Investigation of adsorptive removal of heavy metals onto magnetic core–double shell nanoparticles: kinetic, isotherm, and thermodynamic study

The most perilous environmental hazards arise from the contamination of water by heavy metal ions, owing to the non-biodegradability of these metals, as well as their rapid dissemination throughout components of the environment via the food chain. Nano-based adsorbents have been used for the adsorption removal of many heavy metal cations, but separating and recycling them represent significant difficulties in processing. Magnetic core–double shell nanoparticles provide an attractive solution for processing issues, since they are stable and can be easily separated and recycled. Moreover, the shell thickness, composition, and porosity can be easily tuned. In this work, two samples consisting of magnetic core@TiO2@mesoSiO2 nanoparticles with two shell thicknesses (Mag-T-S-0.2 and Mag-T-S-0.4), along with a magnetic core@SiO2@TiO2 nanoparticle sample (Mag-S-T), were synthesized and characterized by TEM, XRD, magnetic strength measurement and zeta potential. TEM images show the developed core–double shell structure with double shell ranging from 60 to 73 nm. The XRD results indicate the impact of the outer shell on the diffraction pattern. The zeta potential shows that all samples had a negative charge at pH over 4. The magnetic character was suppressed after the formation of the double-shell coating; however, the magnetic core–double shell nanoparticles still had magnetization and could be separated when an external magnetic field was applied. The heavy metal adsorptive ability of Mag-T-S-0.2, Mag-T-S-0.4, and Mag-S-T samples was explored to investigate the effects of shell type and thickness along with kinetic, isotherm, and thermodynamic study. The investigated heavy metals included Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II). The results indicate that, for Mag-T-S-0.2, the equilibrium state occurred after 15 min contact time, with adsorption capacity of 238, 230, 210.6, 181.8, and 245.8 mg/g for Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II), respectively. For Mag-T-S-0.4, the equilibrium state occurred after 15 min contact time, with adsorption capacity of 241, 237.6, 173.8, 189.6, and 257.2 mg g−1, respectively. For Mag-S-T, the equilibrium state occurred after 25 min contact time, with adsorption capacity of 137.8, 131.4, 221, 189.6, and 149.4 mg g−1, respectively. When pseudo-first-order and pseudo-second-order kinetic models were applied to investigate the time interval adsorption data for Mag-T-S-0.2, Mag-T-S-0.4, and Mag-S-T samples, the second-order kinetic model was found to be more suitable for describing the process, indicating a fast adsorption mechanism. The adsorption data did not fit well with the Langmuir model, while they did fit well with the Freundlich model, suggesting heterogeneous material surfaces and multi-layer adsorption. Thermodynamic investigations confirmed the spontaneous nature of adsorptive removal, which helps to promote magnetic core@TiO2@mesoSiO2 and magnetic core@SiO2@TiO2 nanoparticles as effective adsorbents for wastewater treatment.


Introduction
Environmental pollution is a serious day-to-day problem faced by all nations in the world.Common pollutants include toxic organic substances like chlorinated and nonchlorinated aliphatic and aromatic compounds, dyes, pesticides, and inorganic compounds like heavy metals [1,2].Exposure to Cd ions can cause serious health problems such as hepatic, pulmonary, testicular, and renal damage along with dysfunction of the hemopoietic and adrenal systems [3].Exposure to lead ions can lead to significant health issues, include nerve, abdomen, bone, eye, and skin problems.Cu ions are also very hazardous to human health and can cause serious health issues, including abdominal pain, vomiting, nausea, and diarrhea, as well as attention deficit disorder, arthritis, and asthma [4].Ni ions can also cause huge health and pathological complications.An accumulation of Ni cations within the body can lead to severe health problems such as lung fibrosis, respiratory tract cancer, and cardiovascular disease [5,6].It was found that exposure to Mn can induce neurotoxicity, and that an accumulation of Mn ions within the body can cause symptoms that resemble Parkinson's disease [7,8].In addition, exposure to Mn ions can impact liver function, as the highest accumulation of Mn ions will occur in liver tissue, which can ultimately lead to hepatic encephalopathy [9].
