Preferential and enhanced adsorption of methyl green on different greenly synthesized magnetite nanoparticles: investigation of the influence of the mediating plant extract’s acidity

Four magnetite nanoparticle (NP) samples have been greenly synthesized using four aqueous plant extracts, which are Artemisia herba-alba (L), Rosmarinus officinalis (L), Matricaria pubescens (L), and Juniperus phoenicia (L). The pH of these extracts are acidic (5.25, 5.05, 4.63, and 3.69, respectively). The synthesized samples were characterized by XRD, SEM, ATR-FTIR, and UV-Vis. This work aimed to study the preferential and enhanced adsorption of methyl green (MG) on the four greenly synthesized Fe3O4 surfaces by coupling three processes: MG adsorption in ambient dark conditions as the first process, followed by the thermocatalysis of the MG/Fe3O4 residual solution in the second process, and finally photocatalysis by the UV irradiation of MG/Fe3O4 residual solution after carrying out thermocatalysis. The novelty of this study lies in highlighting the influence of the mediating plant extract’s acidity on the magnetite NPs’ physicochemical characteristics, which impact the preferential and enhanced MG adsorption. The studied physicochemical characteristics are the functional hydroxyl group density on the magnetite surface, grain size, and band gap energy. It was found that the plant extract’s acidity has a clear effect on the studied physicochemical properties. The analysis of the FTIR spectra showed that the hydroxyl group densities differ on the four magnetite samples. Furthermore, the calculated grain sizes of the magnetite samples based on XRD spectra data vary from 29.27 to 41.49 nm. The analysis of the UV-Vis spectra of the four magnetite samples showed that the estimated direct band gap energies vary from 2.87 to 2.97 eV. The obtained results showed that the decrease of the mediating plant extract’s acidity leads to an increase in the hydroxyl group density on magnetite surfaces, which resulted in an increase in the MG adsorption capacities and yields in the first process of adsorption. Thus, MG adsorption was more preferred on greenly synthesized magnetite surfaces mediated by plant extracts with low acidity (Artemisia herba-alba (L) and Rosmarinus officinalis (L)). Furthermore, the increase of the plant extract’s acidity leads to a decrease in the particle size and an increase in the band gap energy and, therefore, to the decrease of the electron/hole pair recombination speed upon electron excitation. So, magnetite greenly synthesized from a more acidic mediating plant extract showed higher thermo- and photocatalytic activities for MG adsorption (Juniperus phoenicia (L) and Matricaria pubescens (L)). However, under photocatalysis, the enhancement is even more significant compared to thermocatalysis.


Introduction
Nanomaterials are widely used in the purication of aqueous media. [1][2][3][4] They allow a rapid thermodynamic equilibrium between adsorbent and adsorbate during the adsorption process and the selective removal of pollutants. [5][6][7] Adsorption has been extensively studied as a cost-effective process for removing a wide variety of pollutants from aqueous solutions, such as dyes. [8][9][10] The adsorption ability of iron oxide NPs arises from the intervention of hydroxyl groups during pollutant dissociation. 11 Surface hydroxyl groups, with amphoteric properties, are the functional groups of iron oxide surfaces and they are the chemically reactive entities that behave as the active sites in the adsorption process. These hydroxyl groups may be singly, doubly, and triply coordinated to Fe atoms, with different reactivities. The overall density of these groups depends on both the crystal structure and the extent of the development of the different crystal faces. 12 Photo-and thermocatalysts absorb photons/phonons with an energy equal to or more than the band gap energy between the valence band (VB) and conduction band (CB) of the photoor thermocatalyst. Photon/phonon absorption causes charge separation by exciting electrons from the VB to the CB, followed by the generation of positive holes in the VB. 13,14 These positive holes oxidize adsorbed H 2 O molecules and produce hydroxyl radicals (OHc). Whereas excited electrons reduce the adsorbed O 2 in the CB and produce hydroxyl radicals (OHc). These OHc radicals attack the organic groups of the pollutant and undergo various reactions to convert the organic pollutants into nontoxic and non-hazardous forms or completely degrade them into CO 2 and H 2 O. 14,15 The photo-and thermogenerated electron/hole pairs exhibit a strong tendency to recombine. Recombination lifetime speed is an important factor that inuences the photo-and thermocatalysis efficiency. If the recombination of photo-and thermogenerated charges is slow, then the photo-and thermocatalytic degradation of pollutants is more efficient. 16 Several works have studied the thermo-and photocatalysis of dye adsorption on nanomaterials, and they reported the highefficiency thermo-and photocatalytic activities of nanomaterials. Wu et al. 17 studied the thermocatalysis of methylene blue adsorption on magnetite Fe 3 O 4 @C NPs. They found that an increase of temperature leads to an increase of methylene blue thermodegradation, which indicates the high thermocatalytic activity of the studied nanomaterial. Other authors 18 studied the thermocatalysis of N719 dye on anatase TiO 2 nanosheets with dominant (001) facets and TiO 2 NPs with dominant (101) facets. They found that an increase of temperature leads to an increase of N719 dye thermodegradation on both studied nano-adsorbents due to the thermocatalytic activity of TiO 2 NPs. Farghali et al. 19 studied the thermocatalysis of methylene blue on multi-walled carbon nanotubes decorated with CoFe 2 O 4 NPs by increasing the temperature. They reported that this nanocomposite showed efficient thermocatalytic activity.
