Highly efficient engineered waste eggshell-fly ash for cadmium removal from aqueous solution
Characterization of EFM adsorbentBET analysisThe surface properties of newly prepared adsorbent and its components were investigated through nitrogen adsorption–desorption isotherms. Table 1 presents the results of the adsorbent the textural properties assessment.Table 1 EFM adsorbent and raw materials used (magnetite, eggshell and fly ash)—specific surface area determinate by Brunauer–Emmett–Teller theory (BET).Full size tableFrom the quantitative data reported in Table 1 it can be observed that eggshell BET/N2 surface area is 0.67 m2/g, value similar to that in the literature15,31,40.According to Table 1 the BET/N2 specific surface area for fly ash, is 4.961 m2/g. Apparently the value obtained in this study seems higher than in the data reported in the literature (0.414 m2/g). However, it should be noted that both the type of ash used and the experimental conditions of this study differ from the data reported in the literature16,22.As expected, very different values were obtained for the specific surface of M1 (25.196 m2/g) and respectively 26.866 m2/g for M2, which correspond to the adsorbent prepared in the two molar ratios studied. This difference can be justified by the variation of the ratio between eggshell and fly ash in M1 and M2, respectively.Physical properties of adsorbent (pore size, pore volume) were investigated using the low-temperature (77 K) nitrogen adsorption–desorption isotherms. As presented in Fig. 1 the isotherms of EFM fitted in a type II isotherm with a H3 hysteresis loop, indicating a macroporous structure for the both adsorbent molar ratio (M1 and M2)38.Figure 1The nitrogen adsorption–desorption isotherms for EFM adsorbent.Full size imageXRD studiesThe mineralogical compositions of the raw materials as well as of the adsorbent were studied through XRD analysis. The size of the crystalline domains was evaluated by means of the Debye–Scherrer formula (Eq. 14)41,$$D = frac{0.89lambda }{{beta cos left( theta right)}}$$
(14)
where (lambda) is the X-ray wavelength of Cu K-α ((lambda = 0.15406;{text{nm}})), (beta) is the full width at half maximum in radians and (theta) is the Bragg angle.From the most prominent peak, one gets D = 21.6 nm.The XRD spectrum of magnetite sample (Fig. 2) shows the diffraction peaks of the well crystallized spinel phase magnetite Fe3O4 (COD 9005837) with average crystallite size of 21.6 nm. In the XRD spectrum of eggshell sample (Fig. 3) are recorded the diffraction peaks of the single phase well crystallized calcite CaCO3 (COD 9000965) with mean crystallite size of 125.8 nm15,40.Figure 2XRD spectra of magnetite.Full size imageFigure 3XRD spectra of eggshell.Full size imageThe XRD spectrum (Fig. 4) shows that the fly ash sample has a complex composition. Four crystalline phases have been identified whose characteristics are presented in Table 2.Figure 4XRD spectra of fly ash sample.Full size imageTable 2 The phase compositions of fly ash sample.Full size tableFrom the data presented in the Table 2 indicates that fly ash sample used in the preparation of the adsorbent is not non-hazardous solid waste22.In the XRD spectra of M1 (Fig. 5a) are visible the diffraction peaks characteristic of the crystalline phases existing in the eggshell (calcite CaCO3), the fly ash (but only those of the phase with more intense peaks, quartz SiO2), and magnetite (Fe3O4). Because the eggshell is mixed in a larger proportion than the other two components, the CaCO3 peaks are the most intense.Figure 5(a) XRD spectra of M1. (b) XRD spectra of M2. (c) XRD spectra of magnetite, eggshell, fly ash, M1 and M2.Full size imageAnalyzing the XRD spectrum obtained for M2 (Fig. 5b), the crystalline phases that can be identified are: magnetite Fe3O4, calcite CaCO3, quartz SiO2, and corundum Al2O3. The most intense are the magnetite peaks. In this sample, because the ash is mixed in a larger proportion, the SiO2 peaks are more intense and another crystalline phase present in it (Al2O3) becomes visible in the spectrum.Figure 5c shows the overlapping XRD spectra of M1, M2 and raw materials (magnetite, fly ash and eggshell). In both M1 and M2, the phase peaks of individual components can be observed.However, due to the different materials crystallinity, only the most intense peaks appear by overlap. Also, the phase diffraction lines (Figs. 2, 3, 4, 5) are no longer visible.SEM micrographsThe surface morphology and particle size of