novel adsorbent-photocatalyst materials for wastewater treatment. Herein, novel nanocomposite porous beads were developed using titanium dioxide (TiO 2 ) nanotubes embedded in a sodium alginate (SA)/polyvinylpyrrolidone (PVP) matrix using calcium chloride solution as a crosslinker. The prepared nanocomposite beads’ performance was examined as an adsorbent-photocatalyst for the breakdown of methylene blue in aqueous solutions. Several operation factors inﬂuencing the dye decomposition process, including photocatalyst dosage, illumination time, light intensity, and stability were investigated. The ﬁndings demonstrated that the removal activity of the beads changed with the TiO 2 weight ratio in the composite. It was found that SA/PVP/TiO 2 -3 nanocomposite beads presented the greatest deterioration efﬁciency for methylene blue dye (98.9%). The cycling ability and reusability of the prepared SA/PVP/TiO 2 nanocomposite beads recommend their use as efﬁcient, eco-friendly materials for the treatment of wastewaters contaminated with cationic dyes. 2 nanocomposites was investigated in a dark environment to assess the level of MB dye adsorption in the beads. These data were used to evaluate the photocatalytic activity of SA/PVP/TiO 2 nanocomposites for eliminating MB dye in the presence of visible light. The experiments were carried out using 1 g L − 1 of


Introduction
The release of dye-contaminated wastewaters from different industries [1] into the environment leads to dramatic effects on the living life of our planet, as dyes and their sub-products are often toxic or mutagenic agents [2]. Different techniques are used to treat waters polluted with dyes, such as adsorption [3,4], photocatalysis [5], biological methods [6,7], coagulation, and flocculation [8].
Recently, photocatalysis degradation has been used widely to remove several organic [9] and inorganic [10] wastewater contaminants by transforming them into nonhazardous materials. One of the semiconductor materials most commonly utilized as a photocatalyst to remove various contaminants from wastewaters is titanium dioxide (TiO 2 ), which was studied as a photocatalyst for the first time in 1972 by Fujishima and Honda [11]. TiO 2 is a non-toxic and bio-friendly material, chemically stable, photostable, commercially Figure 1 shows the possible structure of SA/PVP/TiO 2 nanocomposite beads, where the two polymers are crosslinked through acid-base interaction between carboxylic groups of SA and amine groups of PVP, in addition to the hydrogen bonds formed between the oxygenated groups of the polymers' molecules and the TiO 2 doping agent.

Preparation of Polymeric Beads
Each polymer was individually dissolved at 25 °C in deionized water and mixe 2 h in a mixture comprising 90 wt.% SA, 9-7 wt.% PVP, and 1-3 wt.% TiO2 nanotu prepared from previous work [23], to form homogenous solutions. Using a syringe polymer mixture was then added dropwise into a 2% (w/v) CaCl2 solution. After sha the beads, they were rinsed three times with distilled water. Figure 1 shows the possible structure of SA/PVP/TiO2 nanocomposite beads, w the two polymers are crosslinked through acid-base interaction between carbo groups of SA and amine groups of PVP, in addition to the hydrogen bonds formed tween the oxygenated groups of the polymers' molecules and the TiO2 doping agent

Characterization
The crystallographic phases of the produced samples were determined by X powder diffraction (XRD, Shimadzu-7000, Japan). The elements were identified u transmission electron microscopy (TEM, JEM-2100 plus) and scanning electron mi copy (SU-70, Hitachi, Japan) in combination with energy-dispersive X-ray spectros (EDS). A Bruker ALPHA spectrometer (Bruker Corporation, Rheinstetten, Germany) used to perform the Fourier transform infrared (FTIR) study.

Photocatalytic Decay of Methylene Blue
Under illumination of an unfiltered commercial LED visible light, specifically tw W lamps with 1200 lm (Bareeq, Egypt), the photocatalytic degradation of MB dye assessed using two loading ratios of doping agent in the SA/PVP polymer matrix. T cally, 1 g L −1 of the SA/PVP/TiO2 nanocomposite beads was suspended in MB dye solu model wastewater. The suspension was agitated at 25 °C using magnetic stirring u visible light, and samples were taken at regular intervals. A UV-vis spectrophotom (Shimadzu UV-2600, Japan) was used to evaluate the residual MB concentration aft radiation by sampling 3 mL of the reaction mixture at the wavelength of 665 nm. photocatalytic decay of MB was calculated by means of Equation (1), where C0 and C stand for the initial and final MB dye concentration, respectively.