Furthermore, Ge et al. 20 studied the photocatalysis of methylene blue and methyl orange adsorption on iron oxide anchored to single-wall carbon nanotubes by UV irradiation. They reported that the studied adsorbent showed efficient photocatalytic activity. Elhadj et al. 21 studied the photocatalysis of Basic Red 46 dye adsorption over ZnO NPs under solar irradiation. They reported that ZnO NPs exhibit high photocatalytic activity. Moreover, Kumar et al. 22 studied the photodegradation of methylene blue (MB), Congo red (CR), and methylene orange (MO) under sunlight irradiation in the presence of greenly synthesized magnetite mediated by Andean blackberry leaf extract. They reported that the presence of those magnetite NPs accelerated the photodegradation of the three dyes due to their high photocatalytic activity. Sirdeshpande et al. 23 studied the photodegradation of malachite green under sunlight irradiation in the presence of greenly synthesized magnetite using leaf extract of Calliandra haematocephala. They reported that the presence of those magnetite NPs increased the photodegradation of malachite green. Other authors 24 compared the photocatalytic activity of several composites of titanium dioxide containing magnetite NPs with different morphologies and structures in the photodegradation of Rhodamine B by UV irradiation. They reported that the highest dye photodegradation was observed when both spherical and rod-shaped composite structures based on titanium dioxide containing 1 wt% of magnetite NPs were used as a photocatalyst. Jassal et al. 25 studied the thermo-and photodegradation of malachite green (MG) and Eriochrome Black T (EBT) dyes on greenly synthesized potassium zinc hexacyanoferrate nanocubes. They found that this adsorbent acted as a photocatalyst, not a thermocatalyst.
Several parameters can impact photo-and thermocatalysis processes, such as solution pH, adsorbent concentration, dye concentration, solution ionic strength, temperature, 25-28 dye structure properties, 29,30 adsorbent particle size, 31 gap energy, recombination lifetime of the electron/hole pairs, 32,33 adsorbent type, 34,35 light source and time of light exposure. 34 Ullah et al. 32 reported that a Mn 2+ dopant in the ZnO NPs decreased the recombination of the electron/hole pairs, which enhanced the photocatalytic activity efficiency for the removal of dyes. Rafaie et al. 33 studied the photocatalytic properties of ZnO NPs microstructures decorated with Ag NPs for the degradation of methylene blue under UV irradiation. They reported that the Ag NPs played the role of electron sinks and trapped the photogenerated electrons, which increased the electron/hole pair lifetime. As a result, the ZnO-Ag nanostructure exhibited higher photocatalytic activity for the degradation of MB dye.
Saha et al. 5 studied the preferential adsorption of seven different dyes on magnetite NPs. They reported that the magnetite surface preferred adsorbing dyes containing higher OH content. Xiao et al. 36 studied the preferential adsorption of different cationic and anionic dyes on iron NPs. They reported that iron NPs preferred removing cationic dyes over anionic dyes. Madrakian et al. 37 studied the preferential adsorption of seven cationic and anionic dyes on magnetite-coated waste tea. They reported that the adsorption capacities of these NP adsorbents for the adsorption of cationic dyes were more increased compared to those for anionic dyes.
Several factors can inuence the adsorption, such as the solution pH, 12 solution ionic strength, 38 dye concentration, 39 magnetite NP concentration, 5 and hydroxyl group density on the adsorbent surface. 40 The impact of changing plants on greenly synthesized metal oxide NPs' reactivity in dye adsorption has been studied in several works. Huang et al. 41 studied the effect of three different tea extracts (green, oolong, and black teas) on the properties of iron oxide NP surfaces and their reactivities in the removal of methyl green from aqueous solutions. They reported that the plant extract has an effect on the reactivity of the iron oxide NP surfaces, with 81.2%, 75.6%, and 67.1% of methyl green dye being removed by iron oxide NPs synthesized using the extracts of green, oolong, and black teas, respectively. Likewise, Xiao et al. 36 studied the removal of six cationic and anionic dyes. They reported that iron NPs greenly synthesized with tea extract showed preferential adsorption of cationic dyes from an aqueous solution. Other authors 42 synthesized metal oxide NPs using the extracts of owers, bark, and the leaf of Tecoma stans in order to use them in the removal of Congo red (CR) and crystal violet (CV) dyes. They reported that the adsorbent derived from ower extract gave better dye adsorption efficiency than those derived from other extracts. Furthermore, Islam et al. 43 synthesized magnetite NPs using six plant extracts in order to use them in the removal of methyl orange (MO) and crystal violet (CV) dyes. They reported that the plant extract had an effect on the magnetite NPs' surface reactivity in the adsorption, where magnetite NPs synthesized using tea extract showed the highest performance (MO 92.34%, CV 96.1%).