Hence, the need for strict legislation restricting the use of these pollutants and regulating their safe disposal drives the research community to develop clean and green processes for their removal and decomposition before they can enter water bodies.Adsorptive-removal of pollutants is reported as efficient and applicable process for purification purposes [10,11].In this regard, synthesizing nanomaterials with the desired morphology and functionality for wastewater treatment has attracted intense interest recently [12][13][14][15][16][17][18].Among the various morphologies being explored (e.g., nanotubes, nanowires, nanorods), core-shell porous nanostructures have received much attention not only for their unique or enhanced physicochemical properties, but also for their potential widespread application, including drug delivery, water treatment, and catalysis [19,20].
A magnetic core-shell system can provide easy separation of environmental pollutants from aqueous solution by adsorbing them into the shell, then applying a magnetic field [21][22][23].Compared to other subjects of research in the field of water pollutant removal, the magnetic core-shell structure has more merits since it is more stable in a wide pH range and can be functionalized to adsorb single or multiple heavy metals as well as decompose matrix and organic pollutants [22,[24][25][26].Gutierrez et al developed a novel core-shell nanoparticle system with magnetic core and polymer shell for the removal of polychlorinated biphenyls from contaminated water sources [27].Brumovský et al reported the synthesis of zero-valent iron core-shell (Fe/FeS) nanoparticles with enhanced capacity for remediating sites contaminated with trichloroethylene [28].Huang et al proposed a method of selective separation and recovery of valuable metals from complex polluted wastewater.They coated chitosan onto silica core-shell nanoparticles and applied them as adsorbents for selective removal of silver ions (Ag(I)) from complex wastewater wastewater [29].Chang et al synthesized novel magnetic core-shell MnFe 2 O 4 @TiO 2 nanoparticles loaded on reduced graphene oxide (MnFe 2 O 4 @TiO 2 -rGO) and applied them for adsorptive removal of ciprofloxacin and Cu(II) from water [30].Mokadem et al functionalized magnetic nanoparticles and silica-coated magnetic nanoparticles with 1,2,3-triazole and tested them for their ability to remove Pb 2+ , Cu 2+ , and Zn 2+ .The results showed that the adsorption capacity of 1,2,3-triazole magnetic nanoparticles at pH 5.5 and 25 °C for Cu 2+ , Pb 2+ , and Zn 2+ ions was 87.87, 167.78, and 51.20 mg.g-1, respectively [31].Wang et al developed efficient multifunctional BiOI/γ-Fe 2 O 3 core-shell nanoparticles that could simultaneously photocatalytically oxidize and adsorb As(III) cations [32].
Titania nanoparticles have been studied extensively for their photocatalytic decomposition of many types of organic pollutants [33].Titania nanoparticles possess many advantages, such as chemical stability in acidic/basic media, low toxicity, high surface area, and good adsorption capacity [34].All of these advantages make them good candidates for adsorption removal of heavy metal cations.Similarly, silica nanoparticles have been studied thoroughly for the adsorptive removal of organic and inorganic pollutants since they are chemically inert, possess extraordinary surface area, and provide a stable porous structure and ease of functionalization [35][36][37].However, solid/liquid separation and recovery are the main challenges for titania and silica nanoparticles.The formation of a core-shell structure in which magnetic nanoparticles make up the core and titania or silica makes up the shell could solve the difficulties of separation, recovery, and recycling of many adsorbent nanoparticles [38].SiO 2 is ideal candidate for shell formation through sol-gel using Stöber method.TiO 2 represent the second material after silica that can form homogenous coating.Shell formation using these two materials is relay on using controlled hydrolysis of Si and Ti alkoxides.Construction of homogenous shells around magnetic cores using other metal oxide may face challenges due to formation of non-complete island coating.It is quite important to investigate the role of the type of shell on the adsorption characteristics of a core-shell structure [39][40][41][42].
However, magnetic core-double shell configuration could be more efficient for adsorption application since the first shell could provide a protective layer for magnetic nanocores from being leached in acidic medium.On the other hand, the second shell is functional layer where its role is to adsorb the heavy metal cations from aqueous solution.Nevertheless, magnetic core-single shell configuration has a disadvantage that existence of one shell around magnetic nanocores can cause gradual leaching for Fe 3 O 4 that end up with losing the core and the magnetic character.