raw materials and adsorbent were investigated through SEM technique. The micrographs are presented in Figs. 6, 7, 9, 10, 11, 12, 13 and 14.Figure 6Two-dimensional image of the magnetite particle obtained by the SEM technique.Full size imageFigure 7Two-dimensional image of the ash fly particle obtained by the SEM technique.Full size imageThe SEM image of magnetite (Fig. 6) suggest that particles are of nanometric dimensions (with the average size about 21 nm), uniform and with a cubic structure25,42,43.Figure 7 shows that the fly ash particles are of porous spherical shapes with different sizes as well as porous irregularly or angularly shaped particles16,44. As can be seen in (Fig. 7) the surface of the ash sphere is irregular and has streaks due to the mechanical and thermal stress.Figure 8 presents the elemental composition of the ash fly determined through EDX analysis.Figure 8EDS spectra of fly ash sample.Full size imageAccording to the data from EDX (Fig. 8), there are only seven elements which are predominant in sample: aluminum, iron, magnesium, calcium, silica, oxygen and sulphur16,45.SEM micrograph of eggshell sample (Fig. 9a,b) indicates a different size (about 100 nm) irregular crystal on multihole surface structure36,40.Figure 9(a) Two-dimensional image of the eggshell particle obtained by the SEM technique. (b) Two-dimensional image of the eggshell particle obtained by the SEM technique.Full size imageThe morphology of M1 (Fig. 10a,b) indicates the presence of agglomerations of particles of different sizes in the nano field, spherical shape, cubic shaped and irregular crystal structure sizes, suggesting a good connectivity between them.Figure 10(a) Two-dimensional image of M1 particle obtained by the SEM technique (magnitude 3 µm). (b) Two-dimensional image of M1 particle obtained by the SEM technique (magnitude 5 µm).Full size imageAlso, the (Fig. 10b) indicates that the cubic-shaped particles characteristic of magnetite (Fig. 6) loaded into the pores of the ash and eggshell particles.The Fig. 11 shows the live map for M1 and the distribution of the identified elements.Figure 11SEM M1- Live map.Full size imageThe SEM micrograph of M2 (Fig. 12a,b) the same agglomerations of particles of different nano-sizes, spherical shape, cubic shaped and irregular crystal structure sizes are observed as in the case of SEM graph for M1.Figure 12(a) Two-dimensional image of M2 particle obtained by the SEM technique (magnitude 3 µm). (b) Two-dimensional image of M2 particle obtained by the SEM technique (magnitude 5 µm).Full size imageThe Fig. 13 shows the live map for M2 and the distribution of the identified elements.Figure 13M2 SEM—Live map.Full size imageThe comparative analysis of the Fig. 13 showing Live map for M2 and M1 (Fig. 11) highlights the presence of differences regarding the proportion of identification elements in the two samples, due to the different molar ratio between eggshell and ash.In the Fig. 14a can be observed a larger number of spherical particles characteristic of district heating ash, as a result of the change in the ratio between the two wastes (eggshell:ash = 3:1), loaded with magnetite particles. At the same time, in SEM the micrograph for M1 (Fig. 14b) is much more obvious the multihole structure of the eggshell.Figure 14(a) Two-dimensional image of M1 particle obtained by the SEM technique (magnitude 30 µm). (b) Two-dimensional image of M2 particle obtained by the SEM technique (magnitude 50 µm).Full size imageThe analysis of the SEM micrograph (Fig. 14a) of the M1 sample (in which the eggshell component is predominant) indicates that the multi porous structure of the eggshell is loaded with both the cubic-shaped particles of the magnetite and the spherical ones belonging to the ash sample. This aspect is much more visible in the case of SEM micrograph of the M2 sample (Fig. 14b), considering the fact that in this ash is found in the majority proportion (magnetite:eggshell:ash = 1:1:3).This result suggests that through the procedure of mechanical alloying in the mill with high energy balls were achieved simultaneously:
1.
reducing the particle size of magnetite, ash and eggshell;
2.
individual functionalization of each waste (eggshell, ash) with magnetite particles;
3.
a new, nanosized material in which the double functionalization of the eggshell with ash particles functionalized with magnetite was achieved simultaneously with the loading of the pores of the eggshell surface with the magnetite particles.