In this paper, the preferential and enhanced adsorption of MG on four greenly synthesized Fe 3 O 4 NP surfaces has been studied by coupling three processes. The preferential adsorption of MG on the four magnetite surfaces in ambient dark conditions is the rst process, followed by the adsorption enhancement by the thermocatalysis of MG/Fe 3 O 4 residual solutions in dark conditions at the second process, and nally the adsorption enhancement by photocatalysis under UV irradiation (365 nm) in ambient conditions of the MG/Fe 3 O 4 residual solutions aer thermocatalysis. The focus of this study is the investigation of the inuence of the mediating plant extract's acidity on the greenly synthesized magnetite NPs' physicochemical characteristics, which impact the preferential and enhanced MG adsorption. The studied physicochemical characteristics are the functional hydroxyl group density on the magnetite surfaces, grain size, and band gap energy. The mediating plants in the green synthesis are Artemisia herba-alba (L), Matricaria pubescens (L), Juniperus phoenicia (L), and Rosmarinus officinalis (L), and synthesized Fe 3 O 4 samples from their extracts are respectively denoted in this paper as ARM- In preferential MG adsorption, the pseudo-rst-order and pseudo-second-order kinetics of the adsorption, as well as the intra-particle diffusion mechanism, have been analyzed. Under thermocatalysis, the activated thermodynamic parameters of free energy (DG 0 ), entropy (DS 0 ), enthalpy (DH 0 ), and activation energy (E a ) have been analyzed. Under photocatalysis, the pseudo-rst-order kinetics have also been analyzed.
The remainder of this paper is organized as follows. The next section gives a description of the materials used and methods followed during experiments. In Section 3, the obtained results are presented and discussed by analyzing the XRD, SEM, FTIR-ATR, and UV-Vis data. Section 3.7 is then devoted to presenting the effect of the plant extract's acidity on the physicochemical properties of greenly synthesized Fe 3 O 4 in the preferential and enhanced methyl green adsorption. The last section presents the conclusions.

Materials and methods
This section focuses on listing the materials needed and apparatuses used. It also provides details of the methods utilized to perform the adsorption experiments and characterization of the iron oxide NPs.

Methods
In this section, the methods used for solution preparation are described. The protocol used in the adsorption experiments of iron oxide NPs and characterization techniques are described as well.
2.2.1 Batch adsorption experiments of MG on magnetite surfaces in ambient dark conditions. In the rst step, the prepared standard aqueous solutions of MG dye were diluted several times as required. In the second step, 0.0015 g of JUN-Fe 3 O 4 , ROS-Fe 3 O 4 , MAT-Fe 3 O 4 and ARM-Fe 3 O 4 NP powders were added to a volume of 4 ml of the aqueous solution of the dye. The dye solution concentration was 0.0111 mg ml À1 . The ionic strength for all adsorption experiments was kept at 0.1 M by adding an appropriate amount of NaCl (0.023 g). A dilute solution of HCl was used to adjust the dye/Fe 3 O 4 solution pH to 4. This protocol was used to prepare, in total, 44 experiment sets (11 for each magnetite sample). In addition, 4 control experiment sets (without NPs) were also prepared.
All experiment sets were sonicated in an ultrasonic bath for 15 minutes and they were then stirred continually for 60 minutes until a steady state was reached. All adsorption experiments were carried out in ambient dark conditions in batch mode, and they were performed in triplicate for data consistency.
Kinetic experiments were performed by withdrawing samples of the MG/Fe 3 O 4 solutions at regular time intervals to obtain, aer centrifugation, adequate aliquots for the purpose of quantifying the residual dye concentrations and the adsorbed amounts. The concentrations of the aqueous solutions of the residual dye were quantied using a UV-Vis spectrophotometer at an absorbance maxima of MG l max ¼ 249 nm. Furthermore, the adsorbed amounts of MG molecules were calculated from the calibration curve for all adsorption experiments (Y ¼ 42.049X À 0.2885, R 2 ¼ 0.996). In order to obtain the adsorption capacity q eI (mg g À1 ) and the amount of MG cations adsorbed per unit mass (q tI in (mg g À1 )) of magnetite NPs at the equilibrium contact time in the rst process of MG adsorption, the following equations were used: Adsorption yield was calculated using the following equation: where C 0 , C eI , C tI , V and m are, respectively, the initial dye concentration without any treatment (mg ml À1 ), residual dye concentration in the liquid phase at steady state aer the rst process of MG adsorption (mg ml À1 ), residual dye concentration in the liquid phase at steady state aer the accomplishment of the rst process of MG adsorption (mg ml À1 ) at time t, the volume of dye solution (ml), and the amount of magnetite NPs (g).