The aim of this work was to synthesize magnetic core-double shell nanoparticles in order to assess the impact of shell type (TiO 2 or SiO 2 ) and characteristics (texture, thickness, and charge) on the adsorptive removal of heavy metal cations, and to optimize the adsorption parameters for magnetic core@TiO 2 @mesoSiO 2 and magnetic core@SiO 2 @TiO 2 nanoparticles for removal of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) ions.Furthermore, we investigated the kinetic, isotherm, and thermodynamic characteristics in order to assess the heavy metal adsorption process using the developed magnetic core-double shell nanoparticles.

Synthesis of magnetic core-double shell nanoparticles
Magnetic core-double shell nanoparticles were prepared by sol-gel coating of silica and titania layers interchangeably in magnetic cores.Magnetic Fe 3 O 4 nanocores were prepared by solvothermal process as follows: A specific quantity of FeCl 3 •6H 2 O (3.25 g) was dissolved in a particular volume of ethylene glycol (100 ml).Subsequently, predetermined quantities of sodium acetate (6.0 g) and tri-sodium citrate (1.3 g) were introduced.The mixture underwent vigorous and continuous stirring to ensure thorough homogeneity.It was subsequently transferred to a stainless-steel autoclave, which was internally coated with Teflon, and heated at approximately 190 °C for 12 h.Upon reaching ambient temperature, the resulting magnetic core of Fe 3 O 4 underwent 3 cycles of ethanol washing, followed by drying at 60 °C in an oven for approximately 6 h [43].
2.2.1.Synthesis of magnetic core@TiO 2 @mesoSiO 2 nanoparticles The fabrication process of Mag core@TiO 2 @mesoSiO 2 nanoparticles involved coating a titania layer onto the surface of magnetic nanocores, then applying a silica coating as a second layer, and the resulting samples were subjected to calcination.Specifically, 1 ml of magnetic nanocore was dispersed in ethanol (50 ml) and mixed with ammonia solution (0.15, 28%) under ultrasonic stirring.Tetrabutyl titanate was added gradually at volumes of 0.2 and 0.4 ml.The ensuing reaction was allowed to proceed under mechanical stirring for 24 h.The resulting Fe 3 O 4 core-TiO 2 shell was separated from the mother solution and washed multiple times with deionized water and ethanol.The next step in the coating process involved applying meso SiO 2 onto Mag core@TiO 2 , which was conducted according to the following procedure: Mag core@TiO 2 core was dispersed in ultrasonically treated H 2 O/ethanol (35/140 ml), after which 2 ml of NH 4 OH was added.Subsequently, a solution of cetyltrimethylammonium bromide, a cationic surfactant, was added, followed by tetraethyl orthosilicate (1 ml).The reaction was allowed to proceed under mechanical stirring for 6 h.The resulting magnetic core@TiO 2 @mesoSiO 2 nanoparticles were then separated from the mother solution and washed with ethanol and water.Finally, the samples were calcined at 500 °C for 2 h, resulting in the formation of magnetic core@TiO 2 @mesoSiO 2 nanoparticles.Samples prepared with 0.2 and 0.4 ml of tetrabutyl titanate were labeled Mag-T-S-0.2and Mag-T-S-0.4,respectively (table 1).