By modifying the structure of the two wastes from the composition of the newly obtained material (decreasing the number of pores) leads to increased surface areas, confirmed by the results of the BET analysis (Table 1) and implicit sorption sites suggesting an improvement of adsorbent properties.FT-IR studiesFigure 15 shows the IR spectra for EFM adsorbent raw materials (magnetite, fly ash and eggshell).Figure 15IR spectra for adsorbent raw material samples (magnetite, eggshell and fly ash).Full size imageFT-IR spectra for EFM engineered adsorbent are presented in Fig. 16.Figure 16FT-IR spectra of adsorbent (both molar ratios: M1 and M2) and its raw materials.Full size imageThe FT-IR spectra for adsorbent (at the both molar ration: M1 and M2) presents the vibrational bands characteristic of magnetite at 589 and at 432 cm−1associated with Fe–O stretch vibration46. The peaks assigned to the fly ash component: at 588 cm-1 Ca O group, at about 670 cm−1 attributed to the Al–O–Al bending vibration, at 1100 cm−1 is associated with X–O (X = Al, Si) and asymmetric stretching vibrations and band at 830 cm−1 specific to AlO4 coordination16,22,47,48,49. In addition, in the adsorbent FTIR spectra (Fig. 16) were found the characteristic IR bands eggshell component (Fig. 15). Thus, peak at 712 cm−1 (correspond to CaO stretching vibration), peaks at 875 and 1423 are attributed to C–O stretching vibration. The bands at 1798 and 2515 cm−1 are associated with O–C–O and peaks at 2875 respectively at 2981 cm−1 are due to CH– symmetric and asymmetric stretching vibration15,31,50. The position of O–H peak at 3740 cm−1 indicates the presence of moisture and water molecules15,22. As expected, the intensity of the peaks differs in M1 and M2, due to the different molar ratio between two of the raw materials that are part of the adsorbent component (fly ash and eggshell). These results are in close agreement with the literature and theoretical values confirms the presence of magnetite, fly ash and eggshell in adsorbent (at both molar ration: M1 and M2).Thermogravimetric analysisFigure 17a presents the thermal analysis results for fly ash sample.Figure 17(a) Thermogravimetric analysis of the fly ash sample in the range of 30–500 °C with a heating rate of 10 °C/min in open aluminum crucibles in the air atmosphere. (b) Thermogravimetric analysis of the eggshell with a heating rate of 10 °C/min up to 500 °C.Full size imageThe thermal analysis performed in the interval 30–500 °C highlighted two stages of decomposition. The first stage takes place in the range of 30–49 °C with a loss of 0.22% of the sample mass. This decomposition can be attributed to water loss. This process is visible in the DTG curve with a maximum at 45.5 °C, but also on the Heat Flow curve with a maximum at the same temperature and characterized by an exothermic process with ΔH = − 12.44 J/g. The second process presents a continuous thermal decomposition with a maximum observable on the DTG and HF curve at 480 °C, characterized by an exothermic effect. The decomposition does not end in the studied interval. The total weight loss is 2% of the sample mass16.The thermal analysis, in the range of 30–500 °C, performed for the eggshell sample (Fig. 17b), revealed a complex thermal decomposition. This decomposition has several stages that are difficult to separate. It is known that, in addition to inorganic calcium carbonate compounds, in the eggshell are present a multitude of organic components such as: proteins as main constituents, small amounts of carbohydrates and lipids51,52.At the same time, uronic acid is also present, which plays an important role in the resistance of the shell, such as sialic acid in very low concentration and two glycosaminoglycans, including hyaluronic acid, as well as a copolymer consisting of chondroitin sulfate-dermatan sulfate. There is also limited information on variations in nitrogen concentrations and the amino acid composition of the eggshell. A better understanding of the chemicals present in the composition of the eggshell is very important for its application in various fields, including for the purpose of absorbent material.The analysis of the TG curve highlights three hardly separable decomposition stages, the last