2.2.2
Pseudo-rst-order and pseudo-second-order kinetics. The pseudo-rst-order (PFO) of Lagergren 44 and pseudo-secondorder of Ho and Mckay 45 kinetic models were selected to test the adsorption dynamics in this study due to their good applicability in most studies. 46,47 The Lagergren kinetic model assumes that the rate of the occupation of adsorption sites is proportional to the number of unoccupied sites. 48 Lagergren's model (eqn (5)) is suitable for only the initial 20 to 30 minutes of the adsorption action and not for the whole range of contact times. 45 It is generally represented by the following equation: Aer integration by the conditions q tI ¼ 0 at t ¼ 0 and q t ¼ q tI at t ¼ t, then eqn (4) becomes: where K 1 , q tI and q eI are, respectively, the pseudo-rst-order kinetic constant (mn À1 ), adsorbed dye quantity at time t (mg g À1 ) and adsorbed dye quantity at thermodynamic equilibrium in ambient dark conditions (mg g À1 ).
If the active surface of the adsorbent is regarded as invariable, the reaction could be treated as pseudo-rst-order. However, once the active sites have been saturated, the transfer at the pollutant/adsorbent particle interface may be limited by mass transfer. 49 The pseudo-second-order (PSO) model (eqn (6)) is proposed by Ho and McKay. 45 It is based on the adsorption capacity, expressed as follow: where K 2 , q tI and q eI are the pseudo-second-order kinetic constant (mg g À1 mn À1 ), adsorbed dye quantity at time t (mg g À1 ) and adsorbed dye quantity at thermodynamic equilibrium in the rst process (mg g À1 ), respectively. 2.2.3 Intra-particle diffusion kinetics. In order to gain insights into the adsorption mechanisms involved, a homogeneous particle diffusion model (HPDM), as shown in eqn (7), originally proposed by Boyd et al., 50 is used to describe the diffusive adsorption process. In this model, the rate-limiting step is usually described by either an intra-particle diffusion or a lm diffusion mechanism.
where F(t) is the fractional attainment at time t, i.e., F(t) ¼ q tI /q eI , D p (m 2 s À1 ) is the effective diffusion coefficient, r o is the radius of Fe 3 O 4 particles, which are assumed to be spherical, and Z is an integer. For 0 < F(t) < 1, a simplied equation can be obtained for the adsorption on spherical particles: A further formula alteration gives the following: where k p is the diffusion rate constant (1/s) and k p ¼ D p p 2 /r 0 2 .
Eqn (9) was used for the calculation of the effective intra-particle diffusivity (D p (m 2 s À1 )) from the experimental data. In the rst step, a graph of Àln(1 À Additionally, eqn (10) can be used when the rate of adsorption is controlled by liquid lm diffusion. 51 where D f is the lm diffusion coefficient (m 2 s À1 ) in the liquid phase, and C eI (mol l À1 ) and C rI (mol l À1 ) are, respectively, the equilibrium concentrations of MG dye in the solution and solid phases. d is the thickness of the liquid lm, which was assumed to be 10 À5 m according to Yu and Luo. 52 A further formula alteration of eqn (10) gives the following equation: The linearity test of Boyd plots (Àln(1 À F) and Àln(1 À F 2 ) versus time plots) was employed to distinguish between the lm diffusion and particle diffusion-controlled adsorption mechanisms. If the plot of Àln(1 À F) versus time is a straight line passing through the origin, then the adsorption rate is governed by the particle diffusion mechanism; otherwise, if Àln(1 À F 2 ) versus time is a straight line passing through the origin, then the adsorption is governed by lm diffusion. The concentrations of residual MG dye in the liquid phase were quantied using a UV-Vis spectrophotometer at an absorbance maxima of MG l max ¼ 249 nm. Furthermore, the adsorbed amounts of MG molecules were calculated from the calibration curve for all adsorption experiments (Y ¼ 42.049X À 0.2885, R 2 ¼ 0.996). In order to obtain the adsorption capacity q eII (mg g À1 ) of all magnetite samples aer carrying out thermocatalysis in the second process of MG adsorption in dark conditions, the following equation was used: Adsorption yield was calculated using the following equation: where C 0 , C eII , V and m are, respectively, the initial dye concentration without any treatment (mg ml À1 ), residual dye concentration in the liquid phase at steady state in the rst process of MG adsorption (mg ml À1 ), the volume of dye solution (ml), and the amount of magnetite NPs (g). The activated enthalpy (DH 0 ) of MG adsorption on the magnetite NP surface was determined using the Arrhenius equation as follows: where R (1.987 cal mol À1 K À1 ) is the universal gas constant, T is the absolute solution temperature (K), and K d is the distribution coefficient, which can be calculated as: where C aeII (mg ml À1 ) and c eII (mg ml À1 ) are, respectively, the concentration of adsorbed dye on the solid and the dye residual concentration in the liquid phase aer thermocatalysis in dark conditions. The values of activated enthalpy (DH 0 ) and entropy (DS 0 ) were calculated from the slope and intercept of the plot of ln K d versus 1/T. DG 0 was then calculated using the following equation: The free energy change indicates the degree of the spontaneity of the adsorption process and the higher negative value reects more energetically favorable adsorption. The activation energy (DE a ) of MG adsorption on magnetite surface is determined using the following Arrhenius's equation: where K 2 is the distribution coefficient which can be calculated by: where q eII (mg g À1 ) and c eII (mg ml À1 ) are, respectively, the adsorption capacity of the dye on the solid and the dye residual concentration in the liquid phase aer carrying out thermocatalysis in the second process of the MG adsorption in dark conditions. 