2.2.2.Synthesis of magnetic core@SiO 2 @TiO 2 nanoparticles Fe 3 O 4 core magnetic nanoparticles were enveloped by a silica shell using the following methodology: Magnetic cores (1 ml) were ultrasonically dispersed in a mixture of H 2 O and ethanol (35/140 ml).Subsequently, a precise quantity of NH 4 OH (2 ml) was introduced.Following this, a particular volume of tetraethyl orthosilicate was added (1ml) [44].To construct the TiO 2 layer, silica-coated magnetic nanocores were dispersed in ethanol (50 ml) and then ammonia solution (0.15, 28%) was added under ultrasonic stirring.Then tetrabutyl titanate (0.4 ml) was gradually introduced [45].The reaction was allowed to proceed for 24 h under mechanical stirring.The developed magnetic core@SiO 2 @TiO 2 nanoparticles were then extracted from the mother solution and washed multiple times using deionized water and ethanol.The particles were then dried and ultimately underwent calcination at 500 °C for 2 h, leading to the formation of magnetic core@SiO 2 @TiO 2 nanoparticles, labeled Mag-S-T (table 1).40 mA).Nitrogen sorption isotherms were measured at a temperature of 77 K using a Quantachrome NOVA 4200 analyzer (Quantachrome, Boynton Beach, FL, USA).Prior to measurement, the samples underwent degassing in a vacuum at 200 °C for a minimum of 18 h.The specific surface areas were determined by using the Brunauer-Emmett-Teller (BET) method, which relies on an analysis of adsorption data within the relative pressure range of 0.02 to 0.20.Using the Barrett-Joyner-Halenda (BJH) model, pore volume and size distribution were ascertained from the adsorption branches of the isotherms, while total pore volume (Vt) was estimated from the amount of adsorbed material at a relative pressure P/P0 of 0.995.Lastly, the magnetic properties of the samples were determined by using a superconducting quantum interference device (SQUID) magnetometer (Quantum Design, CA, USA).Zeta potential were measured using Malvern zeta nanosizer (Malvern Panalytical, Malvern, UK).

Adsorptive removal of heavy metal cations
A batch process based investigation of magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticles for adsorptive removal of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) was performed as described in our previous works [24,46].In detail, 0.015 g each of Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T adsorbent material was taken in a 50 ml polypropylene tube and mixed with 25 ml of mixed metal solution including Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II), each at 100 ppm.The pH was adjusted to 6 using phosphate buffer.The mixture was shaken for 80 min, then the phases were separated by an external magnetic field.
where C 0 and C e are the primary and final concentration, respectively, of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II); V is the volume of the adsorption solution; and M is the mass of Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T adsorbent material (g).
The described batch process was repeated to study the influence of pH, contact, and equilibrium time on heavy metal cations Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) based on concentration and temperature.
The recycling of magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) after adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) was investigated as described in Habila et al [26] with some modifications.The magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents were regenerated by washing with 2 ml of HCl (0.15 M) twice, then applying them directly to the next usage.The recycling efficiency during recycling was calculated as a ratio based on the first usage (equation ( 2)): where RE % x is the recycling efficiency of cycle x, q x is the adsorption capacity of cycle x, and q 1st is the adsorption capacity of the first cycle = x 1 .( )

Results and discussion
3.1.Characteristics of magnetic core@TiO 2 @mesoSiO 2 and magnetic core@SiO 2 @TiO 2 nanoparticle adsorbents In order to elucidate the formation of the core-shell structure for the prepared nanoparticle adsorbents, figure 1(A) shows the TEM images for magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T).Figure 1(a) shows the formation of the core-shell structure where the magnetic core coating with size ~80 nm has TiO 2 and SiO 2 shells with a total thickness of 60 ± 2.5 nm (table 1). Figure 1(Aa) indicates that due to a low TiO 2 precursor concentration (0.2 ml of tetrabutyl titanate) during the first shell formation step, it was quite difficult to distinguish the formed TiO 2 shell from the second SiO 2 shell.However, the figure also suggests that there was a low agglomeration of magnetic core@TiO 2 @mesoSiO 2 nanoparticles.On the other hand, figure 1(Ab) shows that when the TiO 2 precursor concentration was increased to 0.4 ml of tetrabutyl titanate, more distinguishable shells of TiO 2 and SiO 2 with total thickness of around 73 ± 2.2 nm can be seen.However, the state of agglomeration was found to increase in this sample.At high