of which is characterized by a complex multistage decomposition process. It is observed that in the interval 30–100 °C which can be attributed to dehydration, followed by the loss of crystallization water in the range 100–266 °C (4.8% of the sample mass) and then the complex decomposition of organic components in different stages depending on their stability until at 500 °C15.The last decomposition stage results in the loss of 80% of the total mass of the sample. Over 500 °C the decomposition of the inorganic component takes place, namely Ca carbonate. It can be seen that in the analyzed sample the weight of inorganic component is relatively small, namely 12.2% of the sample mass. During the decomposition stage from the interval 266–500 °C several maxima are observed on the DTG curve, which led us to conclude that simultaneous decompositions of several organic compounds take place, observing maxima at 345, 363, 374, 403, 408, 412, 430, 466 and 470 °C. The same main decomposition steps are faintly visible and the HF curve with processes in most cases exothermic. At temperatures higher than 266 °C and on this curve are visible several processes, most of which are exothermic, which can be attributed to the oxidation of organic compounds and their decomposition. The residue left after the thermogravimetric study (performed up to 500 °C) is calcium carbonate15,53.Subsequently, the mixture of the two wastes was analyzed in the two molar ratios:eggshell:fly ash = 3:1 and eggshell:fly ash = 1:3, respectively.The profile of thermogravimetric analysis for the binary mixture eggshell:ash fly in a 1:3 mass ratio performed in the range of 30–500 °C is depicted in Fig. 18.Figure 18Thermogravimetric analysis for eggshell: fly ash binary mixture in a 1: 3 mass ratios obtained in the range of 30–500 °C.Full size imageThe thermal analysis performed in the case of the binary mixture of eggshell and ash, in a 1:3 molar ratios, highlights the decomposition stages of the two components. Namely, the stage of water loss within the ash is visible, to which is added the loss of moisture observed in the case of the eggshell. On the HF flow is visible the exothermic process with a maximum of 45 °C and a ΔH = − 17.023 J/g which represents a sum of the two processes mentioned above. Two other processes are visible on the TG curve, one in the temperature range, 51–213 °C, with a mass loss of 0.27% of the sample mass. Then followed by a loss of 1.74% in the temperature range 213–405 °C. The thermal decomposition continues even above this temperature and the decomposition process was not completed in the studied temperature range.A thermogravimetric study was performed for the same binary mixture but in the eggshell:fly ash = 3:1 molar ratio. The results are presented in the next figure (Fig. 19).Figure 19Thermogravimetric analysis for eggshell: fly ash binary mixture in a 3:1 mass ratio obtained in the range of 30–500 °C.Full size imageIn the case of the thermal analysis of the binary mixture of eggshell and ash in a 3:1 molar ratio, the decomposition stages and the thermal behaviour of the individual components in correlation with the mixing ratio are very clearly visible.Magnetic measurementsThe magnetic properties of the samples: magnetite, M1 and M2 were investigated with an induction hysteresis-graph at low frequency driving field (50 Hz)54. And the hysteresis loops are presented in Figs. 20, 21 and 22. It was found that the samples reveal ferromagnetic behaviour and from the measured hysteresis loops the saturation magnetization ((sigma_{S})), the coercive field (Hc) and the remnant magnetization ((sigma_{R})) were determined. The results are presented in Table 3.Figure 20The hysteresis loop of sample M2.Full size imageFigure 21The hysteresis loop of sample M1.Full size imageFigure 22The hysteresis loop of magnetite.Full size imageTable 3 The values of coercive field (Hc) and remnant magnetization ((sigma_{R})) of M1, M2 and magnetite sample.Full size tableAs expected, the largest value of the saturation magnetization is that of the sample consisting entirely of magnetite. By diminishing the content of ashes from the thermal power station (from three parts in M2 sample to one part in sample M1) a small increase of