Fabrication and characterisation of magnetic graphene oxide incorporated Fe3O4@polyaniline for the removal of bisphenol A, t-octyl-phenol, and α-naphthol from water

In this study, we fabricated a novel material composed of magnetic graphene oxide incorporated Fe3O4@polyaniline (Fe3O4@PANI-GO) using a modified Hummers’ method, solvothermal, and two-step polymerisation method. The resulting products were characterised by transmission electron microscopy (TEM), Fourier transform infrared spectroscopy (FT-IR), and X-ray diffraction (XRD). The results indicated that magnetic Fe3O4@PANI particles were successfully loaded onto the surface of the graphene oxide. Further Fe3O4@PANI-GO was investigated to remove bisphenol A(BPA), α-naphthol, and t-octyl-phenol (t-OP) from water samples by magnetic solid phase extraction. Under the optimal conditions, the Fe3O4@PANI-GO composite exhibited good adsorption capacity for t-OP, BPA, and α-naphthol, and the adsorption of these followed a pseudo-second-order kinetic model. Adsorption isotherms fit the Langmuir model, and the adsorption was an endothermic and spontaneous process. This work indicated that Fe3O4@PANI-GO earned great application prospect for removing phenolic contaminants from polluted water.

In recent years, magnetic nanoparticles (MNPs) have shown great technological significance in the areas of electronics, catalysis, therapy diagnosis, biosensors, and drug delivery due to their unique superparamagnetic properties 1 . Iron-based materials (e.g. Fe 3 O 4 ) have drawn considerable attention as inorganic supports for the synthesis of organic-inorganic hybrid materials because of their potential application in information storage, drug delivery, targeting, and magnetic separation 2 . Fe 3 O 4 nanoparticles possess many merits such as high surface area, inexpensiveness, easy separation by an external magnetic field, and high reusability, while they also are naturally hydrophilic due to the existence of plentiful hydroxyl groups on the particle surface 3 . However, since magnetite is highly susceptible to oxidation/dissolution, especially in acidic solutions 4,5 , and is easily aggregated, all of these properties cause instability. Thus, magnetic adsorbents are difficult to directly use because of the aggregation and limited adsorption property. An effective strategy is to coat or modify iron oxide nanoparticles with other substrates to enhance the stability of the composite material and improve the special adsorption of target compounds, which is also a successful way to widen the application of the material by coating multifunctional groups on their surfaces.
Polyaniline (PANI) is one of the most technologically important materials based on its environmental stability in a conducting form, unique redox properties, and high conductivity with suitable dopants 6,7 . The physicochemical properties of polyaniline and its potential applications in diverse fields such as battery, sensors, and wave-adsorption [8][9][10] have been reviewed. PANI can be easily synthesised by either chemical or electrochemical methods. Recently, bifunctional Fe 3 O 4 @PANI nanocomposites have attracted intensive attention for application in nanomaterials due to their novel magnetic and conductive properties. Xuan et al. reported the synthesis of Fe 3 O 4 @polyaniline core-shell microspheres with well-defined blackberry-like morphology 6 . Zhao et al. 11 prepared Fe 3 O 4 @PANI composite nanoparticles with a core-shell structure and measured their inductive heat property for localised hyperthermia. PANI was also used to easily and efficiently remove pollutants like heavy metal ions and organic contaminant from aqueous solutions [12][13][14] . In view of PANI polymers having a wealth of amino and benzene ring groups, the material can adsorb organic compounds and metal ions by π-π interaction and electrostatic interaction. Therefore, PANI is expected to be a promising adsorbent for the removal of aromatic compounds in water. However, the mechanical weakness and poor solubility of PANI greatly hinder further experimental investigation and commercial exploitation of the material 15 .