2.2.5 Batch photocatalysis experiments of magnetite NP samples in ambient conditions. In the third MG adsorption process, the study of the photocatalysis of JUN-Fe 3 O 4 , ROS-Fe 3 O 4 , MAT-Fe 3 O 4 and ARM-Fe 3 O 4 NPs under UV irradiation to degrade MG was conducted on all experimental sets containing residual solutions aer carrying out thermocatalysis in dark conditions. All experiment sets were sonicated in an ultrasonic bath for 15 minutes and then they were stirred continuously and exposed to direct UV irradiation (365 nm) in ambient conditions for different times ranging from 60 to 240 minutes. The concentrations of the residual dye aqueous solutions were quantied using a UV-Vis spectrophotometer at an absorbance maxima of MG l max ¼ 249 nm. Furthermore, the adsorbed amounts of MG molecules were calculated from the calibration curve for all adsorption experiments (Y ¼ 42.049X À 0.2885, R 2 ¼ 0.996). In order to obtain the adsorption capacity q eIII (mg g À1 ) by photocatalysis in ambient conditions under the UV irradiation of JUN-Fe 3 O 4 , ROS-Fe 3 O 4 , MAT-Fe 3 O 4 and ARM-Fe 3 O 4 , the following equation was used: Adsorption yield was calculated using the following equation: where C 0 , C eIII , V and m are, respectively, the initial dye concentration without any treatment (mg ml À1 ), the dye residual concentration in liquid phase aer photocatalysis in the third process of MG adsorption in ambient conditions (mg ml À1 ) under UV irradiation, the volume of the dye solution (ml), and the amount of magnetite (g). The degradation kinetics of MG using Fe 3 O 4 NPs can be expressed as a pseudo-rst-order (PFO) reaction as follows: where C 0 , C tIII , and k pd are, respectively, the initial concentration of MG without any treatment (mg g À1 ), the dye residual concentration (mg g À1 ) in the liquid phase at time t aer photocatalysis under UV irradiation, and the PFO photocatalytic degradation rate constant (min À1 ), which can be calculated from the slope of the ln(C 0 /C tIII ) versus t plot.

Results and discussion
3.1 X-ray analysis of the Fe 3 O 4 NPs samples X-ray patterns of all the synthesised samples are presented in Fig. 2. It is found that all synthesized samples have crystalline structures. The X-ray diffraction pattern (A) in Fig. 2 01-076-0958).
The X-ray diffraction pattern (C) in Fig. 2 where D, b, l, and q are the crystallite size, the full width at halfmaximum (FWHM) of the most intense diffraction peak, the Xray wavelength (1.54056Å), and the Bragg angle, respectively.

FTIR-ATR spectroscopy analysis
The FTIR spectra of the synthesized Fe 3 O 4 NPs powders recorded between 4000 and 500 cm À1 are presented in Fig. 3.    shows that all IR spectra (A, B, C, and D) exhibit peaks in different ranges, as summarized in  Fig. 3 shows that the peaks of hydroxyl groups appear with remarkably different areas. The hydroxyl group peak area appears to be the broadest on the ARM-

UV-Vis spectroscopy analysis
The optical absorbance spectra of all Fe 3 O 4 samples were measured in the wavelength range of 200-900 nm. The band gap energies of the Fe 3 O 4 samples were then deduced from those spectra. The band gap (E g ) and the optical absorption coefficient (a) of a semiconductor are related through the known following equation: 55 where a is the linear absorption coefficient of the material, hn is the photon energy, A is a proportionality constant, and the exponent n depends on the nature of electronic transition; it is equal to 1/2 for direct allowed transition and 2 for indirect allowed transition. The E g of the direct transition of all samples were obtained from plotting (ahn) 2 as a function of ahn by the extrapolation of the linear portion of the curve (Fig. 4). However, the E g of the indirect transition of all samples were obtained from plotting (ahn) 1/2 as a function of ahn by the extrapolation of the linear portion of the curve (Fig. 5) Fig. 6b. However, for MAT-Fe 3 O 4 , a decrease in the dimension of the mountain-like structures, with more adherence to its structure, is observed (Fig. 6c). Finally, the ARM-Fe 3 O 4 SEM image contains some big structured single bipyramid crystals, as shown in Fig. 6d.

The analysis of MG adsorption kinetics and thermodynamics
3.5.1 MG adsorption equilibrium in preferential MG adsorption. In all adsorption experiments, the steady-state is reached within 30 minutes, as depicted in Fig. 7. This represents the very fast adsorption kinetics of MG on all four magnetite NP surfaces.
3.5.2 Pseudo-rst-order and pseudo-second-order kinetics in preferential MG adsorption. The results of the pseudo-rstorder kinetics analysis for preferential MG adsorption on all four magnetite NP surfaces (Table 3 and Fig. 8a) indicate good linearity and a good t of the experimental data to this model compared to the pseudo-second-order model, which indicated poor linearity and a poor t of the experimental data (Table 3 and Fig. 8b). The q eI,cal (equilibrium adsorption capacity), computed from the pseudo-rst-order kinetics plots, are also in very close agreement with the empirical q eI,exp , contrary to the q eI,cal calculated from the pseudo-second-order plots (see Table  3). This indicates the best compliance of MG adsorption on all four magnetite NP surfaces with pseudo-rst-order kinetics.