the saturation magnetization was observed, from 14.06 to 15.12 emu/g (see Table 3). This can be explained by the presence of diamagnetic compounds within the ashes of the thermal station, the decrease of which led to the increase in the saturation magnetization of the sample M1, as compared to the sample M2. All three samples have small values of the remnant ratio, (sigma_{R} /sigma_{S}), which is an indication of the ease with which the magnetization reorients to the nearest easy axis magnetization direction after the remove of magnetic field.The dependencies on frequency of the complex magnetic permeability of the samples, (mu left( f right) = mu^{prime}left( f right) – imu^{primeprime}left( f right)), measured at room temperature, over the frequency range 3 kHz to 2 MHz are presented in Fig. 23. The measurements were performed using an Agilent LCR-meter (E-4980A type) in conjunction with a coil containing a vial in which the samples were placed. Details on the method of measurements of the real, (mu^{prime}left( f right)) and imaginary, (mu^{primeprime}left( f right)) components of the complex magnetic permeability are given in a previous study55.Figure 23Frequency dependence of the magnetite, M1, M2 of the complex magnetic permeability.Full size imageIn the frequency range in which the measurements were made, samples M1 and magnetite exhibits visible relaxation peaks of (mu^{primeprime}left( f right)), at the frequency of 30 kHz. Even if the M1 sample and the M2 sample have the same amount of magnetite, due to the diamagnetic compounds in the fly ash, the relaxation peak of the M1 sample is very attenuated (little visible, almost missing).Given the small size of the magnetite particles in the samples (on the order of tens of nanometers), they do not have a multi-domain magnetic structure. Thus, the only magnetic relaxation process, measurable in the radio frequency field, is the Neel relaxation process. The Néel relaxation time, (tau_{N}) is given by Eq. (15)56$$tau_{N} = tau_{0} exp left( {frac{Kv}{{k_{B} T}}} right)$$
(15)
where K is the effective anisotropy constant of the material from which the magnetic particles are made of, v is the magnetic volume of particles, kB is the Boltzmann’s constant, T is the temperature and (tau_{0}) is a constant in order of 10–9 s56.Assuming that the frequency dependence of the complex magnetic permeability,(mu left( f right) = mu^{prime}left( f right) – imu^{primeprime}left( f right)) obeys the Debye dispersion relations, then the frequency corresponding to the maximum of (mu^{primeprime}left( f right)) is correlated with the relaxation time by the relation, (2pi {kern 1pt} ftau = 1). For measurements at room temperature, with f = 30 kHz, the magneto-crystalline anisotropy constant of magnetite, K = 1.1 × 104 J m−3 and (tau = tau_{N}), under assumption of spherical shape of particles, one gets a magnetic diameter of the magnetite particles, d = 18.2 nm. This value compares favourably with the values measured by SEM and X-ray diffraction.Adsorption propertiesEffect of adsorbent dosageFigure 24a and b show the relationships between different material dosage and the cadmium removal efficiency and respectively adsorption capacity.Figure 24(a) The relationship between different material dosage and the cadmium removal efficiency. (b) The relationship between different material dosage and the cadmium adsorption capacity.Full size imageAccording to the Fig. 24a and b, cadmium removal efficiency and adsorption capacity depending on the amount of adsorbent used shows an upward trend (for quantities between 0.05 and 0.25 g), reaches a maximum of 0.25 g adsorbent (99.9% and 75.48 mg/g for M1 and respectively 99.8% and 75.46 mg/g for M2), after which both removal efficiency and heavy metal adsorption capacity gradually decrease with the increase in the adsorbent dose (0.3 g). These results suggest that the increased amount of adsorbent provides a supplement to the free active sites, but after reaching equilibrium, it leads to the formation of agglomerations and consequently to a decrease in the number of available active sites22,23.Effect of initial concentration on cadmium removal efficiencyFigure 25a shows the influence of heavy metal initial concentration on