3.1 Characterization of the prepared nano materiales
The examination of the adsorbents involved a thorough analysis utilizing X-ray diffraction (XRD), referencing ASTM cards for detailed interpretation. Fig. S.3.(a-e) presents the XRD profiles of the synthesized alumina nanoparticles. The corresponding crystal d values, chemical formula, and crystal system were determined by cross-referencing with ASTM cards, providing a comprehensive characterization of the crystalline structure of the synthesized alumina nanoparticles in this study.The characteristic peaks of alumina at 2θ = 19°, 31°, 37°, 39°, 46°, 61°, and 67° denoted the (440), (333), (400), (222), (311), (220), and (111) crystal planes of alumina due to JCPDS card NO. 10–042 51. The XRD pattern of prepared nano γ-alumina from nepheline syenite ore (γA1) is shown in Fig. S.3.(a), which did not match well with the characteristics of the standard sample of γ-alumina (Table S.1). The identification of additional peaks at 2θ values of 7.31°, 28.61°, 28.40°, and 58.13° served as conclusive evidence of impurities within the synthesized material. Specifically, the peak at 2θ = 28.61° was attributed to impurities stemming from nepheline syenite ore, while the remaining peaks were indicative of salt impurities formed during the pH adjustment process through the addition of sodium hydroxide and hydrochloric acid. Notably, the alumina index peaks were discerned at 2θ = 31.75°, 45.51°, and 66.28°, with the latter peak exhibiting a comparatively lower intensity. The characteristic of peaks including the angle of diffraction (θ), the full width at half maximum (FWHM), and the intensity of peak (I) are provided in Table S.2.The approximate crystal size of γA1was calculated according to Debye-Scherrer’s equation (D = kλ/β cos(θ)) 52–55. In this equation D and K stand for the particle size and the constant equal to 0.94, respectively. The wavelength of the X-ray and full width at half maximum of the diffraction peak are λ and β, sequentially. The average crystallite size for γ-alumina (γA1) was calculated as 44.06 nm. Furthermore, the analysis of the XRD pattern indicated that the structure of the prepared γ-alumina was cubic in nature. The XRD pattern related to nano γA2 (Fig. S.3.(b) and Table S.2) demonstrated a good alignment with characteristics of the standard γ-alumina sample (Table S.1). The distinctive diffraction peaks associated with γA2 were observed at 2θ angles of 32.62°, 39.41°, 46.24°, and 66.91°, with specific details provided in Table S.2. The average crystallite size for γA2 was determined to be 10.95 nm through the application of Debye-Scherrer’s equation. Additionally, the nanostructure of γ-alumina denoted as γA2 exhibited a cubic configuration. Comparing the XRD pattern of γA3 (Fig. S.3.(c) and Table S.2) to standard samples (Table S.1) it was deduced that the synthesized sample did not match due to the low intensity of peak at 2θ = 67.55°. The frequent peaks within the 2θ = 31°- 46° range indicated that the crystallization process for the formation of γ-alumina was not completed; hence, further, in this work, the reaction was carried out using a hydrothermal process. The structure of γA3 was cubic and the average crystallite size was 7.27 nm. The X-ray diffraction pattern of the γA4, provided in Fig. S.3.(d) and Table S.2, did not match well with the specifications of the standard γ-alumina sample. This can be justified by the low intensity of peaks at 2θ = 22.94°, 42.41°, 45.81°, and 67.30°, which indicated the incomplete crystallization process. Thus, the initial temperature, reaction time, and calcination temperature was increased further in this work. The average crystallite size of cubic γA4 was found to be 11.69 nm using. The obtained XRD results for γA5 (Fig. S.3.(e) and Table S.2) proved that the synthesized nanoparticle matched well considering the standard γ-alumina sample. The characteristic peaks of alumina were detected at 2θ = 32.12°, 39.60°, 45.63°, and 67.19° and the average crystallite size of these cubic nanoparticles was calculated as 16.51 nm. The obtained XRD spectra and results for γA3, γA4, and γA5 indicated that increasing the initial temperature and reaction resulted in nano γ-alumina with desired purity and crystallization.