Characterization
The crystallographic phases of the produced samples were determined by X-ray powder diffraction (XRD, Shimadzu-7000, Kyoto, Japan). The elements were identified using transmission electron microscopy (TEM, JEM-2100 plus) and scanning electron microscopy (SU-70, Hitachi, Japan) in combination with energy-dispersive X-ray spectroscopy (EDS). A Bruker ALPHA spectrometer (Bruker Corporation, Rheinstetten, Germany) was used to perform the Fourier transform infrared (FTIR) study.

Photocatalytic Decay of Methylene Blue
Under illumination of an unfiltered commercial LED visible light, specifically two 12 W lamps with 1200 lm (Bareeq, Egypt), the photocatalytic degradation of MB dye was assessed using two loading ratios of doping agent in the SA/PVP polymer matrix. Typically, 1 g L −1 of the SA/PVP/TiO 2 nanocomposite beads was suspended in MB dye solution model wastewater. The suspension was agitated at 25 • C using magnetic stirring under visible light, and samples were taken at regular intervals. A UV-vis spectrophotometer (Shimadzu UV-2600, Kyoto, Japan) was used to evaluate the residual MB concentration after irradiation by sampling 3 mL of the reaction mixture at the wavelength of 665 nm. The photocatalytic decay of MB was calculated by means of Equation (1), where C 0 and C stand for the initial and final MB dye concentration, respectively. The photocatalytic efficiency of the produced nanocomposite beads on MB dye degradation was investigated at a pH of 7. This pH value is the most suited for photocatalytic degradation [24] and was set by adding either 0.1 M NaOH or 0.1 M HCl solution.

Kinetic Models
The most widely used kinetic models are the pseudo-first order and pseudo-second order models [1]. The pseudo-first order model is given by Equation (2), where q e and q t represent the MB adsorption capacity (mg g −1 ) (i.e., the quantity of MB dye adsorbed or degraded) at equilibrium and at time t (s), respectively, and K 1 represents the rate constant (s −1 ). The plots of log (q e − q t ) vs. t enable determining the K 1 values. Equation (3) describes the pseudo-second order model, where K 2 (g mg −1 s −1 ) represents the rate constant of the pseudo-second order model. When pseudo-second order kinetics is adopted, the plot of t/q t vs. t should show a linear relation. The slope and intercept of the plot can be used to obtain q e and K 2 , respectively.

Characterization of SA/PVP/TiO 2 Nanocomposite
The TEM image of TiO 2 shown in Figure 2a proves the formation of the tubular shape with nanoscale size. After the TiO 2 nanotubes were embedded in the blended SA/PVP polymers, SA/PVP/TiO 2 nanocomposite beads were obtained, as confirmed from EDS data ( Figure 2b). As shown in the SEM micrograph of Figure 2c, the beads presented a rough and wrinkled morphology. By taking a higher magnification SEM image, a good dispersion of the TiO 2 nanotubes was observed with visible pores in the polymer matrix ( Figure 2d). The photocatalytic efficiency of the produced nanocomposite beads on MB dye degradation was investigated at a pH of 7. This pH value is the most suited for photocatalytic degradation [24] and was set by adding either 0.1 M NaOH or 0.1 M HCl solution.

Kinetic Models
The most widely used kinetic models are the pseudo-first order and pseudo-second order models [1]. The pseudo-first order model is given by Equation (2), log log    (2) where qe and qt represent the MB adsorption capacity (mg g −1 ) (i.e., the quantity of MB dye adsorbed or degraded) at equilibrium and at time t (s), respectively, and K1 represents the rate constant (s −1 ). The plots of log (qe − qt) vs. t enable determining the K1 values. Equation (3) describes the pseudo-second order model, where K2 (g mg −1 s −1 ) represents the rate constant of the pseudo-second order model. When pseudo-second order kinetics is adopted, the plot of t/qt vs. t should show a linear relation. The slope and intercept of the plot can be used to obtain qe and K2, respectively.