cadmium removal efficiency. It can be seen that removal efficiency shows an upward trend simultaneously with the increase of the initial cadmium concentration in the range 0–33.5 mg/L. The maximum removal efficiency (99,9% for M1 and respectively 99.8% for M2) was reached at a concentration of 28.5 mg/L, after which the decrease in cadmium removal efficiency begins.Figure 25(a) Relationship between initial concentration and removal efficiency (%). (b) Relationship between initial concentration and adsorption capacity (mg/g).Full size imageAccording to the Fig. 25b, in the same cadmium concentration range (0–33.5 mg/L), the adsorption capacity shows a similar trend, reaching a maximum at 28.5 mg/L (75.48 mg/g for M1 and respectively 74.46 mg/g for M2), and after which gradually decreases.These results indicate the initially an increase in the concentration of heavy metal causes an increase in the amount of Cd2 + ions and implicitly in the possibility of interaction with the active sites of the EFM adsorbent. And after reaching equilibrium, the amount of available metal ions is disproportionate compared to the decreasing number of free sites in the adsorbent, causing a decrease in the adsorption efficiency of the new engineered magnetic adsorbent used in the study15,57.Effect of pHThe wastewater pH is one of the top parameters with highly influence on the adsorption process efficiency having impact direct on the adsorption rate and adsorption capacity as fluctuations in the pH value of the solute induce changes in the degree of ionization of the adsorptive species and the of adsorbent surface23,57.In this study was investigated the pH influence toward the cadmium removal using the prepared material in the pH range of 3.0–7.0, to avoid the precipitation of Cd(OH)2 at pH values > 715.According to the experimental results presented in Fig. 26a and b, the increase in pH value (between pH 3 and pH 6) leads to a significant increase in adsorption efficiency and adsorption capacity. The adsorption efficiency and adsorption capacity reach a maximum value (99.9% and 75.48 mg/g for M1 and respectively 99.8% and 75.46 mg/g for M2) at pH 6.5, after which it decreases slightly. This could be explained as follows: at low pH values is a competition between protons and Cd2+ to occupy the active sites of the adsorbent, even if they are available in large numbers. An increase in pH simultaneously leads to a decrease in the competition of protons and electrostatic repulsion forces, which induces an increase in cadmium removal efficiency. At pH > 6.5 the removal efficiency begins to decrease as increased hydroxyl ion generation occurs to the detriment of Cd2+ ions. Therefore, the optimal pH 6.5 was chosen for subsequent experiments15,16,25,58,59.Figure 26(a) Effect of pH variation on cadmium removal efficiency. (b) Effect of pH variation on adsorption capacity.Full size imageEffect of contact timeFigure 27a showed the relationship diagram between the contact time and cadmium adsorption capacity.Figure 27(a) Effect pf contact time on cadmium adsorption capacity (mg/g). (b) Effect of contact time on cadmium removal efficiency (%).Full size imageIt can be observed from the Fig. 27a and b that the increase of the contact time determines an increase of the adsorption capacity and of the removal efficiency respectively. Both reached the maxima at 120 min The maximum of cadmium adsorption capacity was 75.48 mg/g for M1 respectively 75.46 mg/g for M2, and the maximum of removal efficiency was 99.9% for M1 and respectively 99.8% for M232.This performance can be attributed to the higher surface, the microporous structure that results from the experimental conditions of this study15.The analysis of this diagram indicates that the adsorption of cadmium takes place in three distinct phases:
1.
0–90 min, characterized by adsorption is fast due to the large number of active sites available on the surface of the adsorbent.
2.
the second phase, 90–120 min, the adsorption is slower which can be attributed to the diminution of the free adsorbent active sites;
3.
phase three:120–330 min, corresponds to the time interval in which there are no more free sites on the surface of the adsorbent and the adsorption has reached equilibrium.