Due to the obtained spectra for γA1 and its impurities, the rest of the samples were further scrutinized using the FT-IR technique. Fig. S.4.(a-d) demonstrates provided FT-IR spectra for γA2, γA3, γA4, and γA5. Fig. S.4 (a) illustrates the FT-IR spectrum corresponding to the nano γ-alumina derived from nepheline syenite ore and lime, denoted as γA2. The discernible peaks within the range of 422.95-1028.71 cm⁻¹ were attributed to the stretching and bending vibrations of aluminum and oxygen bonds. Additionally, bending vibrations related to the O-H bond in aluminum hydroxide, along with signals indicative of alumina formation, were evident at 1420.75-1520.08 cm⁻¹. Bending and stretching vibrations of O-H bonds were further identified at 1642.40-1740.94 cm⁻¹ and 3440.93-3847.47 cm⁻¹, respectively56. For nano γ-alumina without undergoing a hydrothermal process, designated as γA3 and depicted in Fig. S.4 (b), the stretching and bending bonds of Al-O bonds were observable within the range of 611.49-669.18 cm⁻¹ 57. The observed peak at 1461.76 cm⁻¹ was assigned to the bending vibration of the O-H bond in alumina57. The relatively low intensity of the peak associated with alumina formation suggested insufficient time for achieving preferable crystallinity. Additionally, the bending and stretching vibrations of the hydroxyl bond in water were discernible at 1646.01 cm⁻¹ and 3449.67 cm⁻¹, respectively58. These findings provided valuable insights into the molecular characteristics and structural aspects, indicating the temporal considerations for optimal crystalline development. The synthesized nano γ-alumina from aluminum hydroxide using the hydrothermal method (γA4) was scrutinized by FT-IR (Fig. S.4 (c)). The vibration of stretching and bending Al-O bond and bending O-H bond of alumina formation appeared at 611.20 cm− 1 and 1461.54 cm− 1, sequentially. As it was depicted, the time and crystallinity were not enough due to the low intensity. Assigned peaks to bending and stretching bonds of water molecules were at 1646.49 cm− 1 and 3452.49 cm− 1, respectively. Fig. S.4 (d) showcases the FT-IR spectrum of γA5. The vibrational analysis revealed stretching and bending bonds of Al-O within the 580.34-832.06 cm⁻¹ range. Noteworthy peaks in the 1418.45-1520.04 cm⁻¹ range were associated with the bending vibration of the O-H bond during alumina formation56,57. The heightened intensity of these peaks affirmed the adequacy of the reaction duration, indicating the successful synthesis of a crystalline structure. Additionally, bending vibrations of O-H bonds in water molecules were identified in the range of 1647.75-1740.97 cm⁻¹, with stretching bonds observed within the 3449.62-3742.32 cm⁻¹ range. Vibration bonds attributed to ethylene glycol were evident in the 2800–3000 cm⁻¹ range across all examined samples, with a notable decrease in intensity observed for γA5 58,59. This observation suggested that an increase in the initial temperature, reaction time, and calcination temperature led to the completion of the reaction, resulting in the complete combustion of ethylene glycol during elevated calcination temperatures. Furthermore, samples prepared over different durations, including 1 hr, 48 hr, and 5 days, underwent FT-IR analysis. However, the presented sample exhibited the most favorable results, underscoring the optimal conditions for achieving the desired structural characteristics.
The conducted characterizations showed that nano γ-alumina from aluminum hydroxide with extended time was better than the other samples; hence, γA5 was further investigated using SEM to scrutinize the morphology of the surface and the size of the alumina particles (using histogram graph) 60,61. As can be seen in Fig. 1a, the average size of alumina particles was 75.49 nm, and the crystalline size obtained from XRD was 16.51 nm. According to the Fig. 1(b), the prepared alumina had two types of morphology and the increased temperature caused the accumulation of particles and the formation of larger particles. The specific surface area for alumina obtained from nepheline syenite ore and lime was equal to 64.98 m2/g.