concentration of tetrabutyl titanate, the reaction kinetics becomes so fast that the formed titanium alkoxide species (Ti oligomers) are increased rapidly to allow the occurrence of homogenous nucleation (formation of separate TiO 2 nanoparticles) along with heterogeneous nucleation (formation of TiO 2 shell).However, existence of large number of TiO 2 nuclei in vicinity of core-double shell nanoparticles could lead to particle agglomeration [45].Finally, when the shell order was changed, with SiO 2 as the first shell and TiO 2 as the second one (figure 1(Ac)), a clearly distinguished double shell structure could be observed, with a total shell thickness of 72 ± 1.8 nm.Ultimately, magnetic core@SiO 2 @TiO 2 showed the highest agglomeration among the samples.In order to confirm the chemical composition of the core-double shell nanoparticles, EDX analysis was conducted for Mag-T-S-0.To investigate the formed phase within magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents, x-ray diffraction spectroscopy measurement was conducted (figure 2).Figures 2(a)-(b) shows the XRD patterns for magnetic core@TiO 2 @mesoSiO 2 nanoparticles.The labeled peaks refer to Fe 3 O 4 and anatase TiO 2 .The significant noise observed in the XRD pattern is due to its amorphous character of the outer silica shell.Moreover, its position as outer shell provides a shielding effect for x-rays and prevent the full and clear appearance of crystalline TiO 2 middle shell peaks upon subjected to XRD measurement.On the other hand, the XRD pattern for magnetic core@SiO 2 @TiO 2 (figure 2(c)) shows spectra with no noise because the crystalline TiO 2 shell was the outer one while the silica shell was the middle layer, which reduced its effect on the diffraction pattern.To correlate the total shell type and thickness with the separation assisted by a magnetic field, room temperature magnetization was measured for magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents (figure 3).It can be seen that the remnant magnetization values of bare Fe 3 O 4 nanoparticles, Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T were 67.7, 44.7, and 35.9, and 34.5 emu g −1 , respectively.All samples exhibited the typical S-shaped magnetization hysteresis loops, with coercivity and remnant magnetization values of zero, confirming their superparamagnetic behavior.The decreased remnant magnetization value can be attributed to the presence of the non-magnetic double TiO 2 and silica coating layers, which occupies a certain volume of the overall sample.This reduction in remnant magnetization value provides further evidence for the successful formation of double TiO 2 and SiO 2 shells onto Fe 3 O 4 nanocores [47].However, still Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T samples can be separated under the applications of external magnetic field.
The pH of the solution can potentially influence the generated surface charge of adsorbents in an aqueous environment.Consequently, this has an effect on the process of adsorbing heavy metal ions into the adsorbent surface.Therefore, a zeta potential pH curve was constructed for magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) samples at pH values from 2 to 10, as shown in figure 4. Changing the outer shell material show significant difference in the zeta-potential behavior.when the silica was the outer shell, the zeta-potential/pH curve will be as shown in figures 4(a) and (b).On the other hand, upon making TiO2 as outer shell, the zeta-potential behavior will be as shown in figure 4(c).It is clear that Mag-T-S-0.2and Mag-T-S-0.4,with an outer silica layer, have similar curves; both samples had a negative charge from pH 4 and the charge increased slowly with increasing pH of the solution.The point of zero charge (PZC) for Mag-T-S-0.2and Mag-T-S-0.4was 3.66 and 3.47, respectively.However, for Mag-S-T with an outer titania layer, the surface charge was +9 mV at pH 2, then was negative at −7.8 and increased significantly with increasing solution pH.The point of zero charge was 3.05 for the Mag-S-T sample.For comparison purpose zeta  potential measurement was conducted for single shell configuration Mag-S (Fe 3 O4@SiO 2 ) and Mag-T (Fe 3 O 4 @TiO 2 ) samples (figure S1).The point of zero charge (PZC) for Mag-S and Mag-T was 3.40 and 2.95, respectively.Single shell configuration Mag-S and Mag-T samples had a similar zeta-potential curve's trend of core-double shell system.
Textural properties (BET surface area and pore volume) of samples Fe 3 O 4 @TiO 2 @SiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)with porous SiO 2 outer shell were 1133 cm 2 /g, 0.74 CC/g and 1270 cm 2 /g, 0.88 CC/g, respectively (table 1).The textural properties of Fe 3 O 4 @SiO 2 @TiO 2 (Mag-S-T) where crystalline non-porous TiO 2 is outer layer were 52 cm 2 /g, 0.03 CC/g.Such pronounced decline in textural properties with changing the material of outer shell is a clear indication of tuning the properties of nanoparticles with changing the material of outer shell.