3.5.3 Intra-particle diffusion kinetics in preferential MG adsorption. The linearity tests of Boyd plots, Àln(1 À F) and Àln(1 À F 2 ) versus time, are presented in Fig. 9a and b. They show that the kinetic data correlate well with the homogeneous particle diffusion model, as conrmed by the high R 2 values. The results of linear regression analysis for eqn (9) and (11) are presented in Table 4. It was found that the lm diffusion coef-cients D f were in the order of 10 À11 m 2 s À1 , while the intraparticle diffusion coefficients D p were found to be in the order of 10 À19 m 2 s À1 . It is known that the adsorption mechanism is controlled by lm diffusion at D f ranging from 10 À10 to 10 À12 m 2 s À1 , while intra-particle diffusion is the rate-limiting step at D p in the range of 10 À15 to 10 À18 m 2 s À1 . 58 The results found indeed indicate that lm diffusion is the step that controls the adsorption mechanism of MG on Fe 3 O 4 surfaces, which is in agreement with the pseudo-rst-order kinetic model. 3.5.4 Activation thermodynamic parameters of MG adsorption under the thermocatalysis process. The calculated activated enthalpy (DH 0 ), entropy (DS 0 ), and free energy (DG 0 ) are listed in Table 5. DH 0 and DS 0 are respectively calculated from the slopes and intercepts of the Arrhenius linear plots of ln k D versus 1/T (Fig. 10a) 59 As presented in Table 5 (Fig. 10b). The found low E a suggests that MG adsorption on Fe 3 O 4 proceeds with a low energy barrier and can be achieved at relatively low temperatures. As it is known that the activation energy E a of physical adsorption ranges from 1.2 to 12 kcal mol À1 , and from 14.3 to 191 kcal mol À1 for chemical adsorption, 61 the adsorption processes of MG on all four Fe 3 O 4 samples are therefore physical in nature.
3.5.5 Pseudo-rst-order kinetic analysis of MG adsorption under the photocatalysis process. The results of the pseudo-rst-order kinetic analysis of MG adsorption on the four magnetite NP surfaces (Fig. 11) indicate a good linearity of the plots of ln(C 0 /C tIII ) versus time of UV irradiation, as judged from the high correlation coefficients (R 2 > 0.98), which indicate that the rate of MG degradation catalyzed by the Fe 3 O 4 NP samples is able to be tted by a pseudo-rst-order model. The corresponding photodegradation rates (k pd ) of MG by JUN-Fe 3 O 4 , MAT-Fe 3 O 4 , ROS-Fe 3 O 4 , and ARM-Fe 3 O 4 are 0.00132 min À1 , 0.00125 min À1 , 0.00123 min À1 , and 0.00120 min À1 , respectively.

Preferential and enhanced MG adsorption on magnetite surfaces
3.6.1 Preferential MG adsorption. Table 6 shows that, in the rst process of MG adsorption in ambient dark conditions, the adsorption capacity and yield of MG differ depending on the It is known that complexation and electrostatic interactions play important roles in determining the efficiency of adsorption. 6 When Fe 3 O 4 is immersed in the aqueous acidic solution, it develops its surface charge via the protonation and deprotonation of^Fe-OH active sites on its surface according to the following equation: 62 where^Fe-OH 2 + and^Fe-OH 0 are, respectively, the protonated positively charged acid site of the surface with two dissociable H + , and the neutral acid site of the surface with one dissociable H + . pK a1 ¼ 5.1 is the intrinsic acidity constant determined by Davis et al. 62 for Fe 3 O 4 . The binding of MG 2^Fe-OH + dye 2+ 4^(Fe-OH) 2 -dye 2+ (25) where^(Fe-OH) 2 -dye 2+ is a binuclear bonding complex due to hydrogen bonding between MG cations and the surface hydroxyl groups on magnetite NP surfaces. For the four magnetite NPs, the data provided by FTIR analysis (see Section 3.2, Fig. 3) show that the density of OH groups on the ARM-Fe 3 O 4 surface is the highest one, next to ROS-Fe 3 O 4 , then MAT-Fe 3 O 4 , and nally JUN-Fe 3 O 4 . As these hydroxyl groups behave as active sites on the Fe 3 O 4 surfaces, the results found show that MG adsorption yield is more increased on magnetite NP samples that have more OH groups, i.e. morê Fe-OH active sites. 3.6.2 Enhancement of MG adsorption by the thermocatalysis process. To assess the MG adsorption enhancement by   thermocatalysis, the thermocatalytic experiments were conducted on MG/Fe 3 O 4 residual solutions aer MG adsorption in the rst process, so as to give the overall adsorption yield and capacity aer the enhancement, and it also allows the comparison between adsorption yields and capacities before and aer thermocatalysis.    The thermocatalysis effect on MG adsorption on all four magnetite samples was evaluated by assessing the efficiency of the degradation of MG by thermocatalysis in dark conditions in a temperature range from 303.15 K to 318.15 for 20 minutes. Fig. 12a and Table 7 present the comparison of the MG adsorption yield and capacity of the four Fe 3 O 4 surfaces in the rst process of MG adsorption and in the second process in dark conditions under thermocatalysis. The data show that the MG adsorption yield and capacity increase with the increase of temperature in all adsorption experiments, which conrms the endothermic nature of the adsorption processes, as discussed in Section 3.5.4.