3.2 Characterization of nanocomposites
Fn/γA5 nanocomposites were further scrutinized using XRD and the obtained patterns are provided in Fig. 3(a-c). The characteristic peaks of alumina X-ray diffraction patterns of F5/γA5 (Fig.S.5.(c)), and F10/γA5 (Fig.S.5.(b)) nanocomposites, indicated the stability of the crystalline structure of alumina for these materials. The intensity of the peak at 2θ = 67.20° was higher for F5/γA5 compared to F10/γA5, due to the reduced iron content. It was also noteworthy that no shift was detected regarding the alumina characteristic peak of the γA5 sample. Although the intensity of alumina-related peaks decreased by increasing the iron content of the nanocomposite, iron oxide’s characteristic peaks for all nanocomposites were all detected at 2θ = 35.64°, 35.57°, and 35.59° for F30/γA5, F10/γA5, and F5/γA5, respectively. Also, these peaks indicated the high crystallinity of the synthesized iron oxide, which increased with the addition of iron oxide amount in the magnetite-alumina composites. The information about the d values and XRD characteristics of Fn/γA5 nanocomposites are given in Table S.3. According to the obtained values, the displacements in the d values of nanocomposites indicated the presence of iron oxide nanoparticles on the alumina surface. The broad diffraction pattern in the samples was probably due to the nano size of the prepared nanocomposite. The approximate crystallite size of nanocomposites was calculated using the Debye-Scherer Eq. 62.
Magnetic properties are significantly influenced by particle size, with a notable trend towards the emergence of magnetic behaviors as particle size decreases to the nanometer range. In the context of iron oxides, when the size of magnetic nanoparticles surpasses 50 nm, they are classified as paramagnetic iron oxides. Conversely, particles with sizes smaller than 50 nm fall under the category of ultra-fine superparamagnetic particles. In our investigation, all samples underwent examination using a magnet to facilitate the segregation of nanocomposites from water. VSM results of the nanocomposite powder’s magnetization behavior at room temperature is showcased in Fig. S.6.. The inset provided a visual representation of the F10/γA5 nanocomposite dispersed in water, along with its response to an external magnet. The determined saturation magnetization value for the nanocomposite powder stands at 50.03 emu/g for F10/γA5. Due to the significant magnetism exhibited by the material, it can be asserted that the presence of non-magnetic phases in the sample was minimal, if not entirely absent.
3.3 Nitrate removal
3.3.1. pH effect
pH is one of the important factors that can play a crucial role in the adsorption process by affecting the structure of pollutants and the charge of the adsorbent’s surface. This study revealed that the pH of the solution had a noticeable effect on the removal percentage. To systematically explore this impact, 0.1 g of each material was introduced into a nitrate solution (10 ppm), and the solution’s pH was adjusted to 3, 5, 7, and 9. As depicted in Fig. 3, the removal of nitrate exhibited higher efficiency at lower pH values, and gradually diminishing as the pH shifted towards alkaline conditions. Specifically, the minimum and maximum nitrate removal values were observed at pH = 9 and pH = 5, respectively. The heightened removal percentage at lower pH values can be rationalized by the simultaneous increase in H+ ions and a decrease in OH− ions. This led to a positively charged surface, fostering electrostatic attraction forces between the adsorbent and the negatively charged nitrate 63,64. Subsequent investigations, based on the obtained results, were conducted at a pH of 5 due to its optimal performance in nitrate removal.