Optimizing the influence of pH on the capacity for absorptive removal of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II)
The pH of the metal solution is the most important parameter for optimizing the adsorptive removal of heavy metal ions [48][49][50][51].This arises from the fact that pH affects the adsorption process since it impacts the surface charge of adsorbent molecules.The structures of prepared magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents were evaluated for the adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II).The effect of pH was investigated by varying the pH of the metal ion solution from 2 to 7. The capacity for adsorbing Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) onto Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T is presented in figures 5-7, respectively.A weakly acidic medium with a pH of around 6 allowed the highest adsorption of the tested metals for Mag-T-S-0.2and Mag-T-S-0.4,whereas a pH of 4 was the most suitable for Mag-S-T.This may be correlated with the type of outer shell: with SiO 2 as the outer shell, the maximum was obtained at pH 6, while with TiO 2 as the outer shell, the maximum capacity was at pH 4. In addition, the adsorption capacity increased with increasing shell thickness, with higher adsorption observed for Mag-T-S-0.4than Mag-T-S-0.2.In any case, for Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T, the more acidic the medium, the lower the adsorption capacity, which can be attributed to the competition between hydrogen ions and metal cations for adsorption sites in the adsorbent surface.However, increasing the pH of the  solution will lead an increase in the negative charge sites of the adsorbent, which will allow more electrostatic attraction with metal cations and adsorbent sites [52].As indicated in figure 4, the fabricated core-double shell system with silica as the outer shell (Mag-T-S-0.2 and Mag-T-S-0.4)exhibited a smaller range of surface charge variation compared to the system with TiO 2 as the outer shell (Mag-S-T).However, although the highest negatively charged surface for both Mag-T-S-0.2and Mag-T-S-0.4and Mag-S-T was found with an alkaline medium, it was not utilized for pH investigations during adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) due to the precipitation of the metals at higher pH values.For comparison purpose adsorption-pH measurement was conducted for single shell configuration Mag-S and Mag-T samples (figure S2).The results showed that core-double shell configuration possessed superior adsorption capacity compared to coresingle shell system.

Influence of contact time on the uptake of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II)
The time of contact during the adsorption process is a key parameter to achieve efficient removal of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) by the prepared Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T [11,[53][54][55].A time study was conducted to investigate the adsorption capacity of the Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T samples.For Mag-T-S-0.2, the equilibrium state occurred after 15 min contact time, with adsorption capacity of 238, 230, 210.6, 181.8, and 245.8 mg/g for Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II), respectively, as shown in figure 8.For Mag-T-S-0.4, the equilibrium state occurred after 15 min contact time, with adsorption capacity of 241, 237.6, 173.8, 189.6, and 257.2 mg g −1 , respectively, as shown in figure 9.The results of investigating the influence of shell thickness on the adsorption capacity of Mag-T-S-0.2and Mag-T-S-0.4show that increasing shell thickness improved the adsorption capacity for removal of Cd(II), Ni(II), Pb(II), and Cu(II).This improvement may be attributed to the increased adsorbent surface area with increased shell thickness (table 1).For Mag-S-T, the equilibrium state occurred after 25 min contact time, with adsorption capacity of 137.8, 131.4,221, 189.6, and 149.4 mg/g for Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II), respectively, as shown in figure 10.Comparing the performance of Mag-T-S-0.4with Mag-S-T, the adsorption capacity for removal of Cd(II), Ni(II), and Cu(II) was decreased when the outer shell was titania.However, Mag-S-T had a more negative surface, as determined by the zeta potential (figure 4).These results confirm the superior efficiency of Mag-T-S-0.4for the removal of heavy metals, which may be due to the high surface area, as reported in table 1.The fast,  early adsorption that occurred within 15 min for Mag-T-S-0.2,Mag-T-S-0.4and within 25 min for Mag-S-T can be attributed to the ease of diffusion and adsorption of heavy metal cations onto active adsorbent sites.However, as adsorption time increases, the available surface-active adsorbent sites are occupied and the unabsorbed metal cations have to diffuse deeper to available active sites, which slows down the adsorbent rate [56].

Kinetic modeling of adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II)
The progress of the uptake of the investigated heavy metals with high adsorption capacity is considered to be fast [57,58].Herein, kinetic models were applied to assess this fast adsorption process.The fabricated Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T samples exhibited effective adsorptive removal, which could be useful for wastewater treatment with regard to the time-consuming nature of the process [52][53][54][55][56].