Characterization of SA/PVP/TiO2 Nanocomposite
The TEM image of TiO2 shown in Figure 2a proves the formation of the tubular shape with nanoscale size. After the TiO2 nanotubes were embedded in the blended SA/PVP polymers, SA/PVP/TiO2 nanocomposite beads were obtained, as confirmed from EDS data ( Figure 2b). As shown in the SEM micrograph of Figure 2c, the beads presented a rough and wrinkled morphology. By taking a higher magnification SEM image, a good dispersion of the TiO2 nanotubes was observed with visible pores in the polymer matrix ( Figure 2d). The FTIR spectra of TiO 2 nanotubes and SA/PVP/TiO 2 nanocomposite are shown in Figure 3. The band at about 500 cm −1 for the TiO 2 nanotube seen in Figure 3a is characteristic of Ti-O stretching vibration modes. The FTIR spectra of SA/PVP/TiO 2 nanocomposite samples exhibit bands around 1600 cm −1 assigned to O-H stretching mode, as well as absorption bands at 1419 cm −1 ascribed to COO symmetric stretching vibration in SA. The band at 1030 cm −1 corresponds to C-O stretching [25], the band at 2178 cm −1 is related to PVP's C-N bond stretching vibration, and the band located at 2170-2300 cm −1 represents the polymers' C-H bonds' bending vibration [22]. The FTIR spectra of TiO2 nanotubes and SA/PVP/TiO2 nanocomposite are shown in Figure 3. The band at about 500 cm −1 for the TiO2 nanotube seen in Figure 3a is characteristic of Ti-O stretching vibration modes. The FTIR spectra of SA/PVP/TiO2 nanocomposite samples exhibit bands around 1600 cm −1 assigned to O-H stretching mode, as well as absorption bands at 1419 cm −1 ascribed to COO symmetric stretching vibration in SA. The band at 1030 cm −1 corresponds to C-O stretching [25], the band at 2178 cm −1 is related to PVP's C-N bond stretching vibration, and the band located at 2170-2300 cm −1 represents the polymers' C-H bonds' bending vibration [22].
The XRD patterns in Figure 3b show the crystalline features of TiO2 nanotubes, with characteristic peaks at 2θ values of 28, 36, 41, and 54, whereas characteristic spectra of an amorphous structure are obtained for the prepared beads. The amorphous nature of the nanocomposites is related to the low Ti content (e.g., 2.7 wt.% Ti in SA/PVP/TiO2-3), as determined by EDS analysis.

Effect of TiO2 Amount in the SA/PVP Matrix
As the catalyst loading in the SA/PVP/TiO2 nanocomposite has a key role in dye decay efficiency, the effect of the photocatalyst concentration on MB degradation was investigated by increasing the TiO2 amount in the SA/PVP matrix from 1 to 5 wt.%. As seen in Figure 4, the decay efficiency rose when the TiO2 concentration increased from 1 to 3 wt.%, which can be justified by the fact that at low concentrations, more porous empty sites and polymer functional groups, such as COO, are accessible on the beads' external surface to absorb cationic dye molecules via electrostatic attraction. However, the active sites available for the photocatalytic reaction are limited. Thus, by increasing the catalyst loading to 3 wt.%, additional active sites become available for the photocatalytic reaction. This leads to an increase in the hydroxyl ions' adsorption onto the surface of the beads to produce OH • radicals. On the other hand, the photocatalytic activity decreased at a high concentration of the catalyst, as it hampers the dye decay rate as a result of light penetration shortage inside the beads. A second possibility is the agglomeration of the catalyst nanoparticles, resulting in a decrease in the operative surface area of the catalyst, and consequently, a decrease in the decolorization efficiency. The XRD patterns in Figure 3b show the crystalline features of TiO 2 nanotubes, with characteristic peaks at 2θ values of 28, 36, 41, and 54, whereas characteristic spectra of an amorphous structure are obtained for the prepared beads. The amorphous nature of the nanocomposites is related to the low Ti content (e.g., 2.7 wt.% Ti in SA/PVP/TiO 2 -3), as determined by EDS analysis.