Graphene oxide (GO) consists of a hexagonal carbon network bearing hydroxyl and epoxide functional groups on its "basal" plane and is a single sheet of graphite oxide that exhibits good properties for many applications. It can be obtained by exfoliation of graphite oxide 16 whereas the edges are mostly decorated by carboxyl and carbonyl groups 17 . These oxygen-containing functional groups can bind with metal ions and organic contaminants in water. Yang et al. found that the adsorption capacity of Cu(II) on GO was 10 times higher than that of Cu(II) on activated carbon 18 . Chang et al. prepared Fe 3 O 4 /graphene nanocomposites and achieved good adsorption performance for aniline and p-chloroaniline 19 . Xie et al. developed a facile chemical method to produce a superparamagnetic graphene oxide-Fe 3 O 4 hybrid composite and which was successfully utilised for the removal of dyes from aqueous solution with high removal efficiency 20 . However, upon the removal of the hydrophilic functional groups on GO, it can lead to aggregated graphene sheets that are a few layers thick 21 . GO has high hydrophilicity and good dispersibility, making it suitable for direct application as an adsorbent for the separation/ preconcentration of organic contaminants. Some graphene materials need to be centrifuged in the last separation steps in order to prepare graphene composite material with high specific surface area and stability 22 , which is significant for broadening the application of graphene oxide. There exist some reports on the application of PANI in the fabrication of a GO/PANI composite for supercapacitors and high-performance shielding materials 23-25 , etc., which suggests that PANI can be effectively anchored on a magnetic substrate via strong interactions with GO as the intermediate. This process not only reserves the oxygen-containing functional groups of GO, but also enhances the stability of the magnetic composite.
In present study, Fe 3 O 4 @PANI-GO composite was synthesised by decorating GO and PANI, which provided nitrogen-containing functional groups and protected the Fe 3 O 4 nanoparticles. The prepared magnetic composites were investigated as magnetic adsorbents to remove BPA, t-OP, and α-naphthol from aqueous solution. The adsorption parameters of Fe 3 O 4 @PANI-GO for the removal of the three phenols from aqueous solution were investigated. The adsorption kinetics, isotherms, and thermodynamic studies were also performed to demonstrate the mechanism of the composite material toward BPA, α-naphthol, and t-OP.

Results
Morphology and structure.    Figure 1(c,d) shows the TEM images of original GO and Fe 3 O 4 @PANI-GO hybrids. It was also clear that Fe 3 O 4 @PANI particles highly covered the surface of GO nanosheets ( Fig. 1(d)), indicating possible electrostatic attraction between graphene and Fe 3 O 4 @PANI microspheres.
The structure of Fe 3 O 4 @PANI on graphene oxide was corroborated by XRD measurements. As seen in Fig. 2 Figure 2b shows the FT-IR spectra of GO and Fe 3 O 4 @PANI-GO composite material. As GO is concerned, the bands at 1730 and 1070 cm −1 are assigned to the characteristic peaks of C=O and C-O-C, respectively. In the FT-IR spectra of Fe 3 O 4 @PANI-GO, the adsorption bands at 1598 and 1502 cm −1 are attributed to the stretching vibration of C=C/C-C of benzenoid ring and quinoid. The band at 1294 cm −1 is the stretching vibration of C-N, which is the characteristic spectral bands of PANI, while the in-plane bending vibration of C=H is at 1145 cm −1 . As expected, the characteristic peaks of Fe 3 O 4 microspheres appear around 561 cm −1 and are contributed to the Fe-O bond stretching [27][28][29] , and the broad and intense band at 3400 cm −1 is ascribed to the stretching of O-H. In comparison, the same of the peaks of Fe-O and O-H appeared in Fe 3 O 4 @PANI-GO composite material, which indicate that Fe 3 O 4 @PANI was successfully loaded onto graphene oxide.
Optimization of adsorption. The important parameters that affect the adsorption such as amounts of adsorbents, sample pH, HA, ionic strength were optimized. The results showed that best results were obtained with 60 mg of Fe 3 O 4 @PANI-GO dosage at pH6. The salting-out effect and effect of HA were very small (See Fig. 3). The experimental data showed that the adsorption kinetics of BPA, t-OP, and α-naphthol conformed to pseudo-second-order kinetics and the data were exhibited in Fig. 3f  Adsorption isotherm and thermodynamics. The mechanism of adsorption is always in the center of our focusing, and several isotherm models were used to describe the adsorption behavior. The experimental results indicated that Langmuir model fit the adsorption data better than the Freundlich model for the adsorption of BPA, α-naphthol, and t-OP on the Fe 3 O 4 @PANI-GO magnetic composites. The thermodynamic data were calculated and demonstated that the adsorption was a spontaneous and endothermic process. These data were presented in Fig. 4, Tables 2 and 3.