According to the experimental results, optimum time in which the adsorption reaches an equilibrium is 120 min and was selected for the next investigations16,60.Effect of temperature on absorption processTemperature represents a key parameter in adsorption process. Therefore, the influence of temperature on cadmium adsorption on prepared material in the two different molar ratios (both M1 and M2) was investigated in the range of 5–50 °C (278.15–323.15 K). The cadmium removal efficiency and adsorption capacity increase first and then a very slight decrease occurs with the increase of temperature (Fig. 28a,b).Figure 28(a) Relationship between temperature and heavy metal removal efficiency. (b) Relationship between temperature and heavy metal adsorption capacity.Full size imageAt 25 °C the maximum removal efficiency is reached (99.89% for M1 and respectively 99.64% for M2). At the same temperature the heavy metal adsorption capacity is maximum of 75.48 mg/g for M1 and 75.43 mg/g for M2. This can be explained by the fact that within this temperature range indicated the favorability for the heavy metal mobilization and thus contact between cadmium and active sites from adsorbent. The relationship between temperature and cadmium adsorption effect indicates that in the range of 5–25 °C the cadmium absorption on prepared material is an endothermic process (physical adsorption). At 25–50 °C the adsorption process becomes exothermic and chemisorption occurs. However, the removal efficiency remains very high even at a temperature of 50 °C (98.78% for M1 and respectively 97.74% for M2).Comparison of cadmium removal efficiency for with other adsorbentsA comparison of cadmium removal efficiency of the newly engineered adsorbent (EFM) with other adsorbents reported in literature is presented in the next table (Table 4).Table 4 Comparison of the removal efficiency of newly nanosized magnetic adsorbent (at both molar ratios: M1 and M2) with the one reported in the literature (selected study) for some adsorbent materials that use the similar waste.Full size tableThe performance of the nanosized adsorbent EFM (at both molar ratios) can be attributed to the higher surface area, the microporous structure that results from the experimental conditions of this study15.Comparison of cadmium removal efficiency with the raw materialsThe removal efficiency of EFM adsorbent compared to that of its raw materials (fly ash, eggshell and magnetite) was investigated as the effect of contact time on the adsorption process. The relationship between the removal efficiency and contact time is presented in Fig. 29. It can be observed that there is an increase in the efficiency of removing heavy metal for all five investigated adsorbents (eggshell, ash, magnetite, M1 and M2) with a maximum of two hours of contact. According to the experimental results presented in Fig. 29, the best cadmium removal efficiency was obtained for M1 (99.89%) and 99.80% for M2, followed by eggshell (95.23%), fly ash (76.31%) and magnetite (71.44%).Figure 29Relationship between adsorbents removal efficiency and contact time.Full size imageThen, a very slight decrease occurs with the increase in contact time. These results confirm the cadmium removal efficiency dependence on the specific surface area and pores (number of available active sites) of the adsorbent used (Table 1).The maximum cadmium removal efficiencies determined experimentally in this study for the raw materials (eggshell, ash and magnetite) corroborated with the data reported in the literature15,25,33,61.Adsorption IsothermsThe absorption mechanism evaluation can be performed through an isotherm adsorbent study. The equilibrium isotherm plays a key role in the investigation of the adsorption behaviour.Due to their simplicity and convenient accuracy, Langmuir and Freundlich’s models are the most commonly used to adjust an adsorption process.Langmuir models provides information on the interaction between the solute and the monolayer surface of the adsorbent. The main working hypotheses of this model are: (1) adsorbent surface consists of uniform, identical sites distributed on the surface of adsorbent (2) adsorbent process takes place only on the surface of the adsorbent and (3) no contact between adsorbed molecules on the surface of the adsorbent.Freundlich model is appropriate to monolayer and multilayer adsorption processes on multiphase surfaces. This isotherm gives an expression on adsorbent surface heterogeneity and the variation in the heat of adsorption process. The applicability of the Freundlich model is limited by adsorption processes that take place at high pressures, but this restriction does not apply to the Langmuir model.These two adsorption isotherm models were applied in order to identify and implement an optimal model that adequately reproduces the experimental results obtained in this study were employed to study the mechanism of cadmium adsorption on the prepared material60,62.The parameters calculated as well the coefficient of correlation (R2) for both Langmuir and Freundlich models are presented in the Table 5.Table 5 Parameters of adsorption Langmuir