Effect of TiO 2 Amount in the SA/PVP Matrix
As the catalyst loading in the SA/PVP/TiO 2 nanocomposite has a key role in dye decay efficiency, the effect of the photocatalyst concentration on MB degradation was investigated by increasing the TiO 2 amount in the SA/PVP matrix from 1 to 5 wt.%. As seen in Figure 4, the decay efficiency rose when the TiO 2 concentration increased from 1 to 3 wt.%, which can be justified by the fact that at low concentrations, more porous empty sites and polymer functional groups, such as COO, are accessible on the beads' external surface to absorb cationic dye molecules via electrostatic attraction. However, the active sites available for the photocatalytic reaction are limited. Thus, by increasing the catalyst loading to 3 wt.%, additional active sites become available for the photocatalytic reaction. This leads to an increase in the hydroxyl ions' adsorption onto the surface of the beads to produce OH • radicals. On the other hand, the photocatalytic activity decreased at a high concentration of the catalyst, as it hampers the dye decay rate as a result of light penetration shortage inside the beads. A second possibility is the agglomeration of the catalyst nanoparticles, resulting in a decrease in the operative surface area of the catalyst, and consequently, a decrease in the decolorization efficiency.

Effect of Illumination Time on the Decay of MB
The activity of the prepared SA/PVP/TiO 2 nanocomposites was investigated in a dark environment to assess the level of MB dye adsorption in the beads. These data were used to evaluate the photocatalytic activity of SA/PVP/TiO 2 nanocomposites for eliminating MB dye in the presence of visible light. The experiments were carried out using 1 g L −1 of one of the two studied concentrations of doping agent (1 and 3 wt.% of TiO 2 , respectively for SA/PVP/TiO 2 -1 and SA/PVP/TiO 2 -3 nanocomposite beads) in a 500 mL solution containing 50 mg L −1 of MB dye at pH 7. The analysis was done at different time intervals in the dark and under visible light. As illustrated in Table 1, the dark adsorption increased with time and stabilized after 40 min, indicating that the active site and porosity of the SA/PVP blended polymer were saturated with MB molecules. Furthermore, carboxylic groups are the prevalent functional groups in the SA polymer, aiding in the adsorption of the cationic dye molecules.

Effect of Illumination Time on the Decay of MB
The activity of the prepared SA/PVP/TiO2 nanocomposites was investigat environment to assess the level of MB dye adsorption in the beads. These data to evaluate the photocatalytic activity of SA/PVP/TiO2 nanocomposites for elim dye in the presence of visible light. The experiments were carried out using 1 of the two studied concentrations of doping agent (1 and 3 wt.% of TiO2, resp SA/PVP/TiO2-1 and SA/PVP/TiO2-3 nanocomposite beads) in a 500 mL soluti ing 50 mg L −1 of MB dye at pH 7. The analysis was done at different time inte dark and under visible light. As illustrated in Table 1, the dark adsorption incr time and stabilized after 40 min, indicating that the active site and porosity of t blended polymer were saturated with MB molecules. Furthermore, carboxylic the prevalent functional groups in the SA polymer, aiding in the adsorption of dye molecules.   However, under visible light, as can be seen in Figure 5, the results reveal that the photocatalytic activity increased with the duration of the illumination period, with the SA/PVP/TiO 2 -1 and SA/PVP/TiO 2 -3 nanocomposite beads reaching 92% and 98.9% of MB dye decay, respectively, after 120 min of illumination.

Effect of Light Intensity on MB Photocatalytic Degradation
The influence of light intensity on the efficiency of photocatalytic MB dye decay was evaluated using two concentrations of TiO2. The intensity of the light over the system was varied by holding either one or two lamps on the reactor cover. As shown in Figure 6, the efficiency of the system after 120 min was enhanced by increasing the light intensity. This suggests that for higher intensity, the amount of light reaching the photocatalyst particles increases, thus increasing electron stimulation and the efficiency of the system [26].

Effect of Light Intensity on MB Photocatalytic Degradation
The influence of light intensity on the efficiency of photocatalytic MB dye decay was evaluated using two concentrations of TiO 2 . The intensity of the light over the system was varied by holding either one or two lamps on the reactor cover. As shown in Figure 6, the efficiency of the system after 120 min was enhanced by increasing the light intensity. This suggests that for higher intensity, the amount of light reaching the photocatalyst particles increases, thus increasing electron stimulation and the efficiency of the system [26].