3.3.2. Time effect
In this study, nano γA5 and nanocomposites (F5/γA5, F10/γA5) were added to a nitrate solution with an initial concentration of 10 ppm and a pH of 5, each at a dosage of 0.1 g per 100 ml. Figure 4 illustrates the impact of contact time on the nitrate removal process by the investigated samples. Notably, the removal percentage for all samples exhibited an initial upward trend within the first hour of the experiment; however, this percentage gradually declined after the 60 min mark. Based on the obtained results, the reaction equilibrium time for γA5, F5/γA5, and F10/γA5 was determined as 60 min. During the initial phase, active sites on the adsorbent surface were readily available and easily accessible, leading to continuous nitrate adsorption over time. The decrease in nitrate removal percentage following an increase in reaction time can be attributed to several factors inherent in the adsorption process. Initially, active sites on the adsorbent surface are readily available and easily accessible, leading to continuous nitrate adsorption over time. However, as the reaction progresses, several phenomena occur that can lead to a decline in removal efficiency. Firstly, the saturation of active sites plays a significant role. As the reaction proceeds, the active sites on the adsorbent surface become progressively occupied by nitrate ions. Once these sites are saturated, further adsorption becomes limited, resulting in a decrease in removal efficiency. Secondly, mass transfer limitations can impact the process. As the reaction proceeds, a concentration gradient between the solution and the adsorbent surface may develop. This gradient can lead to slower diffusion of nitrate ions from the bulk solution to the active sites on the adsorbent surface, causing a decrease in removal efficiency. Lastly, changes in the adsorbate-adsorbent interaction may occur over time. Factors such as changes in pH, temperature, or the formation of complexes between the adsorbate and the adsorbent material can influence the affinity of nitrate ions for the adsorbent surface, leading to a decrease in removal efficiency. Overall, it was observed that F10/γA5 exhibited superior efficiency than other samples.
3.3.3. Adsorbent dosage effect
To assess the impact of adsorbent dosage on nitrate removal (refer to Fig. 5), solutions containing nitrate with an initial concentration of 10 ppm were prepared, and their pH was adjusted to 5. Subsequently, varying amounts of prepared adsorbents, specifically 0.05, 0.1, and 0.15 g in 100 ml, were individually introduced into each solution. Sampling was then conducted at the equilibrium time. The removal percentage exhibited an increasing trend with the augmentation of adsorbent dosage from 0.05 to 0.1 g. However, it was observed that an additional increase in adsorbent dosage to 0.15 g/100 ml did not yield a significant enhancement in efficiency for all samples. Consequently, 0.1 g/100 ml was identified as the optimal dosage for all three samples, striking a balance between effective nitrate removal and resource utilization.
3.3.4. Nitrate concentration effect
The influence of nitrate’s initial concentration on the removal percentage was investigated utilizing 0.1 g/100 ml of adsorbents in solutions with initial concentrations of 10, 50, and 100 ppm, all maintained at a pH of 5. Following the attainment of equilibrium conditions, samples were collected and tested for the residual nitrate amount, with the comparative results depicted in Fig. 6. The experimental findings revealed that, at higher nitrate concentrations, the removal percentage for all samples increased. This observed phenomenon can be rationalized by the surplus presence of nitrate ions at higher concentrations, facilitating increased contact between active sites and contaminants. Consequently, the adsorption process became more accessible, leading to enhanced effectiveness in nitrate removal.
To facilitate a comparison between the findings of the current study and those of prior research, a summarized table Table 1 is presented, incorporating comparable results from this study and other relevant studies.
Table 1
Comparison of the adsorption capacity and various experimental conditions among multiple adsorbents for nitrate removal.
Material
|
Nitrate Removal (%)
|
Conditions
|
Ref.