The pseudo-first-order kinetic model was applied to investigate the rate of uptake of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) onto magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents.The integrated form of the pseudo-first-order model is stated in equation (3):   )and t.Using the slope and intercept, the values of k 1 and q e were calculated (table 2).The obtained results show poor agreement for the values of experimental and calculated q , e which suggests that the pseudo-first-order model is not suitable for describing the process of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) adsorption onto Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T samples.
The integrated form of the pseudo-second-order kinetic model is stated as equation (4) [54]: Figures 11(B)-13(B) present the plots of t q t and t, from which q e and K can be determined.The calculated and experimental values of q show good correlation, revealing that the pseudo-second-order model is the most consistent for describing the uptake of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) onto Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T adsorbent samples.From the obtained results, the kinetic first-order model showed a bad correlation coefficient as well as significant differences between calculated and experimental qe.However, the pseudo-second-order model was more suitable to describe the adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) onto Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T adsorbent samples based on the correlation coefficient and adsorption capacity values.These results indicate fast uptake of the tested heavy metals onto the prepared adsorbent materials.Table 2. Kinetic constant for pseudo-first-order and pseudo-second-order models for adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) onto Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T.

Pseudo-first-order
Pseudo-second-order q , e exp (mg/g) k 1 (min −1 ) q , e cal (mg/g) R 2 K 2 (g/mg.min)q , e cal (mg/g) R 2 Cd 3.5.Isotherm study of the adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) An isotherm model is the way to express the adsorption relationship between adsorbent and adsorbate.In the Langmuir isotherm model, the adsorbate forms a monolayer on a homogeneous adsorbent surface, and the adsorption energy for the active sites of each adsorbent is similar.On the other hand, in the Freundlich model, the adsorbate forms multiple layers onto the heterogeneous adsorbent surface, where the active sites of the adsorbent possess various affinities for the adsorbate molecules [59].We studied the adsorbed quantity in relation to the concentration of adsorbate solution of the tested heavy metals onto magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents in order to investigate the Langmuir and Freundlich theorems [60][61][62] and determined the adsorption capacity for Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) (figures 14-16).For the Langmuir isotherm, equation (5) was applied: where Q max 0 is the maximum capacity for saturated monolayer adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) onto Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T (mg/g); C e is the concentration of heavy metal ions at equilibrium time (mg/L); q e is the amount of heavy metal uptake at equilibrium (mg/g); and K L is a constant associated with the affinity between adsorbents and heavy metal ions (L/mg).
The correlation coefficient, R 2 , for adsorption of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) onto Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T adsorbents (figures 14(A)-16(A) and table 3) has a lower value, indicating that the Langmuir isotherm cannot be applied to describe the adsorption process.
The Freundlich isotherm model is stated as equation (6): where q e is the amount of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) uptake at equilibrium (mg/g); C e is the concentration of heavy metal ions at equilibrium (mg/L); K f is the Freundlich constant (mg/g)/(mg/L); and n is the Freundlich intensity parameter (dimensionless), which indicates the magnitude of the adsorption driving force or the surface heterogeneity.The values of K f and n can be investigated by plotting log qe versus log C e (figures 14(B)-16(B)).A value of n below 1 indicates favorable adsorption.The results in table 3 show that the adsorption data are well fitted with the Freundlich model, suggesting heterogeneous material surfaces and multilayer adsorption.

Thermodynamic studies
The efficiency of water treatment by adsorption was assessed based on the nature of the adsorption mechanism, which can be chemisorption or physisorption depending on the thermodynamic parameters [63][64][65].In the investigation, ΔG o indicates the possibility of spontaneous adsorption, ΔH o is applied to indicate the nature of the adsorption process, and ΔS o gives information about the degree of randomness [66][67][68][69][70].It was found that the adsorption equilibrium is strongly influenced by the temperature of the system during adsorption of Cd(II),    4 indicate that the adsorption process was spontaneous and exothermic and occurred physically.This finding confirms the applicability of the examined adsorbents, Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T, for the efficient and spontaneous removal of various heavy metal ions from aqueous solution [71,72].