Effect of Light Intensity on MB Photocatalytic Degradation
The influence of light intensity on the efficiency of photocatalytic MB dye decay was evaluated using two concentrations of TiO2. The intensity of the light over the system was varied by holding either one or two lamps on the reactor cover. As shown in Figure 6, the efficiency of the system after 120 min was enhanced by increasing the light intensity. This suggests that for higher intensity, the amount of light reaching the photocatalyst particles increases, thus increasing electron stimulation and the efficiency of the system [26].

Effect of the MB Dye Initial Concentration on its Photocatalytic Decay
The effect of the initial dye concentration on the photocatalytic decay of MB is shown in Figure 7 for both SA/PVP/TiO 2 -1 and SA/PVP/TiO 2 -3 nanocomposite beads. In each experiment, the same quantity of nanocomposite beads (1 g L −1 ) was added to solutions containing different dye concentrations (10, 25, 50, and 100 mg L −1 ). The pH of the dye solution was kept at 7 and the contact time and light intensity were 120 min and 1200 lm, respectively. Figure 7 shows that dye decay tends to increase for higher initial MB concentration, the higher decay being attained at 50 mg L −1 . Further increasing the MB dye concentration negatively affected its decay. This can be justified by the dependence of the photodegradation rate of the dyes on the formation of OH • radicals on the catalyst surface, followed by the reaction of such radicals with the MB dye molecules. The higher dye decay rate observed at lower initial dye concentrations can be related to a higher chance of the dye molecules to react with OH • [27,28]. The following steps describe the mechanism proposed for MB degradation via the nanocomposite beads. First, irradiation with visible light enables electrons in the valence band to move to the conduction band. The surface of the TiO 2 photocatalyst then forms holes and electrons. Electrons react with dissolved O 2 to produce O 2 •− , while holes react with OH − ions to produce OH • . The latter reacts with MB dye to degrade it into non-toxic products, such as CO 2 , and water. Additional OH • is produced in the reaction between H 2 O 2 and e − , further enhancing MB dye degradation [29][30][31].
dation rate of the dyes on the formation of OH • radicals on the catalyst surface, followed by the reaction of such radicals with the MB dye molecules. The higher dye decay rate observed at lower initial dye concentrations can be related to a higher chance of the dye molecules to react with OH • [27,28]. The following steps describe the mechanism proposed for MB degradation via the nanocomposite beads. First, irradiation with visible light enables electrons in the valence band to move to the conduction band. The surface of the TiO2 photocatalyst then forms holes and electrons. Electrons react with dissolved O2 to produce O2 •− , while holes react with OH − ions to produce OH • . The latter reacts with MB dye to degrade it into non-toxic products, such as CO2, and water. Additional OH • is produced in the reaction between H2O2 and e − , further enhancing MB dye degradation [29][30][31].
For higher values of initial MB dye concentration, the photocatalyst activity is diminished. This effect is likely related to a hampering of the reaction between OH • radicals and the dye molecules. As the initial dye concentration increases, the amount of MB molecules adsorbed onto the catalyst surface also increases, consequently hindering OH • formation [32]. Additionally, with the increase of color intensity, the gaps related to the photons entry are inhibited in reaching the photocatalyst surface, restraining the photodegradation of the dye by OH • radicals [31]. Herein, the optimum initial MB dye concentration was found to be 50 mg L −1 , with the SA/PVP/TiO2-3 nanocomposite beads revealing a photodegradation efficiency of 98.9 %.

Kinetic Models
The reaction mechanism between the SA/PVP/TiO2 nanocomposite beads and MB is examined by making use of pseudo-first order and pseudo-second order kinetic models. The reaction rate is usually described by the kinetic model, whereas the dependence of the former on the reacting species concentration defines the reaction order [33,34]. The For higher values of initial MB dye concentration, the photocatalyst activity is diminished. This effect is likely related to a hampering of the reaction between OH • radicals and the dye molecules. As the initial dye concentration increases, the amount of MB molecules adsorbed onto the catalyst surface also increases, consequently hindering OH • formation [32]. Additionally, with the increase of color intensity, the gaps related to the photons entry are inhibited in reaching the photocatalyst surface, restraining the photodegradation of the dye by OH • radicals [31]. Herein, the optimum initial MB dye concentration was found to be 50 mg L −1 , with the SA/PVP/TiO 2 -3 nanocomposite beads revealing a photodegradation efficiency of 98.9 %.