Reusability. As a new adsorbent was concerned, the reusability was often an important parameter. In this study, it was investigated with ten recycles. The results were shown in Fig. 5, and the results indicated that the Fe 3 O 4 @PANI-GO magnetic composite was a good adsorbent with almost no loss of the recovery of BPA, α-naphthol, and t-OP after ten cycles.

Discussion
Adsorption. The adsorption of BPA, t-OP, and α-naphthol was performed by designing a series of experiments. The effect of the amount of adsorbent was investigated with an initial concentration of 5, 5 and 10 mg L −1 for BPA, t-OP and α-naphthol, respectively. ( Fig. 3(a)). It was observed that the removal efficiency of the target compound adsorbed increased when the adsorbent dosage increased from 5 to 60 mg. The removal efficiency reached 91.32, 95.93, and 98.86% for BPA, t-OP, and α-naphthol, respectively. The removal rates of three phenols reached a steady state, and increased very small, and only the removal rate of BPA still had a slight increase, however the increase was very small. Therefore, 60 mg of Fe 3 O 4 @PANI-GO was used.
The effect of salinity was an important parameter and was often optimized parameter, herein it was checked with the concentration of NaCl in the range of 0-25% (w/v). Figure 3(c) shows the results of the effect of ionic strength on the adsorption of BPA, α-naphthol, and t-OP onto Fe3O4@PANI-GO. It was observed that the removal efficiency of BPA increased within the NaCl concentration of 0-10% (w/v) and then decreased to the initial level when the NaCl concentration up to 25%. As α-naphthol was concerned, its removal rate kept constant within NaCl concentration range of 0−15% and then decreased with increase of NaCl concentration up to 25%. For t-OP, no significant influence was observed in the concentration range of 5-25% (w/v).
The presence of humic acid (HA) may have affected the adsorption capacity of phenols due to its competition for surface adsorption sites on the composites. As shown in Fig. 3(d), an interesting phenomenon occurred in which a small amount of humic acid promoted the adsorption of three phenols. The best results were achieved when the concentration of humic acid was 0.1 mgL −1 for these three phenols. The adsorption efficiency of BPA increased to the maximum value of 90.7% with 0.1 mgL −1 humic acid, yet the continuous increase of humic acid concentration resulted in the decrease of BPA adsorption. For α-naphthol and t-OP, no significant influence of HA was observed in the HA concentration range of 0.1-4 mg L −1 . The independence of humic acid concentration on phenols adsorption is important for the application of Fe 3 O 4 @PANI-GO in the removal of some organic pollutants from wastewaters since HA concentration may vary in different samples. Therefore, the effect of HA on the extraction efficiencies of target compounds in real water samples was negligible.
To optimise the pH for maximum adsorption capacity of BPA, α-naphthol, and t-OP on Fe 3 O 4 @PANI-GO magnetic composites, a series of adsorption experiments were carried at various pH values. The effects of pH on adsorption percentages of phenols were investigated from pH2 to pH11. As shown in Fig. 3(e), the Fe 3 O 4 @ PANI-GO composite material adsorbed the phenols effectively in the range of pH 2-7. However, the adsorption rate declined sharply and even decreased to about 4.4% for α-naphthol at pH 9.0 and 11.0 and had no obvious effect for t-OP over the whole pH range. These phenomena could be explained by the net charge of graphene, BPA, and other organic matter at different pH values 30 . The strong adsorption of phenolic organic pollutants on the magnetic nanocomposites might be attributed to physical adsorption: the donor-acceptor interactions between the electrons of the aromatic ring and the graphene sheets 31 . Therefore, the kinetic and isotherm experiments were operated at pH 6.0. Adsorption kinetics. The effect of contact time on the amount of organics adsorbed was investigated, and results are presented in Fig. 3(b). It only took 5 min for BPA and t-OP to attain 59.41% and 78.04%, respectively, and for α-naphthol, the adsorption equilibrium was reached after 300 min. These results demonstrated that a fast adsorption process and the adsorbed amount of these phenols reached equilibrium values very quickly. The time-dependent adsorption capacity was obtained to study the kinetics for the adsorption of these phenols on Fe 3 O 4 @PANI-GO. The kinetics of adsorption is important considering that it controls the process efficiency. The adsorption model that describes the sorption of a solute onto a solid surface can be expressed by pseudo-first-order, pseudo-second-order, or intraparticle diffusion model [27][28][29] . The best-fit model was selected based on the linear regression correlation coefficient values (R 2 ).