Yields and adsorption capacities are increased as follows (the yield aer thermocatalysis is denoted as R II % and the adsorption capacity as q eII ): On ARM-Fe 3 O 4 , the yield increased from R I % ¼ 53.42% to R II % ¼ 65.01%, and the adsorption capacity increased from q eI ¼ 15.81 to q eII ¼ 18.98 mg g À1 . On ROS-Fe 3 O 4 , the yield increased from R I % ¼ 51.90% to R II % ¼ 64.09%, and the adsorption capacity increased from q eI ¼ 15.37 to q eII ¼ 18.80 mg g À1 .
On MAT-Fe 3 O 4 , the yield increased from R I % ¼ 35.91% to R II % ¼ 63.60%, and the adsorption capacity increased from q eI % ¼ 10.83 to q eII % ¼ 18.56 mg g À1 .
On JUN-Fe 3 O 4 , the yield increased from R I % ¼ 19.25% to R II % ¼ 45.59%, and the adsorption capacity increased from q eI ¼ 5.70 to q eII ¼ 13.49 mg g À1 .
As all experiment conditions were kept the same for all adsorption experiments, only the surface properties are responsible for the adsorption enhancement.
The tendency of adsorption capacities and yields on the four magnetite NP surfaces is the same in the rst process of MG adsorption and in the second process of MG adsorption under thermocatalysis. In the rst process of MG adsorption, the highest adsorption capacity was on ARM-Fe 3 Fig. 12b, it is clear that there is a unique difference in the adsorption capacities aer thermocatalysis, denoted as q T e (q T e ¼ q eII Àq eI represents the enhancement in the adsorption capacity by thermocatalysis calculated as the difference between q eI , the adsorption capacity in the rst process of MG adsorption, and q eII , the overall adsorption capacity aer carrying out thermocatalysis at 318.15 K for 20 minutes). Fig. 12b presents q T e for the four MG/Fe 3 O 4 systems. These adsorption capacities are useful to elucidate the enhancement in MG adsorption by thermocatalysis. They show that q T e is the highest on JUN-Fe 3 O 4 (7.79 mg g À1 ) and the lowest on ARM-Fe 3 O 4 (3.17 mg g À1 ). This indicates that the thermocatalytic activity of the JUN-Fe 3 O 4 NPs is the highest and that of ARM-Fe 3 O 4 is the lowest. As all experimental conditions were kept the same for all adsorption experiments on all four magnetite samples, only the magnetite surfaces' properties are responsible for the adsorption enhancement by the thermocatalysis process.
From Fig. 12b and Table 7, it can be seen that the increase in q T e is accompanied by an increase in DS 0 in all the MG-Fe 3 O 4 systems (detailed in Section 3.5.4). This conrms that the increase in q T e has resulted from the change in the surface structure. 60 Thus, the maximum changes occurred in the  3.6.3 Enhancement of MG adsorption by the photocatalysis process. To assess the MG adsorption enhancement by photocatalysis, the photocatalytic experiments were conducted on MG/Fe 3 O 4 residual solutions aer the thermocatalytic experiments, so as to give the overall adsorption yields and capacities aer the enhancement by photocatalysis and allow the comparison between the adsorption yields and capacities before and aer carrying out the photocatalysis process. The impact of the photocatalysis process on MG adsorption on all four magnetite samples was evaluated by assessing the efficiency of the degradation of MG under UV irradiation (365 nm) in a time range from 60 to 240 minutes in ambient conditions. The variation of the MG adsorption yields, as well as the adsorption capacities under photocatalysis, is illustrated in Table 8 and Fig. 13a. They show that the MG adsorption capacities and yields on the four magnetite surfaces are enhanced by photocatalysis, however with remarkably different differences. Table 8 and Fig. 13a show remarkable differences when comparing the adsorption results on the four magnetite samples before carrying out photocatalysis and aer 240 minutes of exposure to UV irradiation in ambient conditions, where the MG adsorption yield and adsorption capacity vary as follows (the yield aer carrying out photocatalysis is denoted as R III % and the adsorption capacity as q eIII ): On ARM-Fe 3 O 4 , the yield increased from R II % ¼ 65.01% to R III % ¼ 71.71%, and the adsorption capacity increased from q eII ¼ 18.98 to q eIII ¼ 21.23 mg g À1 .
On ROS-Fe 3 O 4 , the yield increased from R II % ¼ 64.09% to R III % ¼ 72.07%, and the adsorption capacity increased from q eII ¼ 18.80 to q eIII ¼ 21.33 mg g À1 .