|
F10/γA5
|
74
|
Initial concentration = 100 ppm, pH = 5, contact time = 45 min
|
This study
|
γalumina
|
\(\sim\) 68
|
Initial concentration = 20 ppm, pH = 6.68, contact time = 30 min
|
47
|
Zn/Al chloride layered double hydroxide
|
85
|
Initial concentration = 10 ppm, pH = 6, contact time = 40 min
|
65
|
Fe3O4/MB
|
79
|
Initial concentration = 30 ppm, pH = 7.5, contact time = 90 min
|
63
|
Fe-Zeolite chitosan beads
|
89
|
Initial concentration = 50 ppm, pH = 5, contact time = 60 min
|
58
|
FeMgMn-LDH,
|
\(\sim\) 61
|
Initial concentration = 20 ppm, pH = 7, contact time = 4 h
|
66
|
Alumina magnetic hybrid
|
\(\sim\) 50
|
Initial concentration = 100 ppm, pH = 6, contact time = 80 min
|
67
|
Al2O3/ZrO2/Fe3O4
|
50
|
Initial concentration = 15 ppm, pH = 3, contact time = 24 h
|
68
|
3.3.5. Kinetic study
Enhancing our comprehension of adsorption processes, specifically pore-diffusion and surface-adsorption, can be facilitated through the application of kinetic models. These models not only provide insights into the adsorption mechanisms but also offer valuable information on the controlling factors, such as mass transfer, diffusion control, and chemical reaction. The kinetics of adsorption play a pivotal role in determining optimal operating conditions for full-scale batch processes, and they provide crucial information for modeling and designing the overall process. In this study, we employed pseudo-first-order, pseudo-second-order, and intraparticle diffusion kinetic models, as described by Eqs. 4–6, respectively.
\({\text{q}}_{\text{t}}={\text{q}}_{\text{e}}(1-{\text{e}}^{{\text{k}}_{1}\text{t}})\) Eq. 4
\(\frac{\text{t}}{ {\text{q}}_{\text{t}}}=\frac{1}{{\text{k}}_{2}{\text{q}}_{\text{e}}^{2}}+\frac{\text{t}}{{\text{q}}_{\text{e}}}\) Eq. 5
\({\text{q}}_{\text{t}}={\text{k}}_{3}{\text{t}}^{0.5}+\text{C}\) Eq. 6
Adsorption capacity at equilibrium and the amount of adsorbed nitrate at the contact time of t (min) showed by 𝑞𝑒 (mg/g) and 𝑞𝑡 (mg/g). Constant rates of pseudo-first-order, pseudo-second-order, and Weber and Morris (intraparticle diffusion) models are attributed to 𝑘1(L/min), 𝑘2 (g/mg.min), and 𝑘3(g/mg.min0.5). C (mg/g) represents the thickness and boundary layer and 𝑅2 stands for the regression factor 69,70. All the derived parameters are tabulated in Table 2. It is noteworthy that all tested materials exhibited a good fit with both the Pseudo-second-order and Intra-particle diffusion kinetic models. The adherence to the Pseudo-second-order model suggested that chemisorption acts as the rate-limiting step in the adsorption process. This implied that the rate of adsorption was primarily dependent on the adsorption capacity rather than the concentration of nitrate. The Intra-particle diffusion model predicated on the assumption that the slowest step in the adsorption process was associated with the nitrate (adsorbate) in a liquid film surrounding the adsorbent. The well-established compatibility of the materials with this model indicated the likelihood of intra-particle diffusion playing a significant role in the adsorption mechanism. This underscored the potential involvement of intra-particle diffusion in governing the overall adsorption process, as supported by the model’s favorable alignment with the experimental data 71.
Table 2
Obtained kinetic parameters for γA5, F5/ γA5, and F10/ γA5.
Kinetic model
|
Parameters
|
γA5
|
F5/ γA5
|
F10/ γA5
|
Pseudo-first-order
|
k1 (L/min)
|
0.0928
|
0.0944
|
0.1024
|
qe (mg/g)
|
65.3658
|
67.0005
|
77.4785
|
R2
|
0.8912
|
0.8732
|
0.8875
|
Pseudo-second-order
|
k2 (g/mg.min)
|
0.0397
|
0.0625
|
0.1197
|
qe (mg/g)
|
63.6943
|
70.9220
|
74.6269
|
R2
|
0.9517
|
0.9849
|
0.9863
|
Intra-particle diffusion
|
k3 (g/mg.min0.5)
|
3.6616
|
6.0204
|
7.6928
|
C (mg/g)
|
10.582
|
29.954
|
49.873
|
R2
|
0.9871
|
0.9963
|
0.9837
|