The pseudo-first-order kinetic model is expressed as: The pseudo-second-order kinetic model is present in the following equation as: The intraparticle diffusion model is defined as follows: where q e and q t were the amounts of Fe 3 O 4 @PANI-GO adsorbed (mg g −1 ) at equilibrium and at time t(min) respectively; and k 1 , k 2 , and k i are the rate constants. A good linear relationship between t/qt and t was obtained. The slopes and intercepts of each linear plot in Fig. 3(f) are used to calculate the kinetic parameters for BPA, t-OP, and α-naphthol adsorption, and the results are listed in Table 1. The correlation coefficients, R 2 , of the pseudo-second-order kinetic model for the adsorption of BPA, α-naphthol, and t-OP onto Fe 3 O 4 @PANI-GO were determined to be 0.9991, 0.9958, and 0.9999, respectively, which are much higher than that of the pseudo-first-order and intraparticle diffusion models. Clearly, the pseudo-second-order kinetic curves gave a good fit to the experimental kinetic data with a much higher correlation coefficient (R 2 ). Furthermore, the experimental adsorption capacity (q e , exp) was also in accordance with the calculated adsorption capacity (q e , cal) obtained from the pseudo-second-order model. These results suggest that the pseudo-second-order kinetic model offers a more appropriate description of the adsorption process. Adsorption isotherms. Adsorption isotherms describe the distribution of adsorbed molecules between the liquid phase and solid phase. The adsorption isotherms for the removal of BPA, t-OP, and α-naphthol were studied using an adsorbent dosage of 20-50 mg. Langmuir, Freundlich, Temkin, and Dubinin-Radushkevich isotherm models were used to describe the adsorption process [32][33][34][35] .
The adsorption isotherm of the Langmuir model assumes monolayer adsorption on a perfectly smooth and homogeneous surface. It has been successfully applied to many pollutant adsorption processes from aqueous solution. The equation is expressed as: where q e is the adsorption capacity (mg g −1 ) at the equilibrium point; C e is the equilibrium concentration of BPA, α-naphthol, and t-OP (mg L −1 ); q m represents the maximum adsorption capacity of the adsorbent (mg  Table 3. Thermodynamic parameters for the absorption of phenols onto Fe 3 O 4 @PANI-GO. ; and b is the Langmuir adsorption constant (L mg −1 ).The value of R L indicates the shape of the isotherm to be unfavourable (R L > 1), linear (R L = 1), favourable (0 < R L < 1), or irreversible (R L = 0) in which R L values between 0 and 1 indicate favourable adsorption. Figure 4 exhibits the Langmuir adsorption isotherms of BPA, α-naphthol, and t-OP on as-prepared Fe 3 O 4 @PANI-GO at three different temperatures. The adsorption capacity, q m , and adsorption constant, b, can be determined from the slope and intercept of a linearized plot of C e /q e vs C e , as presented in Table 2. The maximum adsorption capacities (q m ) calculated according to the Langmuir model were 14.43, 13.19, and 24.15 mg g −1 for BPA, α-naphthol, and t-OP on Fe 3 O 4 @PANI-GO composites, respectively. The R L values were obtained in the range of 0.0350-0.5638, thereby confirming that the adsorption is a favourable process. Comparing of the q m values for phenols, the graphene oxide composite has a higher absorbability for t-OP than does BPA and α-naphthol. The Freundlich isotherm is an empirical equation employed for heterogeneous systems and adsorption at multilayers. The equation is expressed as: where k f and n are Freundlich constants that indicate the relative sorption capacity and sorption intensity, respectively. If 1 < n < 10, the adsorption is favourable 31 . Hence, it can be seen ( Table 2) that the adsorption of BPA, α-naphthol, and t-OP on Fe 3 O 4 @PANI-GO composites were favourable in this research.