On MAT-Fe 3 O 4 , the yield increased from R II % ¼ 63.60% to R III % ¼ 72.97%, and the adsorption capacity increased from q eII ¼ 18.56 to q eIII ¼ 21.60 mg g À1 .
On JUN-Fe 3 O 4 , the yield increased from R II % ¼ 45.59% to R III % ¼ 75.23%, and the adsorption capacity increased from q eII ¼ 13.49 to q eIII ¼ 22.27 mg g À1 .
In the rst process of MG adsorption, the highest adsorption capacity was on the ARM-Fe 3  As shown in Fig. 13b, it is evident that there is a clear difference between the adsorption capacities under photocatalysis, denoted as q P e (q P e ¼ q eIII À q eII represents the enhancement in adsorption by photocatalysis, it is calculated from the difference between the overall adsorption capacity q eIII in the third process of the adsorption aer carrying out photocatalysis for 240 minutes, and the adsorption capacity q eII in the second process of MG adsorption aer carrying out thermocatalysis at 318.15 K for 20 minutes) for the four MG/Fe 3 O 4 systems. These q P e are useful to elucidate the MG adsorption enhancement by photocatalysis. They show that the highest one is that of JUN-Fe 3 O 4 (9.48 mg g À1 ) and the lowest one is that of ARM-Fe 3 O 4 (2.27 mg g À1 ). This indicates that the photocatalytic activity of JUN-Fe 3 O 4 is the highest and that of ARM-Fe 3 O 4 is the lowest. As all experimental conditions were kept the same for all adsorption experiments on all four magnetite samples, only the magnetite surfaces' properties are responsible for the adsorption enhancement by the photocatalysis process.
3.7 Inuence of the mediating plant extract's acidity on the preferential and enhanced MG adsorption on magnetite surfaces The results from the analysis of MG adsorption in ambient dark conditions showed that MG was differently adsorbed on the four magnetite surfaces (see Table 6). When comparing the OH group densities on the magnetite surfaces (according to the FTIR spectra analyzed in Section 3.2), it was found that MG adsorption is more preferred on magnetite surfaces that have more OH groups. The adsorption yield was the highest on the ARM- is related to the higher band gap energy value as proof of the quantum size effect. So, the decrease of particle size leads to an increase in the band gap energy. This result is in agreement with that reported by Singh et al. 64 Therefore, one can conclude that the band gap energy increases with the increase of the plant extract's acidity.
The photo-and thermocatalysis adsorption mechanisms are controlled by the photo-and thermogenerated electron/hole pairs, which exhibit a strong tendency to recombine. The lifetime of the electron/hole pairs inuences the photo-and thermocatalytic efficiency. 15,65,66 The results found showed that the thermo-and photocatalytic activities of the magnetite NPs samples differ according to the mediating plant extract's acidity. q P e and q T e are the highest on JUN-Fe 3 O 4 , next on MAT-Fe 3 O 4 , then on ROS-Fe 3 O 4 , and nally on the ARM-Fe 3 O 4 NPs. This indicates that the recombination lifetime of the electron/ hole pairs was more decreased on the JUN-Fe 3 O 4 surface, next on MAT-Fe 3 O 4 , then on ROS-Fe 3 O 4 , and nally on the ARM-Fe 3 O 4 NPs. Seeing that the increase in the direct band gap energy further slows the electron/hole pair recombination, 16 and the band gap energy increases with the increase of the plant extract's acidity, thus one can pronounce that the plant extract has an effect on the recombination lifetime of the electron/hole pairs, where the recombination of the electron/hole pairs is further slowed by the increase of the plant extract's acidity. Thus, the thermo-and photocatalysis enhance the MG adsorption yields and capacities more on magnetite surfaces that are greenly synthesized from more acidic mediating plant extracts. Magnetite NPs greenly synthesized from more acidic mediating plant extracts showed higher thermo-and photocatalytic activities for MG adsorption.

Conclusion
The preferential and enhanced MG adsorption by thermo-and photocatalysis on four greenly synthesized magnetite surfaces has been studied by coupling three processes. In the rst process, MG adsorption on magnetite surfaces was conducted in ambient dark conditions, whereas in the second and third processes, the enhancement by thermo-and photocatalysis were measured in dark conditions and under UV irradiation (365 nm) in ambient conditions, respectively. All four greenly synthesized magnetite samples were characterized by XRD, SEM, ATR-FTIR, and UV-Vis.
The results found showed that: The decrease in the plant extract's acidity leads to the increase of the active site density and, hence, an increase in the MG adsorption yield and capacity.
The mediating plant extract's acidity clearly affects the adsorption enhancement by thermo-and photocatalysis through its effect on the band gap energy of the greenly synthesized magnetite and, consequently, on the recombination lifetime of the electron/hole pairs aer electron excitation.
The band gap energy increases with the increase of the plant extract's acidity, and the recombination speed of the electron/hole pairs is further decreased by the increase of the plant extract's acidity.
Therefore, the thermo-and photocatalysis processes enhance the MG adsorption yield and capacity more on magnetite surfaces that are greenly synthesized from more acidic mediating plant extracts.

Conflicts of interest
There are no conicts to declare.