and Freundlich isotherms for cadmium adsorption.Full size tableAs shown in the Table 5 both models fitted well for the experimental results. The maximum capacities calculated are close to the values for each component of the prepared adsorbent material (magnetite, eggshell and fly ash) and maximum capacities obtained at equilibrium (Table 5)22,23,57,63,64. However, according to the values of the correlation coefficient, R2, the behaviour of cadmium absorption suits better with Freundlich model (the higher correlation coefficient) suggests that the adsorption for cadmium ions was a multi-molecular layer adsorption process. The values for the Langmuir constant, KL, or equilibrium parameter for absorbent (the both molar ratios, M1 respectively M2) falls within the range 0 1 indicated a favourable adsorption process. Moreover, Freundlich dimensionless constant n values having greater than 1 suggests a favourable adsorption process that occurs on the investigated EFM adsorbent heterogeneous surfaces62,65,66.Adsorption kinetic studyThe kinetic models provide information on the efficiency of the adsorbent, the dynamic parameters (rate, time, etc.) of the adsorption process. The cadmium adsorption process on the prepared material was investigated employing linear and non-linear of pseudo-first-order (Eq. 6) pseudo-second-order (Eq. 7) and intraparticle diffusion models (Eq. 8) to fit the obtained experimental adsorption data. The Fig. 30a–c depicted the plots of the first-order, second-order and intraparticle diffusion models for the cadmium adsorption on nano-engineered adsorbent (EFM).Figure 30(a) Pseudo first-order model fitting diagram. (b) Pseudo second-order model fitting diagram. (c) Intraparticle diffusion model fitting diagram.Full size imageThe kinetic parameters were obtained from the slope and intercept of the fitting plots of adsorption reaction models: pseudo first-order model (the correlation between log(qe-qt) against time), respectively the pseudo second-order model (correlation between t/qt function on time) and adsorption diffusion model: intraparticle diffusion model (the plot as function of ({t}^{1/2})).The results of fitting parameters on these kinetic models are presented in Table 6.Table 6 Kinetic parameters for cadmium adsorption on nano-engineered adsorbent (EFM) at both molar rations (M1 and M2).Full size tableAccording to the data obtained in Table 6, the coefficients of adsorption reaction models have both values close to one, slightly differing only at the fourth decimal. It could suggest that cadmium removal is achieved through a physical and chemical adsorption process. It must be noted that were obtained higher values for the correlation coefficient (R2) and the calculated adsorption capacity value is very similar to those determined experimentally in the case of the pseudo-second-order kinetics model. Therefore, pseudo-second-order kinetic model was more suitable to describe the adsorption process. This indicating a chemical adsorption is assumed as the rate-limiting step for the cadmium adsorption on prepared material, involving an electron exchange between adsorbent and adsorbate (cadmium occurs with formation of strong chemical bonding)23.According to the Fig. 30c the allure of intraparticle diffusion model includes three regions. The first region corresponds to a boundary diffusion (cadmium diffusion on the prepared material exterior surface). The second region is related to heavy metal intraparticle diffusion into the pores of nano-engineered adsorbent (EFM). The third region represent the cadmium adoption into the interior site of the EFM. Since, the slope of the three regions gradually decreases (Ki1 > Ki2 > Ki3) is assumed that boundary diffusion is the limiting region, followed by intraparticle diffusion15,67. The results indicate that beginning of the adsorption process cadmium ions can be quickly bound on the prepared material exterior surface. In the intraparticle diffusion process (second region) there is a gradual decrease in adsorption at the sites on the adsorbent surface (adsorption capacity reaches the maximum value). Then, cadmium adsorption takes place on the available sites inside the adsorbent, generating significant mass transfer resistance and reaching the adsorption equilibrium and the adsorption rate gradually decreases68,69,70.The adsorption models used provide information on both the performance of the prepared material and a perspective of the adsorption mechanism.Thermodynamical studyThe Gibbs free energy in adsorption process was calculated according to the corresponding equation (Eq. 10). The thermodynamic parameters ΔS and ΔH were obtained from the slope and intercept of the adsorption thermodynamic curve. The obtained results are presented in next table (Table 7).Table 7 Thermodynamic parameters for the cadmium adsorption on adsorbent.Full size tableFrom these data obtained (Table 7) can be found that the free energy variation value of the adsorption process has negative values (ΔG α lower than 0.05 (α = 0.05), which suggests that between the M1 and M2 there are not statistically significant differences. More