The Temkin isotherm considers the effects of indirect adsorbent/adsorbate interactions on adsorption isotherms. The isotherm assumes that the heat of adsorption of all the molecules in the layer decreases linearly with coverage due to adsorbent-adsorbate interactions. The equation is expressed as: Which can be linearized to: where K T is the constant of Temkin isotherm (g −1 ); and b T is the Temkin isotherm constant related to the heat of adsorption (kJ mol −1 ). The Dubinin-Radushkevich isotherm is used to estimate the characteristic porosity and the apparent free energy of adsorption. The Dubinin-Radushkevich equation is expressed as: where K D (mol 2 kJ −2 ) is the mean free energy E (kJ mol −1 ) of adsorption per molecule of the sorbate when it is transferred to the surface of the solid from infinity in the solution, which was calculated by E = −(2K D ) −0.5 ; and ε is Polanyi potential constant given as RT ln(1 + 1/C e ). The E value can be used to estimate the type of adsorption. If 8 < E < 16 kJ mol −1 , the adsorption can be explained by ion exchange. If E < 8 kJ mol −1 , it is physical absorption, and if E > 16 kJ mol −1 , it is chemical adsorption. In Table 2, the higher correlation coefficients(R 2 ) indicate that the Langmuir model fit the adsorption data better than the Freundlich model for the adsorption of BPA, α-naphthol, and t-OP on the Fe 3 O 4 @PANI-GO magnetic composites. The mechanism of Fe 3 O 4 @PANI-GO adsorption toward the phenols might be based on van der Waals interactions occurring between the hexagonally arrayed carbon atoms in the graphene oxide sheet and the aromatic backbones of the organics. The second reason might be due to the strong π-stacking interaction between the benzene ring of the organics and the large delocalized π-electron system of GO 32 . Thus, it is suggested that the graphene oxide magnetic adsorbent has a higher adsorption capacity for BPA, α-naphthol, and t-OP removal. Further, it is clear that the sorption energy values (E) of the Dubinin-Radushkevich model for BPA, α-naphthol, and t-OP are similar to each other, which are 0.257, 0.010, and 0.265 kJ mol −1 , respectively, at 298k. These indicated that the adsorption was physical adsorption, and the positive values indicated that the adsorption process was endothermic.
Thermodynamic studies. Batch adsorption was performed at different temperatures (298, 308, and 318 K), and the results are summarised in Table 3. The thermodynamic parameters were estimated in order to evaluate the feasibility and exothermic nature of the adsorption process. Free energy of adsorption (∆G°), enthalpy (∆H°), and entropy (∆S°) changes were calculated to predict the nature of adsorption.
The free energy of adsorption (∆G°) can be related to the equilibrium constant, K 0 (L mol −1 ), where K 0 can be obtained from the intercept of the plot ln(q e /C e ) vs. q e . The adsorption standard free energy changes (∆G°) can be calculated according to following equation: where R is the gas universal constant (8.314 J/mol K); K 0 is the equilibrium constant; and T is the absolute temperature. The van't Hoff equation was used to determine K 0 : where ∆H 0 and ∆S 0 values can be obtained from the slope and intercept by plotting lnK 0 vs. 1/T. The thermodynamic parameters are listed in Table 3. The negative values of free energy of adsorption (∆G 0 ) increased when the temperature increased, which indicates that the adsorption process is spontaneous. The positive standard enthalpy change (∆H 0 ) also suggests that the interaction of BPA, α-naphthol, and t-OP adsorbed by Fe 3 O 4 @ PANI-GO is endothermic, which is supported by the increased adsorption of BPA, α-naphthol, and t-OP with the increase in temperature.

Regeneration.
To investigate the possibility of regeneration of the Fe 3 O 4 @PANI-GO adsorbent, desorption experiments were performed in which organic pollutants were easily dissolved in organic solvents. Because of the theory of "similarity and intermiscibility", we selected methanol as the desorption solvent. After adsorption, desorption was carried out by shaking the Fe 3 O 4 @PANI-GO with 3 mL methanol. 5 min was allotted as the extraction and desorption time, and the removal efficiencies of phenols are shown in Fig. 5. From Fig. 5, it is observed that desorption of t-OP was achieved from the solution by using 3 mL methanol, and desorption ratio of 86.03% was obtained over ten adsorption/desorption cycles, which indicated that Fe 3 O 4 @ PANI-GO sorbents were stable during the MSPE procedure and shows good reusability.