The thermodynamic properties have been intensively investigated for heavy metal uptake.For example, Rafatullah et al investigated the influence of temperature on adsorption efficiency for copper, chromium, nickel, and lead removal, and reported that efficiency increased with increasing temperature, confirming the endothermic adsorption mechanism [73].In addition, Ucun et al investigated the influence of temperature on the uptake of Cr(VI) and described an endothermic adsorption process with positive enthalpy value (62.02 kJ mol −1 ) in the temperature range of 25 °C-45 °C, indicating a chemisorption process [65].
3.8.Recycling of magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents By regenerating and reusing the magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents, high recycling efficiency is reported (table 6).For Mag-T-S-0.2 and Mag-T-S-0.4,efficiency is retained above 90% until the third usage.After five uses, Mag-   The achieved performance for removal of heavy metals from aqueous solution are considered highly effective compared to other adsorbents from literature (table 5), however, the transfer to next phase of the application in the real field still need more investigations to assess the effect of harsh conditions of real wastewater samples on the adsorption efficiency, selectivity and recyclability of the prepared core-double shell adsorbents.

Conclusions
In this study, two types of magnetic core-shell structures, magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle adsorbents, were utilized to investigate the effect of shell type and characteristics on the adsorptive removal of Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) for effective wastewater treatment.TEM observations confirmed the formation of core-double shell structure.Remnant magnetization still noticeable after double shell formation.Zeta potential measurement showed that the developed samples possessed negative charge for pH 4. Adsorption process parameters including adsorption pH and contact time were thoroughly investigated, and kinetic, isotherm, and thermodynamic studies were performed.A weakly acidic medium with a pH of around 6 allowed the highest adsorption of the tested metals with Mag-T-S-0.2and Mag-T-S-0.4,while pH 4 was the most suitable for Mag-S-T.Moreover, the adsorption capacity increased with increasing shell thickness, with higher adsorption observed with Mag-T-S-0.4compared to Mag-T-S-0.2.Kinetic modeling showed fast early adsorption taking place within 15 min for Mag-T-S-0.2and Mag-T-S-0.4and within 25 min for Mag-S-T.When the pseudo-first-order and pseudo-second-order kinetic models were applied for different time intervals of the investigated adsorption data (Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T), the second-order kinetic model was found to be more suitable for describing the process, indicating a fast adsorption mechanism.When Langmuir and Freundlich isotherms were applied to the obtained adsorption data, the results showed that the data were better fitted with the Freundlich model than the Langmuir model, suggesting that the adsorption mechanism tended to proceed with heterogeneous material surfaces and multiple layers.Finally, thermodynamic investigations confirmed the spontaneous nature of adsorptive removal, which could help in promoting magnetic core@TiO 2 @mesoSiO 2 (Mag-T-S-0.2 and Mag-T-S-0.4)and magnetic core@SiO 2 @TiO 2 (Mag-S-T) nanoparticle materials as effective adsorbents for wastewater treatment.
The concentration of heavy metal cations Cd(II), Ni(II), Mn(II), Pb(II), and Cu(II) was measured by ICP-MS.Blank samples without Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T were used in all experiments.The adsorption capacity (qe) of Mag-T-S-0.2,Mag-T-S-0.4.and Mag-S-T adsorbents was calculated by using equation (1): 2, Mag-T-S-0.4,and Mag-S-T samples and spectra are shown in figure 1(B).It is clear that Fe, O, Si and Ti peaks can be observed in EDX spectra for Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T samples.EDX results along with TEM images suggested that magnetic nanocores were coated double shells composed of titania and silica.Furthermore, EDX elemental color mapping was performed and overlay images were shown in figure 1(C).The results suggest the formation of core-double configuration with interchangeable TiO 2 and SiO 2 shells.More detailed color mapping of Fe, Ti and Si elementals for Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T samples are shown in figures S3-S5 in supplementary information file.

Table 1 .
Textural and structural properties of magnetic core-double shell nanoparticles.

Table 5 .
Comparison of adsorption capacity of Mag-T-S-0.2,Mag-T-S-0.4,and Mag-S-T adsorbents with other adsorbents from the literature.