Kinetic Models
The reaction mechanism between the SA/PVP/TiO 2 nanocomposite beads and MB is examined by making use of pseudo-first order and pseudo-second order kinetic models. The reaction rate is usually described by the kinetic model, whereas the dependence of the former on the reacting species concentration defines the reaction order [33,34]. The study involved carrying out experiments both in the dark and under light irradiation. Table 2 shows that there are clear differences between the two models in the dark and under irradiation of visible light. In the pseudo-second order model, the rate constant K 2 for SA/PVP/TiO 2 -3 in dark mode is the highest, indicating the chemisorption nature of the MB adsorption process [35]. The MB degradation mechanism begins with the adsorption of the dye on the surface of the nanocomposite by electrostatic interactions [36], followed by its photodegradation. At pH values of 3-7, the beads have a negative surface charge. In addition, TiO 2 contains terminal oxygen atoms that consequently increase the interaction between the beads' surface and nitrogen atoms in the MB molecules [1]. Under the irradiation of light, electronhole pairs are formed in TiO 2 and the generated OH • and O 2 •− radicals are concentrated on the surface [34]. The MB dye is then degraded into smaller molecular fragments, such as CO 2 , H 2 O, and H + , by these hydroxyl radicals or superoxide ion radicals. Table 3 compares the produced nanocomposite beads to other TiO 2 -based nanocomposites that have previously been investigated for the elimination of various organic dyes in the water. When compared to previously reported nanocomposite beads, the removal effectiveness of the herein prepared SA/PVP/TiO 2 -3 nanocomposite beads was almost higher than that of the other TiO 2 -based composites, with the latter also presenting unfavorable synthesis methods and cost.

Reusability of SA/PVP/TiO 2 Nanocomposite
Five consecutive experimental runs were performed under optimal conditions using the same set of beads to evaluate the reusability of SA/PVP/TiO 2 nanocomposites as indicated in Figure 8, which permits the process to be considered a cost-effective degradation process for MB. The SA/PVP/TiO 2 nanocomposite beads were recovered and used five times by washing with 0.1 M HCl solution. The obtained data reveal that the MB decay efficiency remained practically unchanged as the cycle number increased. This result may be due to the stability of TiO 2 nanotubes in the SA/PVP polymer matrix.
Appl. Sci. 2021, 11, x FOR PEER REVIEW 10 of 12 Five consecutive experimental runs were performed under optimal conditions using the same set of beads to evaluate the reusability of SA/PVP/TiO2 nanocomposites as indicated in Figure 8, which permits the process to be considered a cost-effective degradation process for MB. The SA/PVP/TiO2 nanocomposite beads were recovered and used five times by washing with 0.1 M HCl solution. The obtained data reveal that the MB decay efficiency remained practically unchanged as the cycle number increased. This result may be due to the stability of TiO2 nanotubes in the SA/PVP polymer matrix.

Conclusions
TiO2 nanotubes were incorporated into a SA/PVP blend as a doping agent. The syn-

Conclusions
TiO 2 nanotubes were incorporated into a SA/PVP blend as a doping agent. The synthesized SA/PVP/TiO 2 nanocomposite beads were used to degrade MB dye in aqueous solutions when exposed to visible light irradiation using the concept of "absorb and degrade". The process starts with the adsorption of the MB dye molecules on the surface of the SA/PVP/TiO 2 nanocomposite beads. Consequently, the adsorbed dye molecules undergo photocatalytic destruction by the TiO 2 nanotubes. The adsorption mechanism mainly depends on the porosity of the beads and the active sites on their surface, while the photocatalytic activity depends on the TiO 2 . The determined MB removal profiles demonstrated that SA/PVP/TiO 2 -3 nanocomposite beads perform better than SA/PVP/TiO 2 -1 beads. Additionally, it was found that these nanocomposite beads may be easily recovered by simple washing and reused as effective tools for treating wastewaters contaminated with cationic dyes.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.