Conclusions
In summary, we developed a facile method for the preparation of the magnetic composite material, Fe 3 O 4 @ polyaniline with incorporated GO, and TEM, SEM, FT-IR, and XRD investigation revealed the characteristics of the composites. The Langmuir, Freundlich, Temkin, and Dubinin-Radushkevich adsorption models were applied to describe the equilibrium isotherms. The calculated maximum adsorption capacities were 14.43, 13.19, and 24.15 mg g −1 for BPA, α-naphthol, and t-OP on the Fe 3 O 4 @PANI-GO composite, respectively. In addition, the adsorption capacity of Fe 3 O 4 @PANI-GO for t-OP was higher. The negative adsorption standard free energy changes and positive standard enthalpy change indicate that the adsorption was spontaneous and endothermic. Furthermore, owing to the excellent dispersion in water and hydrophilicity of GO, the stability of the hybrid material in water is attributed to GO. Furthermore, the magnetic separation technology provided a rapid and effective method for separating magnetic materials from aqueous phase and displayed good regeneration capacity. This research indicates that Fe 3 O 4 @PANI-GO can be used as an effective sorbent for the simple and rapid removal of organic pollutants from water samples.

Materials and Methods
Chemicals and materials. Graphite

Preparation of graphene oxide (GO) and Fe 3 O 4 nanoparticles.
Graphene oxide (GO) was synthesised from natural graphite powder using a modified Hummers' method 36,37 . Fe 3 O 4 microspheres were synthesised following a solvothermal method reported previously 38,39 . FeCl 3 . 6H 2 O (4.32 g) and sodium acetate (12.0 g) were dispersed in 80 mL ethylene glycol and stirred vigorously for 30 min at room temperature, then transferred to a Teflon-lined stainless steel autoclave (100 mL, capacity) and heated to reflux for 8 h at 200 °C. In this case, FeCl 3 ·6H 2 O was used as the iron source, and NaAc was used as the reductant. NaAc was also used for electrostatic stabilisation to prevent the agglomeration of the particles and to assist in the reduction of Fe 3+ to Fe 3 O 4 40 . The obtained Fe 3 O 4 was washed with deionized water and ethanol several times and then dried at 60 °C for 6 h. Preparation of Fe 3 O 4 @PANI-GO composite material. 0.1 g Fe 3 O 4 was dissolved in 50 mL deionized water containing 1 mL 0.02 M HCl aqueous solution in which HCl was used as a dopant. Then, FeCl 3 .6H 2 O and aniline monomer were added and sonicated for 10 min. After that, the mixture was mechanically stirred for 10 h in an ice bath for short-chain polymerization. APS and GO were then slowly added to the suspension under constant stirring. The long-chain oxidative polymerization was further continued for 12 h. After the reaction, the prepared particles were collected using a magnet and washed with deionized water and ethanol, then the obtained Fe 3 O 4 @PANI-GO composite was dried at 60 °C in a vacuum.
Adsorption experiment. Adsorptions of BPA, t-OP, and α-naphthol by Fe 3 O 4 @PANI-GO magnetic composites were performed by batch adsorption techniques in glass vials at T = 25 ± 1 °C, except for the thermodynamic experiments in which additional temperatures of 35 ± 1 °C and 45 ± 1 °C were used. 25 mL solution spiked with a known initial concentration was shaken with 20 mg magnetic Fe 3 O 4 @PANI-GO on a shaker at 270 rpm.
The effects of pH, ionic strength, and humic acid and the adsorption kinetics and adsorption isotherms were investigated by batch experiments. The initial concentrations of the solutions were 5 mg L −1 for BPA and t-OP, and 10 mg L −1 for α-naphthol in the pH and kinetics experiments. The maximum adsorption capacity and adsorption isotherms were calculated according to different concentrations of organics in the range of 2-35 for α-naphthol and 1-30 mg L −1 for BPA and t-OP, respectively.
After adsorption, the solution was separated by a magnet and analysed using a high-performance liquid chromatography (HPLC, Shimadzu SPD-10A, Wondasil-C18, superb 5 µm) system equipped with a UV-Vis detector. The detection wavelength was set at 225, 225, and 280 nm for BPA, t-OP, and α-naphthol, respectively. The mobile phase was composed of methanol and water (80/20, v/v), and the flow rate was set at 1 mL min −1 . The absorption capacity (q, mg g −1 ) was calculated using Equation (1): where C 0 and C e are the initial and equilibrium concentrations of phenols in the solution (mg L −1 ), respectively; m is the mass of the adsorbent (g); and V (L) is the solution volume. Data Availability. All data generated or analysed during this study are included in this published article.